diff options
Diffstat (limited to 'fs')
518 files changed, 11965 insertions, 13632 deletions
diff --git a/fs/9p/cache.c b/fs/9p/cache.c index 945aa5f02f9b..a9ea73d6dcf3 100644 --- a/fs/9p/cache.c +++ b/fs/9p/cache.c @@ -62,8 +62,8 @@ static uint16_t v9fs_cache_session_get_key(const void *cookie_netfs_data, uint16_t klen = 0; v9ses = (struct v9fs_session_info *)cookie_netfs_data; - P9_DPRINTK(P9_DEBUG_FSC, "session %p buf %p size %u", v9ses, - buffer, bufmax); + p9_debug(P9_DEBUG_FSC, "session %p buf %p size %u\n", + v9ses, buffer, bufmax); if (v9ses->cachetag) klen = strlen(v9ses->cachetag); @@ -72,7 +72,7 @@ static uint16_t v9fs_cache_session_get_key(const void *cookie_netfs_data, return 0; memcpy(buffer, v9ses->cachetag, klen); - P9_DPRINTK(P9_DEBUG_FSC, "cache session tag %s", v9ses->cachetag); + p9_debug(P9_DEBUG_FSC, "cache session tag %s\n", v9ses->cachetag); return klen; } @@ -91,14 +91,14 @@ void v9fs_cache_session_get_cookie(struct v9fs_session_info *v9ses) v9ses->fscache = fscache_acquire_cookie(v9fs_cache_netfs.primary_index, &v9fs_cache_session_index_def, v9ses); - P9_DPRINTK(P9_DEBUG_FSC, "session %p get cookie %p", v9ses, - v9ses->fscache); + p9_debug(P9_DEBUG_FSC, "session %p get cookie %p\n", + v9ses, v9ses->fscache); } void v9fs_cache_session_put_cookie(struct v9fs_session_info *v9ses) { - P9_DPRINTK(P9_DEBUG_FSC, "session %p put cookie %p", v9ses, - v9ses->fscache); + p9_debug(P9_DEBUG_FSC, "session %p put cookie %p\n", + v9ses, v9ses->fscache); fscache_relinquish_cookie(v9ses->fscache, 0); v9ses->fscache = NULL; } @@ -109,8 +109,8 @@ static uint16_t v9fs_cache_inode_get_key(const void *cookie_netfs_data, { const struct v9fs_inode *v9inode = cookie_netfs_data; memcpy(buffer, &v9inode->qid.path, sizeof(v9inode->qid.path)); - P9_DPRINTK(P9_DEBUG_FSC, "inode %p get key %llu", &v9inode->vfs_inode, - v9inode->qid.path); + p9_debug(P9_DEBUG_FSC, "inode %p get key %llu\n", + &v9inode->vfs_inode, v9inode->qid.path); return sizeof(v9inode->qid.path); } @@ -120,8 +120,8 @@ static void v9fs_cache_inode_get_attr(const void *cookie_netfs_data, const struct v9fs_inode *v9inode = cookie_netfs_data; *size = i_size_read(&v9inode->vfs_inode); - P9_DPRINTK(P9_DEBUG_FSC, "inode %p get attr %llu", &v9inode->vfs_inode, - *size); + p9_debug(P9_DEBUG_FSC, "inode %p get attr %llu\n", + &v9inode->vfs_inode, *size); } static uint16_t v9fs_cache_inode_get_aux(const void *cookie_netfs_data, @@ -129,8 +129,8 @@ static uint16_t v9fs_cache_inode_get_aux(const void *cookie_netfs_data, { const struct v9fs_inode *v9inode = cookie_netfs_data; memcpy(buffer, &v9inode->qid.version, sizeof(v9inode->qid.version)); - P9_DPRINTK(P9_DEBUG_FSC, "inode %p get aux %u", &v9inode->vfs_inode, - v9inode->qid.version); + p9_debug(P9_DEBUG_FSC, "inode %p get aux %u\n", + &v9inode->vfs_inode, v9inode->qid.version); return sizeof(v9inode->qid.version); } @@ -206,8 +206,8 @@ void v9fs_cache_inode_get_cookie(struct inode *inode) &v9fs_cache_inode_index_def, v9inode); - P9_DPRINTK(P9_DEBUG_FSC, "inode %p get cookie %p", inode, - v9inode->fscache); + p9_debug(P9_DEBUG_FSC, "inode %p get cookie %p\n", + inode, v9inode->fscache); } void v9fs_cache_inode_put_cookie(struct inode *inode) @@ -216,8 +216,8 @@ void v9fs_cache_inode_put_cookie(struct inode *inode) if (!v9inode->fscache) return; - P9_DPRINTK(P9_DEBUG_FSC, "inode %p put cookie %p", inode, - v9inode->fscache); + p9_debug(P9_DEBUG_FSC, "inode %p put cookie %p\n", + inode, v9inode->fscache); fscache_relinquish_cookie(v9inode->fscache, 0); v9inode->fscache = NULL; @@ -229,8 +229,8 @@ void v9fs_cache_inode_flush_cookie(struct inode *inode) if (!v9inode->fscache) return; - P9_DPRINTK(P9_DEBUG_FSC, "inode %p flush cookie %p", inode, - v9inode->fscache); + p9_debug(P9_DEBUG_FSC, "inode %p flush cookie %p\n", + inode, v9inode->fscache); fscache_relinquish_cookie(v9inode->fscache, 1); v9inode->fscache = NULL; @@ -272,8 +272,8 @@ void v9fs_cache_inode_reset_cookie(struct inode *inode) v9inode->fscache = fscache_acquire_cookie(v9ses->fscache, &v9fs_cache_inode_index_def, v9inode); - P9_DPRINTK(P9_DEBUG_FSC, "inode %p revalidating cookie old %p new %p", - inode, old, v9inode->fscache); + p9_debug(P9_DEBUG_FSC, "inode %p revalidating cookie old %p new %p\n", + inode, old, v9inode->fscache); spin_unlock(&v9inode->fscache_lock); } @@ -323,7 +323,7 @@ int __v9fs_readpage_from_fscache(struct inode *inode, struct page *page) int ret; const struct v9fs_inode *v9inode = V9FS_I(inode); - P9_DPRINTK(P9_DEBUG_FSC, "inode %p page %p", inode, page); + p9_debug(P9_DEBUG_FSC, "inode %p page %p\n", inode, page); if (!v9inode->fscache) return -ENOBUFS; @@ -335,13 +335,13 @@ int __v9fs_readpage_from_fscache(struct inode *inode, struct page *page) switch (ret) { case -ENOBUFS: case -ENODATA: - P9_DPRINTK(P9_DEBUG_FSC, "page/inode not in cache %d", ret); + p9_debug(P9_DEBUG_FSC, "page/inode not in cache %d\n", ret); return 1; case 0: - P9_DPRINTK(P9_DEBUG_FSC, "BIO submitted"); + p9_debug(P9_DEBUG_FSC, "BIO submitted\n"); return ret; default: - P9_DPRINTK(P9_DEBUG_FSC, "ret %d", ret); + p9_debug(P9_DEBUG_FSC, "ret %d\n", ret); return ret; } } @@ -361,7 +361,7 @@ int __v9fs_readpages_from_fscache(struct inode *inode, int ret; const struct v9fs_inode *v9inode = V9FS_I(inode); - P9_DPRINTK(P9_DEBUG_FSC, "inode %p pages %u", inode, *nr_pages); + p9_debug(P9_DEBUG_FSC, "inode %p pages %u\n", inode, *nr_pages); if (!v9inode->fscache) return -ENOBUFS; @@ -373,15 +373,15 @@ int __v9fs_readpages_from_fscache(struct inode *inode, switch (ret) { case -ENOBUFS: case -ENODATA: - P9_DPRINTK(P9_DEBUG_FSC, "pages/inodes not in cache %d", ret); + p9_debug(P9_DEBUG_FSC, "pages/inodes not in cache %d\n", ret); return 1; case 0: BUG_ON(!list_empty(pages)); BUG_ON(*nr_pages != 0); - P9_DPRINTK(P9_DEBUG_FSC, "BIO submitted"); + p9_debug(P9_DEBUG_FSC, "BIO submitted\n"); return ret; default: - P9_DPRINTK(P9_DEBUG_FSC, "ret %d", ret); + p9_debug(P9_DEBUG_FSC, "ret %d\n", ret); return ret; } } @@ -396,9 +396,9 @@ void __v9fs_readpage_to_fscache(struct inode *inode, struct page *page) int ret; const struct v9fs_inode *v9inode = V9FS_I(inode); - P9_DPRINTK(P9_DEBUG_FSC, "inode %p page %p", inode, page); + p9_debug(P9_DEBUG_FSC, "inode %p page %p\n", inode, page); ret = fscache_write_page(v9inode->fscache, page, GFP_KERNEL); - P9_DPRINTK(P9_DEBUG_FSC, "ret = %d", ret); + p9_debug(P9_DEBUG_FSC, "ret = %d\n", ret); if (ret != 0) v9fs_uncache_page(inode, page); } @@ -409,7 +409,7 @@ void __v9fs_readpage_to_fscache(struct inode *inode, struct page *page) void __v9fs_fscache_wait_on_page_write(struct inode *inode, struct page *page) { const struct v9fs_inode *v9inode = V9FS_I(inode); - P9_DPRINTK(P9_DEBUG_FSC, "inode %p page %p", inode, page); + p9_debug(P9_DEBUG_FSC, "inode %p page %p\n", inode, page); if (PageFsCache(page)) fscache_wait_on_page_write(v9inode->fscache, page); } diff --git a/fs/9p/fid.c b/fs/9p/fid.c index 85b67ffa2a43..da8eefbe830d 100644 --- a/fs/9p/fid.c +++ b/fs/9p/fid.c @@ -45,8 +45,8 @@ int v9fs_fid_add(struct dentry *dentry, struct p9_fid *fid) { struct v9fs_dentry *dent; - P9_DPRINTK(P9_DEBUG_VFS, "fid %d dentry %s\n", - fid->fid, dentry->d_name.name); + p9_debug(P9_DEBUG_VFS, "fid %d dentry %s\n", + fid->fid, dentry->d_name.name); dent = dentry->d_fsdata; if (!dent) { @@ -79,8 +79,8 @@ static struct p9_fid *v9fs_fid_find(struct dentry *dentry, u32 uid, int any) struct v9fs_dentry *dent; struct p9_fid *fid, *ret; - P9_DPRINTK(P9_DEBUG_VFS, " dentry: %s (%p) uid %d any %d\n", - dentry->d_name.name, dentry, uid, any); + p9_debug(P9_DEBUG_VFS, " dentry: %s (%p) uid %d any %d\n", + dentry->d_name.name, dentry, uid, any); dent = (struct v9fs_dentry *) dentry->d_fsdata; ret = NULL; if (dent) { diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c index 2b78014a124a..1964f98e74be 100644 --- a/fs/9p/v9fs.c +++ b/fs/9p/v9fs.c @@ -23,6 +23,8 @@ * */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/module.h> #include <linux/errno.h> #include <linux/fs.h> @@ -85,15 +87,15 @@ static int get_cache_mode(char *s) if (!strcmp(s, "loose")) { version = CACHE_LOOSE; - P9_DPRINTK(P9_DEBUG_9P, "Cache mode: loose\n"); + p9_debug(P9_DEBUG_9P, "Cache mode: loose\n"); } else if (!strcmp(s, "fscache")) { version = CACHE_FSCACHE; - P9_DPRINTK(P9_DEBUG_9P, "Cache mode: fscache\n"); + p9_debug(P9_DEBUG_9P, "Cache mode: fscache\n"); } else if (!strcmp(s, "none")) { version = CACHE_NONE; - P9_DPRINTK(P9_DEBUG_9P, "Cache mode: none\n"); + p9_debug(P9_DEBUG_9P, "Cache mode: none\n"); } else - printk(KERN_INFO "9p: Unknown Cache mode %s.\n", s); + pr_info("Unknown Cache mode %s\n", s); return version; } @@ -140,8 +142,8 @@ static int v9fs_parse_options(struct v9fs_session_info *v9ses, char *opts) case Opt_debug: r = match_int(&args[0], &option); if (r < 0) { - P9_DPRINTK(P9_DEBUG_ERROR, - "integer field, but no integer?\n"); + p9_debug(P9_DEBUG_ERROR, + "integer field, but no integer?\n"); ret = r; continue; } @@ -154,8 +156,8 @@ static int v9fs_parse_options(struct v9fs_session_info *v9ses, char *opts) case Opt_dfltuid: r = match_int(&args[0], &option); if (r < 0) { - P9_DPRINTK(P9_DEBUG_ERROR, - "integer field, but no integer?\n"); + p9_debug(P9_DEBUG_ERROR, + "integer field, but no integer?\n"); ret = r; continue; } @@ -164,8 +166,8 @@ static int v9fs_parse_options(struct v9fs_session_info *v9ses, char *opts) case Opt_dfltgid: r = match_int(&args[0], &option); if (r < 0) { - P9_DPRINTK(P9_DEBUG_ERROR, - "integer field, but no integer?\n"); + p9_debug(P9_DEBUG_ERROR, + "integer field, but no integer?\n"); ret = r; continue; } @@ -174,8 +176,8 @@ static int v9fs_parse_options(struct v9fs_session_info *v9ses, char *opts) case Opt_afid: r = match_int(&args[0], &option); if (r < 0) { - P9_DPRINTK(P9_DEBUG_ERROR, - "integer field, but no integer?\n"); + p9_debug(P9_DEBUG_ERROR, + "integer field, but no integer?\n"); ret = r; continue; } @@ -205,8 +207,8 @@ static int v9fs_parse_options(struct v9fs_session_info *v9ses, char *opts) s = match_strdup(&args[0]); if (!s) { ret = -ENOMEM; - P9_DPRINTK(P9_DEBUG_ERROR, - "problem allocating copy of cache arg\n"); + p9_debug(P9_DEBUG_ERROR, + "problem allocating copy of cache arg\n"); goto free_and_return; } ret = get_cache_mode(s); @@ -223,8 +225,8 @@ static int v9fs_parse_options(struct v9fs_session_info *v9ses, char *opts) s = match_strdup(&args[0]); if (!s) { ret = -ENOMEM; - P9_DPRINTK(P9_DEBUG_ERROR, - "problem allocating copy of access arg\n"); + p9_debug(P9_DEBUG_ERROR, + "problem allocating copy of access arg\n"); goto free_and_return; } @@ -240,8 +242,8 @@ static int v9fs_parse_options(struct v9fs_session_info *v9ses, char *opts) v9ses->uid = simple_strtoul(s, &e, 10); if (*e != '\0') { ret = -EINVAL; - printk(KERN_INFO "9p: Unknown access " - "argument %s.\n", s); + pr_info("Unknown access argument %s\n", + s); kfree(s); goto free_and_return; } @@ -254,9 +256,8 @@ static int v9fs_parse_options(struct v9fs_session_info *v9ses, char *opts) #ifdef CONFIG_9P_FS_POSIX_ACL v9ses->flags |= V9FS_POSIX_ACL; #else - P9_DPRINTK(P9_DEBUG_ERROR, - "Not defined CONFIG_9P_FS_POSIX_ACL. " - "Ignoring posixacl option\n"); + p9_debug(P9_DEBUG_ERROR, + "Not defined CONFIG_9P_FS_POSIX_ACL. Ignoring posixacl option\n"); #endif break; @@ -318,7 +319,7 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses, if (IS_ERR(v9ses->clnt)) { retval = PTR_ERR(v9ses->clnt); v9ses->clnt = NULL; - P9_DPRINTK(P9_DEBUG_ERROR, "problem initializing 9p client\n"); + p9_debug(P9_DEBUG_ERROR, "problem initializing 9p client\n"); goto error; } @@ -371,7 +372,7 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses, if (IS_ERR(fid)) { retval = PTR_ERR(fid); fid = NULL; - P9_DPRINTK(P9_DEBUG_ERROR, "cannot attach\n"); + p9_debug(P9_DEBUG_ERROR, "cannot attach\n"); goto error; } @@ -429,7 +430,7 @@ void v9fs_session_close(struct v9fs_session_info *v9ses) */ void v9fs_session_cancel(struct v9fs_session_info *v9ses) { - P9_DPRINTK(P9_DEBUG_ERROR, "cancel session %p\n", v9ses); + p9_debug(P9_DEBUG_ERROR, "cancel session %p\n", v9ses); p9_client_disconnect(v9ses->clnt); } @@ -442,7 +443,7 @@ void v9fs_session_cancel(struct v9fs_session_info *v9ses) { void v9fs_session_begin_cancel(struct v9fs_session_info *v9ses) { - P9_DPRINTK(P9_DEBUG_ERROR, "begin cancel session %p\n", v9ses); + p9_debug(P9_DEBUG_ERROR, "begin cancel session %p\n", v9ses); p9_client_begin_disconnect(v9ses->clnt); } @@ -591,23 +592,23 @@ static void v9fs_cache_unregister(void) static int __init init_v9fs(void) { int err; - printk(KERN_INFO "Installing v9fs 9p2000 file system support\n"); + pr_info("Installing v9fs 9p2000 file system support\n"); /* TODO: Setup list of registered trasnport modules */ err = register_filesystem(&v9fs_fs_type); if (err < 0) { - printk(KERN_ERR "Failed to register filesystem\n"); + pr_err("Failed to register filesystem\n"); return err; } err = v9fs_cache_register(); if (err < 0) { - printk(KERN_ERR "Failed to register v9fs for caching\n"); + pr_err("Failed to register v9fs for caching\n"); goto out_fs_unreg; } err = v9fs_sysfs_init(); if (err < 0) { - printk(KERN_ERR "Failed to register with sysfs\n"); + pr_err("Failed to register with sysfs\n"); goto out_sysfs_cleanup; } diff --git a/fs/9p/v9fs_vfs.h b/fs/9p/v9fs_vfs.h index 410ffd6ceb5f..dc95a252523d 100644 --- a/fs/9p/v9fs_vfs.h +++ b/fs/9p/v9fs_vfs.h @@ -54,9 +54,9 @@ extern struct kmem_cache *v9fs_inode_cache; struct inode *v9fs_alloc_inode(struct super_block *sb); void v9fs_destroy_inode(struct inode *inode); -struct inode *v9fs_get_inode(struct super_block *sb, int mode, dev_t); +struct inode *v9fs_get_inode(struct super_block *sb, umode_t mode, dev_t); int v9fs_init_inode(struct v9fs_session_info *v9ses, - struct inode *inode, int mode, dev_t); + struct inode *inode, umode_t mode, dev_t); void v9fs_evict_inode(struct inode *inode); ino_t v9fs_qid2ino(struct p9_qid *qid); void v9fs_stat2inode(struct p9_wstat *, struct inode *, struct super_block *); diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c index 2524e4cbb8ea..0ad61c6a65a5 100644 --- a/fs/9p/vfs_addr.c +++ b/fs/9p/vfs_addr.c @@ -56,7 +56,7 @@ static int v9fs_fid_readpage(struct p9_fid *fid, struct page *page) struct inode *inode; inode = page->mapping->host; - P9_DPRINTK(P9_DEBUG_VFS, "\n"); + p9_debug(P9_DEBUG_VFS, "\n"); BUG_ON(!PageLocked(page)); @@ -116,14 +116,14 @@ static int v9fs_vfs_readpages(struct file *filp, struct address_space *mapping, struct inode *inode; inode = mapping->host; - P9_DPRINTK(P9_DEBUG_VFS, "inode: %p file: %p\n", inode, filp); + p9_debug(P9_DEBUG_VFS, "inode: %p file: %p\n", inode, filp); ret = v9fs_readpages_from_fscache(inode, mapping, pages, &nr_pages); if (ret == 0) return ret; ret = read_cache_pages(mapping, pages, (void *)v9fs_vfs_readpage, filp); - P9_DPRINTK(P9_DEBUG_VFS, " = %d\n", ret); + p9_debug(P9_DEBUG_VFS, " = %d\n", ret); return ret; } @@ -263,10 +263,9 @@ v9fs_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, * Now that we do caching with cache mode enabled, We need * to support direct IO */ - P9_DPRINTK(P9_DEBUG_VFS, "v9fs_direct_IO: v9fs_direct_IO (%s) " - "off/no(%lld/%lu) EINVAL\n", - iocb->ki_filp->f_path.dentry->d_name.name, - (long long) pos, nr_segs); + p9_debug(P9_DEBUG_VFS, "v9fs_direct_IO: v9fs_direct_IO (%s) off/no(%lld/%lu) EINVAL\n", + iocb->ki_filp->f_path.dentry->d_name.name, + (long long)pos, nr_segs); return -EINVAL; } diff --git a/fs/9p/vfs_dentry.c b/fs/9p/vfs_dentry.c index e022890c6f40..d529437ff442 100644 --- a/fs/9p/vfs_dentry.c +++ b/fs/9p/vfs_dentry.c @@ -53,8 +53,8 @@ static int v9fs_dentry_delete(const struct dentry *dentry) { - P9_DPRINTK(P9_DEBUG_VFS, " dentry: %s (%p)\n", dentry->d_name.name, - dentry); + p9_debug(P9_DEBUG_VFS, " dentry: %s (%p)\n", + dentry->d_name.name, dentry); return 1; } @@ -66,8 +66,8 @@ static int v9fs_dentry_delete(const struct dentry *dentry) */ static int v9fs_cached_dentry_delete(const struct dentry *dentry) { - P9_DPRINTK(P9_DEBUG_VFS, " dentry: %s (%p)\n", - dentry->d_name.name, dentry); + p9_debug(P9_DEBUG_VFS, " dentry: %s (%p)\n", + dentry->d_name.name, dentry); /* Don't cache negative dentries */ if (!dentry->d_inode) @@ -86,8 +86,8 @@ static void v9fs_dentry_release(struct dentry *dentry) struct v9fs_dentry *dent; struct p9_fid *temp, *current_fid; - P9_DPRINTK(P9_DEBUG_VFS, " dentry: %s (%p)\n", dentry->d_name.name, - dentry); + p9_debug(P9_DEBUG_VFS, " dentry: %s (%p)\n", + dentry->d_name.name, dentry); dent = dentry->d_fsdata; if (dent) { list_for_each_entry_safe(current_fid, temp, &dent->fidlist, diff --git a/fs/9p/vfs_dir.c b/fs/9p/vfs_dir.c index 598fff1a54e5..ff911e779651 100644 --- a/fs/9p/vfs_dir.c +++ b/fs/9p/vfs_dir.c @@ -140,7 +140,7 @@ static int v9fs_dir_readdir(struct file *filp, void *dirent, filldir_t filldir) int reclen = 0; struct p9_rdir *rdir; - P9_DPRINTK(P9_DEBUG_VFS, "name %s\n", filp->f_path.dentry->d_name.name); + p9_debug(P9_DEBUG_VFS, "name %s\n", filp->f_path.dentry->d_name.name); fid = filp->private_data; buflen = fid->clnt->msize - P9_IOHDRSZ; @@ -168,7 +168,7 @@ static int v9fs_dir_readdir(struct file *filp, void *dirent, filldir_t filldir) err = p9stat_read(fid->clnt, rdir->buf + rdir->head, rdir->tail - rdir->head, &st); if (err) { - P9_DPRINTK(P9_DEBUG_VFS, "returned %d\n", err); + p9_debug(P9_DEBUG_VFS, "returned %d\n", err); err = -EIO; p9stat_free(&st); goto unlock_and_exit; @@ -213,7 +213,7 @@ static int v9fs_dir_readdir_dotl(struct file *filp, void *dirent, struct p9_dirent curdirent; u64 oldoffset = 0; - P9_DPRINTK(P9_DEBUG_VFS, "name %s\n", filp->f_path.dentry->d_name.name); + p9_debug(P9_DEBUG_VFS, "name %s\n", filp->f_path.dentry->d_name.name); fid = filp->private_data; buflen = fid->clnt->msize - P9_READDIRHDRSZ; @@ -244,7 +244,7 @@ static int v9fs_dir_readdir_dotl(struct file *filp, void *dirent, rdir->tail - rdir->head, &curdirent); if (err < 0) { - P9_DPRINTK(P9_DEBUG_VFS, "returned %d\n", err); + p9_debug(P9_DEBUG_VFS, "returned %d\n", err); err = -EIO; goto unlock_and_exit; } @@ -290,9 +290,8 @@ int v9fs_dir_release(struct inode *inode, struct file *filp) struct p9_fid *fid; fid = filp->private_data; - P9_DPRINTK(P9_DEBUG_VFS, - "v9fs_dir_release: inode: %p filp: %p fid: %d\n", - inode, filp, fid ? fid->fid : -1); + p9_debug(P9_DEBUG_VFS, "inode: %p filp: %p fid: %d\n", + inode, filp, fid ? fid->fid : -1); if (fid) p9_client_clunk(fid); return 0; diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c index 62857a810a79..fc06fd27065e 100644 --- a/fs/9p/vfs_file.c +++ b/fs/9p/vfs_file.c @@ -61,7 +61,7 @@ int v9fs_file_open(struct inode *inode, struct file *file) struct p9_fid *fid; int omode; - P9_DPRINTK(P9_DEBUG_VFS, "inode: %p file: %p\n", inode, file); + p9_debug(P9_DEBUG_VFS, "inode: %p file: %p\n", inode, file); v9inode = V9FS_I(inode); v9ses = v9fs_inode2v9ses(inode); if (v9fs_proto_dotl(v9ses)) @@ -135,7 +135,7 @@ static int v9fs_file_lock(struct file *filp, int cmd, struct file_lock *fl) int res = 0; struct inode *inode = filp->f_path.dentry->d_inode; - P9_DPRINTK(P9_DEBUG_VFS, "filp: %p lock: %p\n", filp, fl); + p9_debug(P9_DEBUG_VFS, "filp: %p lock: %p\n", filp, fl); /* No mandatory locks */ if (__mandatory_lock(inode) && fl->fl_type != F_UNLCK) @@ -204,7 +204,8 @@ static int v9fs_file_do_lock(struct file *filp, int cmd, struct file_lock *fl) break; if (status == P9_LOCK_BLOCKED && !IS_SETLKW(cmd)) break; - schedule_timeout_interruptible(P9_LOCK_TIMEOUT); + if (schedule_timeout_interruptible(P9_LOCK_TIMEOUT) != 0) + break; } /* map 9p status to VFS status */ @@ -304,8 +305,8 @@ static int v9fs_file_lock_dotl(struct file *filp, int cmd, struct file_lock *fl) struct inode *inode = filp->f_path.dentry->d_inode; int ret = -ENOLCK; - P9_DPRINTK(P9_DEBUG_VFS, "filp: %p cmd:%d lock: %p name: %s\n", filp, - cmd, fl, filp->f_path.dentry->d_name.name); + p9_debug(P9_DEBUG_VFS, "filp: %p cmd:%d lock: %p name: %s\n", + filp, cmd, fl, filp->f_path.dentry->d_name.name); /* No mandatory locks */ if (__mandatory_lock(inode) && fl->fl_type != F_UNLCK) @@ -340,8 +341,8 @@ static int v9fs_file_flock_dotl(struct file *filp, int cmd, struct inode *inode = filp->f_path.dentry->d_inode; int ret = -ENOLCK; - P9_DPRINTK(P9_DEBUG_VFS, "filp: %p cmd:%d lock: %p name: %s\n", filp, - cmd, fl, filp->f_path.dentry->d_name.name); + p9_debug(P9_DEBUG_VFS, "filp: %p cmd:%d lock: %p name: %s\n", + filp, cmd, fl, filp->f_path.dentry->d_name.name); /* No mandatory locks */ if (__mandatory_lock(inode) && fl->fl_type != F_UNLCK) @@ -384,8 +385,8 @@ v9fs_fid_readn(struct p9_fid *fid, char *data, char __user *udata, u32 count, { int n, total, size; - P9_DPRINTK(P9_DEBUG_VFS, "fid %d offset %llu count %d\n", fid->fid, - (long long unsigned) offset, count); + p9_debug(P9_DEBUG_VFS, "fid %d offset %llu count %d\n", + fid->fid, (long long unsigned)offset, count); n = 0; total = 0; size = fid->iounit ? fid->iounit : fid->clnt->msize - P9_IOHDRSZ; @@ -443,7 +444,7 @@ v9fs_file_read(struct file *filp, char __user *udata, size_t count, struct p9_fid *fid; size_t size; - P9_DPRINTK(P9_DEBUG_VFS, "count %zu offset %lld\n", count, *offset); + p9_debug(P9_DEBUG_VFS, "count %zu offset %lld\n", count, *offset); fid = filp->private_data; size = fid->iounit ? fid->iounit : fid->clnt->msize - P9_IOHDRSZ; @@ -470,8 +471,8 @@ v9fs_file_write_internal(struct inode *inode, struct p9_fid *fid, loff_t origin = *offset; unsigned long pg_start, pg_end; - P9_DPRINTK(P9_DEBUG_VFS, "data %p count %d offset %x\n", data, - (int)count, (int)*offset); + p9_debug(P9_DEBUG_VFS, "data %p count %d offset %x\n", + data, (int)count, (int)*offset); clnt = fid->clnt; do { @@ -552,7 +553,7 @@ static int v9fs_file_fsync(struct file *filp, loff_t start, loff_t end, return retval; mutex_lock(&inode->i_mutex); - P9_DPRINTK(P9_DEBUG_VFS, "filp %p datasync %x\n", filp, datasync); + p9_debug(P9_DEBUG_VFS, "filp %p datasync %x\n", filp, datasync); fid = filp->private_data; v9fs_blank_wstat(&wstat); @@ -575,8 +576,7 @@ int v9fs_file_fsync_dotl(struct file *filp, loff_t start, loff_t end, return retval; mutex_lock(&inode->i_mutex); - P9_DPRINTK(P9_DEBUG_VFS, "v9fs_file_fsync_dotl: filp %p datasync %x\n", - filp, datasync); + p9_debug(P9_DEBUG_VFS, "filp %p datasync %x\n", filp, datasync); fid = filp->private_data; @@ -607,8 +607,8 @@ v9fs_vm_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) struct inode *inode = filp->f_path.dentry->d_inode; - P9_DPRINTK(P9_DEBUG_VFS, "page %p fid %lx\n", - page, (unsigned long)filp->private_data); + p9_debug(P9_DEBUG_VFS, "page %p fid %lx\n", + page, (unsigned long)filp->private_data); v9inode = V9FS_I(inode); /* make sure the cache has finished storing the page */ diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 879ed8851737..014c8dd62962 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -23,6 +23,8 @@ * */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/module.h> #include <linux/errno.h> #include <linux/fs.h> @@ -59,15 +61,13 @@ static const struct inode_operations v9fs_symlink_inode_operations; * */ -static int unixmode2p9mode(struct v9fs_session_info *v9ses, int mode) +static u32 unixmode2p9mode(struct v9fs_session_info *v9ses, umode_t mode) { int res; res = mode & 0777; if (S_ISDIR(mode)) res |= P9_DMDIR; if (v9fs_proto_dotu(v9ses)) { - if (S_ISLNK(mode)) - res |= P9_DMSYMLINK; if (v9ses->nodev == 0) { if (S_ISSOCK(mode)) res |= P9_DMSOCKET; @@ -85,10 +85,33 @@ static int unixmode2p9mode(struct v9fs_session_info *v9ses, int mode) res |= P9_DMSETGID; if ((mode & S_ISVTX) == S_ISVTX) res |= P9_DMSETVTX; - if ((mode & P9_DMLINK)) - res |= P9_DMLINK; } + return res; +} +/** + * p9mode2perm- convert plan9 mode bits to unix permission bits + * @v9ses: v9fs session information + * @stat: p9_wstat from which mode need to be derived + * + */ +static int p9mode2perm(struct v9fs_session_info *v9ses, + struct p9_wstat *stat) +{ + int res; + int mode = stat->mode; + + res = mode & S_IALLUGO; + if (v9fs_proto_dotu(v9ses)) { + if ((mode & P9_DMSETUID) == P9_DMSETUID) + res |= S_ISUID; + + if ((mode & P9_DMSETGID) == P9_DMSETGID) + res |= S_ISGID; + + if ((mode & P9_DMSETVTX) == P9_DMSETVTX) + res |= S_ISVTX; + } return res; } @@ -99,14 +122,14 @@ static int unixmode2p9mode(struct v9fs_session_info *v9ses, int mode) * @rdev: major number, minor number in case of device files. * */ -static int p9mode2unixmode(struct v9fs_session_info *v9ses, - struct p9_wstat *stat, dev_t *rdev) +static umode_t p9mode2unixmode(struct v9fs_session_info *v9ses, + struct p9_wstat *stat, dev_t *rdev) { int res; - int mode = stat->mode; + u32 mode = stat->mode; - res = mode & S_IALLUGO; *rdev = 0; + res = p9mode2perm(v9ses, stat); if ((mode & P9_DMDIR) == P9_DMDIR) res |= S_IFDIR; @@ -133,24 +156,13 @@ static int p9mode2unixmode(struct v9fs_session_info *v9ses, res |= S_IFBLK; break; default: - P9_DPRINTK(P9_DEBUG_ERROR, - "Unknown special type %c %s\n", type, - stat->extension); + p9_debug(P9_DEBUG_ERROR, "Unknown special type %c %s\n", + type, stat->extension); }; *rdev = MKDEV(major, minor); } else res |= S_IFREG; - if (v9fs_proto_dotu(v9ses)) { - if ((mode & P9_DMSETUID) == P9_DMSETUID) - res |= S_ISUID; - - if ((mode & P9_DMSETGID) == P9_DMSETGID) - res |= S_ISGID; - - if ((mode & P9_DMSETVTX) == P9_DMSETVTX) - res |= S_ISVTX; - } return res; } @@ -251,7 +263,6 @@ struct inode *v9fs_alloc_inode(struct super_block *sb) static void v9fs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(v9fs_inode_cache, V9FS_I(inode)); } @@ -261,7 +272,7 @@ void v9fs_destroy_inode(struct inode *inode) } int v9fs_init_inode(struct v9fs_session_info *v9ses, - struct inode *inode, int mode, dev_t rdev) + struct inode *inode, umode_t mode, dev_t rdev) { int err = 0; @@ -281,8 +292,8 @@ int v9fs_init_inode(struct v9fs_session_info *v9ses, } else if (v9fs_proto_dotu(v9ses)) { inode->i_op = &v9fs_file_inode_operations; } else { - P9_DPRINTK(P9_DEBUG_ERROR, - "special files without extended mode\n"); + p9_debug(P9_DEBUG_ERROR, + "special files without extended mode\n"); err = -EINVAL; goto error; } @@ -307,8 +318,8 @@ int v9fs_init_inode(struct v9fs_session_info *v9ses, break; case S_IFLNK: if (!v9fs_proto_dotu(v9ses) && !v9fs_proto_dotl(v9ses)) { - P9_DPRINTK(P9_DEBUG_ERROR, "extended modes used with " - "legacy protocol.\n"); + p9_debug(P9_DEBUG_ERROR, + "extended modes used with legacy protocol\n"); err = -EINVAL; goto error; } @@ -335,8 +346,8 @@ int v9fs_init_inode(struct v9fs_session_info *v9ses, break; default: - P9_DPRINTK(P9_DEBUG_ERROR, "BAD mode 0x%x S_IFMT 0x%x\n", - mode, mode & S_IFMT); + p9_debug(P9_DEBUG_ERROR, "BAD mode 0x%hx S_IFMT 0x%x\n", + mode, mode & S_IFMT); err = -EINVAL; goto error; } @@ -352,17 +363,18 @@ error: * */ -struct inode *v9fs_get_inode(struct super_block *sb, int mode, dev_t rdev) +struct inode *v9fs_get_inode(struct super_block *sb, umode_t mode, dev_t rdev) { int err; struct inode *inode; struct v9fs_session_info *v9ses = sb->s_fs_info; - P9_DPRINTK(P9_DEBUG_VFS, "super block: %p mode: %o\n", sb, mode); + p9_debug(P9_DEBUG_VFS, "super block: %p mode: %ho\n", sb, mode); inode = new_inode(sb); if (!inode) { - P9_EPRINTK(KERN_WARNING, "Problem allocating inode\n"); + pr_warn("%s (%d): Problem allocating inode\n", + __func__, task_pid_nr(current)); return ERR_PTR(-ENOMEM); } err = v9fs_init_inode(v9ses, inode, mode, rdev); @@ -492,7 +504,8 @@ static struct inode *v9fs_qid_iget(struct super_block *sb, int new) { dev_t rdev; - int retval, umode; + int retval; + umode_t umode; unsigned long i_ino; struct inode *inode; struct v9fs_session_info *v9ses = sb->s_fs_info; @@ -578,15 +591,15 @@ static int v9fs_remove(struct inode *dir, struct dentry *dentry, int flags) struct p9_fid *v9fid, *dfid; struct v9fs_session_info *v9ses; - P9_DPRINTK(P9_DEBUG_VFS, "inode: %p dentry: %p rmdir: %x\n", - dir, dentry, flags); + p9_debug(P9_DEBUG_VFS, "inode: %p dentry: %p rmdir: %x\n", + dir, dentry, flags); v9ses = v9fs_inode2v9ses(dir); inode = dentry->d_inode; dfid = v9fs_fid_lookup(dentry->d_parent); if (IS_ERR(dfid)) { retval = PTR_ERR(dfid); - P9_DPRINTK(P9_DEBUG_VFS, "fid lookup failed %d\n", retval); + p9_debug(P9_DEBUG_VFS, "fid lookup failed %d\n", retval); return retval; } if (v9fs_proto_dotl(v9ses)) @@ -635,7 +648,7 @@ v9fs_create(struct v9fs_session_info *v9ses, struct inode *dir, struct p9_fid *dfid, *ofid, *fid; struct inode *inode; - P9_DPRINTK(P9_DEBUG_VFS, "name %s\n", dentry->d_name.name); + p9_debug(P9_DEBUG_VFS, "name %s\n", dentry->d_name.name); err = 0; ofid = NULL; @@ -644,7 +657,7 @@ v9fs_create(struct v9fs_session_info *v9ses, struct inode *dir, dfid = v9fs_fid_lookup(dentry->d_parent); if (IS_ERR(dfid)) { err = PTR_ERR(dfid); - P9_DPRINTK(P9_DEBUG_VFS, "fid lookup failed %d\n", err); + p9_debug(P9_DEBUG_VFS, "fid lookup failed %d\n", err); return ERR_PTR(err); } @@ -652,36 +665,41 @@ v9fs_create(struct v9fs_session_info *v9ses, struct inode *dir, ofid = p9_client_walk(dfid, 0, NULL, 1); if (IS_ERR(ofid)) { err = PTR_ERR(ofid); - P9_DPRINTK(P9_DEBUG_VFS, "p9_client_walk failed %d\n", err); + p9_debug(P9_DEBUG_VFS, "p9_client_walk failed %d\n", err); return ERR_PTR(err); } err = p9_client_fcreate(ofid, name, perm, mode, extension); if (err < 0) { - P9_DPRINTK(P9_DEBUG_VFS, "p9_client_fcreate failed %d\n", err); + p9_debug(P9_DEBUG_VFS, "p9_client_fcreate failed %d\n", err); goto error; } - /* now walk from the parent so we can get unopened fid */ - fid = p9_client_walk(dfid, 1, &name, 1); - if (IS_ERR(fid)) { - err = PTR_ERR(fid); - P9_DPRINTK(P9_DEBUG_VFS, "p9_client_walk failed %d\n", err); - fid = NULL; - goto error; - } - - /* instantiate inode and assign the unopened fid to the dentry */ - inode = v9fs_get_new_inode_from_fid(v9ses, fid, dir->i_sb); - if (IS_ERR(inode)) { - err = PTR_ERR(inode); - P9_DPRINTK(P9_DEBUG_VFS, "inode creation failed %d\n", err); - goto error; + if (!(perm & P9_DMLINK)) { + /* now walk from the parent so we can get unopened fid */ + fid = p9_client_walk(dfid, 1, &name, 1); + if (IS_ERR(fid)) { + err = PTR_ERR(fid); + p9_debug(P9_DEBUG_VFS, + "p9_client_walk failed %d\n", err); + fid = NULL; + goto error; + } + /* + * instantiate inode and assign the unopened fid to the dentry + */ + inode = v9fs_get_new_inode_from_fid(v9ses, fid, dir->i_sb); + if (IS_ERR(inode)) { + err = PTR_ERR(inode); + p9_debug(P9_DEBUG_VFS, + "inode creation failed %d\n", err); + goto error; + } + err = v9fs_fid_add(dentry, fid); + if (err < 0) + goto error; + d_instantiate(dentry, inode); } - err = v9fs_fid_add(dentry, fid); - if (err < 0) - goto error; - d_instantiate(dentry, inode); return ofid; error: if (ofid) @@ -703,7 +721,7 @@ error: */ static int -v9fs_vfs_create(struct inode *dir, struct dentry *dentry, int mode, +v9fs_vfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { int err; @@ -786,14 +804,14 @@ error: * */ -static int v9fs_vfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int v9fs_vfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { int err; u32 perm; struct p9_fid *fid; struct v9fs_session_info *v9ses; - P9_DPRINTK(P9_DEBUG_VFS, "name %s\n", dentry->d_name.name); + p9_debug(P9_DEBUG_VFS, "name %s\n", dentry->d_name.name); err = 0; v9ses = v9fs_inode2v9ses(dir); perm = unixmode2p9mode(v9ses, mode | S_IFDIR); @@ -831,8 +849,8 @@ struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, char *name; int result = 0; - P9_DPRINTK(P9_DEBUG_VFS, "dir: %p dentry: (%s) %p nameidata: %p\n", - dir, dentry->d_name.name, dentry, nameidata); + p9_debug(P9_DEBUG_VFS, "dir: %p dentry: (%s) %p nameidata: %p\n", + dir, dentry->d_name.name, dentry, nameidata); if (dentry->d_name.len > NAME_MAX) return ERR_PTR(-ENAMETOOLONG); @@ -938,7 +956,7 @@ v9fs_vfs_rename(struct inode *old_dir, struct dentry *old_dentry, struct p9_fid *newdirfid; struct p9_wstat wstat; - P9_DPRINTK(P9_DEBUG_VFS, "\n"); + p9_debug(P9_DEBUG_VFS, "\n"); retval = 0; old_inode = old_dentry->d_inode; new_inode = new_dentry->d_inode; @@ -974,8 +992,7 @@ v9fs_vfs_rename(struct inode *old_dir, struct dentry *old_dentry, * 9P .u can only handle file rename in the same directory */ - P9_DPRINTK(P9_DEBUG_ERROR, - "old dir and new dir are different\n"); + p9_debug(P9_DEBUG_ERROR, "old dir and new dir are different\n"); retval = -EXDEV; goto clunk_newdir; } @@ -1031,7 +1048,7 @@ v9fs_vfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct p9_fid *fid; struct p9_wstat *st; - P9_DPRINTK(P9_DEBUG_VFS, "dentry: %p\n", dentry); + p9_debug(P9_DEBUG_VFS, "dentry: %p\n", dentry); err = -EPERM; v9ses = v9fs_dentry2v9ses(dentry); if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) { @@ -1068,7 +1085,7 @@ static int v9fs_vfs_setattr(struct dentry *dentry, struct iattr *iattr) struct p9_fid *fid; struct p9_wstat wstat; - P9_DPRINTK(P9_DEBUG_VFS, "\n"); + p9_debug(P9_DEBUG_VFS, "\n"); retval = inode_change_ok(dentry->d_inode, iattr); if (retval) return retval; @@ -1131,7 +1148,7 @@ void v9fs_stat2inode(struct p9_wstat *stat, struct inode *inode, struct super_block *sb) { - mode_t mode; + umode_t mode; char ext[32]; char tag_name[14]; unsigned int i_nlink; @@ -1167,7 +1184,7 @@ v9fs_stat2inode(struct p9_wstat *stat, struct inode *inode, set_nlink(inode, i_nlink); } } - mode = stat->mode & S_IALLUGO; + mode = p9mode2perm(v9ses, stat); mode |= inode->i_mode & ~S_IALLUGO; inode->i_mode = mode; i_size_write(inode, stat->length); @@ -1213,7 +1230,7 @@ static int v9fs_readlink(struct dentry *dentry, char *buffer, int buflen) struct p9_fid *fid; struct p9_wstat *st; - P9_DPRINTK(P9_DEBUG_VFS, " %s\n", dentry->d_name.name); + p9_debug(P9_DEBUG_VFS, " %s\n", dentry->d_name.name); retval = -EPERM; v9ses = v9fs_dentry2v9ses(dentry); fid = v9fs_fid_lookup(dentry); @@ -1235,8 +1252,8 @@ static int v9fs_readlink(struct dentry *dentry, char *buffer, int buflen) /* copy extension buffer into buffer */ strncpy(buffer, st->extension, buflen); - P9_DPRINTK(P9_DEBUG_VFS, - "%s -> %s (%s)\n", dentry->d_name.name, st->extension, buffer); + p9_debug(P9_DEBUG_VFS, "%s -> %s (%s)\n", + dentry->d_name.name, st->extension, buffer); retval = strnlen(buffer, buflen); done: @@ -1257,7 +1274,7 @@ static void *v9fs_vfs_follow_link(struct dentry *dentry, struct nameidata *nd) int len = 0; char *link = __getname(); - P9_DPRINTK(P9_DEBUG_VFS, "%s n", dentry->d_name.name); + p9_debug(P9_DEBUG_VFS, "%s\n", dentry->d_name.name); if (!link) link = ERR_PTR(-ENOMEM); @@ -1288,8 +1305,8 @@ v9fs_vfs_put_link(struct dentry *dentry, struct nameidata *nd, void *p) { char *s = nd_get_link(nd); - P9_DPRINTK(P9_DEBUG_VFS, " %s %s\n", dentry->d_name.name, - IS_ERR(s) ? "<error>" : s); + p9_debug(P9_DEBUG_VFS, " %s %s\n", + dentry->d_name.name, IS_ERR(s) ? "<error>" : s); if (!IS_ERR(s)) __putname(s); } @@ -1304,19 +1321,17 @@ v9fs_vfs_put_link(struct dentry *dentry, struct nameidata *nd, void *p) */ static int v9fs_vfs_mkspecial(struct inode *dir, struct dentry *dentry, - int mode, const char *extension) + u32 perm, const char *extension) { - u32 perm; struct p9_fid *fid; struct v9fs_session_info *v9ses; v9ses = v9fs_inode2v9ses(dir); if (!v9fs_proto_dotu(v9ses)) { - P9_DPRINTK(P9_DEBUG_ERROR, "not extended\n"); + p9_debug(P9_DEBUG_ERROR, "not extended\n"); return -EPERM; } - perm = unixmode2p9mode(v9ses, mode); fid = v9fs_create(v9ses, dir, dentry, (char *) extension, perm, P9_OREAD); if (IS_ERR(fid)) @@ -1340,10 +1355,10 @@ static int v9fs_vfs_mkspecial(struct inode *dir, struct dentry *dentry, static int v9fs_vfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) { - P9_DPRINTK(P9_DEBUG_VFS, " %lu,%s,%s\n", dir->i_ino, - dentry->d_name.name, symname); + p9_debug(P9_DEBUG_VFS, " %lu,%s,%s\n", + dir->i_ino, dentry->d_name.name, symname); - return v9fs_vfs_mkspecial(dir, dentry, S_IFLNK, symname); + return v9fs_vfs_mkspecial(dir, dentry, P9_DMSYMLINK, symname); } /** @@ -1362,9 +1377,8 @@ v9fs_vfs_link(struct dentry *old_dentry, struct inode *dir, char *name; struct p9_fid *oldfid; - P9_DPRINTK(P9_DEBUG_VFS, - " %lu,%s,%s\n", dir->i_ino, dentry->d_name.name, - old_dentry->d_name.name); + p9_debug(P9_DEBUG_VFS, " %lu,%s,%s\n", + dir->i_ino, dentry->d_name.name, old_dentry->d_name.name); oldfid = v9fs_fid_clone(old_dentry); if (IS_ERR(oldfid)) @@ -1398,14 +1412,16 @@ clunk_fid: */ static int -v9fs_vfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev) +v9fs_vfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) { + struct v9fs_session_info *v9ses = v9fs_inode2v9ses(dir); int retval; char *name; + u32 perm; - P9_DPRINTK(P9_DEBUG_VFS, - " %lu,%s mode: %x MAJOR: %u MINOR: %u\n", dir->i_ino, - dentry->d_name.name, mode, MAJOR(rdev), MINOR(rdev)); + p9_debug(P9_DEBUG_VFS, " %lu,%s mode: %hx MAJOR: %u MINOR: %u\n", + dir->i_ino, dentry->d_name.name, mode, + MAJOR(rdev), MINOR(rdev)); if (!new_valid_dev(rdev)) return -EINVAL; @@ -1427,7 +1443,8 @@ v9fs_vfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev) return -EINVAL; } - retval = v9fs_vfs_mkspecial(dir, dentry, mode, name); + perm = unixmode2p9mode(v9ses, mode); + retval = v9fs_vfs_mkspecial(dir, dentry, perm, name); __putname(name); return retval; diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c index 0b5745e21946..a1e6c990cd41 100644 --- a/fs/9p/vfs_inode_dotl.c +++ b/fs/9p/vfs_inode_dotl.c @@ -48,7 +48,7 @@ #include "acl.h" static int -v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, int omode, +v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, umode_t omode, dev_t rdev); /** @@ -253,7 +253,7 @@ int v9fs_open_to_dotl_flags(int flags) */ static int -v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, int omode, +v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, umode_t omode, struct nameidata *nd) { int err = 0; @@ -283,13 +283,13 @@ v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, int omode, } name = (char *) dentry->d_name.name; - P9_DPRINTK(P9_DEBUG_VFS, "v9fs_vfs_create_dotl: name:%s flags:0x%x " - "mode:0x%x\n", name, flags, omode); + p9_debug(P9_DEBUG_VFS, "name:%s flags:0x%x mode:0x%hx\n", + name, flags, omode); dfid = v9fs_fid_lookup(dentry->d_parent); if (IS_ERR(dfid)) { err = PTR_ERR(dfid); - P9_DPRINTK(P9_DEBUG_VFS, "fid lookup failed %d\n", err); + p9_debug(P9_DEBUG_VFS, "fid lookup failed %d\n", err); return err; } @@ -297,7 +297,7 @@ v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, int omode, ofid = p9_client_walk(dfid, 0, NULL, 1); if (IS_ERR(ofid)) { err = PTR_ERR(ofid); - P9_DPRINTK(P9_DEBUG_VFS, "p9_client_walk failed %d\n", err); + p9_debug(P9_DEBUG_VFS, "p9_client_walk failed %d\n", err); return err; } @@ -307,16 +307,15 @@ v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, int omode, /* Update mode based on ACL value */ err = v9fs_acl_mode(dir, &mode, &dacl, &pacl); if (err) { - P9_DPRINTK(P9_DEBUG_VFS, - "Failed to get acl values in creat %d\n", err); + p9_debug(P9_DEBUG_VFS, "Failed to get acl values in creat %d\n", + err); goto error; } err = p9_client_create_dotl(ofid, name, v9fs_open_to_dotl_flags(flags), mode, gid, &qid); if (err < 0) { - P9_DPRINTK(P9_DEBUG_VFS, - "p9_client_open_dotl failed in creat %d\n", - err); + p9_debug(P9_DEBUG_VFS, "p9_client_open_dotl failed in creat %d\n", + err); goto error; } v9fs_invalidate_inode_attr(dir); @@ -325,14 +324,14 @@ v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, int omode, fid = p9_client_walk(dfid, 1, &name, 1); if (IS_ERR(fid)) { err = PTR_ERR(fid); - P9_DPRINTK(P9_DEBUG_VFS, "p9_client_walk failed %d\n", err); + p9_debug(P9_DEBUG_VFS, "p9_client_walk failed %d\n", err); fid = NULL; goto error; } inode = v9fs_get_new_inode_from_fid(v9ses, fid, dir->i_sb); if (IS_ERR(inode)) { err = PTR_ERR(inode); - P9_DPRINTK(P9_DEBUG_VFS, "inode creation failed %d\n", err); + p9_debug(P9_DEBUG_VFS, "inode creation failed %d\n", err); goto error; } err = v9fs_fid_add(dentry, fid); @@ -395,7 +394,7 @@ err_clunk_old_fid: */ static int v9fs_vfs_mkdir_dotl(struct inode *dir, - struct dentry *dentry, int omode) + struct dentry *dentry, umode_t omode) { int err; struct v9fs_session_info *v9ses; @@ -408,7 +407,7 @@ static int v9fs_vfs_mkdir_dotl(struct inode *dir, struct dentry *dir_dentry; struct posix_acl *dacl = NULL, *pacl = NULL; - P9_DPRINTK(P9_DEBUG_VFS, "name %s\n", dentry->d_name.name); + p9_debug(P9_DEBUG_VFS, "name %s\n", dentry->d_name.name); err = 0; v9ses = v9fs_inode2v9ses(dir); @@ -420,7 +419,7 @@ static int v9fs_vfs_mkdir_dotl(struct inode *dir, dfid = v9fs_fid_lookup(dir_dentry); if (IS_ERR(dfid)) { err = PTR_ERR(dfid); - P9_DPRINTK(P9_DEBUG_VFS, "fid lookup failed %d\n", err); + p9_debug(P9_DEBUG_VFS, "fid lookup failed %d\n", err); dfid = NULL; goto error; } @@ -430,8 +429,8 @@ static int v9fs_vfs_mkdir_dotl(struct inode *dir, /* Update mode based on ACL value */ err = v9fs_acl_mode(dir, &mode, &dacl, &pacl); if (err) { - P9_DPRINTK(P9_DEBUG_VFS, - "Failed to get acl values in mkdir %d\n", err); + p9_debug(P9_DEBUG_VFS, "Failed to get acl values in mkdir %d\n", + err); goto error; } name = (char *) dentry->d_name.name; @@ -444,8 +443,8 @@ static int v9fs_vfs_mkdir_dotl(struct inode *dir, fid = p9_client_walk(dfid, 1, &name, 1); if (IS_ERR(fid)) { err = PTR_ERR(fid); - P9_DPRINTK(P9_DEBUG_VFS, "p9_client_walk failed %d\n", - err); + p9_debug(P9_DEBUG_VFS, "p9_client_walk failed %d\n", + err); fid = NULL; goto error; } @@ -453,8 +452,8 @@ static int v9fs_vfs_mkdir_dotl(struct inode *dir, inode = v9fs_get_new_inode_from_fid(v9ses, fid, dir->i_sb); if (IS_ERR(inode)) { err = PTR_ERR(inode); - P9_DPRINTK(P9_DEBUG_VFS, "inode creation failed %d\n", - err); + p9_debug(P9_DEBUG_VFS, "inode creation failed %d\n", + err); goto error; } err = v9fs_fid_add(dentry, fid); @@ -495,7 +494,7 @@ v9fs_vfs_getattr_dotl(struct vfsmount *mnt, struct dentry *dentry, struct p9_fid *fid; struct p9_stat_dotl *st; - P9_DPRINTK(P9_DEBUG_VFS, "dentry: %p\n", dentry); + p9_debug(P9_DEBUG_VFS, "dentry: %p\n", dentry); err = -EPERM; v9ses = v9fs_dentry2v9ses(dentry); if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) { @@ -523,6 +522,46 @@ v9fs_vfs_getattr_dotl(struct vfsmount *mnt, struct dentry *dentry, return 0; } +/* + * Attribute flags. + */ +#define P9_ATTR_MODE (1 << 0) +#define P9_ATTR_UID (1 << 1) +#define P9_ATTR_GID (1 << 2) +#define P9_ATTR_SIZE (1 << 3) +#define P9_ATTR_ATIME (1 << 4) +#define P9_ATTR_MTIME (1 << 5) +#define P9_ATTR_CTIME (1 << 6) +#define P9_ATTR_ATIME_SET (1 << 7) +#define P9_ATTR_MTIME_SET (1 << 8) + +struct dotl_iattr_map { + int iattr_valid; + int p9_iattr_valid; +}; + +static int v9fs_mapped_iattr_valid(int iattr_valid) +{ + int i; + int p9_iattr_valid = 0; + struct dotl_iattr_map dotl_iattr_map[] = { + { ATTR_MODE, P9_ATTR_MODE }, + { ATTR_UID, P9_ATTR_UID }, + { ATTR_GID, P9_ATTR_GID }, + { ATTR_SIZE, P9_ATTR_SIZE }, + { ATTR_ATIME, P9_ATTR_ATIME }, + { ATTR_MTIME, P9_ATTR_MTIME }, + { ATTR_CTIME, P9_ATTR_CTIME }, + { ATTR_ATIME_SET, P9_ATTR_ATIME_SET }, + { ATTR_MTIME_SET, P9_ATTR_MTIME_SET }, + }; + for (i = 0; i < ARRAY_SIZE(dotl_iattr_map); i++) { + if (iattr_valid & dotl_iattr_map[i].iattr_valid) + p9_iattr_valid |= dotl_iattr_map[i].p9_iattr_valid; + } + return p9_iattr_valid; +} + /** * v9fs_vfs_setattr_dotl - set file metadata * @dentry: file whose metadata to set @@ -537,13 +576,13 @@ int v9fs_vfs_setattr_dotl(struct dentry *dentry, struct iattr *iattr) struct p9_fid *fid; struct p9_iattr_dotl p9attr; - P9_DPRINTK(P9_DEBUG_VFS, "\n"); + p9_debug(P9_DEBUG_VFS, "\n"); retval = inode_change_ok(dentry->d_inode, iattr); if (retval) return retval; - p9attr.valid = iattr->ia_valid; + p9attr.valid = v9fs_mapped_iattr_valid(iattr->ia_valid); p9attr.mode = iattr->ia_mode; p9attr.uid = iattr->ia_uid; p9attr.gid = iattr->ia_gid; @@ -594,7 +633,7 @@ int v9fs_vfs_setattr_dotl(struct dentry *dentry, struct iattr *iattr) void v9fs_stat2inode_dotl(struct p9_stat_dotl *stat, struct inode *inode) { - mode_t mode; + umode_t mode; struct v9fs_inode *v9inode = V9FS_I(inode); if ((stat->st_result_mask & P9_STATS_BASIC) == P9_STATS_BASIC) { @@ -670,14 +709,13 @@ v9fs_vfs_symlink_dotl(struct inode *dir, struct dentry *dentry, struct v9fs_session_info *v9ses; name = (char *) dentry->d_name.name; - P9_DPRINTK(P9_DEBUG_VFS, "v9fs_vfs_symlink_dotl : %lu,%s,%s\n", - dir->i_ino, name, symname); + p9_debug(P9_DEBUG_VFS, "%lu,%s,%s\n", dir->i_ino, name, symname); v9ses = v9fs_inode2v9ses(dir); dfid = v9fs_fid_lookup(dentry->d_parent); if (IS_ERR(dfid)) { err = PTR_ERR(dfid); - P9_DPRINTK(P9_DEBUG_VFS, "fid lookup failed %d\n", err); + p9_debug(P9_DEBUG_VFS, "fid lookup failed %d\n", err); return err; } @@ -687,7 +725,7 @@ v9fs_vfs_symlink_dotl(struct inode *dir, struct dentry *dentry, err = p9_client_symlink(dfid, name, (char *)symname, gid, &qid); if (err < 0) { - P9_DPRINTK(P9_DEBUG_VFS, "p9_client_symlink failed %d\n", err); + p9_debug(P9_DEBUG_VFS, "p9_client_symlink failed %d\n", err); goto error; } @@ -697,8 +735,8 @@ v9fs_vfs_symlink_dotl(struct inode *dir, struct dentry *dentry, fid = p9_client_walk(dfid, 1, &name, 1); if (IS_ERR(fid)) { err = PTR_ERR(fid); - P9_DPRINTK(P9_DEBUG_VFS, "p9_client_walk failed %d\n", - err); + p9_debug(P9_DEBUG_VFS, "p9_client_walk failed %d\n", + err); fid = NULL; goto error; } @@ -707,8 +745,8 @@ v9fs_vfs_symlink_dotl(struct inode *dir, struct dentry *dentry, inode = v9fs_get_new_inode_from_fid(v9ses, fid, dir->i_sb); if (IS_ERR(inode)) { err = PTR_ERR(inode); - P9_DPRINTK(P9_DEBUG_VFS, "inode creation failed %d\n", - err); + p9_debug(P9_DEBUG_VFS, "inode creation failed %d\n", + err); goto error; } err = v9fs_fid_add(dentry, fid); @@ -751,9 +789,8 @@ v9fs_vfs_link_dotl(struct dentry *old_dentry, struct inode *dir, struct p9_fid *dfid, *oldfid; struct v9fs_session_info *v9ses; - P9_DPRINTK(P9_DEBUG_VFS, "dir ino: %lu, old_name: %s, new_name: %s\n", - dir->i_ino, old_dentry->d_name.name, - dentry->d_name.name); + p9_debug(P9_DEBUG_VFS, "dir ino: %lu, old_name: %s, new_name: %s\n", + dir->i_ino, old_dentry->d_name.name, dentry->d_name.name); v9ses = v9fs_inode2v9ses(dir); dir_dentry = v9fs_dentry_from_dir_inode(dir); @@ -770,7 +807,7 @@ v9fs_vfs_link_dotl(struct dentry *old_dentry, struct inode *dir, err = p9_client_link(dfid, oldfid, (char *)dentry->d_name.name); if (err < 0) { - P9_DPRINTK(P9_DEBUG_VFS, "p9_client_link failed %d\n", err); + p9_debug(P9_DEBUG_VFS, "p9_client_link failed %d\n", err); return err; } @@ -799,7 +836,7 @@ v9fs_vfs_link_dotl(struct dentry *old_dentry, struct inode *dir, * */ static int -v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, int omode, +v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, umode_t omode, dev_t rdev) { int err; @@ -813,9 +850,9 @@ v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, int omode, struct dentry *dir_dentry; struct posix_acl *dacl = NULL, *pacl = NULL; - P9_DPRINTK(P9_DEBUG_VFS, - " %lu,%s mode: %x MAJOR: %u MINOR: %u\n", dir->i_ino, - dentry->d_name.name, omode, MAJOR(rdev), MINOR(rdev)); + p9_debug(P9_DEBUG_VFS, " %lu,%s mode: %hx MAJOR: %u MINOR: %u\n", + dir->i_ino, dentry->d_name.name, omode, + MAJOR(rdev), MINOR(rdev)); if (!new_valid_dev(rdev)) return -EINVAL; @@ -825,7 +862,7 @@ v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, int omode, dfid = v9fs_fid_lookup(dir_dentry); if (IS_ERR(dfid)) { err = PTR_ERR(dfid); - P9_DPRINTK(P9_DEBUG_VFS, "fid lookup failed %d\n", err); + p9_debug(P9_DEBUG_VFS, "fid lookup failed %d\n", err); dfid = NULL; goto error; } @@ -835,8 +872,8 @@ v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, int omode, /* Update mode based on ACL value */ err = v9fs_acl_mode(dir, &mode, &dacl, &pacl); if (err) { - P9_DPRINTK(P9_DEBUG_VFS, - "Failed to get acl values in mknod %d\n", err); + p9_debug(P9_DEBUG_VFS, "Failed to get acl values in mknod %d\n", + err); goto error; } name = (char *) dentry->d_name.name; @@ -851,8 +888,8 @@ v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, int omode, fid = p9_client_walk(dfid, 1, &name, 1); if (IS_ERR(fid)) { err = PTR_ERR(fid); - P9_DPRINTK(P9_DEBUG_VFS, "p9_client_walk failed %d\n", - err); + p9_debug(P9_DEBUG_VFS, "p9_client_walk failed %d\n", + err); fid = NULL; goto error; } @@ -860,8 +897,8 @@ v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, int omode, inode = v9fs_get_new_inode_from_fid(v9ses, fid, dir->i_sb); if (IS_ERR(inode)) { err = PTR_ERR(inode); - P9_DPRINTK(P9_DEBUG_VFS, "inode creation failed %d\n", - err); + p9_debug(P9_DEBUG_VFS, "inode creation failed %d\n", + err); goto error; } err = v9fs_fid_add(dentry, fid); @@ -905,7 +942,7 @@ v9fs_vfs_follow_link_dotl(struct dentry *dentry, struct nameidata *nd) char *link = __getname(); char *target; - P9_DPRINTK(P9_DEBUG_VFS, "%s\n", dentry->d_name.name); + p9_debug(P9_DEBUG_VFS, "%s\n", dentry->d_name.name); if (!link) { link = ERR_PTR(-ENOMEM); diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c index c70251d47ed1..7b0cd87b07c2 100644 --- a/fs/9p/vfs_super.c +++ b/fs/9p/vfs_super.c @@ -117,11 +117,11 @@ static struct dentry *v9fs_mount(struct file_system_type *fs_type, int flags, struct inode *inode = NULL; struct dentry *root = NULL; struct v9fs_session_info *v9ses = NULL; - int mode = S_IRWXUGO | S_ISVTX; + umode_t mode = S_IRWXUGO | S_ISVTX; struct p9_fid *fid; int retval = 0; - P9_DPRINTK(P9_DEBUG_VFS, " \n"); + p9_debug(P9_DEBUG_VFS, "\n"); v9ses = kzalloc(sizeof(struct v9fs_session_info), GFP_KERNEL); if (!v9ses) @@ -191,7 +191,7 @@ static struct dentry *v9fs_mount(struct file_system_type *fs_type, int flags, goto release_sb; v9fs_fid_add(root, fid); - P9_DPRINTK(P9_DEBUG_VFS, " simple set mount, return 0\n"); + p9_debug(P9_DEBUG_VFS, " simple set mount, return 0\n"); return dget(sb->s_root); clunk_fid: @@ -223,7 +223,7 @@ static void v9fs_kill_super(struct super_block *s) { struct v9fs_session_info *v9ses = s->s_fs_info; - P9_DPRINTK(P9_DEBUG_VFS, " %p\n", s); + p9_debug(P9_DEBUG_VFS, " %p\n", s); kill_anon_super(s); @@ -231,7 +231,7 @@ static void v9fs_kill_super(struct super_block *s) v9fs_session_close(v9ses); kfree(v9ses); s->s_fs_info = NULL; - P9_DPRINTK(P9_DEBUG_VFS, "exiting kill_super\n"); + p9_debug(P9_DEBUG_VFS, "exiting kill_super\n"); } static void @@ -303,7 +303,7 @@ static int v9fs_write_inode(struct inode *inode, * send an fsync request to server irrespective of * wbc->sync_mode. */ - P9_DPRINTK(P9_DEBUG_VFS, "%s: inode %p\n", __func__, inode); + p9_debug(P9_DEBUG_VFS, "%s: inode %p\n", __func__, inode); v9inode = V9FS_I(inode); if (!v9inode->writeback_fid) return 0; @@ -326,7 +326,7 @@ static int v9fs_write_inode_dotl(struct inode *inode, * send an fsync request to server irrespective of * wbc->sync_mode. */ - P9_DPRINTK(P9_DEBUG_VFS, "%s: inode %p\n", __func__, inode); + p9_debug(P9_DEBUG_VFS, "%s: inode %p\n", __func__, inode); v9inode = V9FS_I(inode); if (!v9inode->writeback_fid) return 0; diff --git a/fs/9p/xattr.c b/fs/9p/xattr.c index d288773871b3..29653b70a9c3 100644 --- a/fs/9p/xattr.c +++ b/fs/9p/xattr.c @@ -32,8 +32,8 @@ ssize_t v9fs_fid_xattr_get(struct p9_fid *fid, const char *name, attr_fid = p9_client_xattrwalk(fid, name, &attr_size); if (IS_ERR(attr_fid)) { retval = PTR_ERR(attr_fid); - P9_DPRINTK(P9_DEBUG_VFS, - "p9_client_attrwalk failed %zd\n", retval); + p9_debug(P9_DEBUG_VFS, "p9_client_attrwalk failed %zd\n", + retval); attr_fid = NULL; goto error; } @@ -87,8 +87,8 @@ ssize_t v9fs_xattr_get(struct dentry *dentry, const char *name, { struct p9_fid *fid; - P9_DPRINTK(P9_DEBUG_VFS, "%s: name = %s value_len = %zu\n", - __func__, name, buffer_size); + p9_debug(P9_DEBUG_VFS, "name = %s value_len = %zu\n", + name, buffer_size); fid = v9fs_fid_lookup(dentry); if (IS_ERR(fid)) return PTR_ERR(fid); @@ -115,8 +115,8 @@ int v9fs_xattr_set(struct dentry *dentry, const char *name, int retval, msize, write_count; struct p9_fid *fid = NULL; - P9_DPRINTK(P9_DEBUG_VFS, "%s: name = %s value_len = %zu flags = %d\n", - __func__, name, value_len, flags); + p9_debug(P9_DEBUG_VFS, "name = %s value_len = %zu flags = %d\n", + name, value_len, flags); fid = v9fs_fid_clone(dentry); if (IS_ERR(fid)) { @@ -129,8 +129,8 @@ int v9fs_xattr_set(struct dentry *dentry, const char *name, */ retval = p9_client_xattrcreate(fid, name, value_len, flags); if (retval < 0) { - P9_DPRINTK(P9_DEBUG_VFS, - "p9_client_xattrcreate failed %d\n", retval); + p9_debug(P9_DEBUG_VFS, "p9_client_xattrcreate failed %d\n", + retval); goto error; } msize = fid->clnt->msize; diff --git a/fs/Kconfig b/fs/Kconfig index 5f4c45d4aa10..d621f02a3f9e 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -218,6 +218,8 @@ source "fs/exofs/Kconfig" endif # MISC_FILESYSTEMS +source "fs/exofs/Kconfig.ore" + menuconfig NETWORK_FILESYSTEMS bool "Network File Systems" default y @@ -266,14 +268,6 @@ source "fs/9p/Kconfig" endif # NETWORK_FILESYSTEMS -if BLOCK -menu "Partition Types" - -source "fs/partitions/Kconfig" - -endmenu -endif - source "fs/nls/Kconfig" source "fs/dlm/Kconfig" diff --git a/fs/Kconfig.binfmt b/fs/Kconfig.binfmt index 79e2ca7973b7..e95d1b64082c 100644 --- a/fs/Kconfig.binfmt +++ b/fs/Kconfig.binfmt @@ -27,6 +27,9 @@ config COMPAT_BINFMT_ELF bool depends on COMPAT && BINFMT_ELF +config ARCH_BINFMT_ELF_RANDOMIZE_PIE + bool + config BINFMT_ELF_FDPIC bool "Kernel support for FDPIC ELF binaries" default y diff --git a/fs/Makefile b/fs/Makefile index d2c3353d5477..93804d4d66e1 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -19,6 +19,8 @@ else obj-y += no-block.o endif +obj-$(CONFIG_PROC_FS) += proc_namespace.o + obj-$(CONFIG_BLK_DEV_INTEGRITY) += bio-integrity.o obj-y += notify/ obj-$(CONFIG_EPOLL) += eventpoll.o @@ -52,7 +54,6 @@ obj-$(CONFIG_FHANDLE) += fhandle.o obj-y += quota/ obj-$(CONFIG_PROC_FS) += proc/ -obj-y += partitions/ obj-$(CONFIG_SYSFS) += sysfs/ obj-$(CONFIG_CONFIGFS_FS) += configfs/ obj-y += devpts/ diff --git a/fs/adfs/super.c b/fs/adfs/super.c index c8bf36a1996a..8e3b36ace305 100644 --- a/fs/adfs/super.c +++ b/fs/adfs/super.c @@ -126,9 +126,9 @@ static void adfs_put_super(struct super_block *sb) sb->s_fs_info = NULL; } -static int adfs_show_options(struct seq_file *seq, struct vfsmount *mnt) +static int adfs_show_options(struct seq_file *seq, struct dentry *root) { - struct adfs_sb_info *asb = ADFS_SB(mnt->mnt_sb); + struct adfs_sb_info *asb = ADFS_SB(root->d_sb); if (asb->s_uid != 0) seq_printf(seq, ",uid=%u", asb->s_uid); diff --git a/fs/affs/affs.h b/fs/affs/affs.h index c2b9c79eb64e..45a0ce45d7b4 100644 --- a/fs/affs/affs.h +++ b/fs/affs/affs.h @@ -136,7 +136,7 @@ extern int affs_remove_header(struct dentry *dentry); extern u32 affs_checksum_block(struct super_block *sb, struct buffer_head *bh); extern void affs_fix_checksum(struct super_block *sb, struct buffer_head *bh); extern void secs_to_datestamp(time_t secs, struct affs_date *ds); -extern mode_t prot_to_mode(u32 prot); +extern umode_t prot_to_mode(u32 prot); extern void mode_to_prot(struct inode *inode); extern void affs_error(struct super_block *sb, const char *function, const char *fmt, ...); extern void affs_warning(struct super_block *sb, const char *function, const char *fmt, ...); @@ -156,8 +156,8 @@ extern void affs_free_bitmap(struct super_block *sb); extern int affs_hash_name(struct super_block *sb, const u8 *name, unsigned int len); extern struct dentry *affs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *); extern int affs_unlink(struct inode *dir, struct dentry *dentry); -extern int affs_create(struct inode *dir, struct dentry *dentry, int mode, struct nameidata *); -extern int affs_mkdir(struct inode *dir, struct dentry *dentry, int mode); +extern int affs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *); +extern int affs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode); extern int affs_rmdir(struct inode *dir, struct dentry *dentry); extern int affs_link(struct dentry *olddentry, struct inode *dir, struct dentry *dentry); diff --git a/fs/affs/amigaffs.c b/fs/affs/amigaffs.c index de37ec842340..52a6407682e6 100644 --- a/fs/affs/amigaffs.c +++ b/fs/affs/amigaffs.c @@ -390,10 +390,10 @@ secs_to_datestamp(time_t secs, struct affs_date *ds) ds->ticks = cpu_to_be32(secs * 50); } -mode_t +umode_t prot_to_mode(u32 prot) { - int mode = 0; + umode_t mode = 0; if (!(prot & FIBF_NOWRITE)) mode |= S_IWUSR; @@ -421,7 +421,7 @@ void mode_to_prot(struct inode *inode) { u32 prot = AFFS_I(inode)->i_protect; - mode_t mode = inode->i_mode; + umode_t mode = inode->i_mode; if (!(mode & S_IXUSR)) prot |= FIBF_NOEXECUTE; diff --git a/fs/affs/namei.c b/fs/affs/namei.c index 780a11dc6318..47806940aac0 100644 --- a/fs/affs/namei.c +++ b/fs/affs/namei.c @@ -255,13 +255,13 @@ affs_unlink(struct inode *dir, struct dentry *dentry) } int -affs_create(struct inode *dir, struct dentry *dentry, int mode, struct nameidata *nd) +affs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { struct super_block *sb = dir->i_sb; struct inode *inode; int error; - pr_debug("AFFS: create(%lu,\"%.*s\",0%o)\n",dir->i_ino,(int)dentry->d_name.len, + pr_debug("AFFS: create(%lu,\"%.*s\",0%ho)\n",dir->i_ino,(int)dentry->d_name.len, dentry->d_name.name,mode); inode = affs_new_inode(dir); @@ -285,12 +285,12 @@ affs_create(struct inode *dir, struct dentry *dentry, int mode, struct nameidata } int -affs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +affs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct inode *inode; int error; - pr_debug("AFFS: mkdir(%lu,\"%.*s\",0%o)\n",dir->i_ino, + pr_debug("AFFS: mkdir(%lu,\"%.*s\",0%ho)\n",dir->i_ino, (int)dentry->d_name.len,dentry->d_name.name,mode); inode = affs_new_inode(dir); diff --git a/fs/affs/super.c b/fs/affs/super.c index b31507d0f9b9..8ba73fed7964 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -98,7 +98,6 @@ static struct inode *affs_alloc_inode(struct super_block *sb) static void affs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(affs_inode_cachep, AFFS_I(inode)); } diff --git a/fs/afs/dir.c b/fs/afs/dir.c index 1b0b19550015..e22dc4b4a503 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -28,9 +28,9 @@ static int afs_d_delete(const struct dentry *dentry); static void afs_d_release(struct dentry *dentry); static int afs_lookup_filldir(void *_cookie, const char *name, int nlen, loff_t fpos, u64 ino, unsigned dtype); -static int afs_create(struct inode *dir, struct dentry *dentry, int mode, +static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd); -static int afs_mkdir(struct inode *dir, struct dentry *dentry, int mode); +static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode); static int afs_rmdir(struct inode *dir, struct dentry *dentry); static int afs_unlink(struct inode *dir, struct dentry *dentry); static int afs_link(struct dentry *from, struct inode *dir, @@ -764,7 +764,7 @@ static void afs_d_release(struct dentry *dentry) /* * create a directory on an AFS filesystem */ -static int afs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct afs_file_status status; struct afs_callback cb; @@ -777,7 +777,7 @@ static int afs_mkdir(struct inode *dir, struct dentry *dentry, int mode) dvnode = AFS_FS_I(dir); - _enter("{%x:%u},{%s},%o", + _enter("{%x:%u},{%s},%ho", dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name, mode); ret = -ENAMETOOLONG; @@ -948,7 +948,7 @@ error: /* * create a regular file on an AFS filesystem */ -static int afs_create(struct inode *dir, struct dentry *dentry, int mode, +static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { struct afs_file_status status; @@ -962,7 +962,7 @@ static int afs_create(struct inode *dir, struct dentry *dentry, int mode, dvnode = AFS_FS_I(dir); - _enter("{%x:%u},{%s},%o,", + _enter("{%x:%u},{%s},%ho,", dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name, mode); ret = -ENAMETOOLONG; diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c index aa59184151d0..8f4ce2658b7d 100644 --- a/fs/afs/mntpt.c +++ b/fs/afs/mntpt.c @@ -242,7 +242,7 @@ struct vfsmount *afs_d_automount(struct path *path) { struct vfsmount *newmnt; - _enter("{%s,%s}", path->mnt->mnt_devname, path->dentry->d_name.name); + _enter("{%s}", path->dentry->d_name.name); newmnt = afs_mntpt_do_automount(path->dentry); if (IS_ERR(newmnt)) @@ -252,7 +252,7 @@ struct vfsmount *afs_d_automount(struct path *path) mnt_set_expiry(newmnt, &afs_vfsmounts); queue_delayed_work(afs_wq, &afs_mntpt_expiry_timer, afs_mntpt_expiry_timeout * HZ); - _leave(" = %p {%s}", newmnt, newmnt->mnt_devname); + _leave(" = %p", newmnt); return newmnt; } diff --git a/fs/afs/super.c b/fs/afs/super.c index 356dcf0929e8..983ec59fc80d 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -495,7 +495,6 @@ static void afs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); struct afs_vnode *vnode = AFS_FS_I(inode); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(afs_inode_cachep, vnode); } @@ -476,14 +476,21 @@ static void kiocb_batch_init(struct kiocb_batch *batch, long total) batch->count = total; } -static void kiocb_batch_free(struct kiocb_batch *batch) +static void kiocb_batch_free(struct kioctx *ctx, struct kiocb_batch *batch) { struct kiocb *req, *n; + if (list_empty(&batch->head)) + return; + + spin_lock_irq(&ctx->ctx_lock); list_for_each_entry_safe(req, n, &batch->head, ki_batch) { list_del(&req->ki_batch); + list_del(&req->ki_list); kmem_cache_free(kiocb_cachep, req); + ctx->reqs_active--; } + spin_unlock_irq(&ctx->ctx_lock); } /* @@ -1742,7 +1749,7 @@ long do_io_submit(aio_context_t ctx_id, long nr, } blk_finish_plug(&plug); - kiocb_batch_free(&batch); + kiocb_batch_free(ctx, &batch); put_ioctx(ctx); return i ? i : ret; } diff --git a/fs/attr.c b/fs/attr.c index 7ee7ba488313..95053ad8abcc 100644 --- a/fs/attr.c +++ b/fs/attr.c @@ -166,7 +166,7 @@ EXPORT_SYMBOL(setattr_copy); int notify_change(struct dentry * dentry, struct iattr * attr) { struct inode *inode = dentry->d_inode; - mode_t mode = inode->i_mode; + umode_t mode = inode->i_mode; int error; struct timespec now; unsigned int ia_valid = attr->ia_valid; @@ -177,7 +177,7 @@ int notify_change(struct dentry * dentry, struct iattr * attr) } if ((ia_valid & ATTR_MODE)) { - mode_t amode = attr->ia_mode; + umode_t amode = attr->ia_mode; /* Flag setting protected by i_mutex */ if (is_sxid(amode)) inode->i_flags &= ~S_NOSEC; diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index 326dc08d3e3f..d8d8e7ba6a1e 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -116,6 +116,7 @@ struct autofs_sb_info { int needs_reghost; struct super_block *sb; struct mutex wq_mutex; + struct mutex pipe_mutex; spinlock_t fs_lock; struct autofs_wait_queue *queues; /* Wait queue pointer */ spinlock_t lookup_lock; @@ -155,7 +156,7 @@ static inline int autofs4_ispending(struct dentry *dentry) return 0; } -struct inode *autofs4_get_inode(struct super_block *, mode_t); +struct inode *autofs4_get_inode(struct super_block *, umode_t); void autofs4_free_ino(struct autofs_info *); /* Expiration */ diff --git a/fs/autofs4/dev-ioctl.c b/fs/autofs4/dev-ioctl.c index 509fe1eb66ae..76741d8d7786 100644 --- a/fs/autofs4/dev-ioctl.c +++ b/fs/autofs4/dev-ioctl.c @@ -194,7 +194,7 @@ static int find_autofs_mount(const char *pathname, return err; err = -ENOENT; while (path.dentry == path.mnt->mnt_root) { - if (path.mnt->mnt_sb->s_magic == AUTOFS_SUPER_MAGIC) { + if (path.dentry->d_sb->s_magic == AUTOFS_SUPER_MAGIC) { if (test(&path, data)) { path_get(&path); if (!err) /* already found some */ @@ -212,7 +212,7 @@ static int find_autofs_mount(const char *pathname, static int test_by_dev(struct path *path, void *p) { - return path->mnt->mnt_sb->s_dev == *(dev_t *)p; + return path->dentry->d_sb->s_dev == *(dev_t *)p; } static int test_by_type(struct path *path, void *p) @@ -538,11 +538,11 @@ static int autofs_dev_ioctl_ismountpoint(struct file *fp, err = find_autofs_mount(name, &path, test_by_type, &type); if (err) goto out; - devid = new_encode_dev(path.mnt->mnt_sb->s_dev); + devid = new_encode_dev(path.dentry->d_sb->s_dev); err = 0; if (path.mnt->mnt_root == path.dentry) { err = 1; - magic = path.mnt->mnt_sb->s_magic; + magic = path.dentry->d_sb->s_magic; } } else { dev_t dev = sbi->sb->s_dev; @@ -556,7 +556,7 @@ static int autofs_dev_ioctl_ismountpoint(struct file *fp, err = have_submounts(path.dentry); if (follow_down_one(&path)) - magic = path.mnt->mnt_sb->s_magic; + magic = path.dentry->d_sb->s_magic; } param->ismountpoint.out.devid = devid; diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index 8179f1ab8175..e16980b00b8d 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -70,10 +70,10 @@ out_kill_sb: kill_litter_super(sb); } -static int autofs4_show_options(struct seq_file *m, struct vfsmount *mnt) +static int autofs4_show_options(struct seq_file *m, struct dentry *root) { - struct autofs_sb_info *sbi = autofs4_sbi(mnt->mnt_sb); - struct inode *root_inode = mnt->mnt_sb->s_root->d_inode; + struct autofs_sb_info *sbi = autofs4_sbi(root->d_sb); + struct inode *root_inode = root->d_sb->s_root->d_inode; if (!sbi) return 0; @@ -225,6 +225,7 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent) sbi->min_proto = 0; sbi->max_proto = 0; mutex_init(&sbi->wq_mutex); + mutex_init(&sbi->pipe_mutex); spin_lock_init(&sbi->fs_lock); sbi->queues = NULL; spin_lock_init(&sbi->lookup_lock); @@ -326,7 +327,7 @@ fail_unlock: return -EINVAL; } -struct inode *autofs4_get_inode(struct super_block *sb, mode_t mode) +struct inode *autofs4_get_inode(struct super_block *sb, umode_t mode) { struct inode *inode = new_inode(sb); diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index f55ae23b137e..75e5f1c8e028 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -26,7 +26,7 @@ static int autofs4_dir_symlink(struct inode *,struct dentry *,const char *); static int autofs4_dir_unlink(struct inode *,struct dentry *); static int autofs4_dir_rmdir(struct inode *,struct dentry *); -static int autofs4_dir_mkdir(struct inode *,struct dentry *,int); +static int autofs4_dir_mkdir(struct inode *,struct dentry *,umode_t); static long autofs4_root_ioctl(struct file *,unsigned int,unsigned long); #ifdef CONFIG_COMPAT static long autofs4_root_compat_ioctl(struct file *,unsigned int,unsigned long); @@ -699,7 +699,7 @@ static int autofs4_dir_rmdir(struct inode *dir, struct dentry *dentry) return 0; } -static int autofs4_dir_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int autofs4_dir_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct autofs_sb_info *sbi = autofs4_sbi(dir->i_sb); struct autofs_info *ino = autofs4_dentry_ino(dentry); diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index e1fbdeef85db..da8876d38a7b 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -56,26 +56,27 @@ void autofs4_catatonic_mode(struct autofs_sb_info *sbi) mutex_unlock(&sbi->wq_mutex); } -static int autofs4_write(struct file *file, const void *addr, int bytes) +static int autofs4_write(struct autofs_sb_info *sbi, + struct file *file, const void *addr, int bytes) { unsigned long sigpipe, flags; mm_segment_t fs; const char *data = (const char *)addr; ssize_t wr = 0; - /** WARNING: this is not safe for writing more than PIPE_BUF bytes! **/ - sigpipe = sigismember(¤t->pending.signal, SIGPIPE); /* Save pointer to user space and point back to kernel space */ fs = get_fs(); set_fs(KERNEL_DS); + mutex_lock(&sbi->pipe_mutex); while (bytes && (wr = file->f_op->write(file,data,bytes,&file->f_pos)) > 0) { data += wr; bytes -= wr; } + mutex_unlock(&sbi->pipe_mutex); set_fs(fs); @@ -110,6 +111,13 @@ static void autofs4_notify_daemon(struct autofs_sb_info *sbi, pkt.hdr.proto_version = sbi->version; pkt.hdr.type = type; + mutex_lock(&sbi->wq_mutex); + + /* Check if we have become catatonic */ + if (sbi->catatonic) { + mutex_unlock(&sbi->wq_mutex); + return; + } switch (type) { /* Kernel protocol v4 missing and expire packets */ case autofs_ptype_missing: @@ -163,22 +171,18 @@ static void autofs4_notify_daemon(struct autofs_sb_info *sbi, } default: printk("autofs4_notify_daemon: bad type %d!\n", type); + mutex_unlock(&sbi->wq_mutex); return; } - /* Check if we have become catatonic */ - mutex_lock(&sbi->wq_mutex); - if (!sbi->catatonic) { - pipe = sbi->pipe; - get_file(pipe); - } + pipe = sbi->pipe; + get_file(pipe); + mutex_unlock(&sbi->wq_mutex); - if (pipe) { - if (autofs4_write(pipe, &pkt, pktsz)) - autofs4_catatonic_mode(sbi); - fput(pipe); - } + if (autofs4_write(sbi, pipe, &pkt, pktsz)) + autofs4_catatonic_mode(sbi); + fput(pipe); } static int autofs4_getpath(struct autofs_sb_info *sbi, @@ -257,6 +261,9 @@ static int validate_request(struct autofs_wait_queue **wait, struct autofs_wait_queue *wq; struct autofs_info *ino; + if (sbi->catatonic) + return -ENOENT; + /* Wait in progress, continue; */ wq = autofs4_find_wait(sbi, qstr); if (wq) { @@ -289,6 +296,9 @@ static int validate_request(struct autofs_wait_queue **wait, if (mutex_lock_interruptible(&sbi->wq_mutex)) return -EINTR; + if (sbi->catatonic) + return -ENOENT; + wq = autofs4_find_wait(sbi, qstr); if (wq) { *wait = wq; @@ -389,7 +399,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, ret = validate_request(&wq, sbi, &qstr, dentry, notify); if (ret <= 0) { - if (ret == 0) + if (ret != -EINTR) mutex_unlock(&sbi->wq_mutex); kfree(qstr.name); return ret; diff --git a/fs/bad_inode.c b/fs/bad_inode.c index 9205cf25f1c6..22e9a78872ff 100644 --- a/fs/bad_inode.c +++ b/fs/bad_inode.c @@ -173,7 +173,7 @@ static const struct file_operations bad_file_ops = }; static int bad_inode_create (struct inode *dir, struct dentry *dentry, - int mode, struct nameidata *nd) + umode_t mode, struct nameidata *nd) { return -EIO; } @@ -202,7 +202,7 @@ static int bad_inode_symlink (struct inode *dir, struct dentry *dentry, } static int bad_inode_mkdir(struct inode *dir, struct dentry *dentry, - int mode) + umode_t mode) { return -EIO; } @@ -213,7 +213,7 @@ static int bad_inode_rmdir (struct inode *dir, struct dentry *dentry) } static int bad_inode_mknod (struct inode *dir, struct dentry *dentry, - int mode, dev_t rdev) + umode_t mode, dev_t rdev) { return -EIO; } diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index 8342ca67abcd..6e6d536767fe 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -286,7 +286,6 @@ befs_alloc_inode(struct super_block *sb) static void befs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(befs_inode_cachep, BEFS_I(inode)); } diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c index 9cc074019479..d12c7966db27 100644 --- a/fs/bfs/dir.c +++ b/fs/bfs/dir.c @@ -84,7 +84,7 @@ const struct file_operations bfs_dir_operations = { extern void dump_imap(const char *, struct super_block *); -static int bfs_create(struct inode *dir, struct dentry *dentry, int mode, +static int bfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { int err; diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c index 697af5bf70b3..b0391bc402b1 100644 --- a/fs/bfs/inode.c +++ b/fs/bfs/inode.c @@ -251,7 +251,6 @@ static struct inode *bfs_alloc_inode(struct super_block *sb) static void bfs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(bfs_inode_cachep, BFS_I(inode)); } diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 21ac5ee4b43f..bcb884e2d613 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -794,7 +794,7 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) * default mmap base, as well as whatever program they * might try to exec. This is because the brk will * follow the loader, and is not movable. */ -#if defined(CONFIG_X86) || defined(CONFIG_ARM) +#ifdef CONFIG_ARCH_BINFMT_ELF_RANDOMIZE_PIE /* Memory randomization might have been switched off * in runtime via sysctl. * If that is the case, retain the original non-zero diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index 1e9edbdeda7e..a9198dfd5f85 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -560,7 +560,7 @@ static ssize_t bm_entry_write(struct file *file, const char __user *buffer, break; case 2: set_bit(Enabled, &e->flags); break; - case 3: root = dget(file->f_path.mnt->mnt_sb->s_root); + case 3: root = dget(file->f_path.dentry->d_sb->s_root); mutex_lock(&root->d_inode->i_mutex); kill_node(e); @@ -587,7 +587,7 @@ static ssize_t bm_register_write(struct file *file, const char __user *buffer, Node *e; struct inode *inode; struct dentry *root, *dentry; - struct super_block *sb = file->f_path.mnt->mnt_sb; + struct super_block *sb = file->f_path.dentry->d_sb; int err = 0; e = create_entry(buffer, count); @@ -666,7 +666,7 @@ static ssize_t bm_status_write(struct file * file, const char __user * buffer, switch (res) { case 1: enabled = 0; break; case 2: enabled = 1; break; - case 3: root = dget(file->f_path.mnt->mnt_sb->s_root); + case 3: root = dget(file->f_path.dentry->d_sb->s_root); mutex_lock(&root->d_inode->i_mutex); while (!list_empty(&entries)) @@ -337,7 +337,7 @@ static void bio_fs_destructor(struct bio *bio) * RETURNS: * Pointer to new bio on success, NULL on failure. */ -struct bio *bio_alloc(gfp_t gfp_mask, int nr_iovecs) +struct bio *bio_alloc(gfp_t gfp_mask, unsigned int nr_iovecs) { struct bio *bio = bio_alloc_bioset(gfp_mask, nr_iovecs, fs_bio_set); @@ -365,7 +365,7 @@ static void bio_kmalloc_destructor(struct bio *bio) * %__GFP_WAIT, the allocation is guaranteed to succeed. * **/ -struct bio *bio_kmalloc(gfp_t gfp_mask, int nr_iovecs) +struct bio *bio_kmalloc(gfp_t gfp_mask, unsigned int nr_iovecs) { struct bio *bio; @@ -696,7 +696,8 @@ static void bio_free_map_data(struct bio_map_data *bmd) kfree(bmd); } -static struct bio_map_data *bio_alloc_map_data(int nr_segs, int iov_count, +static struct bio_map_data *bio_alloc_map_data(int nr_segs, + unsigned int iov_count, gfp_t gfp_mask) { struct bio_map_data *bmd; diff --git a/fs/block_dev.c b/fs/block_dev.c index b07f1da1de4e..0e575d1304b4 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -17,6 +17,7 @@ #include <linux/module.h> #include <linux/blkpg.h> #include <linux/buffer_head.h> +#include <linux/swap.h> #include <linux/pagevec.h> #include <linux/writeback.h> #include <linux/mpage.h> @@ -24,7 +25,7 @@ #include <linux/uio.h> #include <linux/namei.h> #include <linux/log2.h> -#include <linux/kmemleak.h> +#include <linux/cleancache.h> #include <asm/uaccess.h> #include "internal.h" @@ -82,13 +83,35 @@ static sector_t max_block(struct block_device *bdev) } /* Kill _all_ buffers and pagecache , dirty or not.. */ -static void kill_bdev(struct block_device *bdev) +void kill_bdev(struct block_device *bdev) { - if (bdev->bd_inode->i_mapping->nrpages == 0) + struct address_space *mapping = bdev->bd_inode->i_mapping; + + if (mapping->nrpages == 0) return; + invalidate_bh_lrus(); - truncate_inode_pages(bdev->bd_inode->i_mapping, 0); + truncate_inode_pages(mapping, 0); } +EXPORT_SYMBOL(kill_bdev); + +/* Invalidate clean unused buffers and pagecache. */ +void invalidate_bdev(struct block_device *bdev) +{ + struct address_space *mapping = bdev->bd_inode->i_mapping; + + if (mapping->nrpages == 0) + return; + + invalidate_bh_lrus(); + lru_add_drain_all(); /* make sure all lru add caches are flushed */ + invalidate_mapping_pages(mapping, 0, -1); + /* 99% of the time, we don't need to flush the cleancache on the bdev. + * But, for the strange corners, lets be cautious + */ + cleancache_flush_inode(mapping); +} +EXPORT_SYMBOL(invalidate_bdev); int set_blocksize(struct block_device *bdev, int size) { @@ -425,7 +448,6 @@ static void bdev_i_callback(struct rcu_head *head) struct inode *inode = container_of(head, struct inode, i_rcu); struct bdev_inode *bdi = BDEV_I(inode); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(bdev_cachep, bdi); } @@ -493,12 +515,12 @@ static struct file_system_type bd_type = { .kill_sb = kill_anon_super, }; -struct super_block *blockdev_superblock __read_mostly; +static struct super_block *blockdev_superblock __read_mostly; void __init bdev_cache_init(void) { int err; - struct vfsmount *bd_mnt; + static struct vfsmount *bd_mnt; bdev_cachep = kmem_cache_create("bdev_cache", sizeof(struct bdev_inode), 0, (SLAB_HWCACHE_ALIGN|SLAB_RECLAIM_ACCOUNT| @@ -510,12 +532,7 @@ void __init bdev_cache_init(void) bd_mnt = kern_mount(&bd_type); if (IS_ERR(bd_mnt)) panic("Cannot create bdev pseudo-fs"); - /* - * This vfsmount structure is only used to obtain the - * blockdev_superblock, so tell kmemleak not to report it. - */ - kmemleak_not_leak(bd_mnt); - blockdev_superblock = bd_mnt->mnt_sb; /* For writeback */ + blockdev_superblock = bd_mnt->mnt_sb; /* For writeback */ } /* @@ -639,6 +656,11 @@ static struct block_device *bd_acquire(struct inode *inode) return bdev; } +static inline int sb_is_blkdev_sb(struct super_block *sb) +{ + return sb == blockdev_superblock; +} + /* Call when you free inode */ void bd_forget(struct inode *inode) @@ -1117,6 +1139,7 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) mutex_lock_nested(&bdev->bd_mutex, for_part); if (!bdev->bd_openers) { bdev->bd_disk = disk; + bdev->bd_queue = disk->queue; bdev->bd_contains = bdev; if (!partno) { struct backing_dev_info *bdi; @@ -1137,6 +1160,7 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) disk_put_part(bdev->bd_part); bdev->bd_part = NULL; bdev->bd_disk = NULL; + bdev->bd_queue = NULL; mutex_unlock(&bdev->bd_mutex); disk_unblock_events(disk); put_disk(disk); @@ -1210,6 +1234,7 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) disk_put_part(bdev->bd_part); bdev->bd_disk = NULL; bdev->bd_part = NULL; + bdev->bd_queue = NULL; bdev_inode_switch_bdi(bdev->bd_inode, &default_backing_dev_info); if (bdev != bdev->bd_contains) __blkdev_put(bdev->bd_contains, mode, 1); diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c index 7ec14097fef1..0cc20b35c1c4 100644 --- a/fs/btrfs/async-thread.c +++ b/fs/btrfs/async-thread.c @@ -64,6 +64,8 @@ struct btrfs_worker_thread { int idle; }; +static int __btrfs_start_workers(struct btrfs_workers *workers); + /* * btrfs_start_workers uses kthread_run, which can block waiting for memory * for a very long time. It will actually throttle on page writeback, @@ -88,27 +90,10 @@ static void start_new_worker_func(struct btrfs_work *work) { struct worker_start *start; start = container_of(work, struct worker_start, work); - btrfs_start_workers(start->queue, 1); + __btrfs_start_workers(start->queue); kfree(start); } -static int start_new_worker(struct btrfs_workers *queue) -{ - struct worker_start *start; - int ret; - - start = kzalloc(sizeof(*start), GFP_NOFS); - if (!start) - return -ENOMEM; - - start->work.func = start_new_worker_func; - start->queue = queue; - ret = btrfs_queue_worker(queue->atomic_worker_start, &start->work); - if (ret) - kfree(start); - return ret; -} - /* * helper function to move a thread onto the idle list after it * has finished some requests. @@ -153,12 +138,20 @@ static void check_busy_worker(struct btrfs_worker_thread *worker) static void check_pending_worker_creates(struct btrfs_worker_thread *worker) { struct btrfs_workers *workers = worker->workers; + struct worker_start *start; unsigned long flags; rmb(); if (!workers->atomic_start_pending) return; + start = kzalloc(sizeof(*start), GFP_NOFS); + if (!start) + return; + + start->work.func = start_new_worker_func; + start->queue = workers; + spin_lock_irqsave(&workers->lock, flags); if (!workers->atomic_start_pending) goto out; @@ -170,10 +163,11 @@ static void check_pending_worker_creates(struct btrfs_worker_thread *worker) workers->num_workers_starting += 1; spin_unlock_irqrestore(&workers->lock, flags); - start_new_worker(workers); + btrfs_queue_worker(workers->atomic_worker_start, &start->work); return; out: + kfree(start); spin_unlock_irqrestore(&workers->lock, flags); } @@ -331,7 +325,7 @@ again: run_ordered_completions(worker->workers, work); check_pending_worker_creates(worker); - + cond_resched(); } spin_lock_irq(&worker->lock); @@ -340,7 +334,7 @@ again: if (freezing(current)) { worker->working = 0; spin_unlock_irq(&worker->lock); - refrigerator(); + try_to_freeze(); } else { spin_unlock_irq(&worker->lock); if (!kthread_should_stop()) { @@ -462,56 +456,55 @@ void btrfs_init_workers(struct btrfs_workers *workers, char *name, int max, * starts new worker threads. This does not enforce the max worker * count in case you need to temporarily go past it. */ -static int __btrfs_start_workers(struct btrfs_workers *workers, - int num_workers) +static int __btrfs_start_workers(struct btrfs_workers *workers) { struct btrfs_worker_thread *worker; int ret = 0; - int i; - for (i = 0; i < num_workers; i++) { - worker = kzalloc(sizeof(*worker), GFP_NOFS); - if (!worker) { - ret = -ENOMEM; - goto fail; - } + worker = kzalloc(sizeof(*worker), GFP_NOFS); + if (!worker) { + ret = -ENOMEM; + goto fail; + } - INIT_LIST_HEAD(&worker->pending); - INIT_LIST_HEAD(&worker->prio_pending); - INIT_LIST_HEAD(&worker->worker_list); - spin_lock_init(&worker->lock); - - atomic_set(&worker->num_pending, 0); - atomic_set(&worker->refs, 1); - worker->workers = workers; - worker->task = kthread_run(worker_loop, worker, - "btrfs-%s-%d", workers->name, - workers->num_workers + i); - if (IS_ERR(worker->task)) { - ret = PTR_ERR(worker->task); - kfree(worker); - goto fail; - } - spin_lock_irq(&workers->lock); - list_add_tail(&worker->worker_list, &workers->idle_list); - worker->idle = 1; - workers->num_workers++; - workers->num_workers_starting--; - WARN_ON(workers->num_workers_starting < 0); - spin_unlock_irq(&workers->lock); + INIT_LIST_HEAD(&worker->pending); + INIT_LIST_HEAD(&worker->prio_pending); + INIT_LIST_HEAD(&worker->worker_list); + spin_lock_init(&worker->lock); + + atomic_set(&worker->num_pending, 0); + atomic_set(&worker->refs, 1); + worker->workers = workers; + worker->task = kthread_run(worker_loop, worker, + "btrfs-%s-%d", workers->name, + workers->num_workers + 1); + if (IS_ERR(worker->task)) { + ret = PTR_ERR(worker->task); + kfree(worker); + goto fail; } + spin_lock_irq(&workers->lock); + list_add_tail(&worker->worker_list, &workers->idle_list); + worker->idle = 1; + workers->num_workers++; + workers->num_workers_starting--; + WARN_ON(workers->num_workers_starting < 0); + spin_unlock_irq(&workers->lock); + return 0; fail: - btrfs_stop_workers(workers); + spin_lock_irq(&workers->lock); + workers->num_workers_starting--; + spin_unlock_irq(&workers->lock); return ret; } -int btrfs_start_workers(struct btrfs_workers *workers, int num_workers) +int btrfs_start_workers(struct btrfs_workers *workers) { spin_lock_irq(&workers->lock); - workers->num_workers_starting += num_workers; + workers->num_workers_starting++; spin_unlock_irq(&workers->lock); - return __btrfs_start_workers(workers, num_workers); + return __btrfs_start_workers(workers); } /* @@ -568,9 +561,10 @@ static struct btrfs_worker_thread *find_worker(struct btrfs_workers *workers) struct btrfs_worker_thread *worker; unsigned long flags; struct list_head *fallback; + int ret; -again: spin_lock_irqsave(&workers->lock, flags); +again: worker = next_worker(workers); if (!worker) { @@ -584,7 +578,10 @@ again: workers->num_workers_starting++; spin_unlock_irqrestore(&workers->lock, flags); /* we're below the limit, start another worker */ - __btrfs_start_workers(workers, 1); + ret = __btrfs_start_workers(workers); + spin_lock_irqsave(&workers->lock, flags); + if (ret) + goto fallback; goto again; } } @@ -665,7 +662,7 @@ void btrfs_set_work_high_prio(struct btrfs_work *work) /* * places a struct btrfs_work into the pending queue of one of the kthreads */ -int btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work) +void btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work) { struct btrfs_worker_thread *worker; unsigned long flags; @@ -673,7 +670,7 @@ int btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work) /* don't requeue something already on a list */ if (test_and_set_bit(WORK_QUEUED_BIT, &work->flags)) - goto out; + return; worker = find_worker(workers); if (workers->ordered) { @@ -712,7 +709,4 @@ int btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work) if (wake) wake_up_process(worker->task); spin_unlock_irqrestore(&worker->lock, flags); - -out: - return 0; } diff --git a/fs/btrfs/async-thread.h b/fs/btrfs/async-thread.h index 5077746cf85e..f34cc31fa3c9 100644 --- a/fs/btrfs/async-thread.h +++ b/fs/btrfs/async-thread.h @@ -109,8 +109,8 @@ struct btrfs_workers { char *name; }; -int btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work); -int btrfs_start_workers(struct btrfs_workers *workers, int num_workers); +void btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work); +int btrfs_start_workers(struct btrfs_workers *workers); int btrfs_stop_workers(struct btrfs_workers *workers); void btrfs_init_workers(struct btrfs_workers *workers, char *name, int max, struct btrfs_workers *async_starter); diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index 8855aad3929c..22c64fff1bd5 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -683,7 +683,7 @@ static int inode_to_path(u64 inum, struct btrfs_inode_ref *iref, return PTR_ERR(fspath); if (fspath > fspath_min) { - ipath->fspath->val[i] = (u64)fspath; + ipath->fspath->val[i] = (u64)(unsigned long)fspath; ++ipath->fspath->elem_cnt; ipath->fspath->bytes_left = fspath - fspath_min; } else { diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index 0fe615e4ea38..dede441bdeee 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -514,10 +514,25 @@ static inline int should_cow_block(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct extent_buffer *buf) { + /* ensure we can see the force_cow */ + smp_rmb(); + + /* + * We do not need to cow a block if + * 1) this block is not created or changed in this transaction; + * 2) this block does not belong to TREE_RELOC tree; + * 3) the root is not forced COW. + * + * What is forced COW: + * when we create snapshot during commiting the transaction, + * after we've finished coping src root, we must COW the shared + * block to ensure the metadata consistency. + */ if (btrfs_header_generation(buf) == trans->transid && !btrfs_header_flag(buf, BTRFS_HEADER_FLAG_WRITTEN) && !(root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID && - btrfs_header_flag(buf, BTRFS_HEADER_FLAG_RELOC))) + btrfs_header_flag(buf, BTRFS_HEADER_FLAG_RELOC)) && + !root->force_cow) return 0; return 1; } diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index b9ba59ff9292..67385033323d 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -848,7 +848,8 @@ struct btrfs_free_cluster { enum btrfs_caching_type { BTRFS_CACHE_NO = 0, BTRFS_CACHE_STARTED = 1, - BTRFS_CACHE_FINISHED = 2, + BTRFS_CACHE_FAST = 2, + BTRFS_CACHE_FINISHED = 3, }; enum btrfs_disk_cache_state { @@ -1271,6 +1272,8 @@ struct btrfs_root { * for stat. It may be used for more later */ dev_t anon_dev; + + int force_cow; }; struct btrfs_ioctl_defrag_range_args { @@ -2366,6 +2369,9 @@ int btrfs_block_rsv_check(struct btrfs_root *root, int btrfs_block_rsv_refill(struct btrfs_root *root, struct btrfs_block_rsv *block_rsv, u64 min_reserved); +int btrfs_block_rsv_refill_noflush(struct btrfs_root *root, + struct btrfs_block_rsv *block_rsv, + u64 min_reserved); int btrfs_block_rsv_migrate(struct btrfs_block_rsv *src_rsv, struct btrfs_block_rsv *dst_rsv, u64 num_bytes); @@ -2686,7 +2692,8 @@ int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf); int btrfs_readpage(struct file *file, struct page *page); void btrfs_evict_inode(struct inode *inode); int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc); -void btrfs_dirty_inode(struct inode *inode, int flags); +int btrfs_dirty_inode(struct inode *inode); +int btrfs_update_time(struct file *file); struct inode *btrfs_alloc_inode(struct super_block *sb); void btrfs_destroy_inode(struct inode *inode); int btrfs_drop_inode(struct inode *inode); diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index 5b163572e0ca..9c1eccc2c503 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -640,8 +640,8 @@ static int btrfs_delayed_inode_reserve_metadata( * Now if src_rsv == delalloc_block_rsv we'll let it just steal since * we're accounted for. */ - if (!trans->bytes_reserved && - src_rsv != &root->fs_info->delalloc_block_rsv) { + if (!src_rsv || (!trans->bytes_reserved && + src_rsv != &root->fs_info->delalloc_block_rsv)) { ret = btrfs_block_rsv_add_noflush(root, dst_rsv, num_bytes); /* * Since we're under a transaction reserve_metadata_bytes could diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 62afe5c5694e..d8525662ca7a 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -620,7 +620,7 @@ out: static int btree_io_failed_hook(struct bio *failed_bio, struct page *page, u64 start, u64 end, - u64 mirror_num, struct extent_state *state) + int mirror_num, struct extent_state *state) { struct extent_io_tree *tree; unsigned long len; @@ -872,7 +872,8 @@ static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio, #ifdef CONFIG_MIGRATION static int btree_migratepage(struct address_space *mapping, - struct page *newpage, struct page *page) + struct page *newpage, struct page *page, + enum migrate_mode mode) { /* * we can't safely write a btree page from here, @@ -887,7 +888,7 @@ static int btree_migratepage(struct address_space *mapping, if (page_has_private(page) && !try_to_release_page(page, GFP_KERNEL)) return -EAGAIN; - return migrate_page(mapping, newpage, page); + return migrate_page(mapping, newpage, page, mode); } #endif @@ -1579,9 +1580,7 @@ static int cleaner_kthread(void *arg) btrfs_run_defrag_inodes(root->fs_info); } - if (freezing(current)) { - refrigerator(); - } else { + if (!try_to_freeze()) { set_current_state(TASK_INTERRUPTIBLE); if (!kthread_should_stop()) schedule(); @@ -1635,9 +1634,7 @@ sleep: wake_up_process(root->fs_info->cleaner_kthread); mutex_unlock(&root->fs_info->transaction_kthread_mutex); - if (freezing(current)) { - refrigerator(); - } else { + if (!try_to_freeze()) { set_current_state(TASK_INTERRUPTIBLE); if (!kthread_should_stop() && !btrfs_transaction_blocked(root->fs_info)) @@ -2194,19 +2191,27 @@ struct btrfs_root *open_ctree(struct super_block *sb, fs_info->endio_meta_write_workers.idle_thresh = 2; fs_info->readahead_workers.idle_thresh = 2; - btrfs_start_workers(&fs_info->workers, 1); - btrfs_start_workers(&fs_info->generic_worker, 1); - btrfs_start_workers(&fs_info->submit_workers, 1); - btrfs_start_workers(&fs_info->delalloc_workers, 1); - btrfs_start_workers(&fs_info->fixup_workers, 1); - btrfs_start_workers(&fs_info->endio_workers, 1); - btrfs_start_workers(&fs_info->endio_meta_workers, 1); - btrfs_start_workers(&fs_info->endio_meta_write_workers, 1); - btrfs_start_workers(&fs_info->endio_write_workers, 1); - btrfs_start_workers(&fs_info->endio_freespace_worker, 1); - btrfs_start_workers(&fs_info->delayed_workers, 1); - btrfs_start_workers(&fs_info->caching_workers, 1); - btrfs_start_workers(&fs_info->readahead_workers, 1); + /* + * btrfs_start_workers can really only fail because of ENOMEM so just + * return -ENOMEM if any of these fail. + */ + ret = btrfs_start_workers(&fs_info->workers); + ret |= btrfs_start_workers(&fs_info->generic_worker); + ret |= btrfs_start_workers(&fs_info->submit_workers); + ret |= btrfs_start_workers(&fs_info->delalloc_workers); + ret |= btrfs_start_workers(&fs_info->fixup_workers); + ret |= btrfs_start_workers(&fs_info->endio_workers); + ret |= btrfs_start_workers(&fs_info->endio_meta_workers); + ret |= btrfs_start_workers(&fs_info->endio_meta_write_workers); + ret |= btrfs_start_workers(&fs_info->endio_write_workers); + ret |= btrfs_start_workers(&fs_info->endio_freespace_worker); + ret |= btrfs_start_workers(&fs_info->delayed_workers); + ret |= btrfs_start_workers(&fs_info->caching_workers); + ret |= btrfs_start_workers(&fs_info->readahead_workers); + if (ret) { + ret = -ENOMEM; + goto fail_sb_buffer; + } fs_info->bdi.ra_pages *= btrfs_super_num_devices(disk_super); fs_info->bdi.ra_pages = max(fs_info->bdi.ra_pages, @@ -2573,22 +2578,10 @@ static int write_dev_supers(struct btrfs_device *device, int errors = 0; u32 crc; u64 bytenr; - int last_barrier = 0; if (max_mirrors == 0) max_mirrors = BTRFS_SUPER_MIRROR_MAX; - /* make sure only the last submit_bh does a barrier */ - if (do_barriers) { - for (i = 0; i < max_mirrors; i++) { - bytenr = btrfs_sb_offset(i); - if (bytenr + BTRFS_SUPER_INFO_SIZE >= - device->total_bytes) - break; - last_barrier = i; - } - } - for (i = 0; i < max_mirrors; i++) { bytenr = btrfs_sb_offset(i); if (bytenr + BTRFS_SUPER_INFO_SIZE >= device->total_bytes) @@ -2634,17 +2627,136 @@ static int write_dev_supers(struct btrfs_device *device, bh->b_end_io = btrfs_end_buffer_write_sync; } - if (i == last_barrier && do_barriers) - ret = submit_bh(WRITE_FLUSH_FUA, bh); - else - ret = submit_bh(WRITE_SYNC, bh); - + /* + * we fua the first super. The others we allow + * to go down lazy. + */ + ret = submit_bh(WRITE_FUA, bh); if (ret) errors++; } return errors < i ? 0 : -1; } +/* + * endio for the write_dev_flush, this will wake anyone waiting + * for the barrier when it is done + */ +static void btrfs_end_empty_barrier(struct bio *bio, int err) +{ + if (err) { + if (err == -EOPNOTSUPP) + set_bit(BIO_EOPNOTSUPP, &bio->bi_flags); + clear_bit(BIO_UPTODATE, &bio->bi_flags); + } + if (bio->bi_private) + complete(bio->bi_private); + bio_put(bio); +} + +/* + * trigger flushes for one the devices. If you pass wait == 0, the flushes are + * sent down. With wait == 1, it waits for the previous flush. + * + * any device where the flush fails with eopnotsupp are flagged as not-barrier + * capable + */ +static int write_dev_flush(struct btrfs_device *device, int wait) +{ + struct bio *bio; + int ret = 0; + + if (device->nobarriers) + return 0; + + if (wait) { + bio = device->flush_bio; + if (!bio) + return 0; + + wait_for_completion(&device->flush_wait); + + if (bio_flagged(bio, BIO_EOPNOTSUPP)) { + printk("btrfs: disabling barriers on dev %s\n", + device->name); + device->nobarriers = 1; + } + if (!bio_flagged(bio, BIO_UPTODATE)) { + ret = -EIO; + } + + /* drop the reference from the wait == 0 run */ + bio_put(bio); + device->flush_bio = NULL; + + return ret; + } + + /* + * one reference for us, and we leave it for the + * caller + */ + device->flush_bio = NULL;; + bio = bio_alloc(GFP_NOFS, 0); + if (!bio) + return -ENOMEM; + + bio->bi_end_io = btrfs_end_empty_barrier; + bio->bi_bdev = device->bdev; + init_completion(&device->flush_wait); + bio->bi_private = &device->flush_wait; + device->flush_bio = bio; + + bio_get(bio); + submit_bio(WRITE_FLUSH, bio); + + return 0; +} + +/* + * send an empty flush down to each device in parallel, + * then wait for them + */ +static int barrier_all_devices(struct btrfs_fs_info *info) +{ + struct list_head *head; + struct btrfs_device *dev; + int errors = 0; + int ret; + + /* send down all the barriers */ + head = &info->fs_devices->devices; + list_for_each_entry_rcu(dev, head, dev_list) { + if (!dev->bdev) { + errors++; + continue; + } + if (!dev->in_fs_metadata || !dev->writeable) + continue; + + ret = write_dev_flush(dev, 0); + if (ret) + errors++; + } + + /* wait for all the barriers */ + list_for_each_entry_rcu(dev, head, dev_list) { + if (!dev->bdev) { + errors++; + continue; + } + if (!dev->in_fs_metadata || !dev->writeable) + continue; + + ret = write_dev_flush(dev, 1); + if (ret) + errors++; + } + if (errors) + return -EIO; + return 0; +} + int write_all_supers(struct btrfs_root *root, int max_mirrors) { struct list_head *head; @@ -2666,6 +2778,10 @@ int write_all_supers(struct btrfs_root *root, int max_mirrors) mutex_lock(&root->fs_info->fs_devices->device_list_mutex); head = &root->fs_info->fs_devices->devices; + + if (do_barriers) + barrier_all_devices(root->fs_info); + list_for_each_entry_rcu(dev, head, dev_list) { if (!dev->bdev) { total_errors++; diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index b232150b5b6b..f5fbe576d2ba 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -467,13 +467,59 @@ static int cache_block_group(struct btrfs_block_group_cache *cache, struct btrfs_root *root, int load_cache_only) { + DEFINE_WAIT(wait); struct btrfs_fs_info *fs_info = cache->fs_info; struct btrfs_caching_control *caching_ctl; int ret = 0; - smp_mb(); - if (cache->cached != BTRFS_CACHE_NO) + caching_ctl = kzalloc(sizeof(*caching_ctl), GFP_NOFS); + BUG_ON(!caching_ctl); + + INIT_LIST_HEAD(&caching_ctl->list); + mutex_init(&caching_ctl->mutex); + init_waitqueue_head(&caching_ctl->wait); + caching_ctl->block_group = cache; + caching_ctl->progress = cache->key.objectid; + atomic_set(&caching_ctl->count, 1); + caching_ctl->work.func = caching_thread; + + spin_lock(&cache->lock); + /* + * This should be a rare occasion, but this could happen I think in the + * case where one thread starts to load the space cache info, and then + * some other thread starts a transaction commit which tries to do an + * allocation while the other thread is still loading the space cache + * info. The previous loop should have kept us from choosing this block + * group, but if we've moved to the state where we will wait on caching + * block groups we need to first check if we're doing a fast load here, + * so we can wait for it to finish, otherwise we could end up allocating + * from a block group who's cache gets evicted for one reason or + * another. + */ + while (cache->cached == BTRFS_CACHE_FAST) { + struct btrfs_caching_control *ctl; + + ctl = cache->caching_ctl; + atomic_inc(&ctl->count); + prepare_to_wait(&ctl->wait, &wait, TASK_UNINTERRUPTIBLE); + spin_unlock(&cache->lock); + + schedule(); + + finish_wait(&ctl->wait, &wait); + put_caching_control(ctl); + spin_lock(&cache->lock); + } + + if (cache->cached != BTRFS_CACHE_NO) { + spin_unlock(&cache->lock); + kfree(caching_ctl); return 0; + } + WARN_ON(cache->caching_ctl); + cache->caching_ctl = caching_ctl; + cache->cached = BTRFS_CACHE_FAST; + spin_unlock(&cache->lock); /* * We can't do the read from on-disk cache during a commit since we need @@ -484,56 +530,51 @@ static int cache_block_group(struct btrfs_block_group_cache *cache, if (trans && (!trans->transaction->in_commit) && (root && root != root->fs_info->tree_root) && btrfs_test_opt(root, SPACE_CACHE)) { - spin_lock(&cache->lock); - if (cache->cached != BTRFS_CACHE_NO) { - spin_unlock(&cache->lock); - return 0; - } - cache->cached = BTRFS_CACHE_STARTED; - spin_unlock(&cache->lock); - ret = load_free_space_cache(fs_info, cache); spin_lock(&cache->lock); if (ret == 1) { + cache->caching_ctl = NULL; cache->cached = BTRFS_CACHE_FINISHED; cache->last_byte_to_unpin = (u64)-1; } else { - cache->cached = BTRFS_CACHE_NO; + if (load_cache_only) { + cache->caching_ctl = NULL; + cache->cached = BTRFS_CACHE_NO; + } else { + cache->cached = BTRFS_CACHE_STARTED; + } } spin_unlock(&cache->lock); + wake_up(&caching_ctl->wait); if (ret == 1) { + put_caching_control(caching_ctl); free_excluded_extents(fs_info->extent_root, cache); return 0; } + } else { + /* + * We are not going to do the fast caching, set cached to the + * appropriate value and wakeup any waiters. + */ + spin_lock(&cache->lock); + if (load_cache_only) { + cache->caching_ctl = NULL; + cache->cached = BTRFS_CACHE_NO; + } else { + cache->cached = BTRFS_CACHE_STARTED; + } + spin_unlock(&cache->lock); + wake_up(&caching_ctl->wait); } - if (load_cache_only) - return 0; - - caching_ctl = kzalloc(sizeof(*caching_ctl), GFP_NOFS); - BUG_ON(!caching_ctl); - - INIT_LIST_HEAD(&caching_ctl->list); - mutex_init(&caching_ctl->mutex); - init_waitqueue_head(&caching_ctl->wait); - caching_ctl->block_group = cache; - caching_ctl->progress = cache->key.objectid; - /* one for caching kthread, one for caching block group list */ - atomic_set(&caching_ctl->count, 2); - caching_ctl->work.func = caching_thread; - - spin_lock(&cache->lock); - if (cache->cached != BTRFS_CACHE_NO) { - spin_unlock(&cache->lock); - kfree(caching_ctl); + if (load_cache_only) { + put_caching_control(caching_ctl); return 0; } - cache->caching_ctl = caching_ctl; - cache->cached = BTRFS_CACHE_STARTED; - spin_unlock(&cache->lock); down_write(&fs_info->extent_commit_sem); + atomic_inc(&caching_ctl->count); list_add_tail(&caching_ctl->list, &fs_info->caching_block_groups); up_write(&fs_info->extent_commit_sem); @@ -2781,7 +2822,7 @@ out_free: btrfs_release_path(path); out: spin_lock(&block_group->lock); - if (!ret) + if (!ret && dcs == BTRFS_DC_SETUP) block_group->cache_generation = trans->transid; block_group->disk_cache_state = dcs; spin_unlock(&block_group->lock); @@ -3847,9 +3888,9 @@ int btrfs_block_rsv_check(struct btrfs_root *root, return ret; } -int btrfs_block_rsv_refill(struct btrfs_root *root, - struct btrfs_block_rsv *block_rsv, - u64 min_reserved) +static inline int __btrfs_block_rsv_refill(struct btrfs_root *root, + struct btrfs_block_rsv *block_rsv, + u64 min_reserved, int flush) { u64 num_bytes = 0; int ret = -ENOSPC; @@ -3868,7 +3909,7 @@ int btrfs_block_rsv_refill(struct btrfs_root *root, if (!ret) return 0; - ret = reserve_metadata_bytes(root, block_rsv, num_bytes, 1); + ret = reserve_metadata_bytes(root, block_rsv, num_bytes, flush); if (!ret) { block_rsv_add_bytes(block_rsv, num_bytes, 0); return 0; @@ -3877,6 +3918,20 @@ int btrfs_block_rsv_refill(struct btrfs_root *root, return ret; } +int btrfs_block_rsv_refill(struct btrfs_root *root, + struct btrfs_block_rsv *block_rsv, + u64 min_reserved) +{ + return __btrfs_block_rsv_refill(root, block_rsv, min_reserved, 1); +} + +int btrfs_block_rsv_refill_noflush(struct btrfs_root *root, + struct btrfs_block_rsv *block_rsv, + u64 min_reserved) +{ + return __btrfs_block_rsv_refill(root, block_rsv, min_reserved, 0); +} + int btrfs_block_rsv_migrate(struct btrfs_block_rsv *src_rsv, struct btrfs_block_rsv *dst_rsv, u64 num_bytes) @@ -4149,12 +4204,17 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes) struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_block_rsv *block_rsv = &root->fs_info->delalloc_block_rsv; u64 to_reserve = 0; + u64 csum_bytes; unsigned nr_extents = 0; + int extra_reserve = 0; int flush = 1; int ret; + /* Need to be holding the i_mutex here if we aren't free space cache */ if (btrfs_is_free_space_inode(root, inode)) flush = 0; + else + WARN_ON(!mutex_is_locked(&inode->i_mutex)); if (flush && btrfs_transaction_in_commit(root->fs_info)) schedule_timeout(1); @@ -4165,11 +4225,9 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes) BTRFS_I(inode)->outstanding_extents++; if (BTRFS_I(inode)->outstanding_extents > - BTRFS_I(inode)->reserved_extents) { + BTRFS_I(inode)->reserved_extents) nr_extents = BTRFS_I(inode)->outstanding_extents - BTRFS_I(inode)->reserved_extents; - BTRFS_I(inode)->reserved_extents += nr_extents; - } /* * Add an item to reserve for updating the inode when we complete the @@ -4177,11 +4235,12 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes) */ if (!BTRFS_I(inode)->delalloc_meta_reserved) { nr_extents++; - BTRFS_I(inode)->delalloc_meta_reserved = 1; + extra_reserve = 1; } to_reserve = btrfs_calc_trans_metadata_size(root, nr_extents); to_reserve += calc_csum_metadata_size(inode, num_bytes, 1); + csum_bytes = BTRFS_I(inode)->csum_bytes; spin_unlock(&BTRFS_I(inode)->lock); ret = reserve_metadata_bytes(root, block_rsv, to_reserve, flush); @@ -4191,22 +4250,35 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes) spin_lock(&BTRFS_I(inode)->lock); dropped = drop_outstanding_extent(inode); - to_free = calc_csum_metadata_size(inode, num_bytes, 0); - spin_unlock(&BTRFS_I(inode)->lock); - to_free += btrfs_calc_trans_metadata_size(root, dropped); - /* - * Somebody could have come in and twiddled with the - * reservation, so if we have to free more than we would have - * reserved from this reservation go ahead and release those - * bytes. + * If the inodes csum_bytes is the same as the original + * csum_bytes then we know we haven't raced with any free()ers + * so we can just reduce our inodes csum bytes and carry on. + * Otherwise we have to do the normal free thing to account for + * the case that the free side didn't free up its reserve + * because of this outstanding reservation. */ - to_free -= to_reserve; + if (BTRFS_I(inode)->csum_bytes == csum_bytes) + calc_csum_metadata_size(inode, num_bytes, 0); + else + to_free = calc_csum_metadata_size(inode, num_bytes, 0); + spin_unlock(&BTRFS_I(inode)->lock); + if (dropped) + to_free += btrfs_calc_trans_metadata_size(root, dropped); + if (to_free) btrfs_block_rsv_release(root, block_rsv, to_free); return ret; } + spin_lock(&BTRFS_I(inode)->lock); + if (extra_reserve) { + BTRFS_I(inode)->delalloc_meta_reserved = 1; + nr_extents--; + } + BTRFS_I(inode)->reserved_extents += nr_extents; + spin_unlock(&BTRFS_I(inode)->lock); + block_rsv_add_bytes(block_rsv, to_reserve, 1); return 0; @@ -5052,11 +5124,11 @@ static noinline int find_free_extent(struct btrfs_trans_handle *trans, struct btrfs_root *root = orig_root->fs_info->extent_root; struct btrfs_free_cluster *last_ptr = NULL; struct btrfs_block_group_cache *block_group = NULL; + struct btrfs_block_group_cache *used_block_group; int empty_cluster = 2 * 1024 * 1024; int allowed_chunk_alloc = 0; int done_chunk_alloc = 0; struct btrfs_space_info *space_info; - int last_ptr_loop = 0; int loop = 0; int index = 0; int alloc_type = (data & BTRFS_BLOCK_GROUP_DATA) ? @@ -5118,6 +5190,7 @@ static noinline int find_free_extent(struct btrfs_trans_handle *trans, ideal_cache: block_group = btrfs_lookup_block_group(root->fs_info, search_start); + used_block_group = block_group; /* * we don't want to use the block group if it doesn't match our * allocation bits, or if its not cached. @@ -5155,6 +5228,7 @@ search: u64 offset; int cached; + used_block_group = block_group; btrfs_get_block_group(block_group); search_start = block_group->key.objectid; @@ -5178,13 +5252,15 @@ search: } have_block_group: - if (unlikely(block_group->cached == BTRFS_CACHE_NO)) { + cached = block_group_cache_done(block_group); + if (unlikely(!cached)) { u64 free_percent; + found_uncached_bg = true; ret = cache_block_group(block_group, trans, orig_root, 1); if (block_group->cached == BTRFS_CACHE_FINISHED) - goto have_block_group; + goto alloc; free_percent = btrfs_block_group_used(&block_group->item); free_percent *= 100; @@ -5206,7 +5282,6 @@ have_block_group: orig_root, 0); BUG_ON(ret); } - found_uncached_bg = true; /* * If loop is set for cached only, try the next block @@ -5216,94 +5291,80 @@ have_block_group: goto loop; } - cached = block_group_cache_done(block_group); - if (unlikely(!cached)) - found_uncached_bg = true; - +alloc: if (unlikely(block_group->ro)) goto loop; spin_lock(&block_group->free_space_ctl->tree_lock); if (cached && block_group->free_space_ctl->free_space < - num_bytes + empty_size) { + num_bytes + empty_cluster + empty_size) { spin_unlock(&block_group->free_space_ctl->tree_lock); goto loop; } spin_unlock(&block_group->free_space_ctl->tree_lock); /* - * Ok we want to try and use the cluster allocator, so lets look - * there, unless we are on LOOP_NO_EMPTY_SIZE, since we will - * have tried the cluster allocator plenty of times at this - * point and not have found anything, so we are likely way too - * fragmented for the clustering stuff to find anything, so lets - * just skip it and let the allocator find whatever block it can - * find + * Ok we want to try and use the cluster allocator, so + * lets look there */ - if (last_ptr && loop < LOOP_NO_EMPTY_SIZE) { + if (last_ptr) { /* * the refill lock keeps out other * people trying to start a new cluster */ spin_lock(&last_ptr->refill_lock); - if (last_ptr->block_group && - (last_ptr->block_group->ro || - !block_group_bits(last_ptr->block_group, data))) { - offset = 0; + used_block_group = last_ptr->block_group; + if (used_block_group != block_group && + (!used_block_group || + used_block_group->ro || + !block_group_bits(used_block_group, data))) { + used_block_group = block_group; goto refill_cluster; } - offset = btrfs_alloc_from_cluster(block_group, last_ptr, - num_bytes, search_start); + if (used_block_group != block_group) + btrfs_get_block_group(used_block_group); + + offset = btrfs_alloc_from_cluster(used_block_group, + last_ptr, num_bytes, used_block_group->key.objectid); if (offset) { /* we have a block, we're done */ spin_unlock(&last_ptr->refill_lock); goto checks; } - spin_lock(&last_ptr->lock); - /* - * whoops, this cluster doesn't actually point to - * this block group. Get a ref on the block - * group is does point to and try again - */ - if (!last_ptr_loop && last_ptr->block_group && - last_ptr->block_group != block_group && - index <= - get_block_group_index(last_ptr->block_group)) { - - btrfs_put_block_group(block_group); - block_group = last_ptr->block_group; - btrfs_get_block_group(block_group); - spin_unlock(&last_ptr->lock); - spin_unlock(&last_ptr->refill_lock); - - last_ptr_loop = 1; - search_start = block_group->key.objectid; - /* - * we know this block group is properly - * in the list because - * btrfs_remove_block_group, drops the - * cluster before it removes the block - * group from the list - */ - goto have_block_group; + WARN_ON(last_ptr->block_group != used_block_group); + if (used_block_group != block_group) { + btrfs_put_block_group(used_block_group); + used_block_group = block_group; } - spin_unlock(&last_ptr->lock); refill_cluster: + BUG_ON(used_block_group != block_group); + /* If we are on LOOP_NO_EMPTY_SIZE, we can't + * set up a new clusters, so lets just skip it + * and let the allocator find whatever block + * it can find. If we reach this point, we + * will have tried the cluster allocator + * plenty of times and not have found + * anything, so we are likely way too + * fragmented for the clustering stuff to find + * anything. */ + if (loop >= LOOP_NO_EMPTY_SIZE) { + spin_unlock(&last_ptr->refill_lock); + goto unclustered_alloc; + } + /* * this cluster didn't work out, free it and * start over */ btrfs_return_cluster_to_free_space(NULL, last_ptr); - last_ptr_loop = 0; - /* allocate a cluster in this block group */ ret = btrfs_find_space_cluster(trans, root, block_group, last_ptr, - offset, num_bytes, + search_start, num_bytes, empty_cluster + empty_size); if (ret == 0) { /* @@ -5339,6 +5400,7 @@ refill_cluster: goto loop; } +unclustered_alloc: offset = btrfs_find_space_for_alloc(block_group, search_start, num_bytes, empty_size); /* @@ -5365,14 +5427,14 @@ checks: search_start = stripe_align(root, offset); /* move on to the next group */ if (search_start + num_bytes >= search_end) { - btrfs_add_free_space(block_group, offset, num_bytes); + btrfs_add_free_space(used_block_group, offset, num_bytes); goto loop; } /* move on to the next group */ if (search_start + num_bytes > - block_group->key.objectid + block_group->key.offset) { - btrfs_add_free_space(block_group, offset, num_bytes); + used_block_group->key.objectid + used_block_group->key.offset) { + btrfs_add_free_space(used_block_group, offset, num_bytes); goto loop; } @@ -5380,14 +5442,14 @@ checks: ins->offset = num_bytes; if (offset < search_start) - btrfs_add_free_space(block_group, offset, + btrfs_add_free_space(used_block_group, offset, search_start - offset); BUG_ON(offset > search_start); - ret = btrfs_update_reserved_bytes(block_group, num_bytes, + ret = btrfs_update_reserved_bytes(used_block_group, num_bytes, alloc_type); if (ret == -EAGAIN) { - btrfs_add_free_space(block_group, offset, num_bytes); + btrfs_add_free_space(used_block_group, offset, num_bytes); goto loop; } @@ -5396,15 +5458,19 @@ checks: ins->offset = num_bytes; if (offset < search_start) - btrfs_add_free_space(block_group, offset, + btrfs_add_free_space(used_block_group, offset, search_start - offset); BUG_ON(offset > search_start); + if (used_block_group != block_group) + btrfs_put_block_group(used_block_group); btrfs_put_block_group(block_group); break; loop: failed_cluster_refill = false; failed_alloc = false; BUG_ON(index != get_block_group_index(block_group)); + if (used_block_group != block_group) + btrfs_put_block_group(used_block_group); btrfs_put_block_group(block_group); } up_read(&space_info->groups_sem); diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 1f87c4d0e7a0..49f3c9dc09f4 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -935,8 +935,10 @@ again: node = tree_search(tree, start); if (!node) { prealloc = alloc_extent_state_atomic(prealloc); - if (!prealloc) - return -ENOMEM; + if (!prealloc) { + err = -ENOMEM; + goto out; + } err = insert_state(tree, prealloc, start, end, &bits); prealloc = NULL; BUG_ON(err == -EEXIST); @@ -992,8 +994,10 @@ hit_next: */ if (state->start < start) { prealloc = alloc_extent_state_atomic(prealloc); - if (!prealloc) - return -ENOMEM; + if (!prealloc) { + err = -ENOMEM; + goto out; + } err = split_state(tree, state, prealloc, start); BUG_ON(err == -EEXIST); prealloc = NULL; @@ -1024,8 +1028,10 @@ hit_next: this_end = last_start - 1; prealloc = alloc_extent_state_atomic(prealloc); - if (!prealloc) - return -ENOMEM; + if (!prealloc) { + err = -ENOMEM; + goto out; + } /* * Avoid to free 'prealloc' if it can be merged with @@ -1051,8 +1057,10 @@ hit_next: */ if (state->start <= end && state->end > end) { prealloc = alloc_extent_state_atomic(prealloc); - if (!prealloc) - return -ENOMEM; + if (!prealloc) { + err = -ENOMEM; + goto out; + } err = split_state(tree, state, prealloc, end + 1); BUG_ON(err == -EEXIST); @@ -2285,16 +2293,22 @@ static void end_bio_extent_readpage(struct bio *bio, int err) clean_io_failure(start, page); } if (!uptodate) { - u64 failed_mirror; - failed_mirror = (u64)bio->bi_bdev; - if (tree->ops && tree->ops->readpage_io_failed_hook) - ret = tree->ops->readpage_io_failed_hook( - bio, page, start, end, - failed_mirror, state); - else - ret = bio_readpage_error(bio, page, start, end, - failed_mirror, NULL); + int failed_mirror; + failed_mirror = (int)(unsigned long)bio->bi_bdev; + /* + * The generic bio_readpage_error handles errors the + * following way: If possible, new read requests are + * created and submitted and will end up in + * end_bio_extent_readpage as well (if we're lucky, not + * in the !uptodate case). In that case it returns 0 and + * we just go on with the next page in our bio. If it + * can't handle the error it will return -EIO and we + * remain responsible for that page. + */ + ret = bio_readpage_error(bio, page, start, end, + failed_mirror, NULL); if (ret == 0) { +error_handled: uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags); if (err) @@ -2302,6 +2316,13 @@ static void end_bio_extent_readpage(struct bio *bio, int err) uncache_state(&cached); continue; } + if (tree->ops && tree->ops->readpage_io_failed_hook) { + ret = tree->ops->readpage_io_failed_hook( + bio, page, start, end, + failed_mirror, state); + if (ret == 0) + goto error_handled; + } } if (uptodate) { @@ -3366,6 +3387,9 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, return -ENOMEM; path->leave_spinning = 1; + start = ALIGN(start, BTRFS_I(inode)->root->sectorsize); + len = ALIGN(len, BTRFS_I(inode)->root->sectorsize); + /* * lookup the last file extent. We're not using i_size here * because there might be preallocation past i_size @@ -3413,7 +3437,7 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, lock_extent_bits(&BTRFS_I(inode)->io_tree, start, start + len, 0, &cached_state, GFP_NOFS); - em = get_extent_skip_holes(inode, off, last_for_get_extent, + em = get_extent_skip_holes(inode, start, last_for_get_extent, get_extent); if (!em) goto out; diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index feb9be0e23bc..7604c3001322 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -70,7 +70,7 @@ struct extent_io_ops { unsigned long bio_flags); int (*readpage_io_hook)(struct page *page, u64 start, u64 end); int (*readpage_io_failed_hook)(struct bio *bio, struct page *page, - u64 start, u64 end, u64 failed_mirror, + u64 start, u64 end, int failed_mirror, struct extent_state *state); int (*writepage_io_failed_hook)(struct bio *bio, struct page *page, u64 start, u64 end, diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index dafdfa059bf6..034d98503229 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1081,7 +1081,7 @@ static noinline int prepare_pages(struct btrfs_root *root, struct file *file, again: for (i = 0; i < num_pages; i++) { pages[i] = find_or_create_page(inode->i_mapping, index + i, - mask); + mask | __GFP_WRITE); if (!pages[i]) { faili = i - 1; err = -ENOMEM; @@ -1136,7 +1136,8 @@ again: GFP_NOFS); } for (i = 0; i < num_pages; i++) { - clear_page_dirty_for_io(pages[i]); + if (clear_page_dirty_for_io(pages[i])) + account_page_redirty(pages[i]); set_page_extent_mapped(pages[i]); WARN_ON(!PageLocked(pages[i])); } @@ -1167,6 +1168,8 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file, nrptrs = min((iov_iter_count(i) + PAGE_CACHE_SIZE - 1) / PAGE_CACHE_SIZE, PAGE_CACHE_SIZE / (sizeof(struct page *))); + nrptrs = min(nrptrs, current->nr_dirtied_pause - current->nr_dirtied); + nrptrs = max(nrptrs, 8); pages = kmalloc(nrptrs * sizeof(struct page *), GFP_KERNEL); if (!pages) return -ENOMEM; @@ -1387,7 +1390,11 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb, goto out; } - file_update_time(file); + err = btrfs_update_time(file); + if (err) { + mutex_unlock(&inode->i_mutex); + goto out; + } BTRFS_I(inode)->sequence++; start_pos = round_down(pos, root->sectorsize); diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index 181760f9d2ab..9a897bf79538 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -351,6 +351,11 @@ static int io_ctl_prepare_pages(struct io_ctl *io_ctl, struct inode *inode, } } + for (i = 0; i < io_ctl->num_pages; i++) { + clear_page_dirty_for_io(io_ctl->pages[i]); + set_page_extent_mapped(io_ctl->pages[i]); + } + return 0; } @@ -418,7 +423,7 @@ static void io_ctl_set_crc(struct io_ctl *io_ctl, int index) } if (index == 0) - offset = sizeof(u32) * io_ctl->num_pages;; + offset = sizeof(u32) * io_ctl->num_pages; crc = btrfs_csum_data(io_ctl->root, io_ctl->orig + offset, crc, PAGE_CACHE_SIZE - offset); @@ -1465,6 +1470,7 @@ static void add_new_bitmap(struct btrfs_free_space_ctl *ctl, { info->offset = offset_to_bitmap(ctl, offset); info->bytes = 0; + INIT_LIST_HEAD(&info->list); link_free_space(ctl, info); ctl->total_bitmaps++; @@ -1844,7 +1850,13 @@ again: info = tree_search_offset(ctl, offset_to_bitmap(ctl, offset), 1, 0); if (!info) { - WARN_ON(1); + /* the tree logging code might be calling us before we + * have fully loaded the free space rbtree for this + * block group. So it is possible the entry won't + * be in the rbtree yet at all. The caching code + * will make sure not to put it in the rbtree if + * the logging code has pinned it. + */ goto out_lock; } } @@ -2308,6 +2320,7 @@ again: if (!found) { start = i; + cluster->max_size = 0; found = true; } @@ -2451,16 +2464,23 @@ setup_cluster_bitmap(struct btrfs_block_group_cache *block_group, { struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; struct btrfs_free_space *entry; - struct rb_node *node; int ret = -ENOSPC; + u64 bitmap_offset = offset_to_bitmap(ctl, offset); if (ctl->total_bitmaps == 0) return -ENOSPC; /* - * First check our cached list of bitmaps and see if there is an entry - * here that will work. + * The bitmap that covers offset won't be in the list unless offset + * is just its start offset. */ + entry = list_first_entry(bitmaps, struct btrfs_free_space, list); + if (entry->offset != bitmap_offset) { + entry = tree_search_offset(ctl, bitmap_offset, 1, 0); + if (entry && list_empty(&entry->list)) + list_add(&entry->list, bitmaps); + } + list_for_each_entry(entry, bitmaps, list) { if (entry->bytes < min_bytes) continue; @@ -2471,38 +2491,10 @@ setup_cluster_bitmap(struct btrfs_block_group_cache *block_group, } /* - * If we do have entries on our list and we are here then we didn't find - * anything, so go ahead and get the next entry after the last entry in - * this list and start the search from there. + * The bitmaps list has all the bitmaps that record free space + * starting after offset, so no more search is required. */ - if (!list_empty(bitmaps)) { - entry = list_entry(bitmaps->prev, struct btrfs_free_space, - list); - node = rb_next(&entry->offset_index); - if (!node) - return -ENOSPC; - entry = rb_entry(node, struct btrfs_free_space, offset_index); - goto search; - } - - entry = tree_search_offset(ctl, offset_to_bitmap(ctl, offset), 0, 1); - if (!entry) - return -ENOSPC; - -search: - node = &entry->offset_index; - do { - entry = rb_entry(node, struct btrfs_free_space, offset_index); - node = rb_next(&entry->offset_index); - if (!entry->bitmap) - continue; - if (entry->bytes < min_bytes) - continue; - ret = btrfs_bitmap_cluster(block_group, entry, cluster, offset, - bytes, min_bytes); - } while (ret && node); - - return ret; + return -ENOSPC; } /* @@ -2520,8 +2512,8 @@ int btrfs_find_space_cluster(struct btrfs_trans_handle *trans, u64 offset, u64 bytes, u64 empty_size) { struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; - struct list_head bitmaps; struct btrfs_free_space *entry, *tmp; + LIST_HEAD(bitmaps); u64 min_bytes; int ret; @@ -2560,7 +2552,6 @@ int btrfs_find_space_cluster(struct btrfs_trans_handle *trans, goto out; } - INIT_LIST_HEAD(&bitmaps); ret = setup_cluster_no_bitmap(block_group, cluster, &bitmaps, offset, bytes, min_bytes); if (ret) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 116ab67a06df..81b235a61f8c 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -38,6 +38,7 @@ #include <linux/falloc.h> #include <linux/slab.h> #include <linux/ratelimit.h> +#include <linux/mount.h> #include "compat.h" #include "ctree.h" #include "disk-io.h" @@ -1943,7 +1944,7 @@ enum btrfs_orphan_cleanup_state { }; /* - * This is called in transaction commmit time. If there are no orphan + * This is called in transaction commit time. If there are no orphan * files in the subvolume, it removes orphan item and frees block_rsv * structure. */ @@ -2031,7 +2032,7 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode) /* insert an orphan item to track this unlinked/truncated file */ if (insert >= 1) { ret = btrfs_insert_orphan_item(trans, root, btrfs_ino(inode)); - BUG_ON(ret); + BUG_ON(ret && ret != -EEXIST); } /* insert an orphan item to track subvolume contains orphan files */ @@ -2158,6 +2159,38 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) if (ret && ret != -ESTALE) goto out; + if (ret == -ESTALE && root == root->fs_info->tree_root) { + struct btrfs_root *dead_root; + struct btrfs_fs_info *fs_info = root->fs_info; + int is_dead_root = 0; + + /* + * this is an orphan in the tree root. Currently these + * could come from 2 sources: + * a) a snapshot deletion in progress + * b) a free space cache inode + * We need to distinguish those two, as the snapshot + * orphan must not get deleted. + * find_dead_roots already ran before us, so if this + * is a snapshot deletion, we should find the root + * in the dead_roots list + */ + spin_lock(&fs_info->trans_lock); + list_for_each_entry(dead_root, &fs_info->dead_roots, + root_list) { + if (dead_root->root_key.objectid == + found_key.objectid) { + is_dead_root = 1; + break; + } + } + spin_unlock(&fs_info->trans_lock); + if (is_dead_root) { + /* prevent this orphan from being found again */ + key.offset = found_key.objectid - 1; + continue; + } + } /* * Inode is already gone but the orphan item is still there, * kill the orphan item. @@ -2191,7 +2224,14 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) continue; } nr_truncate++; + /* + * Need to hold the imutex for reservation purposes, not + * a huge deal here but I have a WARN_ON in + * btrfs_delalloc_reserve_space to catch offenders. + */ + mutex_lock(&inode->i_mutex); ret = btrfs_truncate(inode); + mutex_unlock(&inode->i_mutex); } else { nr_unlink++; } @@ -3327,7 +3367,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) u64 hint_byte = 0; hole_size = last_byte - cur_offset; - trans = btrfs_start_transaction(root, 2); + trans = btrfs_start_transaction(root, 3); if (IS_ERR(trans)) { err = PTR_ERR(trans); break; @@ -3337,6 +3377,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) cur_offset + hole_size, &hint_byte, 1); if (err) { + btrfs_update_inode(trans, root, inode); btrfs_end_transaction(trans, root); break; } @@ -3346,6 +3387,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) 0, hole_size, 0, hole_size, 0, 0, 0); if (err) { + btrfs_update_inode(trans, root, inode); btrfs_end_transaction(trans, root); break; } @@ -3353,6 +3395,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) btrfs_drop_extent_cache(inode, hole_start, last_byte - 1, 0); + btrfs_update_inode(trans, root, inode); btrfs_end_transaction(trans, root); } free_extent_map(em); @@ -3370,6 +3413,8 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) static int btrfs_setsize(struct inode *inode, loff_t newsize) { + struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_trans_handle *trans; loff_t oldsize = i_size_read(inode); int ret; @@ -3377,16 +3422,19 @@ static int btrfs_setsize(struct inode *inode, loff_t newsize) return 0; if (newsize > oldsize) { - i_size_write(inode, newsize); - btrfs_ordered_update_i_size(inode, i_size_read(inode), NULL); truncate_pagecache(inode, oldsize, newsize); ret = btrfs_cont_expand(inode, oldsize, newsize); - if (ret) { - btrfs_setsize(inode, oldsize); + if (ret) return ret; - } - mark_inode_dirty(inode); + trans = btrfs_start_transaction(root, 1); + if (IS_ERR(trans)) + return PTR_ERR(trans); + + i_size_write(inode, newsize); + btrfs_ordered_update_i_size(inode, i_size_read(inode), NULL); + ret = btrfs_update_inode(trans, root, inode); + btrfs_end_transaction_throttle(trans, root); } else { /* @@ -3426,9 +3474,9 @@ static int btrfs_setattr(struct dentry *dentry, struct iattr *attr) if (attr->ia_valid) { setattr_copy(inode, attr); - mark_inode_dirty(inode); + err = btrfs_dirty_inode(inode); - if (attr->ia_valid & ATTR_MODE) + if (!err && attr->ia_valid & ATTR_MODE) err = btrfs_acl_chmod(inode); } @@ -3490,7 +3538,7 @@ void btrfs_evict_inode(struct inode *inode) * doing the truncate. */ while (1) { - ret = btrfs_block_rsv_refill(root, rsv, min_size); + ret = btrfs_block_rsv_refill_noflush(root, rsv, min_size); /* * Try and steal from the global reserve since we will @@ -4204,42 +4252,80 @@ int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc) * FIXME, needs more benchmarking...there are no reasons other than performance * to keep or drop this code. */ -void btrfs_dirty_inode(struct inode *inode, int flags) +int btrfs_dirty_inode(struct inode *inode) { struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_trans_handle *trans; int ret; if (BTRFS_I(inode)->dummy_inode) - return; + return 0; trans = btrfs_join_transaction(root); - BUG_ON(IS_ERR(trans)); + if (IS_ERR(trans)) + return PTR_ERR(trans); ret = btrfs_update_inode(trans, root, inode); if (ret && ret == -ENOSPC) { /* whoops, lets try again with the full transaction */ btrfs_end_transaction(trans, root); trans = btrfs_start_transaction(root, 1); - if (IS_ERR(trans)) { - printk_ratelimited(KERN_ERR "btrfs: fail to " - "dirty inode %llu error %ld\n", - (unsigned long long)btrfs_ino(inode), - PTR_ERR(trans)); - return; - } + if (IS_ERR(trans)) + return PTR_ERR(trans); ret = btrfs_update_inode(trans, root, inode); - if (ret) { - printk_ratelimited(KERN_ERR "btrfs: fail to " - "dirty inode %llu error %d\n", - (unsigned long long)btrfs_ino(inode), - ret); - } } btrfs_end_transaction(trans, root); if (BTRFS_I(inode)->delayed_node) btrfs_balance_delayed_items(root); + + return ret; +} + +/* + * This is a copy of file_update_time. We need this so we can return error on + * ENOSPC for updating the inode in the case of file write and mmap writes. + */ +int btrfs_update_time(struct file *file) +{ + struct inode *inode = file->f_path.dentry->d_inode; + struct timespec now; + int ret; + enum { S_MTIME = 1, S_CTIME = 2, S_VERSION = 4 } sync_it = 0; + + /* First try to exhaust all avenues to not sync */ + if (IS_NOCMTIME(inode)) + return 0; + + now = current_fs_time(inode->i_sb); + if (!timespec_equal(&inode->i_mtime, &now)) + sync_it = S_MTIME; + + if (!timespec_equal(&inode->i_ctime, &now)) + sync_it |= S_CTIME; + + if (IS_I_VERSION(inode)) + sync_it |= S_VERSION; + + if (!sync_it) + return 0; + + /* Finally allowed to write? Takes lock. */ + if (mnt_want_write_file(file)) + return 0; + + /* Only change inode inside the lock region */ + if (sync_it & S_VERSION) + inode_inc_iversion(inode); + if (sync_it & S_CTIME) + inode->i_ctime = now; + if (sync_it & S_MTIME) + inode->i_mtime = now; + ret = btrfs_dirty_inode(inode); + if (!ret) + mark_inode_dirty_sync(inode); + mnt_drop_write(file->f_path.mnt); + return ret; } /* @@ -4326,8 +4412,8 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct inode *dir, const char *name, int name_len, - u64 ref_objectid, u64 objectid, int mode, - u64 *index) + u64 ref_objectid, u64 objectid, + umode_t mode, u64 *index) { struct inode *inode; struct btrfs_inode_item *inode_item; @@ -4504,17 +4590,13 @@ static int btrfs_add_nondir(struct btrfs_trans_handle *trans, int err = btrfs_add_link(trans, dir, inode, dentry->d_name.name, dentry->d_name.len, backref, index); - if (!err) { - d_instantiate(dentry, inode); - return 0; - } if (err > 0) err = -EEXIST; return err; } static int btrfs_mknod(struct inode *dir, struct dentry *dentry, - int mode, dev_t rdev) + umode_t mode, dev_t rdev) { struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(dir)->root; @@ -4555,13 +4637,21 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry, goto out_unlock; } + /* + * If the active LSM wants to access the inode during + * d_instantiate it needs these. Smack checks to see + * if the filesystem supports xattrs by looking at the + * ops vector. + */ + + inode->i_op = &btrfs_special_inode_operations; err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index); if (err) drop_inode = 1; else { - inode->i_op = &btrfs_special_inode_operations; init_special_inode(inode, inode->i_mode, rdev); btrfs_update_inode(trans, root, inode); + d_instantiate(dentry, inode); } out_unlock: nr = trans->blocks_used; @@ -4575,7 +4665,7 @@ out_unlock: } static int btrfs_create(struct inode *dir, struct dentry *dentry, - int mode, struct nameidata *nd) + umode_t mode, struct nameidata *nd) { struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(dir)->root; @@ -4613,15 +4703,23 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry, goto out_unlock; } + /* + * If the active LSM wants to access the inode during + * d_instantiate it needs these. Smack checks to see + * if the filesystem supports xattrs by looking at the + * ops vector. + */ + inode->i_fop = &btrfs_file_operations; + inode->i_op = &btrfs_file_inode_operations; + err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index); if (err) drop_inode = 1; else { inode->i_mapping->a_ops = &btrfs_aops; inode->i_mapping->backing_dev_info = &root->fs_info->bdi; - inode->i_fop = &btrfs_file_operations; - inode->i_op = &btrfs_file_inode_operations; BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops; + d_instantiate(dentry, inode); } out_unlock: nr = trans->blocks_used; @@ -4679,6 +4777,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *parent = dentry->d_parent; err = btrfs_update_inode(trans, root, inode); BUG_ON(err); + d_instantiate(dentry, inode); btrfs_log_new_name(trans, inode, NULL, parent); } @@ -4693,7 +4792,7 @@ fail: return err; } -static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct inode *inode = NULL; struct btrfs_trans_handle *trans; @@ -6303,7 +6402,12 @@ int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) u64 page_start; u64 page_end; + /* Need this to keep space reservations serialized */ + mutex_lock(&inode->i_mutex); ret = btrfs_delalloc_reserve_space(inode, PAGE_CACHE_SIZE); + mutex_unlock(&inode->i_mutex); + if (!ret) + ret = btrfs_update_time(vma->vm_file); if (ret) { if (ret == -ENOMEM) ret = VM_FAULT_OOM; @@ -6515,8 +6619,9 @@ static int btrfs_truncate(struct inode *inode) /* Just need the 1 for updating the inode */ trans = btrfs_start_transaction(root, 1); if (IS_ERR(trans)) { - err = PTR_ERR(trans); - goto out; + ret = err = PTR_ERR(trans); + trans = NULL; + break; } } @@ -6656,7 +6761,6 @@ struct inode *btrfs_alloc_inode(struct super_block *sb) static void btrfs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(btrfs_inode_cachep, BTRFS_I(inode)); } @@ -6794,11 +6898,13 @@ static int btrfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) { struct inode *inode = dentry->d_inode; + u32 blocksize = inode->i_sb->s_blocksize; + generic_fillattr(inode, stat); stat->dev = BTRFS_I(inode)->root->anon_dev; stat->blksize = PAGE_CACHE_SIZE; - stat->blocks = (inode_get_bytes(inode) + - BTRFS_I(inode)->delalloc_bytes) >> 9; + stat->blocks = (ALIGN(inode_get_bytes(inode), blocksize) + + ALIGN(BTRFS_I(inode)->delalloc_bytes, blocksize)) >> 9; return 0; } @@ -7074,14 +7180,21 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, goto out_unlock; } + /* + * If the active LSM wants to access the inode during + * d_instantiate it needs these. Smack checks to see + * if the filesystem supports xattrs by looking at the + * ops vector. + */ + inode->i_fop = &btrfs_file_operations; + inode->i_op = &btrfs_file_inode_operations; + err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index); if (err) drop_inode = 1; else { inode->i_mapping->a_ops = &btrfs_aops; inode->i_mapping->backing_dev_info = &root->fs_info->bdi; - inode->i_fop = &btrfs_file_operations; - inode->i_op = &btrfs_file_inode_operations; BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops; } if (drop_inode) @@ -7130,6 +7243,8 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, drop_inode = 1; out_unlock: + if (!err) + d_instantiate(dentry, inode); nr = trans->blocks_used; btrfs_end_transaction_throttle(trans, root); if (drop_inode) { @@ -7351,6 +7466,7 @@ static const struct inode_operations btrfs_symlink_inode_operations = { .follow_link = page_follow_link_light, .put_link = page_put_link, .getattr = btrfs_getattr, + .setattr = btrfs_setattr, .permission = btrfs_permission, .setxattr = btrfs_setxattr, .getxattr = btrfs_getxattr, diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 4a34c472f126..5441ff1480fd 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -201,7 +201,7 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) } } - ret = mnt_want_write(file->f_path.mnt); + ret = mnt_want_write_file(file); if (ret) goto out_unlock; @@ -252,14 +252,14 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) trans = btrfs_join_transaction(root); BUG_ON(IS_ERR(trans)); + btrfs_update_iflags(inode); + inode->i_ctime = CURRENT_TIME; ret = btrfs_update_inode(trans, root, inode); BUG_ON(ret); - btrfs_update_iflags(inode); - inode->i_ctime = CURRENT_TIME; btrfs_end_transaction(trans, root); - mnt_drop_write(file->f_path.mnt); + mnt_drop_write_file(file); ret = 0; out_unlock: @@ -858,8 +858,10 @@ static int cluster_pages_for_defrag(struct inode *inode, return 0; file_end = (isize - 1) >> PAGE_CACHE_SHIFT; + mutex_lock(&inode->i_mutex); ret = btrfs_delalloc_reserve_space(inode, num_pages << PAGE_CACHE_SHIFT); + mutex_unlock(&inode->i_mutex); if (ret) return ret; again: @@ -1216,12 +1218,12 @@ static noinline int btrfs_ioctl_resize(struct btrfs_root *root, *devstr = '\0'; devstr = vol_args->name; devid = simple_strtoull(devstr, &end, 10); - printk(KERN_INFO "resizing devid %llu\n", + printk(KERN_INFO "btrfs: resizing devid %llu\n", (unsigned long long)devid); } device = btrfs_find_device(root, devid, NULL, NULL); if (!device) { - printk(KERN_INFO "resizer unable to find device %llu\n", + printk(KERN_INFO "btrfs: resizer unable to find device %llu\n", (unsigned long long)devid); ret = -EINVAL; goto out_unlock; @@ -1267,7 +1269,7 @@ static noinline int btrfs_ioctl_resize(struct btrfs_root *root, do_div(new_size, root->sectorsize); new_size *= root->sectorsize; - printk(KERN_INFO "new size for %s is %llu\n", + printk(KERN_INFO "btrfs: new size for %s is %llu\n", device->name, (unsigned long long)new_size); if (new_size > old_size) { @@ -1278,7 +1280,7 @@ static noinline int btrfs_ioctl_resize(struct btrfs_root *root, } ret = btrfs_grow_device(trans, device, new_size); btrfs_commit_transaction(trans, root); - } else { + } else if (new_size < old_size) { ret = btrfs_shrink_device(device, new_size); } @@ -1853,7 +1855,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, goto out; } - err = mnt_want_write(file->f_path.mnt); + err = mnt_want_write_file(file); if (err) goto out; @@ -1969,7 +1971,7 @@ out_dput: dput(dentry); out_unlock_dir: mutex_unlock(&dir->i_mutex); - mnt_drop_write(file->f_path.mnt); + mnt_drop_write_file(file); out: kfree(vol_args); return err; @@ -1985,7 +1987,7 @@ static int btrfs_ioctl_defrag(struct file *file, void __user *argp) if (btrfs_root_readonly(root)) return -EROFS; - ret = mnt_want_write(file->f_path.mnt); + ret = mnt_want_write_file(file); if (ret) return ret; @@ -2038,7 +2040,7 @@ static int btrfs_ioctl_defrag(struct file *file, void __user *argp) ret = -EINVAL; } out: - mnt_drop_write(file->f_path.mnt); + mnt_drop_write_file(file); return ret; } @@ -2193,7 +2195,7 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, if (btrfs_root_readonly(root)) return -EROFS; - ret = mnt_want_write(file->f_path.mnt); + ret = mnt_want_write_file(file); if (ret) return ret; @@ -2508,7 +2510,7 @@ out_unlock: out_fput: fput(src_file); out_drop_write: - mnt_drop_write(file->f_path.mnt); + mnt_drop_write_file(file); return ret; } @@ -2547,7 +2549,7 @@ static long btrfs_ioctl_trans_start(struct file *file) if (btrfs_root_readonly(root)) goto out; - ret = mnt_want_write(file->f_path.mnt); + ret = mnt_want_write_file(file); if (ret) goto out; @@ -2563,7 +2565,7 @@ static long btrfs_ioctl_trans_start(struct file *file) out_drop: atomic_dec(&root->fs_info->open_ioctl_trans); - mnt_drop_write(file->f_path.mnt); + mnt_drop_write_file(file); out: return ret; } @@ -2798,7 +2800,7 @@ long btrfs_ioctl_trans_end(struct file *file) atomic_dec(&root->fs_info->open_ioctl_trans); - mnt_drop_write(file->f_path.mnt); + mnt_drop_write_file(file); return 0; } @@ -2930,11 +2932,13 @@ static long btrfs_ioctl_ino_to_path(struct btrfs_root *root, void __user *arg) goto out; for (i = 0; i < ipath->fspath->elem_cnt; ++i) { - rel_ptr = ipath->fspath->val[i] - (u64)ipath->fspath->val; + rel_ptr = ipath->fspath->val[i] - + (u64)(unsigned long)ipath->fspath->val; ipath->fspath->val[i] = rel_ptr; } - ret = copy_to_user((void *)ipa->fspath, (void *)ipath->fspath, size); + ret = copy_to_user((void *)(unsigned long)ipa->fspath, + (void *)(unsigned long)ipath->fspath, size); if (ret) { ret = -EFAULT; goto out; @@ -3017,7 +3021,8 @@ static long btrfs_ioctl_logical_to_ino(struct btrfs_root *root, if (ret < 0) goto out; - ret = copy_to_user((void *)loi->inodes, (void *)inodes, size); + ret = copy_to_user((void *)(unsigned long)loi->inodes, + (void *)(unsigned long)inodes, size); if (ret) ret = -EFAULT; diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index dff29d5e151a..cfb55434a469 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -2947,7 +2947,9 @@ static int relocate_file_extent_cluster(struct inode *inode, index = (cluster->start - offset) >> PAGE_CACHE_SHIFT; last_index = (cluster->end - offset) >> PAGE_CACHE_SHIFT; while (index <= last_index) { + mutex_lock(&inode->i_mutex); ret = btrfs_delalloc_reserve_metadata(inode, PAGE_CACHE_SIZE); + mutex_unlock(&inode->i_mutex); if (ret) goto out; diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index f4190f22edfb..ddf2c90d3fc0 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -256,6 +256,11 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root, void *ctx) btrfs_release_path(swarn->path); ipath = init_ipath(4096, local_root, swarn->path); + if (IS_ERR(ipath)) { + ret = PTR_ERR(ipath); + ipath = NULL; + goto err; + } ret = paths_from_inode(inum, ipath); if (ret < 0) @@ -272,7 +277,7 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root, void *ctx) swarn->logical, swarn->dev->name, (unsigned long long)swarn->sector, root, inum, offset, min(isize - offset, (u64)PAGE_SIZE), nlink, - (char *)ipath->fspath->val[i]); + (char *)(unsigned long)ipath->fspath->val[i]); free_ipath(ipath); return 0; @@ -1530,18 +1535,22 @@ static noinline_for_stack int scrub_supers(struct scrub_dev *sdev) static noinline_for_stack int scrub_workers_get(struct btrfs_root *root) { struct btrfs_fs_info *fs_info = root->fs_info; + int ret = 0; mutex_lock(&fs_info->scrub_lock); if (fs_info->scrub_workers_refcnt == 0) { btrfs_init_workers(&fs_info->scrub_workers, "scrub", fs_info->thread_pool_size, &fs_info->generic_worker); fs_info->scrub_workers.idle_thresh = 4; - btrfs_start_workers(&fs_info->scrub_workers, 1); + ret = btrfs_start_workers(&fs_info->scrub_workers); + if (ret) + goto out; } ++fs_info->scrub_workers_refcnt; +out: mutex_unlock(&fs_info->scrub_lock); - return 0; + return ret; } static noinline_for_stack void scrub_workers_put(struct btrfs_root *root) diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 8bd9d6d0e07a..ae488aa1966a 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -40,7 +40,7 @@ #include <linux/magic.h> #include <linux/slab.h> #include <linux/cleancache.h> -#include <linux/mnt_namespace.h> +#include <linux/ratelimit.h> #include "compat.h" #include "delayed-inode.h" #include "ctree.h" @@ -661,9 +661,9 @@ int btrfs_sync_fs(struct super_block *sb, int wait) return ret; } -static int btrfs_show_options(struct seq_file *seq, struct vfsmount *vfs) +static int btrfs_show_options(struct seq_file *seq, struct dentry *dentry) { - struct btrfs_root *root = btrfs_sb(vfs->mnt_sb); + struct btrfs_root *root = btrfs_sb(dentry->d_sb); struct btrfs_fs_info *info = root->fs_info; char *compress_type; @@ -825,13 +825,9 @@ static char *setup_root_args(char *args) static struct dentry *mount_subvol(const char *subvol_name, int flags, const char *device_name, char *data) { - struct super_block *s; struct dentry *root; struct vfsmount *mnt; - struct mnt_namespace *ns_private; char *newargs; - struct path path; - int error; newargs = setup_root_args(data); if (!newargs) @@ -842,39 +838,17 @@ static struct dentry *mount_subvol(const char *subvol_name, int flags, if (IS_ERR(mnt)) return ERR_CAST(mnt); - ns_private = create_mnt_ns(mnt); - if (IS_ERR(ns_private)) { - mntput(mnt); - return ERR_CAST(ns_private); - } + root = mount_subtree(mnt, subvol_name); - /* - * This will trigger the automount of the subvol so we can just - * drop the mnt we have here and return the dentry that we - * found. - */ - error = vfs_path_lookup(mnt->mnt_root, mnt, subvol_name, - LOOKUP_FOLLOW, &path); - put_mnt_ns(ns_private); - if (error) - return ERR_PTR(error); - - if (!is_subvolume_inode(path.dentry->d_inode)) { - path_put(&path); - mntput(mnt); - error = -EINVAL; + if (!IS_ERR(root) && !is_subvolume_inode(root->d_inode)) { + struct super_block *s = root->d_sb; + dput(root); + root = ERR_PTR(-EINVAL); + deactivate_locked_super(s); printk(KERN_ERR "btrfs: '%s' is not a valid subvolume\n", subvol_name); - return ERR_PTR(-EINVAL); } - /* Get a ref to the sb and the dentry we found and return it */ - s = path.mnt->mnt_sb; - atomic_inc(&s->s_active); - root = dget(path.dentry); - path_put(&path); - down_write(&s->s_umount); - return root; } @@ -1079,11 +1053,11 @@ static int btrfs_calc_avail_data_space(struct btrfs_root *root, u64 *free_bytes) u64 avail_space; u64 used_space; u64 min_stripe_size; - int min_stripes = 1; + int min_stripes = 1, num_stripes = 1; int i = 0, nr_devices; int ret; - nr_devices = fs_info->fs_devices->rw_devices; + nr_devices = fs_info->fs_devices->open_devices; BUG_ON(!nr_devices); devices_info = kmalloc(sizeof(*devices_info) * nr_devices, @@ -1093,20 +1067,24 @@ static int btrfs_calc_avail_data_space(struct btrfs_root *root, u64 *free_bytes) /* calc min stripe number for data space alloction */ type = btrfs_get_alloc_profile(root, 1); - if (type & BTRFS_BLOCK_GROUP_RAID0) + if (type & BTRFS_BLOCK_GROUP_RAID0) { min_stripes = 2; - else if (type & BTRFS_BLOCK_GROUP_RAID1) + num_stripes = nr_devices; + } else if (type & BTRFS_BLOCK_GROUP_RAID1) { min_stripes = 2; - else if (type & BTRFS_BLOCK_GROUP_RAID10) + num_stripes = 2; + } else if (type & BTRFS_BLOCK_GROUP_RAID10) { min_stripes = 4; + num_stripes = 4; + } if (type & BTRFS_BLOCK_GROUP_DUP) min_stripe_size = 2 * BTRFS_STRIPE_LEN; else min_stripe_size = BTRFS_STRIPE_LEN; - list_for_each_entry(device, &fs_devices->alloc_list, dev_alloc_list) { - if (!device->in_fs_metadata) + list_for_each_entry(device, &fs_devices->devices, dev_list) { + if (!device->in_fs_metadata || !device->bdev) continue; avail_space = device->total_bytes - device->bytes_used; @@ -1167,13 +1145,16 @@ static int btrfs_calc_avail_data_space(struct btrfs_root *root, u64 *free_bytes) i = nr_devices - 1; avail_space = 0; while (nr_devices >= min_stripes) { + if (num_stripes > nr_devices) + num_stripes = nr_devices; + if (devices_info[i].max_avail >= min_stripe_size) { int j; u64 alloc_size; - avail_space += devices_info[i].max_avail * min_stripes; + avail_space += devices_info[i].max_avail * num_stripes; alloc_size = devices_info[i].max_avail; - for (j = i + 1 - min_stripes; j <= i; j++) + for (j = i + 1 - num_stripes; j <= i; j++) devices_info[j].max_avail -= alloc_size; } i--; @@ -1290,6 +1271,16 @@ static int btrfs_unfreeze(struct super_block *sb) return 0; } +static void btrfs_fs_dirty_inode(struct inode *inode, int flags) +{ + int ret; + + ret = btrfs_dirty_inode(inode); + if (ret) + printk_ratelimited(KERN_ERR "btrfs: fail to dirty inode %Lu " + "error %d\n", btrfs_ino(inode), ret); +} + static const struct super_operations btrfs_super_ops = { .drop_inode = btrfs_drop_inode, .evict_inode = btrfs_evict_inode, @@ -1297,7 +1288,7 @@ static const struct super_operations btrfs_super_ops = { .sync_fs = btrfs_sync_fs, .show_options = btrfs_show_options, .write_inode = btrfs_write_inode, - .dirty_inode = btrfs_dirty_inode, + .dirty_inode = btrfs_fs_dirty_inode, .alloc_inode = btrfs_alloc_inode, .destroy_inode = btrfs_destroy_inode, .statfs = btrfs_statfs, diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 6a0574e923bc..81376d94cd3c 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -785,6 +785,10 @@ static noinline int commit_fs_roots(struct btrfs_trans_handle *trans, btrfs_save_ino_cache(root, trans); + /* see comments in should_cow_block() */ + root->force_cow = 0; + smp_wmb(); + if (root->commit_root != root->node) { mutex_lock(&root->fs_commit_mutex); switch_commit_root(root); @@ -947,6 +951,10 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, btrfs_tree_unlock(old); free_extent_buffer(old); + /* see comments in should_cow_block() */ + root->force_cow = 1; + smp_wmb(); + btrfs_set_root_node(new_root_item, tmp); /* record when the snapshot was created in key.offset */ key.offset = trans->transid; diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index c37433d3cd82..f4b839fd3c9d 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -295,6 +295,12 @@ loop_lock: btrfs_requeue_work(&device->work); goto done; } + /* unplug every 64 requests just for good measure */ + if (batch_run % 64 == 0) { + blk_finish_plug(&plug); + blk_start_plug(&plug); + sync_pending = 0; + } } cond_resched(); @@ -1611,7 +1617,7 @@ int btrfs_init_new_device(struct btrfs_root *root, char *device_path) if ((sb->s_flags & MS_RDONLY) && !root->fs_info->fs_devices->seeding) return -EINVAL; - bdev = blkdev_get_by_path(device_path, FMODE_EXCL, + bdev = blkdev_get_by_path(device_path, FMODE_WRITE | FMODE_EXCL, root->fs_info->bdev_holder); if (IS_ERR(bdev)) return PTR_ERR(bdev); @@ -3258,7 +3264,7 @@ static void btrfs_end_bio(struct bio *bio, int err) */ if (atomic_read(&bbio->error) > bbio->max_errors) { err = -EIO; - } else if (err) { + } else { /* * this bio is actually up to date, we didn't * go over the max number of errors diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index ab5b1c49f352..78f2d4d4f37f 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -100,6 +100,12 @@ struct btrfs_device { struct reada_zone *reada_curr_zone; struct radix_tree_root reada_zones; struct radix_tree_root reada_extents; + + /* for sending down flush barriers */ + struct bio *flush_bio; + struct completion flush_wait; + int nobarriers; + }; struct btrfs_fs_devices { diff --git a/fs/buffer.c b/fs/buffer.c index 19d8eb7fdc81..1a30db77af32 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -41,7 +41,6 @@ #include <linux/bitops.h> #include <linux/mpage.h> #include <linux/bit_spinlock.h> -#include <linux/cleancache.h> static int fsync_buffers_list(spinlock_t *lock, struct list_head *list); @@ -231,55 +230,6 @@ out: return ret; } -/* If invalidate_buffers() will trash dirty buffers, it means some kind - of fs corruption is going on. Trashing dirty data always imply losing - information that was supposed to be just stored on the physical layer - by the user. - - Thus invalidate_buffers in general usage is not allwowed to trash - dirty buffers. For example ioctl(FLSBLKBUF) expects dirty data to - be preserved. These buffers are simply skipped. - - We also skip buffers which are still in use. For example this can - happen if a userspace program is reading the block device. - - NOTE: In the case where the user removed a removable-media-disk even if - there's still dirty data not synced on disk (due a bug in the device driver - or due an error of the user), by not destroying the dirty buffers we could - generate corruption also on the next media inserted, thus a parameter is - necessary to handle this case in the most safe way possible (trying - to not corrupt also the new disk inserted with the data belonging to - the old now corrupted disk). Also for the ramdisk the natural thing - to do in order to release the ramdisk memory is to destroy dirty buffers. - - These are two special cases. Normal usage imply the device driver - to issue a sync on the device (without waiting I/O completion) and - then an invalidate_buffers call that doesn't trash dirty buffers. - - For handling cache coherency with the blkdev pagecache the 'update' case - is been introduced. It is needed to re-read from disk any pinned - buffer. NOTE: re-reading from disk is destructive so we can do it only - when we assume nobody is changing the buffercache under our I/O and when - we think the disk contains more recent information than the buffercache. - The update == 1 pass marks the buffers we need to update, the update == 2 - pass does the actual I/O. */ -void invalidate_bdev(struct block_device *bdev) -{ - struct address_space *mapping = bdev->bd_inode->i_mapping; - - if (mapping->nrpages == 0) - return; - - invalidate_bh_lrus(); - lru_add_drain_all(); /* make sure all lru add caches are flushed */ - invalidate_mapping_pages(mapping, 0, -1); - /* 99% of the time, we don't need to flush the cleancache on the bdev. - * But, for the strange corners, lets be cautious - */ - cleancache_flush_inode(mapping); -} -EXPORT_SYMBOL(invalidate_bdev); - /* * Kick the writeback threads then try to free up some ZONE_NORMAL memory. */ diff --git a/fs/cachefiles/interface.c b/fs/cachefiles/interface.c index 1064805e653b..67bef6d01484 100644 --- a/fs/cachefiles/interface.c +++ b/fs/cachefiles/interface.c @@ -11,7 +11,6 @@ #include <linux/slab.h> #include <linux/mount.h> -#include <linux/buffer_head.h> #include "internal.h" #define list_to_page(head) (list_entry((head)->prev, struct page, lru)) diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 4144caf2f9d3..173b1d22e59b 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -87,7 +87,7 @@ static int ceph_set_page_dirty(struct page *page) snapc = ceph_get_snap_context(ci->i_snap_realm->cached_context); /* dirty the head */ - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_head_snapc == NULL) ci->i_head_snapc = ceph_get_snap_context(snapc); ++ci->i_wrbuffer_ref_head; @@ -100,7 +100,7 @@ static int ceph_set_page_dirty(struct page *page) ci->i_wrbuffer_ref-1, ci->i_wrbuffer_ref_head-1, ci->i_wrbuffer_ref, ci->i_wrbuffer_ref_head, snapc, snapc->seq, snapc->num_snaps); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); /* now adjust page */ spin_lock_irq(&mapping->tree_lock); @@ -391,7 +391,7 @@ static struct ceph_snap_context *get_oldest_context(struct inode *inode, struct ceph_snap_context *snapc = NULL; struct ceph_cap_snap *capsnap = NULL; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) { dout(" cap_snap %p snapc %p has %d dirty pages\n", capsnap, capsnap->context, capsnap->dirty_pages); @@ -407,7 +407,7 @@ static struct ceph_snap_context *get_oldest_context(struct inode *inode, dout(" head snapc %p has %d dirty pages\n", snapc, ci->i_wrbuffer_ref_head); } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return snapc; } diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 0f327c6c9679..b60fc8bfb3e9 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -309,7 +309,7 @@ void ceph_reservation_status(struct ceph_fs_client *fsc, /* * Find ceph_cap for given mds, if any. * - * Called with i_lock held. + * Called with i_ceph_lock held. */ static struct ceph_cap *__get_cap_for_mds(struct ceph_inode_info *ci, int mds) { @@ -332,9 +332,9 @@ struct ceph_cap *ceph_get_cap_for_mds(struct ceph_inode_info *ci, int mds) { struct ceph_cap *cap; - spin_lock(&ci->vfs_inode.i_lock); + spin_lock(&ci->i_ceph_lock); cap = __get_cap_for_mds(ci, mds); - spin_unlock(&ci->vfs_inode.i_lock); + spin_unlock(&ci->i_ceph_lock); return cap; } @@ -361,15 +361,16 @@ static int __ceph_get_cap_mds(struct ceph_inode_info *ci) int ceph_get_cap_mds(struct inode *inode) { + struct ceph_inode_info *ci = ceph_inode(inode); int mds; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); mds = __ceph_get_cap_mds(ceph_inode(inode)); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return mds; } /* - * Called under i_lock. + * Called under i_ceph_lock. */ static void __insert_cap_node(struct ceph_inode_info *ci, struct ceph_cap *new) @@ -415,7 +416,7 @@ static void __cap_set_timeouts(struct ceph_mds_client *mdsc, * * If I_FLUSH is set, leave the inode at the front of the list. * - * Caller holds i_lock + * Caller holds i_ceph_lock * -> we take mdsc->cap_delay_lock */ static void __cap_delay_requeue(struct ceph_mds_client *mdsc, @@ -457,7 +458,7 @@ static void __cap_delay_requeue_front(struct ceph_mds_client *mdsc, /* * Cancel delayed work on cap. * - * Caller must hold i_lock. + * Caller must hold i_ceph_lock. */ static void __cap_delay_cancel(struct ceph_mds_client *mdsc, struct ceph_inode_info *ci) @@ -532,14 +533,14 @@ int ceph_add_cap(struct inode *inode, wanted |= ceph_caps_for_mode(fmode); retry: - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); cap = __get_cap_for_mds(ci, mds); if (!cap) { if (new_cap) { cap = new_cap; new_cap = NULL; } else { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); new_cap = get_cap(mdsc, caps_reservation); if (new_cap == NULL) return -ENOMEM; @@ -625,7 +626,7 @@ retry: if (fmode >= 0) __ceph_get_fmode(ci, fmode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); wake_up_all(&ci->i_cap_wq); return 0; } @@ -792,7 +793,7 @@ int ceph_caps_revoking(struct ceph_inode_info *ci, int mask) struct rb_node *p; int ret = 0; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); for (p = rb_first(&ci->i_caps); p; p = rb_next(p)) { cap = rb_entry(p, struct ceph_cap, ci_node); if (__cap_is_valid(cap) && @@ -801,7 +802,7 @@ int ceph_caps_revoking(struct ceph_inode_info *ci, int mask) break; } } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); dout("ceph_caps_revoking %p %s = %d\n", inode, ceph_cap_string(mask), ret); return ret; @@ -855,7 +856,7 @@ int __ceph_caps_mds_wanted(struct ceph_inode_info *ci) } /* - * called under i_lock + * called under i_ceph_lock */ static int __ceph_is_any_caps(struct ceph_inode_info *ci) { @@ -865,7 +866,7 @@ static int __ceph_is_any_caps(struct ceph_inode_info *ci) /* * Remove a cap. Take steps to deal with a racing iterate_session_caps. * - * caller should hold i_lock. + * caller should hold i_ceph_lock. * caller will not hold session s_mutex if called from destroy_inode. */ void __ceph_remove_cap(struct ceph_cap *cap) @@ -927,7 +928,7 @@ static int send_cap_msg(struct ceph_mds_session *session, u64 size, u64 max_size, struct timespec *mtime, struct timespec *atime, u64 time_warp_seq, - uid_t uid, gid_t gid, mode_t mode, + uid_t uid, gid_t gid, umode_t mode, u64 xattr_version, struct ceph_buffer *xattrs_buf, u64 follows) @@ -1028,7 +1029,7 @@ static void __queue_cap_release(struct ceph_mds_session *session, /* * Queue cap releases when an inode is dropped from our cache. Since - * inode is about to be destroyed, there is no need for i_lock. + * inode is about to be destroyed, there is no need for i_ceph_lock. */ void ceph_queue_caps_release(struct inode *inode) { @@ -1049,7 +1050,7 @@ void ceph_queue_caps_release(struct inode *inode) /* * Send a cap msg on the given inode. Update our caps state, then - * drop i_lock and send the message. + * drop i_ceph_lock and send the message. * * Make note of max_size reported/requested from mds, revoked caps * that have now been implemented. @@ -1061,13 +1062,13 @@ void ceph_queue_caps_release(struct inode *inode) * Return non-zero if delayed release, or we experienced an error * such that the caller should requeue + retry later. * - * called with i_lock, then drops it. + * called with i_ceph_lock, then drops it. * caller should hold snap_rwsem (read), s_mutex. */ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap, int op, int used, int want, int retain, int flushing, unsigned *pflush_tid) - __releases(cap->ci->vfs_inode->i_lock) + __releases(cap->ci->i_ceph_lock) { struct ceph_inode_info *ci = cap->ci; struct inode *inode = &ci->vfs_inode; @@ -1077,7 +1078,7 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap, u64 size, max_size; struct timespec mtime, atime; int wake = 0; - mode_t mode; + umode_t mode; uid_t uid; gid_t gid; struct ceph_mds_session *session; @@ -1170,7 +1171,7 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap, xattr_version = ci->i_xattrs.version; } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); ret = send_cap_msg(session, ceph_vino(inode).ino, cap_id, op, keep, want, flushing, seq, flush_tid, issue_seq, mseq, @@ -1198,13 +1199,13 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap, * Unless @again is true, skip cap_snaps that were already sent to * the MDS (i.e., during this session). * - * Called under i_lock. Takes s_mutex as needed. + * Called under i_ceph_lock. Takes s_mutex as needed. */ void __ceph_flush_snaps(struct ceph_inode_info *ci, struct ceph_mds_session **psession, int again) - __releases(ci->vfs_inode->i_lock) - __acquires(ci->vfs_inode->i_lock) + __releases(ci->i_ceph_lock) + __acquires(ci->i_ceph_lock) { struct inode *inode = &ci->vfs_inode; int mds; @@ -1261,7 +1262,7 @@ retry: session = NULL; } if (!session) { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); mutex_lock(&mdsc->mutex); session = __ceph_lookup_mds_session(mdsc, mds); mutex_unlock(&mdsc->mutex); @@ -1275,7 +1276,7 @@ retry: * deletion or migration. retry, and we'll * get a better @mds value next time. */ - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); goto retry; } @@ -1285,7 +1286,7 @@ retry: list_del_init(&capsnap->flushing_item); list_add_tail(&capsnap->flushing_item, &session->s_cap_snaps_flushing); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); dout("flush_snaps %p cap_snap %p follows %lld tid %llu\n", inode, capsnap, capsnap->follows, capsnap->flush_tid); @@ -1302,7 +1303,7 @@ retry: next_follows = capsnap->follows + 1; ceph_put_cap_snap(capsnap); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); goto retry; } @@ -1322,11 +1323,9 @@ out: static void ceph_flush_snaps(struct ceph_inode_info *ci) { - struct inode *inode = &ci->vfs_inode; - - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); __ceph_flush_snaps(ci, NULL, 0); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } /* @@ -1373,7 +1372,7 @@ int __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask) * Add dirty inode to the flushing list. Assigned a seq number so we * can wait for caps to flush without starving. * - * Called under i_lock. + * Called under i_ceph_lock. */ static int __mark_caps_flushing(struct inode *inode, struct ceph_mds_session *session) @@ -1421,9 +1420,9 @@ static int try_nonblocking_invalidate(struct inode *inode) struct ceph_inode_info *ci = ceph_inode(inode); u32 invalidating_gen = ci->i_rdcache_gen; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); invalidate_mapping_pages(&inode->i_data, 0, -1); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (inode->i_data.nrpages == 0 && invalidating_gen == ci->i_rdcache_gen) { @@ -1470,7 +1469,7 @@ void ceph_check_caps(struct ceph_inode_info *ci, int flags, if (mdsc->stopping) is_delayed = 1; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_ceph_flags & CEPH_I_FLUSH) flags |= CHECK_CAPS_FLUSH; @@ -1480,7 +1479,7 @@ void ceph_check_caps(struct ceph_inode_info *ci, int flags, __ceph_flush_snaps(ci, &session, 0); goto retry_locked; retry: - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); retry_locked: file_wanted = __ceph_caps_file_wanted(ci); used = __ceph_caps_used(ci); @@ -1634,7 +1633,7 @@ ack: if (mutex_trylock(&session->s_mutex) == 0) { dout("inverting session/ino locks on %p\n", session); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (took_snap_rwsem) { up_read(&mdsc->snap_rwsem); took_snap_rwsem = 0; @@ -1648,7 +1647,7 @@ ack: if (down_read_trylock(&mdsc->snap_rwsem) == 0) { dout("inverting snap/in locks on %p\n", inode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); down_read(&mdsc->snap_rwsem); took_snap_rwsem = 1; goto retry; @@ -1664,10 +1663,10 @@ ack: mds = cap->mds; /* remember mds, so we don't repeat */ sent++; - /* __send_cap drops i_lock */ + /* __send_cap drops i_ceph_lock */ delayed += __send_cap(mdsc, cap, CEPH_CAP_OP_UPDATE, used, want, retain, flushing, NULL); - goto retry; /* retake i_lock and restart our cap scan. */ + goto retry; /* retake i_ceph_lock and restart our cap scan. */ } /* @@ -1681,7 +1680,7 @@ ack: else if (!is_delayed || force_requeue) __cap_delay_requeue(mdsc, ci); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (queue_invalidate) ceph_queue_invalidate(inode); @@ -1704,7 +1703,7 @@ static int try_flush_caps(struct inode *inode, struct ceph_mds_session *session, int flushing = 0; retry: - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_ceph_flags & CEPH_I_NOFLUSH) { dout("try_flush_caps skipping %p I_NOFLUSH set\n", inode); goto out; @@ -1716,7 +1715,7 @@ retry: int delayed; if (!session) { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); session = cap->session; mutex_lock(&session->s_mutex); goto retry; @@ -1727,18 +1726,18 @@ retry: flushing = __mark_caps_flushing(inode, session); - /* __send_cap drops i_lock */ + /* __send_cap drops i_ceph_lock */ delayed = __send_cap(mdsc, cap, CEPH_CAP_OP_FLUSH, used, want, cap->issued | cap->implemented, flushing, flush_tid); if (!delayed) goto out_unlocked; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); __cap_delay_requeue(mdsc, ci); } out: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); out_unlocked: if (session && unlock_session) mutex_unlock(&session->s_mutex); @@ -1753,7 +1752,7 @@ static int caps_are_flushed(struct inode *inode, unsigned tid) struct ceph_inode_info *ci = ceph_inode(inode); int i, ret = 1; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); for (i = 0; i < CEPH_CAP_BITS; i++) if ((ci->i_flushing_caps & (1 << i)) && ci->i_cap_flush_tid[i] <= tid) { @@ -1761,7 +1760,7 @@ static int caps_are_flushed(struct inode *inode, unsigned tid) ret = 0; break; } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return ret; } @@ -1868,10 +1867,10 @@ int ceph_write_inode(struct inode *inode, struct writeback_control *wbc) struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (__ceph_caps_dirty(ci)) __cap_delay_requeue_front(mdsc, ci); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } return err; } @@ -1894,7 +1893,7 @@ static void kick_flushing_capsnaps(struct ceph_mds_client *mdsc, struct inode *inode = &ci->vfs_inode; struct ceph_cap *cap; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); cap = ci->i_auth_cap; if (cap && cap->session == session) { dout("kick_flushing_caps %p cap %p capsnap %p\n", inode, @@ -1904,7 +1903,7 @@ static void kick_flushing_capsnaps(struct ceph_mds_client *mdsc, pr_err("%p auth cap %p not mds%d ???\n", inode, cap, session->s_mds); } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } } @@ -1921,7 +1920,7 @@ void ceph_kick_flushing_caps(struct ceph_mds_client *mdsc, struct ceph_cap *cap; int delayed = 0; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); cap = ci->i_auth_cap; if (cap && cap->session == session) { dout("kick_flushing_caps %p cap %p %s\n", inode, @@ -1932,14 +1931,14 @@ void ceph_kick_flushing_caps(struct ceph_mds_client *mdsc, cap->issued | cap->implemented, ci->i_flushing_caps, NULL); if (delayed) { - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); __cap_delay_requeue(mdsc, ci); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } } else { pr_err("%p auth cap %p not mds%d ???\n", inode, cap, session->s_mds); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } } } @@ -1952,7 +1951,7 @@ static void kick_flushing_inode_caps(struct ceph_mds_client *mdsc, struct ceph_cap *cap; int delayed = 0; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); cap = ci->i_auth_cap; dout("kick_flushing_inode_caps %p flushing %s flush_seq %lld\n", inode, ceph_cap_string(ci->i_flushing_caps), ci->i_cap_flush_seq); @@ -1964,12 +1963,12 @@ static void kick_flushing_inode_caps(struct ceph_mds_client *mdsc, cap->issued | cap->implemented, ci->i_flushing_caps, NULL); if (delayed) { - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); __cap_delay_requeue(mdsc, ci); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } } else { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } } @@ -1978,7 +1977,7 @@ static void kick_flushing_inode_caps(struct ceph_mds_client *mdsc, * Take references to capabilities we hold, so that we don't release * them to the MDS prematurely. * - * Protected by i_lock. + * Protected by i_ceph_lock. */ static void __take_cap_refs(struct ceph_inode_info *ci, int got) { @@ -2016,7 +2015,7 @@ static int try_get_cap_refs(struct ceph_inode_info *ci, int need, int want, dout("get_cap_refs %p need %s want %s\n", inode, ceph_cap_string(need), ceph_cap_string(want)); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); /* make sure file is actually open */ file_wanted = __ceph_caps_file_wanted(ci); @@ -2077,7 +2076,7 @@ static int try_get_cap_refs(struct ceph_inode_info *ci, int need, int want, ceph_cap_string(have), ceph_cap_string(need)); } out: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); dout("get_cap_refs %p ret %d got %s\n", inode, ret, ceph_cap_string(*got)); return ret; @@ -2094,7 +2093,7 @@ static void check_max_size(struct inode *inode, loff_t endoff) int check = 0; /* do we need to explicitly request a larger max_size? */ - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if ((endoff >= ci->i_max_size || endoff > (inode->i_size << 1)) && endoff > ci->i_wanted_max_size) { @@ -2103,7 +2102,7 @@ static void check_max_size(struct inode *inode, loff_t endoff) ci->i_wanted_max_size = endoff; check = 1; } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (check) ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL); } @@ -2140,9 +2139,9 @@ retry: */ void ceph_get_cap_refs(struct ceph_inode_info *ci, int caps) { - spin_lock(&ci->vfs_inode.i_lock); + spin_lock(&ci->i_ceph_lock); __take_cap_refs(ci, caps); - spin_unlock(&ci->vfs_inode.i_lock); + spin_unlock(&ci->i_ceph_lock); } /* @@ -2160,7 +2159,7 @@ void ceph_put_cap_refs(struct ceph_inode_info *ci, int had) int last = 0, put = 0, flushsnaps = 0, wake = 0; struct ceph_cap_snap *capsnap; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (had & CEPH_CAP_PIN) --ci->i_pin_ref; if (had & CEPH_CAP_FILE_RD) @@ -2193,7 +2192,7 @@ void ceph_put_cap_refs(struct ceph_inode_info *ci, int had) } } } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); dout("put_cap_refs %p had %s%s%s\n", inode, ceph_cap_string(had), last ? " last" : "", put ? " put" : ""); @@ -2225,7 +2224,7 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr, int found = 0; struct ceph_cap_snap *capsnap = NULL; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); ci->i_wrbuffer_ref -= nr; last = !ci->i_wrbuffer_ref; @@ -2274,7 +2273,7 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr, } } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (last) { ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL); @@ -2291,7 +2290,7 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr, * Handle a cap GRANT message from the MDS. (Note that a GRANT may * actually be a revocation if it specifies a smaller cap set.) * - * caller holds s_mutex and i_lock, we drop both. + * caller holds s_mutex and i_ceph_lock, we drop both. * * return value: * 0 - ok @@ -2302,7 +2301,7 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant, struct ceph_mds_session *session, struct ceph_cap *cap, struct ceph_buffer *xattr_buf) - __releases(inode->i_lock) + __releases(ci->i_ceph_lock) { struct ceph_inode_info *ci = ceph_inode(inode); int mds = session->s_mds; @@ -2453,7 +2452,7 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant, } BUG_ON(cap->issued & ~cap->implemented); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (writeback) /* * queue inode for writeback: we can't actually call @@ -2483,7 +2482,7 @@ static void handle_cap_flush_ack(struct inode *inode, u64 flush_tid, struct ceph_mds_caps *m, struct ceph_mds_session *session, struct ceph_cap *cap) - __releases(inode->i_lock) + __releases(ci->i_ceph_lock) { struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc; @@ -2539,7 +2538,7 @@ static void handle_cap_flush_ack(struct inode *inode, u64 flush_tid, wake_up_all(&ci->i_cap_wq); out: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (drop) iput(inode); } @@ -2562,7 +2561,7 @@ static void handle_cap_flushsnap_ack(struct inode *inode, u64 flush_tid, dout("handle_cap_flushsnap_ack inode %p ci %p mds%d follows %lld\n", inode, ci, session->s_mds, follows); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) { if (capsnap->follows == follows) { if (capsnap->flush_tid != flush_tid) { @@ -2585,7 +2584,7 @@ static void handle_cap_flushsnap_ack(struct inode *inode, u64 flush_tid, capsnap, capsnap->follows); } } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (drop) iput(inode); } @@ -2598,7 +2597,7 @@ static void handle_cap_flushsnap_ack(struct inode *inode, u64 flush_tid, static void handle_cap_trunc(struct inode *inode, struct ceph_mds_caps *trunc, struct ceph_mds_session *session) - __releases(inode->i_lock) + __releases(ci->i_ceph_lock) { struct ceph_inode_info *ci = ceph_inode(inode); int mds = session->s_mds; @@ -2617,7 +2616,7 @@ static void handle_cap_trunc(struct inode *inode, inode, mds, seq, truncate_size, truncate_seq); queue_trunc = ceph_fill_file_size(inode, issued, truncate_seq, truncate_size, size); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (queue_trunc) ceph_queue_vmtruncate(inode); @@ -2646,7 +2645,7 @@ static void handle_cap_export(struct inode *inode, struct ceph_mds_caps *ex, dout("handle_cap_export inode %p ci %p mds%d mseq %d\n", inode, ci, mds, mseq); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); /* make sure we haven't seen a higher mseq */ for (p = rb_first(&ci->i_caps); p; p = rb_next(p)) { @@ -2690,7 +2689,7 @@ static void handle_cap_export(struct inode *inode, struct ceph_mds_caps *ex, } /* else, we already released it */ - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } /* @@ -2745,9 +2744,9 @@ static void handle_cap_import(struct ceph_mds_client *mdsc, up_read(&mdsc->snap_rwsem); /* make sure we re-request max_size, if necessary */ - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); ci->i_requested_max_size = 0; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } /* @@ -2762,6 +2761,7 @@ void ceph_handle_caps(struct ceph_mds_session *session, struct ceph_mds_client *mdsc = session->s_mdsc; struct super_block *sb = mdsc->fsc->sb; struct inode *inode; + struct ceph_inode_info *ci; struct ceph_cap *cap; struct ceph_mds_caps *h; int mds = session->s_mds; @@ -2815,6 +2815,7 @@ void ceph_handle_caps(struct ceph_mds_session *session, /* lookup ino */ inode = ceph_find_inode(sb, vino); + ci = ceph_inode(inode); dout(" op %s ino %llx.%llx inode %p\n", ceph_cap_op_name(op), vino.ino, vino.snap, inode); if (!inode) { @@ -2844,16 +2845,16 @@ void ceph_handle_caps(struct ceph_mds_session *session, } /* the rest require a cap */ - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); cap = __get_cap_for_mds(ceph_inode(inode), mds); if (!cap) { dout(" no cap on %p ino %llx.%llx from mds%d\n", inode, ceph_ino(inode), ceph_snap(inode), mds); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); goto flush_cap_releases; } - /* note that each of these drops i_lock for us */ + /* note that each of these drops i_ceph_lock for us */ switch (op) { case CEPH_CAP_OP_REVOKE: case CEPH_CAP_OP_GRANT: @@ -2869,7 +2870,7 @@ void ceph_handle_caps(struct ceph_mds_session *session, break; default: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); pr_err("ceph_handle_caps: unknown cap op %d %s\n", op, ceph_cap_op_name(op)); } @@ -2962,13 +2963,13 @@ void ceph_put_fmode(struct ceph_inode_info *ci, int fmode) struct inode *inode = &ci->vfs_inode; int last = 0; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); dout("put_fmode %p fmode %d %d -> %d\n", inode, fmode, ci->i_nr_by_mode[fmode], ci->i_nr_by_mode[fmode]-1); BUG_ON(ci->i_nr_by_mode[fmode] == 0); if (--ci->i_nr_by_mode[fmode] == 0) last++; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (last && ci->i_vino.snap == CEPH_NOSNAP) ceph_check_caps(ci, 0, NULL); @@ -2991,7 +2992,7 @@ int ceph_encode_inode_release(void **p, struct inode *inode, int used, dirty; int ret = 0; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); used = __ceph_caps_used(ci); dirty = __ceph_caps_dirty(ci); @@ -3046,7 +3047,7 @@ int ceph_encode_inode_release(void **p, struct inode *inode, inode, cap, ceph_cap_string(cap->issued)); } } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return ret; } @@ -3061,7 +3062,7 @@ int ceph_encode_dentry_release(void **p, struct dentry *dentry, /* * force an record for the directory caps if we have a dentry lease. - * this is racy (can't take i_lock and d_lock together), but it + * this is racy (can't take i_ceph_lock and d_lock together), but it * doesn't have to be perfect; the mds will revoke anything we don't * release. */ diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index 2abd0dfad7f8..618246bc2196 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -281,18 +281,18 @@ static int ceph_readdir(struct file *filp, void *dirent, filldir_t filldir) } /* can we use the dcache? */ - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if ((filp->f_pos == 2 || fi->dentry) && !ceph_test_mount_opt(fsc, NOASYNCREADDIR) && ceph_snap(inode) != CEPH_SNAPDIR && ceph_dir_test_complete(inode) && __ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1)) { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); err = __dcache_readdir(filp, dirent, filldir); if (err != -EAGAIN) return err; } else { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } if (fi->dentry) { err = note_last_dentry(fi, fi->dentry->d_name.name, @@ -428,12 +428,12 @@ more: * were released during the whole readdir, and we should have * the complete dir contents in our cache. */ - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_release_count == fi->dir_release_count) { ceph_dir_set_complete(inode); ci->i_max_offset = filp->f_pos; } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); dout("readdir %p filp %p done.\n", inode, filp); return 0; @@ -607,7 +607,7 @@ static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry, struct ceph_inode_info *ci = ceph_inode(dir); struct ceph_dentry_info *di = ceph_dentry(dentry); - spin_lock(&dir->i_lock); + spin_lock(&ci->i_ceph_lock); dout(" dir %p flags are %d\n", dir, ci->i_ceph_flags); if (strncmp(dentry->d_name.name, fsc->mount_options->snapdir_name, @@ -615,13 +615,13 @@ static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry, !is_root_ceph_dentry(dir, dentry) && ceph_dir_test_complete(dir) && (__ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1))) { - spin_unlock(&dir->i_lock); + spin_unlock(&ci->i_ceph_lock); dout(" dir %p complete, -ENOENT\n", dir); d_add(dentry, NULL); di->lease_shared_gen = ci->i_shared_gen; return NULL; } - spin_unlock(&dir->i_lock); + spin_unlock(&ci->i_ceph_lock); } op = ceph_snap(dir) == CEPH_SNAPDIR ? @@ -666,7 +666,7 @@ int ceph_handle_notrace_create(struct inode *dir, struct dentry *dentry) } static int ceph_mknod(struct inode *dir, struct dentry *dentry, - int mode, dev_t rdev) + umode_t mode, dev_t rdev) { struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb); struct ceph_mds_client *mdsc = fsc->mdsc; @@ -676,7 +676,7 @@ static int ceph_mknod(struct inode *dir, struct dentry *dentry, if (ceph_snap(dir) != CEPH_NOSNAP) return -EROFS; - dout("mknod in dir %p dentry %p mode 0%o rdev %d\n", + dout("mknod in dir %p dentry %p mode 0%ho rdev %d\n", dir, dentry, mode, rdev); req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_MKNOD, USE_AUTH_MDS); if (IS_ERR(req)) { @@ -699,7 +699,7 @@ static int ceph_mknod(struct inode *dir, struct dentry *dentry, return err; } -static int ceph_create(struct inode *dir, struct dentry *dentry, int mode, +static int ceph_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { dout("create in dir %p dentry %p name '%.*s'\n", @@ -753,7 +753,7 @@ static int ceph_symlink(struct inode *dir, struct dentry *dentry, return err; } -static int ceph_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int ceph_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb); struct ceph_mds_client *mdsc = fsc->mdsc; @@ -767,7 +767,7 @@ static int ceph_mkdir(struct inode *dir, struct dentry *dentry, int mode) dout("mksnap dir %p snap '%.*s' dn %p\n", dir, dentry->d_name.len, dentry->d_name.name, dentry); } else if (ceph_snap(dir) == CEPH_NOSNAP) { - dout("mkdir dir %p dn %p mode 0%o\n", dir, dentry, mode); + dout("mkdir dir %p dn %p mode 0%ho\n", dir, dentry, mode); op = CEPH_MDS_OP_MKDIR; } else { goto out; @@ -841,12 +841,12 @@ static int drop_caps_for_unlink(struct inode *inode) struct ceph_inode_info *ci = ceph_inode(inode); int drop = CEPH_CAP_LINK_SHARED | CEPH_CAP_LINK_EXCL; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (inode->i_nlink == 1) { drop |= ~(__ceph_caps_wanted(ci) | CEPH_CAP_PIN); ci->i_ceph_flags |= CEPH_I_NODELAY; } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return drop; } @@ -870,7 +870,7 @@ static int ceph_unlink(struct inode *dir, struct dentry *dentry) } else if (ceph_snap(dir) == CEPH_NOSNAP) { dout("unlink/rmdir dir %p dn %p inode %p\n", dir, dentry, inode); - op = ((dentry->d_inode->i_mode & S_IFMT) == S_IFDIR) ? + op = S_ISDIR(dentry->d_inode->i_mode) ? CEPH_MDS_OP_RMDIR : CEPH_MDS_OP_UNLINK; } else goto out; @@ -973,7 +973,7 @@ static int dentry_lease_is_valid(struct dentry *dentry) spin_lock(&dentry->d_lock); di = ceph_dentry(dentry); - if (di && di->lease_session) { + if (di->lease_session) { s = di->lease_session; spin_lock(&s->s_cap_lock); gen = s->s_cap_gen; @@ -1015,10 +1015,10 @@ static int dir_lease_is_valid(struct inode *dir, struct dentry *dentry) struct ceph_dentry_info *di = ceph_dentry(dentry); int valid = 0; - spin_lock(&dir->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_shared_gen == di->lease_shared_gen) valid = __ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1); - spin_unlock(&dir->i_lock); + spin_unlock(&ci->i_ceph_lock); dout("dir_lease_is_valid dir %p v%u dentry %p v%u = %d\n", dir, (unsigned)ci->i_shared_gen, dentry, (unsigned)di->lease_shared_gen, valid); @@ -1072,13 +1072,11 @@ static void ceph_d_release(struct dentry *dentry) struct ceph_dentry_info *di = ceph_dentry(dentry); dout("d_release %p\n", dentry); - if (di) { - ceph_dentry_lru_del(dentry); - if (di->lease_session) - ceph_put_mds_session(di->lease_session); - kmem_cache_free(ceph_dentry_cachep, di); - dentry->d_fsdata = NULL; - } + ceph_dentry_lru_del(dentry); + if (di->lease_session) + ceph_put_mds_session(di->lease_session); + kmem_cache_free(ceph_dentry_cachep, di); + dentry->d_fsdata = NULL; } static int ceph_snapdir_d_revalidate(struct dentry *dentry, @@ -1094,42 +1092,38 @@ static int ceph_snapdir_d_revalidate(struct dentry *dentry, /* * Set/clear/test dir complete flag on the dir's dentry. */ -static struct dentry * __d_find_any_alias(struct inode *inode) -{ - struct dentry *alias; - - if (list_empty(&inode->i_dentry)) - return NULL; - alias = list_first_entry(&inode->i_dentry, struct dentry, d_alias); - return alias; -} - void ceph_dir_set_complete(struct inode *inode) { - struct dentry *dentry = __d_find_any_alias(inode); + struct dentry *dentry = d_find_any_alias(inode); - if (dentry && ceph_dentry(dentry)) { + if (dentry && ceph_dentry(dentry) && + ceph_test_mount_opt(ceph_sb_to_client(dentry->d_sb), DCACHE)) { dout(" marking %p (%p) complete\n", inode, dentry); set_bit(CEPH_D_COMPLETE, &ceph_dentry(dentry)->flags); } + dput(dentry); } void ceph_dir_clear_complete(struct inode *inode) { - struct dentry *dentry = __d_find_any_alias(inode); + struct dentry *dentry = d_find_any_alias(inode); if (dentry && ceph_dentry(dentry)) { - dout(" marking %p (%p) NOT complete\n", inode, dentry); - clear_bit(CEPH_D_COMPLETE, &ceph_dentry(dentry)->flags); + dout(" marking %p (%p) complete\n", inode, dentry); + set_bit(CEPH_D_COMPLETE, &ceph_dentry(dentry)->flags); } + dput(dentry); } bool ceph_dir_test_complete(struct inode *inode) { - struct dentry *dentry = __d_find_any_alias(inode); + struct dentry *dentry = d_find_any_alias(inode); - if (dentry && ceph_dentry(dentry)) - return test_bit(CEPH_D_COMPLETE, &ceph_dentry(dentry)->flags); + if (dentry && ceph_dentry(dentry)) { + dout(" marking %p (%p) NOT complete\n", inode, dentry); + clear_bit(CEPH_D_COMPLETE, &ceph_dentry(dentry)->flags); + } + dput(dentry); return false; } @@ -1143,7 +1137,7 @@ static void ceph_d_prune(struct dentry *dentry) { struct ceph_dentry_info *di; - dout("d_release %p\n", dentry); + dout("ceph_d_prune %p\n", dentry); /* do we have a valid parent? */ if (!dentry->d_parent || IS_ROOT(dentry)) @@ -1243,6 +1237,7 @@ static int ceph_dir_fsync(struct file *file, loff_t start, loff_t end, do { ceph_mdsc_get_request(req); spin_unlock(&ci->i_unsafe_lock); + dout("dir_fsync %p wait on tid %llu (until %llu)\n", inode, req->r_tid, last_tid); if (req->r_timeout) { @@ -1255,9 +1250,9 @@ static int ceph_dir_fsync(struct file *file, loff_t start, loff_t end, } else { wait_for_completion(&req->r_safe_completion); } - spin_lock(&ci->i_unsafe_lock); ceph_mdsc_put_request(req); + spin_lock(&ci->i_unsafe_lock); if (ret || list_empty(head)) break; req = list_entry(head->next, @@ -1282,13 +1277,11 @@ void ceph_dentry_lru_add(struct dentry *dn) dout("dentry_lru_add %p %p '%.*s'\n", di, dn, dn->d_name.len, dn->d_name.name); - if (di) { - mdsc = ceph_sb_to_client(dn->d_sb)->mdsc; - spin_lock(&mdsc->dentry_lru_lock); - list_add_tail(&di->lru, &mdsc->dentry_lru); - mdsc->num_dentry++; - spin_unlock(&mdsc->dentry_lru_lock); - } + mdsc = ceph_sb_to_client(dn->d_sb)->mdsc; + spin_lock(&mdsc->dentry_lru_lock); + list_add_tail(&di->lru, &mdsc->dentry_lru); + mdsc->num_dentry++; + spin_unlock(&mdsc->dentry_lru_lock); } void ceph_dentry_lru_touch(struct dentry *dn) @@ -1298,12 +1291,10 @@ void ceph_dentry_lru_touch(struct dentry *dn) dout("dentry_lru_touch %p %p '%.*s' (offset %lld)\n", di, dn, dn->d_name.len, dn->d_name.name, di->offset); - if (di) { - mdsc = ceph_sb_to_client(dn->d_sb)->mdsc; - spin_lock(&mdsc->dentry_lru_lock); - list_move_tail(&di->lru, &mdsc->dentry_lru); - spin_unlock(&mdsc->dentry_lru_lock); - } + mdsc = ceph_sb_to_client(dn->d_sb)->mdsc; + spin_lock(&mdsc->dentry_lru_lock); + list_move_tail(&di->lru, &mdsc->dentry_lru); + spin_unlock(&mdsc->dentry_lru_lock); } void ceph_dentry_lru_del(struct dentry *dn) @@ -1313,13 +1304,11 @@ void ceph_dentry_lru_del(struct dentry *dn) dout("dentry_lru_del %p %p '%.*s'\n", di, dn, dn->d_name.len, dn->d_name.name); - if (di) { - mdsc = ceph_sb_to_client(dn->d_sb)->mdsc; - spin_lock(&mdsc->dentry_lru_lock); - list_del_init(&di->lru); - mdsc->num_dentry--; - spin_unlock(&mdsc->dentry_lru_lock); - } + mdsc = ceph_sb_to_client(dn->d_sb)->mdsc; + spin_lock(&mdsc->dentry_lru_lock); + list_del_init(&di->lru); + mdsc->num_dentry--; + spin_unlock(&mdsc->dentry_lru_lock); } /* diff --git a/fs/ceph/export.c b/fs/ceph/export.c index 9fbcdecaaccd..fbb2a643ef10 100644 --- a/fs/ceph/export.c +++ b/fs/ceph/export.c @@ -56,9 +56,7 @@ static int ceph_encode_fh(struct dentry *dentry, u32 *rawfh, int *max_len, return -EINVAL; spin_lock(&dentry->d_lock); - parent = dget(dentry->d_parent); - spin_unlock(&dentry->d_lock); - + parent = dentry->d_parent; if (*max_len >= connected_handle_length) { dout("encode_fh %p connectable\n", dentry); cfh->ino = ceph_ino(dentry->d_inode); @@ -81,7 +79,7 @@ static int ceph_encode_fh(struct dentry *dentry, u32 *rawfh, int *max_len, *max_len = handle_length; type = 255; } - dput(parent); + spin_unlock(&dentry->d_lock); return type; } diff --git a/fs/ceph/file.c b/fs/ceph/file.c index ce549d31eeb7..ed72428d9c75 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -147,9 +147,9 @@ int ceph_open(struct inode *inode, struct file *file) /* trivially open snapdir */ if (ceph_snap(inode) == CEPH_SNAPDIR) { - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); __ceph_get_fmode(ci, fmode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return ceph_init_file(inode, file, fmode); } @@ -158,7 +158,7 @@ int ceph_open(struct inode *inode, struct file *file) * write) or any MDS (for read). Update wanted set * asynchronously. */ - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (__ceph_is_any_real_caps(ci) && (((fmode & CEPH_FILE_MODE_WR) == 0) || ci->i_auth_cap)) { int mds_wanted = __ceph_caps_mds_wanted(ci); @@ -168,7 +168,7 @@ int ceph_open(struct inode *inode, struct file *file) inode, fmode, ceph_cap_string(wanted), ceph_cap_string(issued)); __ceph_get_fmode(ci, fmode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); /* adjust wanted? */ if ((issued & wanted) != wanted && @@ -180,10 +180,10 @@ int ceph_open(struct inode *inode, struct file *file) } else if (ceph_snap(inode) != CEPH_NOSNAP && (ci->i_snap_caps & wanted) == wanted) { __ceph_get_fmode(ci, fmode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return ceph_init_file(inode, file, fmode); } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); dout("open fmode %d wants %s\n", fmode, ceph_cap_string(wanted)); req = prepare_open_request(inode->i_sb, flags, 0); @@ -743,9 +743,9 @@ retry_snap: */ int dirty; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); ceph_put_cap_refs(ci, got); ret = generic_file_aio_write(iocb, iov, nr_segs, pos); @@ -764,9 +764,9 @@ retry_snap: if (ret >= 0) { int dirty; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (dirty) __mark_inode_dirty(inode, dirty); } @@ -797,7 +797,8 @@ static loff_t ceph_llseek(struct file *file, loff_t offset, int origin) mutex_lock(&inode->i_mutex); __ceph_do_pending_vmtruncate(inode); - if (origin != SEEK_CUR || origin != SEEK_SET) { + + if (origin == SEEK_END || origin == SEEK_DATA || origin == SEEK_HOLE) { ret = ceph_do_getattr(inode, CEPH_STAT_CAP_SIZE); if (ret < 0) { offset = ret; diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index e392bfce84a3..2c489378b4cd 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -297,6 +297,8 @@ struct inode *ceph_alloc_inode(struct super_block *sb) dout("alloc_inode %p\n", &ci->vfs_inode); + spin_lock_init(&ci->i_ceph_lock); + ci->i_version = 0; ci->i_time_warp_seq = 0; ci->i_ceph_flags = 0; @@ -382,7 +384,6 @@ static void ceph_i_callback(struct rcu_head *head) struct inode *inode = container_of(head, struct inode, i_rcu); struct ceph_inode_info *ci = ceph_inode(inode); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(ceph_inode_cachep, ci); } @@ -583,7 +584,7 @@ static int fill_inode(struct inode *inode, iinfo->xattr_len); } - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); /* * provided version will be odd if inode value is projected, @@ -680,7 +681,7 @@ static int fill_inode(struct inode *inode, char *sym; BUG_ON(symlen != inode->i_size); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); err = -ENOMEM; sym = kmalloc(symlen+1, GFP_NOFS); @@ -689,7 +690,7 @@ static int fill_inode(struct inode *inode, memcpy(sym, iinfo->symlink, symlen); sym[symlen] = 0; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (!ci->i_symlink) ci->i_symlink = sym; else @@ -715,7 +716,7 @@ static int fill_inode(struct inode *inode, } no_change: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); /* queue truncate if we saw i_size decrease */ if (queue_trunc) @@ -750,13 +751,13 @@ no_change: info->cap.flags, caps_reservation); } else { - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); dout(" %p got snap_caps %s\n", inode, ceph_cap_string(le32_to_cpu(info->cap.caps))); ci->i_snap_caps |= le32_to_cpu(info->cap.caps); if (cap_fmode >= 0) __ceph_get_fmode(ci, cap_fmode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } } else if (cap_fmode >= 0) { pr_warning("mds issued no caps on %llx.%llx\n", @@ -849,19 +850,21 @@ static void ceph_set_dentry_offset(struct dentry *dn) { struct dentry *dir = dn->d_parent; struct inode *inode = dir->d_inode; + struct ceph_inode_info *ci; struct ceph_dentry_info *di; BUG_ON(!inode); + ci = ceph_inode(inode); di = ceph_dentry(dn); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (!ceph_dir_test_complete(inode)) { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return; } di->offset = ceph_inode(inode)->i_max_offset++; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); spin_lock(&dir->d_lock); spin_lock_nested(&dn->d_lock, DENTRY_D_LOCK_NESTED); @@ -1308,7 +1311,7 @@ int ceph_inode_set_size(struct inode *inode, loff_t size) struct ceph_inode_info *ci = ceph_inode(inode); int ret = 0; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); dout("set_size %p %llu -> %llu\n", inode, inode->i_size, size); inode->i_size = size; inode->i_blocks = (size + (1 << 9) - 1) >> 9; @@ -1318,7 +1321,7 @@ int ceph_inode_set_size(struct inode *inode, loff_t size) (ci->i_reported_size << 1) < ci->i_max_size) ret = 1; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return ret; } @@ -1328,12 +1331,13 @@ int ceph_inode_set_size(struct inode *inode, loff_t size) */ void ceph_queue_writeback(struct inode *inode) { + ihold(inode); if (queue_work(ceph_inode_to_client(inode)->wb_wq, &ceph_inode(inode)->i_wb_work)) { dout("ceph_queue_writeback %p\n", inode); - ihold(inode); } else { dout("ceph_queue_writeback %p failed\n", inode); + iput(inode); } } @@ -1353,12 +1357,13 @@ static void ceph_writeback_work(struct work_struct *work) */ void ceph_queue_invalidate(struct inode *inode) { + ihold(inode); if (queue_work(ceph_inode_to_client(inode)->pg_inv_wq, &ceph_inode(inode)->i_pg_inv_work)) { dout("ceph_queue_invalidate %p\n", inode); - ihold(inode); } else { dout("ceph_queue_invalidate %p failed\n", inode); + iput(inode); } } @@ -1374,20 +1379,20 @@ static void ceph_invalidate_work(struct work_struct *work) u32 orig_gen; int check = 0; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); dout("invalidate_pages %p gen %d revoking %d\n", inode, ci->i_rdcache_gen, ci->i_rdcache_revoking); if (ci->i_rdcache_revoking != ci->i_rdcache_gen) { /* nevermind! */ - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); goto out; } orig_gen = ci->i_rdcache_gen; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); truncate_inode_pages(&inode->i_data, 0); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (orig_gen == ci->i_rdcache_gen && orig_gen == ci->i_rdcache_revoking) { dout("invalidate_pages %p gen %d successful\n", inode, @@ -1399,7 +1404,7 @@ static void ceph_invalidate_work(struct work_struct *work) inode, orig_gen, ci->i_rdcache_gen, ci->i_rdcache_revoking); } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (check) ceph_check_caps(ci, 0, NULL); @@ -1434,13 +1439,14 @@ void ceph_queue_vmtruncate(struct inode *inode) { struct ceph_inode_info *ci = ceph_inode(inode); + ihold(inode); if (queue_work(ceph_sb_to_client(inode->i_sb)->trunc_wq, &ci->i_vmtruncate_work)) { dout("ceph_queue_vmtruncate %p\n", inode); - ihold(inode); } else { dout("ceph_queue_vmtruncate %p failed, pending=%d\n", inode, ci->i_truncate_pending); + iput(inode); } } @@ -1457,10 +1463,10 @@ void __ceph_do_pending_vmtruncate(struct inode *inode) int wrbuffer_refs, wake = 0; retry: - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_truncate_pending == 0) { dout("__do_pending_vmtruncate %p none pending\n", inode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return; } @@ -1471,7 +1477,7 @@ retry: if (ci->i_wrbuffer_ref_head < ci->i_wrbuffer_ref) { dout("__do_pending_vmtruncate %p flushing snaps first\n", inode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); filemap_write_and_wait_range(&inode->i_data, 0, inode->i_sb->s_maxbytes); goto retry; @@ -1481,15 +1487,15 @@ retry: wrbuffer_refs = ci->i_wrbuffer_ref; dout("__do_pending_vmtruncate %p (%d) to %lld\n", inode, ci->i_truncate_pending, to); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); truncate_inode_pages(inode->i_mapping, to); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); ci->i_truncate_pending--; if (ci->i_truncate_pending == 0) wake = 1; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (wrbuffer_refs == 0) ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL); @@ -1544,7 +1550,7 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr) if (IS_ERR(req)) return PTR_ERR(req); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); issued = __ceph_caps_issued(ci, NULL); dout("setattr %p issued %s\n", inode, ceph_cap_string(issued)); @@ -1692,7 +1698,7 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr) } release &= issued; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (inode_dirty_flags) __mark_inode_dirty(inode, inode_dirty_flags); @@ -1714,7 +1720,7 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr) __ceph_do_pending_vmtruncate(inode); return err; out: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); ceph_mdsc_put_request(req); return err; } diff --git a/fs/ceph/ioctl.c b/fs/ceph/ioctl.c index 5a14c29cbba6..790914a598dd 100644 --- a/fs/ceph/ioctl.c +++ b/fs/ceph/ioctl.c @@ -241,11 +241,11 @@ static long ceph_ioctl_lazyio(struct file *file) struct ceph_inode_info *ci = ceph_inode(inode); if ((fi->fmode & CEPH_FILE_MODE_LAZY) == 0) { - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); ci->i_nr_by_mode[fi->fmode]--; fi->fmode |= CEPH_FILE_MODE_LAZY; ci->i_nr_by_mode[fi->fmode]++; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); dout("ioctl_layzio: file %p marked lazy\n", file); ceph_check_caps(ci, 0, NULL); diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 264ab701154f..23ab6a3f1825 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -732,21 +732,21 @@ static int __choose_mds(struct ceph_mds_client *mdsc, } } - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); cap = NULL; if (mode == USE_AUTH_MDS) cap = ci->i_auth_cap; if (!cap && !RB_EMPTY_ROOT(&ci->i_caps)) cap = rb_entry(rb_first(&ci->i_caps), struct ceph_cap, ci_node); if (!cap) { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); goto random; } mds = cap->session->s_mds; dout("choose_mds %p %llx.%llx mds%d (%scap %p)\n", inode, ceph_vinop(inode), mds, cap == ci->i_auth_cap ? "auth " : "", cap); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return mds; random: @@ -951,7 +951,7 @@ static int remove_session_caps_cb(struct inode *inode, struct ceph_cap *cap, dout("removing cap %p, ci is %p, inode is %p\n", cap, ci, &ci->vfs_inode); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); __ceph_remove_cap(cap); if (!__ceph_is_any_real_caps(ci)) { struct ceph_mds_client *mdsc = @@ -984,7 +984,7 @@ static int remove_session_caps_cb(struct inode *inode, struct ceph_cap *cap, } spin_unlock(&mdsc->cap_dirty_lock); } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); while (drop--) iput(inode); return 0; @@ -1015,10 +1015,10 @@ static int wake_up_session_cb(struct inode *inode, struct ceph_cap *cap, wake_up_all(&ci->i_cap_wq); if (arg) { - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); ci->i_wanted_max_size = 0; ci->i_requested_max_size = 0; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } return 0; } @@ -1151,7 +1151,7 @@ static int trim_caps_cb(struct inode *inode, struct ceph_cap *cap, void *arg) if (session->s_trim_caps <= 0) return -1; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); mine = cap->issued | cap->implemented; used = __ceph_caps_used(ci); oissued = __ceph_caps_issued_other(ci, cap); @@ -1170,7 +1170,7 @@ static int trim_caps_cb(struct inode *inode, struct ceph_cap *cap, void *arg) __ceph_remove_cap(cap); } else { /* try to drop referring dentries */ - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); d_prune_aliases(inode); dout("trim_caps_cb %p cap %p pruned, count now %d\n", inode, cap, atomic_read(&inode->i_count)); @@ -1178,7 +1178,7 @@ static int trim_caps_cb(struct inode *inode, struct ceph_cap *cap, void *arg) } out: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return 0; } @@ -1296,7 +1296,7 @@ static int check_cap_flush(struct ceph_mds_client *mdsc, u64 want_flush_seq) i_flushing_item); struct inode *inode = &ci->vfs_inode; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_cap_flush_seq <= want_flush_seq) { dout("check_cap_flush still flushing %p " "seq %lld <= %lld to mds%d\n", inode, @@ -1304,7 +1304,7 @@ static int check_cap_flush(struct ceph_mds_client *mdsc, u64 want_flush_seq) session->s_mds); ret = 0; } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } mutex_unlock(&session->s_mutex); ceph_put_mds_session(session); @@ -1495,6 +1495,7 @@ retry: pos, temp); } else if (stop_on_nosnap && inode && ceph_snap(inode) == CEPH_NOSNAP) { + spin_unlock(&temp->d_lock); break; } else { pos -= temp->d_name.len; @@ -2011,10 +2012,10 @@ void ceph_invalidate_dir_request(struct ceph_mds_request *req) struct ceph_inode_info *ci = ceph_inode(inode); dout("invalidate_dir_request %p (D_COMPLETE, lease(s))\n", inode); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); ceph_dir_clear_complete(inode); ci->i_release_count++; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (req->r_dentry) ceph_invalidate_dentry_lease(req->r_dentry); @@ -2422,7 +2423,7 @@ static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap, if (err) goto out_free; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); cap->seq = 0; /* reset cap seq */ cap->issue_seq = 0; /* and issue_seq */ @@ -2445,7 +2446,7 @@ static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap, rec.v1.pathbase = cpu_to_le64(pathbase); reclen = sizeof(rec.v1); } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (recon_state->flock) { int num_fcntl_locks, num_flock_locks; @@ -2771,7 +2772,7 @@ static void handle_lease(struct ceph_mds_client *mdsc, di = ceph_dentry(dentry); switch (h->action) { case CEPH_MDS_LEASE_REVOKE: - if (di && di->lease_session == session) { + if (di->lease_session == session) { if (ceph_seq_cmp(di->lease_seq, seq) > 0) h->seq = cpu_to_le32(di->lease_seq); __ceph_mdsc_drop_dentry_lease(dentry); @@ -2780,7 +2781,7 @@ static void handle_lease(struct ceph_mds_client *mdsc, break; case CEPH_MDS_LEASE_RENEW: - if (di && di->lease_session == session && + if (di->lease_session == session && di->lease_gen == session->s_cap_gen && di->lease_renew_from && di->lease_renew_after == 0) { diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index 4bb239921dbd..a50ca0e39475 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -20,7 +20,7 @@ * * mdsc->snap_rwsem * - * inode->i_lock + * ci->i_ceph_lock * mdsc->snap_flush_lock * mdsc->cap_delay_lock * diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c index e26437191333..a559c80f127a 100644 --- a/fs/ceph/snap.c +++ b/fs/ceph/snap.c @@ -446,7 +446,7 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci) return; } - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); used = __ceph_caps_used(ci); dirty = __ceph_caps_dirty(ci); @@ -528,7 +528,7 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci) kfree(capsnap); } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } /* @@ -537,7 +537,7 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci) * * If capsnap can now be flushed, add to snap_flush list, and return 1. * - * Caller must hold i_lock. + * Caller must hold i_ceph_lock. */ int __ceph_finish_cap_snap(struct ceph_inode_info *ci, struct ceph_cap_snap *capsnap) @@ -739,9 +739,9 @@ static void flush_snaps(struct ceph_mds_client *mdsc) inode = &ci->vfs_inode; ihold(inode); spin_unlock(&mdsc->snap_flush_lock); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); __ceph_flush_snaps(ci, &session, 0); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); iput(inode); spin_lock(&mdsc->snap_flush_lock); } @@ -847,7 +847,7 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc, continue; ci = ceph_inode(inode); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (!ci->i_snap_realm) goto skip_inode; /* @@ -876,7 +876,7 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc, oldrealm = ci->i_snap_realm; ci->i_snap_realm = realm; spin_unlock(&realm->inodes_with_caps_lock); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); ceph_get_snap_realm(mdsc, realm); ceph_put_snap_realm(mdsc, oldrealm); @@ -885,7 +885,7 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc, continue; skip_inode: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); iput(inode); } diff --git a/fs/ceph/super.c b/fs/ceph/super.c index a90846fac759..00de2c9568cd 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -131,6 +131,8 @@ enum { Opt_rbytes, Opt_norbytes, Opt_noasyncreaddir, + Opt_dcache, + Opt_nodcache, Opt_ino32, }; @@ -152,6 +154,8 @@ static match_table_t fsopt_tokens = { {Opt_rbytes, "rbytes"}, {Opt_norbytes, "norbytes"}, {Opt_noasyncreaddir, "noasyncreaddir"}, + {Opt_dcache, "dcache"}, + {Opt_nodcache, "nodcache"}, {Opt_ino32, "ino32"}, {-1, NULL} }; @@ -231,6 +235,12 @@ static int parse_fsopt_token(char *c, void *private) case Opt_noasyncreaddir: fsopt->flags |= CEPH_MOUNT_OPT_NOASYNCREADDIR; break; + case Opt_dcache: + fsopt->flags |= CEPH_MOUNT_OPT_DCACHE; + break; + case Opt_nodcache: + fsopt->flags &= ~CEPH_MOUNT_OPT_DCACHE; + break; case Opt_ino32: fsopt->flags |= CEPH_MOUNT_OPT_INO32; break; @@ -341,11 +351,11 @@ out: /** * ceph_show_options - Show mount options in /proc/mounts * @m: seq_file to write to - * @mnt: mount descriptor + * @root: root of that (sub)tree */ -static int ceph_show_options(struct seq_file *m, struct vfsmount *mnt) +static int ceph_show_options(struct seq_file *m, struct dentry *root) { - struct ceph_fs_client *fsc = ceph_sb_to_client(mnt->mnt_sb); + struct ceph_fs_client *fsc = ceph_sb_to_client(root->d_sb); struct ceph_mount_options *fsopt = fsc->mount_options; struct ceph_options *opt = fsc->client->options; @@ -377,13 +387,17 @@ static int ceph_show_options(struct seq_file *m, struct vfsmount *mnt) seq_puts(m, ",norbytes"); if (fsopt->flags & CEPH_MOUNT_OPT_NOASYNCREADDIR) seq_puts(m, ",noasyncreaddir"); + if (fsopt->flags & CEPH_MOUNT_OPT_DCACHE) + seq_puts(m, ",dcache"); + else + seq_puts(m, ",nodcache"); if (fsopt->wsize) seq_printf(m, ",wsize=%d", fsopt->wsize); if (fsopt->rsize != CEPH_RSIZE_DEFAULT) seq_printf(m, ",rsize=%d", fsopt->rsize); if (fsopt->rasize != CEPH_RASIZE_DEFAULT) - seq_printf(m, ",rasize=%d", fsopt->rsize); + seq_printf(m, ",rasize=%d", fsopt->rasize); if (fsopt->congestion_kb != default_congestion_kb()) seq_printf(m, ",write_congestion_kb=%d", fsopt->congestion_kb); if (fsopt->caps_wanted_delay_min != CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT) @@ -636,17 +650,26 @@ static struct dentry *open_root_dentry(struct ceph_fs_client *fsc, req->r_num_caps = 2; err = ceph_mdsc_do_request(mdsc, NULL, req); if (err == 0) { - dout("open_root_inode success\n"); - if (ceph_ino(req->r_target_inode) == CEPH_INO_ROOT && - fsc->sb->s_root == NULL) - root = d_alloc_root(req->r_target_inode); - else - root = d_obtain_alias(req->r_target_inode); + struct inode *inode = req->r_target_inode; req->r_target_inode = NULL; + dout("open_root_inode success\n"); + if (ceph_ino(inode) == CEPH_INO_ROOT && + fsc->sb->s_root == NULL) { + root = d_alloc_root(inode); + if (!root) { + iput(inode); + root = ERR_PTR(-ENOMEM); + goto out; + } + } else { + root = d_obtain_alias(inode); + } + ceph_init_dentry(root); dout("open_root_inode success, root dentry is %p\n", root); } else { root = ERR_PTR(err); } +out: ceph_mdsc_put_request(req); return root; } diff --git a/fs/ceph/super.h b/fs/ceph/super.h index 01bf189e08a9..1421f3d875a2 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -28,6 +28,7 @@ #define CEPH_MOUNT_OPT_RBYTES (1<<5) /* dir st_bytes = rbytes */ #define CEPH_MOUNT_OPT_NOASYNCREADDIR (1<<7) /* no dcache readdir */ #define CEPH_MOUNT_OPT_INO32 (1<<8) /* 32 bit inos */ +#define CEPH_MOUNT_OPT_DCACHE (1<<9) /* use dcache for readdir etc */ #define CEPH_MOUNT_OPT_DEFAULT (CEPH_MOUNT_OPT_RBYTES) @@ -136,7 +137,7 @@ struct ceph_cap_snap { int issued, dirty; struct ceph_snap_context *context; - mode_t mode; + umode_t mode; uid_t uid; gid_t gid; @@ -220,7 +221,7 @@ struct ceph_dentry_info { * The locking for D_COMPLETE is a bit odd: * - we can clear it at almost any time (see ceph_d_prune) * - it is only meaningful if: - * - we hold dir inode i_lock + * - we hold dir inode i_ceph_lock * - we hold dir FILE_SHARED caps * - the dentry D_COMPLETE is set */ @@ -250,6 +251,8 @@ struct ceph_inode_xattrs_info { struct ceph_inode_info { struct ceph_vino i_vino; /* ceph ino + snap */ + spinlock_t i_ceph_lock; + u64 i_version; u32 i_time_warp_seq; @@ -271,7 +274,7 @@ struct ceph_inode_info { struct ceph_inode_xattrs_info i_xattrs; - /* capabilities. protected _both_ by i_lock and cap->session's + /* capabilities. protected _both_ by i_ceph_lock and cap->session's * s_mutex. */ struct rb_root i_caps; /* cap list */ struct ceph_cap *i_auth_cap; /* authoritative cap, if any */ @@ -437,18 +440,18 @@ static inline void ceph_i_clear(struct inode *inode, unsigned mask) { struct ceph_inode_info *ci = ceph_inode(inode); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); ci->i_ceph_flags &= ~mask; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } static inline void ceph_i_set(struct inode *inode, unsigned mask) { struct ceph_inode_info *ci = ceph_inode(inode); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); ci->i_ceph_flags |= mask; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } static inline bool ceph_i_test(struct inode *inode, unsigned mask) @@ -456,9 +459,9 @@ static inline bool ceph_i_test(struct inode *inode, unsigned mask) struct ceph_inode_info *ci = ceph_inode(inode); bool r; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); r = (ci->i_ceph_flags & mask) == mask; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return r; } @@ -508,9 +511,9 @@ extern int __ceph_caps_issued_other(struct ceph_inode_info *ci, static inline int ceph_caps_issued(struct ceph_inode_info *ci) { int issued; - spin_lock(&ci->vfs_inode.i_lock); + spin_lock(&ci->i_ceph_lock); issued = __ceph_caps_issued(ci, NULL); - spin_unlock(&ci->vfs_inode.i_lock); + spin_unlock(&ci->i_ceph_lock); return issued; } @@ -518,9 +521,9 @@ static inline int ceph_caps_issued_mask(struct ceph_inode_info *ci, int mask, int touch) { int r; - spin_lock(&ci->vfs_inode.i_lock); + spin_lock(&ci->i_ceph_lock); r = __ceph_caps_issued_mask(ci, mask, touch); - spin_unlock(&ci->vfs_inode.i_lock); + spin_unlock(&ci->i_ceph_lock); return r; } @@ -743,10 +746,9 @@ extern int ceph_add_cap(struct inode *inode, extern void __ceph_remove_cap(struct ceph_cap *cap); static inline void ceph_remove_cap(struct ceph_cap *cap) { - struct inode *inode = &cap->ci->vfs_inode; - spin_lock(&inode->i_lock); + spin_lock(&cap->ci->i_ceph_lock); __ceph_remove_cap(cap); - spin_unlock(&inode->i_lock); + spin_unlock(&cap->ci->i_ceph_lock); } extern void ceph_put_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap); diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c index 96c6739a0280..857214ae8c08 100644 --- a/fs/ceph/xattr.c +++ b/fs/ceph/xattr.c @@ -343,8 +343,8 @@ void __ceph_destroy_xattrs(struct ceph_inode_info *ci) } static int __build_xattrs(struct inode *inode) - __releases(inode->i_lock) - __acquires(inode->i_lock) + __releases(ci->i_ceph_lock) + __acquires(ci->i_ceph_lock) { u32 namelen; u32 numattr = 0; @@ -372,7 +372,7 @@ start: end = p + ci->i_xattrs.blob->vec.iov_len; ceph_decode_32_safe(&p, end, numattr, bad); xattr_version = ci->i_xattrs.version; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); xattrs = kcalloc(numattr, sizeof(struct ceph_xattr *), GFP_NOFS); @@ -387,7 +387,7 @@ start: goto bad_lock; } - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_xattrs.version != xattr_version) { /* lost a race, retry */ for (i = 0; i < numattr; i++) @@ -418,7 +418,7 @@ start: return err; bad_lock: - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); bad: if (xattrs) { for (i = 0; i < numattr; i++) @@ -512,7 +512,7 @@ ssize_t ceph_getxattr(struct dentry *dentry, const char *name, void *value, if (vxattrs) vxattr = ceph_match_vxattr(vxattrs, name); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); dout("getxattr %p ver=%lld index_ver=%lld\n", inode, ci->i_xattrs.version, ci->i_xattrs.index_version); @@ -520,14 +520,14 @@ ssize_t ceph_getxattr(struct dentry *dentry, const char *name, void *value, (ci->i_xattrs.index_version >= ci->i_xattrs.version)) { goto get_xattr; } else { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); /* get xattrs from mds (if we don't already have them) */ err = ceph_do_getattr(inode, CEPH_STAT_CAP_XATTR); if (err) return err; } - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (vxattr && vxattr->readonly) { err = vxattr->getxattr_cb(ci, value, size); @@ -558,7 +558,7 @@ get_xattr: memcpy(value, xattr->val, xattr->val_len); out: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return err; } @@ -573,7 +573,7 @@ ssize_t ceph_listxattr(struct dentry *dentry, char *names, size_t size) u32 len; int i; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); dout("listxattr %p ver=%lld index_ver=%lld\n", inode, ci->i_xattrs.version, ci->i_xattrs.index_version); @@ -581,13 +581,13 @@ ssize_t ceph_listxattr(struct dentry *dentry, char *names, size_t size) (ci->i_xattrs.index_version >= ci->i_xattrs.version)) { goto list_xattr; } else { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); err = ceph_do_getattr(inode, CEPH_STAT_CAP_XATTR); if (err) return err; } - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); err = __build_xattrs(inode); if (err < 0) @@ -619,7 +619,7 @@ list_xattr: } out: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return err; } @@ -739,7 +739,7 @@ int ceph_setxattr(struct dentry *dentry, const char *name, if (!xattr) goto out; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); retry: issued = __ceph_caps_issued(ci, NULL); if (!(issued & CEPH_CAP_XATTR_EXCL)) @@ -752,12 +752,12 @@ retry: required_blob_size > ci->i_xattrs.prealloc_blob->alloc_len) { struct ceph_buffer *blob = NULL; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); dout(" preaallocating new blob size=%d\n", required_blob_size); blob = ceph_buffer_new(required_blob_size, GFP_NOFS); if (!blob) goto out; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_xattrs.prealloc_blob) ceph_buffer_put(ci->i_xattrs.prealloc_blob); ci->i_xattrs.prealloc_blob = blob; @@ -770,13 +770,13 @@ retry: dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL); ci->i_xattrs.dirty = true; inode->i_ctime = CURRENT_TIME; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (dirty) __mark_inode_dirty(inode, dirty); return err; do_sync: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); err = ceph_sync_setxattr(dentry, name, value, size, flags); out: kfree(newname); @@ -818,6 +818,7 @@ int ceph_removexattr(struct dentry *dentry, const char *name) struct ceph_vxattr_cb *vxattrs = ceph_inode_vxattrs(inode); int issued; int err; + int required_blob_size; int dirty; if (ceph_snap(inode) != CEPH_NOSNAP) @@ -833,26 +834,47 @@ int ceph_removexattr(struct dentry *dentry, const char *name) return -EOPNOTSUPP; } - spin_lock(&inode->i_lock); + err = -ENOMEM; + spin_lock(&ci->i_ceph_lock); __build_xattrs(inode); +retry: issued = __ceph_caps_issued(ci, NULL); dout("removexattr %p issued %s\n", inode, ceph_cap_string(issued)); if (!(issued & CEPH_CAP_XATTR_EXCL)) goto do_sync; + required_blob_size = __get_required_blob_size(ci, 0, 0); + + if (!ci->i_xattrs.prealloc_blob || + required_blob_size > ci->i_xattrs.prealloc_blob->alloc_len) { + struct ceph_buffer *blob; + + spin_unlock(&ci->i_ceph_lock); + dout(" preaallocating new blob size=%d\n", required_blob_size); + blob = ceph_buffer_new(required_blob_size, GFP_NOFS); + if (!blob) + goto out; + spin_lock(&ci->i_ceph_lock); + if (ci->i_xattrs.prealloc_blob) + ceph_buffer_put(ci->i_xattrs.prealloc_blob); + ci->i_xattrs.prealloc_blob = blob; + goto retry; + } + err = __remove_xattr_by_name(ceph_inode(inode), name); dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL); ci->i_xattrs.dirty = true; inode->i_ctime = CURRENT_TIME; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (dirty) __mark_inode_dirty(inode, dirty); return err; do_sync: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); err = ceph_send_removexattr(dentry, name); +out: return err; } diff --git a/fs/char_dev.c b/fs/char_dev.c index dca9e5e0f73b..3f152b92a94a 100644 --- a/fs/char_dev.c +++ b/fs/char_dev.c @@ -272,7 +272,7 @@ int __register_chrdev(unsigned int major, unsigned int baseminor, cd = __register_chrdev_region(major, baseminor, count, name); if (IS_ERR(cd)) return PTR_ERR(cd); - + cdev = cdev_alloc(); if (!cdev) goto out2; @@ -280,7 +280,7 @@ int __register_chrdev(unsigned int major, unsigned int baseminor, cdev->owner = fops->owner; cdev->ops = fops; kobject_set_name(&cdev->kobj, "%s", name); - + err = cdev_add(cdev, MKDEV(cd->major, baseminor), count); if (err) goto out; @@ -405,7 +405,7 @@ static int chrdev_open(struct inode *inode, struct file *filp) goto out_cdev_put; if (filp->f_op->open) { - ret = filp->f_op->open(inode,filp); + ret = filp->f_op->open(inode, filp); if (ret) goto out_cdev_put; } diff --git a/fs/cifs/cifs_fs_sb.h b/fs/cifs/cifs_fs_sb.h index 500d65859279..c865bfdfe819 100644 --- a/fs/cifs/cifs_fs_sb.h +++ b/fs/cifs/cifs_fs_sb.h @@ -59,8 +59,8 @@ struct cifs_sb_info { gid_t mnt_gid; uid_t mnt_backupuid; gid_t mnt_backupgid; - mode_t mnt_file_mode; - mode_t mnt_dir_mode; + umode_t mnt_file_mode; + umode_t mnt_dir_mode; unsigned int mnt_cifs_flags; char *mountdata; /* options received at mount time or via DFS refs */ struct backing_dev_info bdi; diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 8f1fe324162b..b1fd382d1952 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -343,9 +343,9 @@ cifs_show_security(struct seq_file *s, struct TCP_Server_Info *server) * ones are. */ static int -cifs_show_options(struct seq_file *s, struct vfsmount *m) +cifs_show_options(struct seq_file *s, struct dentry *root) { - struct cifs_sb_info *cifs_sb = CIFS_SB(m->mnt_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(root->d_sb); struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); struct sockaddr *srcaddr; srcaddr = (struct sockaddr *)&tcon->ses->server->srcaddr; @@ -393,7 +393,7 @@ cifs_show_options(struct seq_file *s, struct vfsmount *m) cifs_show_address(s, tcon->ses->server); if (!tcon->unix_ext) - seq_printf(s, ",file_mode=0%o,dir_mode=0%o", + seq_printf(s, ",file_mode=0%ho,dir_mode=0%ho", cifs_sb->mnt_file_mode, cifs_sb->mnt_dir_mode); if (tcon->seal) @@ -430,7 +430,7 @@ cifs_show_options(struct seq_file *s, struct vfsmount *m) seq_printf(s, ",cifsacl"); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DYNPERM) seq_printf(s, ",dynperm"); - if (m->mnt_sb->s_flags & MS_POSIXACL) + if (root->d_sb->s_flags & MS_POSIXACL) seq_printf(s, ",acl"); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MF_SYMLINKS) seq_printf(s, ",mfsymlinks"); @@ -488,7 +488,7 @@ static void cifs_umount_begin(struct super_block *sb) } #ifdef CONFIG_CIFS_STATS2 -static int cifs_show_stats(struct seq_file *s, struct vfsmount *mnt) +static int cifs_show_stats(struct seq_file *s, struct dentry *root) { /* BB FIXME */ return 0; diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 30ff56005d8f..fe5ecf1b422a 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -44,14 +44,14 @@ extern const struct address_space_operations cifs_addr_ops_smallbuf; /* Functions related to inodes */ extern const struct inode_operations cifs_dir_inode_ops; extern struct inode *cifs_root_iget(struct super_block *); -extern int cifs_create(struct inode *, struct dentry *, int, +extern int cifs_create(struct inode *, struct dentry *, umode_t, struct nameidata *); extern struct dentry *cifs_lookup(struct inode *, struct dentry *, struct nameidata *); extern int cifs_unlink(struct inode *dir, struct dentry *dentry); extern int cifs_hardlink(struct dentry *, struct inode *, struct dentry *); -extern int cifs_mknod(struct inode *, struct dentry *, int, dev_t); -extern int cifs_mkdir(struct inode *, struct dentry *, int); +extern int cifs_mknod(struct inode *, struct dentry *, umode_t, dev_t); +extern int cifs_mkdir(struct inode *, struct dentry *, umode_t); extern int cifs_rmdir(struct inode *, struct dentry *); extern int cifs_rename(struct inode *, struct dentry *, struct inode *, struct dentry *); diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 8238aa13e01c..ba53c1c6c6cc 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -169,8 +169,8 @@ struct smb_vol { gid_t linux_gid; uid_t backupuid; gid_t backupgid; - mode_t file_mode; - mode_t dir_mode; + umode_t file_mode; + umode_t dir_mode; unsigned secFlg; bool retry:1; bool intr:1; diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index d6a972df0338..4666780f315d 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -282,7 +282,7 @@ static int coalesce_t2(struct smb_hdr *psecond, struct smb_hdr *pTargetSMB) byte_count = be32_to_cpu(pTargetSMB->smb_buf_length); byte_count += total_in_buf2; /* don't allow buffer to overflow */ - if (byte_count > CIFSMaxBufSize) + if (byte_count > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) return -ENOBUFS; pTargetSMB->smb_buf_length = cpu_to_be32(byte_count); @@ -441,6 +441,8 @@ cifs_readv_from_socket(struct TCP_Server_Info *server, struct kvec *iov_orig, smb_msg.msg_controllen = 0; for (total_read = 0; to_read; total_read += length, to_read -= length) { + try_to_freeze(); + if (server_unresponsive(server)) { total_read = -EAGAIN; break; @@ -2120,7 +2122,7 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb_vol *volume_info) warned_on_ntlm = true; cERROR(1, "default security mechanism requested. The default " "security mechanism will be upgraded from ntlm to " - "ntlmv2 in kernel release 3.2"); + "ntlmv2 in kernel release 3.3"); } ses->overrideSecFlg = volume_info->secFlg; @@ -2817,7 +2819,7 @@ void cifs_setup_cifs_sb(struct smb_vol *pvolume_info, cifs_sb->mnt_backupgid = pvolume_info->backupgid; cifs_sb->mnt_file_mode = pvolume_info->file_mode; cifs_sb->mnt_dir_mode = pvolume_info->dir_mode; - cFYI(1, "file mode: 0x%x dir mode: 0x%x", + cFYI(1, "file mode: 0x%hx dir mode: 0x%hx", cifs_sb->mnt_file_mode, cifs_sb->mnt_dir_mode); cifs_sb->actimeo = pvolume_info->actimeo; diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index d7eeb9d3ed6f..df8fecb5b993 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -136,7 +136,7 @@ cifs_bp_rename_retry: /* Inode operations in similar order to how they appear in Linux file fs.h */ int -cifs_create(struct inode *inode, struct dentry *direntry, int mode, +cifs_create(struct inode *inode, struct dentry *direntry, umode_t mode, struct nameidata *nd) { int rc = -ENOENT; @@ -355,7 +355,7 @@ cifs_create_out: return rc; } -int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode, +int cifs_mknod(struct inode *inode, struct dentry *direntry, umode_t mode, dev_t device_number) { int rc = -EPERM; diff --git a/fs/cifs/file.c b/fs/cifs/file.c index cf0b1539b321..4dd9283885e7 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -702,6 +702,13 @@ cifs_find_lock_conflict(struct cifsInodeInfo *cinode, struct cifsLockInfo *lock, lock->type, lock->netfid, conf_lock); } +/* + * Check if there is another lock that prevents us to set the lock (mandatory + * style). If such a lock exists, update the flock structure with its + * properties. Otherwise, set the flock type to F_UNLCK if we can cache brlocks + * or leave it the same if we can't. Returns 0 if we don't need to request to + * the server or 1 otherwise. + */ static int cifs_lock_test(struct cifsInodeInfo *cinode, __u64 offset, __u64 length, __u8 type, __u16 netfid, struct file_lock *flock) @@ -739,6 +746,12 @@ cifs_lock_add(struct cifsInodeInfo *cinode, struct cifsLockInfo *lock) mutex_unlock(&cinode->lock_mutex); } +/* + * Set the byte-range lock (mandatory style). Returns: + * 1) 0, if we set the lock and don't need to request to the server; + * 2) 1, if no locks prevent us but we need to request to the server; + * 3) -EACCESS, if there is a lock that prevents us and wait is false. + */ static int cifs_lock_add_if(struct cifsInodeInfo *cinode, struct cifsLockInfo *lock, bool wait) @@ -778,6 +791,13 @@ try_again: return rc; } +/* + * Check if there is another lock that prevents us to set the lock (posix + * style). If such a lock exists, update the flock structure with its + * properties. Otherwise, set the flock type to F_UNLCK if we can cache brlocks + * or leave it the same if we can't. Returns 0 if we don't need to request to + * the server or 1 otherwise. + */ static int cifs_posix_lock_test(struct file *file, struct file_lock *flock) { @@ -800,6 +820,12 @@ cifs_posix_lock_test(struct file *file, struct file_lock *flock) return rc; } +/* + * Set the byte-range lock (posix style). Returns: + * 1) 0, if we set the lock and don't need to request to the server; + * 2) 1, if we need to request to the server; + * 3) <0, if the error occurs while setting the lock. + */ static int cifs_posix_lock_set(struct file *file, struct file_lock *flock) { diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index e851d5b8931e..a5f54b7d9822 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -1264,7 +1264,7 @@ unlink_out: return rc; } -int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) +int cifs_mkdir(struct inode *inode, struct dentry *direntry, umode_t mode) { int rc = 0, tmprc; int xid; @@ -1275,7 +1275,7 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) struct inode *newinode = NULL; struct cifs_fattr fattr; - cFYI(1, "In cifs_mkdir, mode = 0x%x inode = 0x%p", mode, inode); + cFYI(1, "In cifs_mkdir, mode = 0x%hx inode = 0x%p", mode, inode); cifs_sb = CIFS_SB(inode->i_sb); tlink = cifs_sb_tlink(cifs_sb); diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 5de03ec20144..a090bbe6ee29 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -554,7 +554,10 @@ static int find_cifs_entry(const int xid, struct cifs_tcon *pTcon, rc); return rc; } - cifs_save_resume_key(cifsFile->srch_inf.last_entry, cifsFile); + /* FindFirst/Next set last_entry to NULL on malformed reply */ + if (cifsFile->srch_inf.last_entry) + cifs_save_resume_key(cifsFile->srch_inf.last_entry, + cifsFile); } while ((index_to_find >= cifsFile->srch_inf.index_of_last_entry) && @@ -562,7 +565,10 @@ static int find_cifs_entry(const int xid, struct cifs_tcon *pTcon, cFYI(1, "calling findnext2"); rc = CIFSFindNext(xid, pTcon, cifsFile->netfid, &cifsFile->srch_inf); - cifs_save_resume_key(cifsFile->srch_inf.last_entry, cifsFile); + /* FindFirst/Next set last_entry to NULL on malformed reply */ + if (cifsFile->srch_inf.last_entry) + cifs_save_resume_key(cifsFile->srch_inf.last_entry, + cifsFile); if (rc) return -ENOENT; } diff --git a/fs/cifs/smbencrypt.c b/fs/cifs/smbencrypt.c index 7cacba12b8f1..80d850881938 100644 --- a/fs/cifs/smbencrypt.c +++ b/fs/cifs/smbencrypt.c @@ -209,7 +209,7 @@ E_md4hash(const unsigned char *passwd, unsigned char *p16, { int rc; int len; - __u16 wpwd[129]; + __le16 wpwd[129]; /* Password cannot be longer than 128 characters */ if (passwd) /* Password must be converted to NT unicode */ @@ -219,8 +219,8 @@ E_md4hash(const unsigned char *passwd, unsigned char *p16, *wpwd = 0; /* Ensure string is null terminated */ } - rc = mdfour(p16, (unsigned char *) wpwd, len * sizeof(__u16)); - memset(wpwd, 0, 129 * sizeof(__u16)); + rc = mdfour(p16, (unsigned char *) wpwd, len * sizeof(__le16)); + memset(wpwd, 0, 129 * sizeof(__le16)); return rc; } diff --git a/fs/coda/cnode.c b/fs/coda/cnode.c index 6475877b0763..911cf30d057d 100644 --- a/fs/coda/cnode.c +++ b/fs/coda/cnode.c @@ -88,24 +88,21 @@ struct inode * coda_iget(struct super_block * sb, struct CodaFid * fid, - link the two up if this is needed - fill in the attributes */ -int coda_cnode_make(struct inode **inode, struct CodaFid *fid, struct super_block *sb) +struct inode *coda_cnode_make(struct CodaFid *fid, struct super_block *sb) { struct coda_vattr attr; + struct inode *inode; int error; /* We get inode numbers from Venus -- see venus source */ error = venus_getattr(sb, fid, &attr); - if ( error ) { - *inode = NULL; - return error; - } + if (error) + return ERR_PTR(error); - *inode = coda_iget(sb, fid, &attr); - if ( IS_ERR(*inode) ) { + inode = coda_iget(sb, fid, &attr); + if (IS_ERR(inode)) printk("coda_cnode_make: coda_iget failed\n"); - return PTR_ERR(*inode); - } - return 0; + return inode; } @@ -156,19 +153,16 @@ struct inode *coda_fid_to_inode(struct CodaFid *fid, struct super_block *sb) } /* the CONTROL inode is made without asking attributes from Venus */ -int coda_cnode_makectl(struct inode **inode, struct super_block *sb) +struct inode *coda_cnode_makectl(struct super_block *sb) { - int error = -ENOMEM; - - *inode = new_inode(sb); - if (*inode) { - (*inode)->i_ino = CTL_INO; - (*inode)->i_op = &coda_ioctl_inode_operations; - (*inode)->i_fop = &coda_ioctl_operations; - (*inode)->i_mode = 0444; - error = 0; + struct inode *inode = new_inode(sb); + if (inode) { + inode->i_ino = CTL_INO; + inode->i_op = &coda_ioctl_inode_operations; + inode->i_fop = &coda_ioctl_operations; + inode->i_mode = 0444; + return inode; } - - return error; + return ERR_PTR(-ENOMEM); } diff --git a/fs/coda/coda_fs_i.h b/fs/coda/coda_fs_i.h index e35071b1de0e..b24fdfd8a3f0 100644 --- a/fs/coda/coda_fs_i.h +++ b/fs/coda/coda_fs_i.h @@ -49,9 +49,9 @@ struct coda_file_info { #define C_DYING 0x4 /* from venus (which died) */ #define C_PURGE 0x8 -int coda_cnode_make(struct inode **, struct CodaFid *, struct super_block *); +struct inode *coda_cnode_make(struct CodaFid *, struct super_block *); struct inode *coda_iget(struct super_block *sb, struct CodaFid *fid, struct coda_vattr *attr); -int coda_cnode_makectl(struct inode **inode, struct super_block *sb); +struct inode *coda_cnode_makectl(struct super_block *sb); struct inode *coda_fid_to_inode(struct CodaFid *fid, struct super_block *sb); void coda_replace_fid(struct inode *, struct CodaFid *, struct CodaFid *); diff --git a/fs/coda/dir.c b/fs/coda/dir.c index 28e7e135cfab..177515829062 100644 --- a/fs/coda/dir.c +++ b/fs/coda/dir.c @@ -30,14 +30,14 @@ #include "coda_int.h" /* dir inode-ops */ -static int coda_create(struct inode *dir, struct dentry *new, int mode, struct nameidata *nd); +static int coda_create(struct inode *dir, struct dentry *new, umode_t mode, struct nameidata *nd); static struct dentry *coda_lookup(struct inode *dir, struct dentry *target, struct nameidata *nd); static int coda_link(struct dentry *old_dentry, struct inode *dir_inode, struct dentry *entry); static int coda_unlink(struct inode *dir_inode, struct dentry *entry); static int coda_symlink(struct inode *dir_inode, struct dentry *entry, const char *symname); -static int coda_mkdir(struct inode *dir_inode, struct dentry *entry, int mode); +static int coda_mkdir(struct inode *dir_inode, struct dentry *entry, umode_t mode); static int coda_rmdir(struct inode *dir_inode, struct dentry *entry); static int coda_rename(struct inode *old_inode, struct dentry *old_dentry, struct inode *new_inode, struct dentry *new_dentry); @@ -96,12 +96,11 @@ const struct file_operations coda_dir_operations = { /* access routines: lookup, readlink, permission */ static struct dentry *coda_lookup(struct inode *dir, struct dentry *entry, struct nameidata *nd) { - struct inode *inode = NULL; - struct CodaFid resfid = { { 0, } }; - int type = 0; - int error = 0; + struct super_block *sb = dir->i_sb; const char *name = entry->d_name.name; size_t length = entry->d_name.len; + struct inode *inode; + int type = 0; if (length > CODA_MAXNAMLEN) { printk(KERN_ERR "name too long: lookup, %s (%*s)\n", @@ -111,23 +110,21 @@ static struct dentry *coda_lookup(struct inode *dir, struct dentry *entry, struc /* control object, create inode on the fly */ if (coda_isroot(dir) && coda_iscontrol(name, length)) { - error = coda_cnode_makectl(&inode, dir->i_sb); + inode = coda_cnode_makectl(sb); type = CODA_NOCACHE; - goto exit; + } else { + struct CodaFid fid = { { 0, } }; + int error = venus_lookup(sb, coda_i2f(dir), name, length, + &type, &fid); + inode = !error ? coda_cnode_make(&fid, sb) : ERR_PTR(error); } - error = venus_lookup(dir->i_sb, coda_i2f(dir), name, length, - &type, &resfid); - if (!error) - error = coda_cnode_make(&inode, &resfid, dir->i_sb); - - if (error && error != -ENOENT) - return ERR_PTR(error); - -exit: - if (inode && (type & CODA_NOCACHE)) + if (!IS_ERR(inode) && (type & CODA_NOCACHE)) coda_flag_inode(inode, C_VATTR | C_PURGE); + if (inode == ERR_PTR(-ENOENT)) + inode = NULL; + return d_splice_alias(inode, entry); } @@ -191,7 +188,7 @@ static inline void coda_dir_drop_nlink(struct inode *dir) } /* creation routines: create, mknod, mkdir, link, symlink */ -static int coda_create(struct inode *dir, struct dentry *de, int mode, struct nameidata *nd) +static int coda_create(struct inode *dir, struct dentry *de, umode_t mode, struct nameidata *nd) { int error; const char *name=de->d_name.name; @@ -223,7 +220,7 @@ err_out: return error; } -static int coda_mkdir(struct inode *dir, struct dentry *de, int mode) +static int coda_mkdir(struct inode *dir, struct dentry *de, umode_t mode) { struct inode *inode; struct coda_vattr attrs; diff --git a/fs/coda/inode.c b/fs/coda/inode.c index 871b27715465..5e2e1b3f068d 100644 --- a/fs/coda/inode.c +++ b/fs/coda/inode.c @@ -58,7 +58,6 @@ static struct inode *coda_alloc_inode(struct super_block *sb) static void coda_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(coda_inode_cachep, ITOC(inode)); } @@ -205,10 +204,12 @@ static int coda_fill_super(struct super_block *sb, void *data, int silent) printk("coda_read_super: rootfid is %s\n", coda_f2s(&fid)); /* make root inode */ - error = coda_cnode_make(&root, &fid, sb); - if ( error || !root ) { - printk("Failure of coda_cnode_make for root: error %d\n", error); - goto error; + root = coda_cnode_make(&fid, sb); + if (IS_ERR(root)) { + error = PTR_ERR(root); + printk("Failure of coda_cnode_make for root: error %d\n", error); + root = NULL; + goto error; } printk("coda_read_super: rootinode is %ld dev %s\n", diff --git a/fs/compat.c b/fs/compat.c index c98787536bb8..fa9d721ecfee 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -342,16 +342,9 @@ asmlinkage long compat_sys_fstatfs64(unsigned int fd, compat_size_t sz, struct c */ asmlinkage long compat_sys_ustat(unsigned dev, struct compat_ustat __user *u) { - struct super_block *sb; struct compat_ustat tmp; struct kstatfs sbuf; - int err; - - sb = user_get_super(new_decode_dev(dev)); - if (!sb) - return -EINVAL; - err = statfs_by_dentry(sb->s_root, &sbuf); - drop_super(sb); + int err = vfs_ustat(new_decode_dev(dev), &sbuf); if (err) return err; @@ -1288,7 +1281,7 @@ compat_sys_vmsplice(int fd, const struct compat_iovec __user *iov32, * O_LARGEFILE flag. */ asmlinkage long -compat_sys_open(const char __user *filename, int flags, int mode) +compat_sys_open(const char __user *filename, int flags, umode_t mode) { return do_sys_open(AT_FDCWD, filename, flags, mode); } @@ -1298,7 +1291,7 @@ compat_sys_open(const char __user *filename, int flags, int mode) * O_LARGEFILE flag. */ asmlinkage long -compat_sys_openat(unsigned int dfd, const char __user *filename, int flags, int mode) +compat_sys_openat(unsigned int dfd, const char __user *filename, int flags, umode_t mode) { return do_sys_open(dfd, filename, flags, mode); } diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index 51352de88ef1..a26bea10e81b 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -105,6 +105,7 @@ #include <linux/hiddev.h> +#define __DVB_CORE__ #include <linux/dvb/audio.h> #include <linux/dvb/dmx.h> #include <linux/dvb/frontend.h> @@ -1506,35 +1507,6 @@ static long do_ioctl_trans(int fd, unsigned int cmd, return -ENOIOCTLCMD; } -static void compat_ioctl_error(struct file *filp, unsigned int fd, - unsigned int cmd, unsigned long arg) -{ - char buf[10]; - char *fn = "?"; - char *path; - - /* find the name of the device. */ - path = (char *)__get_free_page(GFP_KERNEL); - if (path) { - fn = d_path(&filp->f_path, path, PAGE_SIZE); - if (IS_ERR(fn)) - fn = "?"; - } - - sprintf(buf,"'%c'", (cmd>>_IOC_TYPESHIFT) & _IOC_TYPEMASK); - if (!isprint(buf[1])) - sprintf(buf, "%02x", buf[1]); - compat_printk("ioctl32(%s:%d): Unknown cmd fd(%d) " - "cmd(%08x){t:%s;sz:%u} arg(%08x) on %s\n", - current->comm, current->pid, - (int)fd, (unsigned int)cmd, buf, - (cmd >> _IOC_SIZESHIFT) & _IOC_SIZEMASK, - (unsigned int)arg, fn); - - if (path) - free_page((unsigned long)path); -} - static int compat_ioctl_check_table(unsigned int xcmd) { int i; @@ -1621,13 +1593,8 @@ asmlinkage long compat_sys_ioctl(unsigned int fd, unsigned int cmd, goto found_handler; error = do_ioctl_trans(fd, cmd, arg, filp); - if (error == -ENOIOCTLCMD) { - static int count; - - if (++count <= 50) - compat_ioctl_error(filp, fd, cmd, arg); - error = -EINVAL; - } + if (error == -ENOIOCTLCMD) + error = -ENOTTY; goto out_fput; diff --git a/fs/configfs/configfs_internal.h b/fs/configfs/configfs_internal.h index 82bda8fdfc1c..ede857d20a04 100644 --- a/fs/configfs/configfs_internal.h +++ b/fs/configfs/configfs_internal.h @@ -63,8 +63,8 @@ extern struct kmem_cache *configfs_dir_cachep; extern int configfs_is_root(struct config_item *item); -extern struct inode * configfs_new_inode(mode_t mode, struct configfs_dirent *); -extern int configfs_create(struct dentry *, int mode, int (*init)(struct inode *)); +extern struct inode * configfs_new_inode(umode_t mode, struct configfs_dirent *); +extern int configfs_create(struct dentry *, umode_t mode, int (*init)(struct inode *)); extern int configfs_inode_init(void); extern void configfs_inode_exit(void); diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c index 9a37a9b6de3a..5ddd7ebd9dcd 100644 --- a/fs/configfs/dir.c +++ b/fs/configfs/dir.c @@ -311,8 +311,8 @@ static int configfs_create_dir(struct config_item * item, struct dentry *dentry) if (item->ci_parent) parent = item->ci_parent->ci_dentry; - else if (configfs_mount && configfs_mount->mnt_sb) - parent = configfs_mount->mnt_sb->s_root; + else if (configfs_mount) + parent = configfs_mount->mnt_root; else return -EFAULT; @@ -1170,7 +1170,7 @@ void configfs_undepend_item(struct configfs_subsystem *subsys, } EXPORT_SYMBOL(configfs_undepend_item); -static int configfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int configfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { int ret = 0; int module_got = 0; diff --git a/fs/configfs/inode.c b/fs/configfs/inode.c index ca418aaf6352..3ee36d418863 100644 --- a/fs/configfs/inode.c +++ b/fs/configfs/inode.c @@ -116,7 +116,7 @@ int configfs_setattr(struct dentry * dentry, struct iattr * iattr) return error; } -static inline void set_default_inode_attr(struct inode * inode, mode_t mode) +static inline void set_default_inode_attr(struct inode * inode, umode_t mode) { inode->i_mode = mode; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; @@ -132,7 +132,7 @@ static inline void set_inode_attr(struct inode * inode, struct iattr * iattr) inode->i_ctime = iattr->ia_ctime; } -struct inode * configfs_new_inode(mode_t mode, struct configfs_dirent * sd) +struct inode *configfs_new_inode(umode_t mode, struct configfs_dirent * sd) { struct inode * inode = new_inode(configfs_sb); if (inode) { @@ -185,7 +185,7 @@ static void configfs_set_inode_lock_class(struct configfs_dirent *sd, #endif /* CONFIG_LOCKDEP */ -int configfs_create(struct dentry * dentry, int mode, int (*init)(struct inode *)) +int configfs_create(struct dentry * dentry, umode_t mode, int (*init)(struct inode *)) { int error = 0; struct inode * inode = NULL; @@ -292,7 +292,7 @@ int __init configfs_inode_init(void) return bdi_init(&configfs_backing_dev_info); } -void __exit configfs_inode_exit(void) +void configfs_inode_exit(void) { bdi_destroy(&configfs_backing_dev_info); } diff --git a/fs/configfs/mount.c b/fs/configfs/mount.c index ecc62178beda..276e15cafd58 100644 --- a/fs/configfs/mount.c +++ b/fs/configfs/mount.c @@ -143,28 +143,26 @@ static int __init configfs_init(void) goto out; config_kobj = kobject_create_and_add("config", kernel_kobj); - if (!config_kobj) { - kmem_cache_destroy(configfs_dir_cachep); - configfs_dir_cachep = NULL; - goto out; - } + if (!config_kobj) + goto out2; + + err = configfs_inode_init(); + if (err) + goto out3; err = register_filesystem(&configfs_fs_type); - if (err) { - printk(KERN_ERR "configfs: Unable to register filesystem!\n"); - kobject_put(config_kobj); - kmem_cache_destroy(configfs_dir_cachep); - configfs_dir_cachep = NULL; - goto out; - } + if (err) + goto out4; - err = configfs_inode_init(); - if (err) { - unregister_filesystem(&configfs_fs_type); - kobject_put(config_kobj); - kmem_cache_destroy(configfs_dir_cachep); - configfs_dir_cachep = NULL; - } + return 0; +out4: + printk(KERN_ERR "configfs: Unable to register filesystem!\n"); + configfs_inode_exit(); +out3: + kobject_put(config_kobj); +out2: + kmem_cache_destroy(configfs_dir_cachep); + configfs_dir_cachep = NULL; out: return err; } diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index 739fb59bcdc2..a2ee8f9f5a38 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -20,7 +20,6 @@ #include <linux/cramfs_fs.h> #include <linux/slab.h> #include <linux/cramfs_fs_sb.h> -#include <linux/buffer_head.h> #include <linux/vfs.h> #include <linux/mutex.h> @@ -378,7 +377,7 @@ static int cramfs_readdir(struct file *filp, void *dirent, filldir_t filldir) unsigned long nextoffset; char *name; ino_t ino; - mode_t mode; + umode_t mode; int namelen, error; mutex_lock(&read_mutex); diff --git a/fs/dcache.c b/fs/dcache.c index a901c6901bce..16a53cc2cc02 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -36,7 +36,9 @@ #include <linux/bit_spinlock.h> #include <linux/rculist_bl.h> #include <linux/prefetch.h> +#include <linux/ratelimit.h> #include "internal.h" +#include "mount.h" /* * Usage: @@ -241,6 +243,7 @@ static void dentry_lru_add(struct dentry *dentry) static void __dentry_lru_del(struct dentry *dentry) { list_del_init(&dentry->d_lru); + dentry->d_flags &= ~DCACHE_SHRINK_LIST; dentry->d_sb->s_nr_dentry_unused--; dentry_stat.nr_unused--; } @@ -274,15 +277,15 @@ static void dentry_lru_prune(struct dentry *dentry) } } -static void dentry_lru_move_tail(struct dentry *dentry) +static void dentry_lru_move_list(struct dentry *dentry, struct list_head *list) { spin_lock(&dcache_lru_lock); if (list_empty(&dentry->d_lru)) { - list_add_tail(&dentry->d_lru, &dentry->d_sb->s_dentry_lru); + list_add_tail(&dentry->d_lru, list); dentry->d_sb->s_nr_dentry_unused++; dentry_stat.nr_unused++; } else { - list_move_tail(&dentry->d_lru, &dentry->d_sb->s_dentry_lru); + list_move_tail(&dentry->d_lru, list); } spin_unlock(&dcache_lru_lock); } @@ -768,14 +771,18 @@ static void shrink_dentry_list(struct list_head *list) } /** - * __shrink_dcache_sb - shrink the dentry LRU on a given superblock - * @sb: superblock to shrink dentry LRU. - * @count: number of entries to prune - * @flags: flags to control the dentry processing + * prune_dcache_sb - shrink the dcache + * @sb: superblock + * @count: number of entries to try to free + * + * Attempt to shrink the superblock dcache LRU by @count entries. This is + * done when we need more memory an called from the superblock shrinker + * function. * - * If flags contains DCACHE_REFERENCED reference dentries will not be pruned. + * This function may fail to free any resources if all the dentries are in + * use. */ -static void __shrink_dcache_sb(struct super_block *sb, int count, int flags) +void prune_dcache_sb(struct super_block *sb, int count) { struct dentry *dentry; LIST_HEAD(referenced); @@ -794,18 +801,13 @@ relock: goto relock; } - /* - * If we are honouring the DCACHE_REFERENCED flag and the - * dentry has this flag set, don't free it. Clear the flag - * and put it back on the LRU. - */ - if (flags & DCACHE_REFERENCED && - dentry->d_flags & DCACHE_REFERENCED) { + if (dentry->d_flags & DCACHE_REFERENCED) { dentry->d_flags &= ~DCACHE_REFERENCED; list_move(&dentry->d_lru, &referenced); spin_unlock(&dentry->d_lock); } else { list_move_tail(&dentry->d_lru, &tmp); + dentry->d_flags |= DCACHE_SHRINK_LIST; spin_unlock(&dentry->d_lock); if (!--count) break; @@ -820,23 +822,6 @@ relock: } /** - * prune_dcache_sb - shrink the dcache - * @sb: superblock - * @nr_to_scan: number of entries to try to free - * - * Attempt to shrink the superblock dcache LRU by @nr_to_scan entries. This is - * done when we need more memory an called from the superblock shrinker - * function. - * - * This function may fail to free any resources if all the dentries are in - * use. - */ -void prune_dcache_sb(struct super_block *sb, int nr_to_scan) -{ - __shrink_dcache_sb(sb, nr_to_scan, DCACHE_REFERENCED); -} - -/** * shrink_dcache_sb - shrink dcache for a superblock * @sb: superblock * @@ -1090,7 +1075,7 @@ EXPORT_SYMBOL(have_submounts); * drop the lock and return early due to latency * constraints. */ -static int select_parent(struct dentry * parent) +static int select_parent(struct dentry *parent, struct list_head *dispose) { struct dentry *this_parent; struct list_head *next; @@ -1112,17 +1097,21 @@ resume: spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); - /* - * move only zero ref count dentries to the end - * of the unused list for prune_dcache + /* + * move only zero ref count dentries to the dispose list. + * + * Those which are presently on the shrink list, being processed + * by shrink_dentry_list(), shouldn't be moved. Otherwise the + * loop in shrink_dcache_parent() might not make any progress + * and loop forever. */ - if (!dentry->d_count) { - dentry_lru_move_tail(dentry); - found++; - } else { + if (dentry->d_count) { dentry_lru_del(dentry); + } else if (!(dentry->d_flags & DCACHE_SHRINK_LIST)) { + dentry_lru_move_list(dentry, dispose); + dentry->d_flags |= DCACHE_SHRINK_LIST; + found++; } - /* * We can return to the caller if we have found some (this * ensures forward progress). We'll be coming back to find @@ -1179,14 +1168,13 @@ rename_retry: * * Prune the dcache to remove unused children of the parent dentry. */ - void shrink_dcache_parent(struct dentry * parent) { - struct super_block *sb = parent->d_sb; + LIST_HEAD(dispose); int found; - while ((found = select_parent(parent)) != 0) - __shrink_dcache_sb(sb, found, 0); + while ((found = select_parent(parent, &dispose)) != 0) + shrink_dentry_list(&dispose); } EXPORT_SYMBOL(shrink_dcache_parent); @@ -1459,6 +1447,23 @@ struct dentry * d_alloc_root(struct inode * root_inode) } EXPORT_SYMBOL(d_alloc_root); +struct dentry *d_make_root(struct inode *root_inode) +{ + struct dentry *res = NULL; + + if (root_inode) { + static const struct qstr name = { .name = "/", .len = 1 }; + + res = __d_alloc(root_inode->i_sb, &name); + if (res) + d_instantiate(res, root_inode); + else + iput(root_inode); + } + return res; +} +EXPORT_SYMBOL(d_make_root); + static struct dentry * __d_find_any_alias(struct inode *inode) { struct dentry *alias; @@ -1470,7 +1475,14 @@ static struct dentry * __d_find_any_alias(struct inode *inode) return alias; } -static struct dentry * d_find_any_alias(struct inode *inode) +/** + * d_find_any_alias - find any alias for a given inode + * @inode: inode to find an alias for + * + * If any aliases exist for the given inode, take and return a + * reference for one of them. If no aliases exist, return %NULL. + */ +struct dentry *d_find_any_alias(struct inode *inode) { struct dentry *de; @@ -1479,7 +1491,7 @@ static struct dentry * d_find_any_alias(struct inode *inode) spin_unlock(&inode->i_lock); return de; } - +EXPORT_SYMBOL(d_find_any_alias); /** * d_obtain_alias - find or allocate a dentry for a given inode @@ -2383,8 +2395,16 @@ struct dentry *d_materialise_unique(struct dentry *dentry, struct inode *inode) actual = __d_unalias(inode, dentry, alias); } write_sequnlock(&rename_lock); - if (IS_ERR(actual)) + if (IS_ERR(actual)) { + if (PTR_ERR(actual) == -ELOOP) + pr_warn_ratelimited( + "VFS: Lookup of '%s' in %s %s" + " would have caused loop\n", + dentry->d_name.name, + inode->i_sb->s_type->name, + inode->i_sb->s_id); dput(alias); + } goto out_nolock; } } @@ -2430,20 +2450,19 @@ static int prepend_name(char **buffer, int *buflen, struct qstr *name) /** * prepend_path - Prepend path string to a buffer * @path: the dentry/vfsmount to report - * @root: root vfsmnt/dentry (may be modified by this function) + * @root: root vfsmnt/dentry * @buffer: pointer to the end of the buffer * @buflen: pointer to buffer length * * Caller holds the rename_lock. - * - * If path is not reachable from the supplied root, then the value of - * root is changed (without modifying refcounts). */ -static int prepend_path(const struct path *path, struct path *root, +static int prepend_path(const struct path *path, + const struct path *root, char **buffer, int *buflen) { struct dentry *dentry = path->dentry; struct vfsmount *vfsmnt = path->mnt; + struct mount *mnt = real_mount(vfsmnt); bool slash = false; int error = 0; @@ -2453,11 +2472,11 @@ static int prepend_path(const struct path *path, struct path *root, if (dentry == vfsmnt->mnt_root || IS_ROOT(dentry)) { /* Global root? */ - if (vfsmnt->mnt_parent == vfsmnt) { + if (!mnt_has_parent(mnt)) goto global_root; - } - dentry = vfsmnt->mnt_mountpoint; - vfsmnt = vfsmnt->mnt_parent; + dentry = mnt->mnt_mountpoint; + mnt = mnt->mnt_parent; + vfsmnt = &mnt->mnt; continue; } parent = dentry->d_parent; @@ -2474,10 +2493,10 @@ static int prepend_path(const struct path *path, struct path *root, dentry = parent; } -out: if (!error && !slash) error = prepend(buffer, buflen, "/", 1); +out: br_read_unlock(vfsmount_lock); return error; @@ -2491,15 +2510,17 @@ global_root: WARN(1, "Root dentry has weird name <%.*s>\n", (int) dentry->d_name.len, dentry->d_name.name); } - root->mnt = vfsmnt; - root->dentry = dentry; + if (!slash) + error = prepend(buffer, buflen, "/", 1); + if (!error) + error = real_mount(vfsmnt)->mnt_ns ? 1 : 2; goto out; } /** * __d_path - return the path of a dentry * @path: the dentry/vfsmount to report - * @root: root vfsmnt/dentry (may be modified by this function) + * @root: root vfsmnt/dentry * @buf: buffer to return value in * @buflen: buffer length * @@ -2510,10 +2531,10 @@ global_root: * * "buflen" should be positive. * - * If path is not reachable from the supplied root, then the value of - * root is changed (without modifying refcounts). + * If the path is not reachable from the supplied root, return %NULL. */ -char *__d_path(const struct path *path, struct path *root, +char *__d_path(const struct path *path, + const struct path *root, char *buf, int buflen) { char *res = buf + buflen; @@ -2524,7 +2545,28 @@ char *__d_path(const struct path *path, struct path *root, error = prepend_path(path, root, &res, &buflen); write_sequnlock(&rename_lock); - if (error) + if (error < 0) + return ERR_PTR(error); + if (error > 0) + return NULL; + return res; +} + +char *d_absolute_path(const struct path *path, + char *buf, int buflen) +{ + struct path root = {}; + char *res = buf + buflen; + int error; + + prepend(&res, &buflen, "\0", 1); + write_seqlock(&rename_lock); + error = prepend_path(path, &root, &res, &buflen); + write_sequnlock(&rename_lock); + + if (error > 1) + error = -EINVAL; + if (error < 0) return ERR_PTR(error); return res; } @@ -2532,8 +2574,9 @@ char *__d_path(const struct path *path, struct path *root, /* * same as __d_path but appends "(deleted)" for unlinked files. */ -static int path_with_deleted(const struct path *path, struct path *root, - char **buf, int *buflen) +static int path_with_deleted(const struct path *path, + const struct path *root, + char **buf, int *buflen) { prepend(buf, buflen, "\0", 1); if (d_unlinked(path->dentry)) { @@ -2570,7 +2613,6 @@ char *d_path(const struct path *path, char *buf, int buflen) { char *res = buf + buflen; struct path root; - struct path tmp; int error; /* @@ -2585,9 +2627,8 @@ char *d_path(const struct path *path, char *buf, int buflen) get_fs_root(current->fs, &root); write_seqlock(&rename_lock); - tmp = root; - error = path_with_deleted(path, &tmp, &res, &buflen); - if (error) + error = path_with_deleted(path, &root, &res, &buflen); + if (error < 0) res = ERR_PTR(error); write_sequnlock(&rename_lock); path_put(&root); @@ -2608,7 +2649,6 @@ char *d_path_with_unreachable(const struct path *path, char *buf, int buflen) { char *res = buf + buflen; struct path root; - struct path tmp; int error; if (path->dentry->d_op && path->dentry->d_op->d_dname) @@ -2616,9 +2656,8 @@ char *d_path_with_unreachable(const struct path *path, char *buf, int buflen) get_fs_root(current->fs, &root); write_seqlock(&rename_lock); - tmp = root; - error = path_with_deleted(path, &tmp, &res, &buflen); - if (!error && !path_equal(&tmp, &root)) + error = path_with_deleted(path, &root, &res, &buflen); + if (error > 0) error = prepend_unreachable(&res, &buflen); write_sequnlock(&rename_lock); path_put(&root); @@ -2749,19 +2788,18 @@ SYSCALL_DEFINE2(getcwd, char __user *, buf, unsigned long, size) write_seqlock(&rename_lock); if (!d_unlinked(pwd.dentry)) { unsigned long len; - struct path tmp = root; char *cwd = page + PAGE_SIZE; int buflen = PAGE_SIZE; prepend(&cwd, &buflen, "\0", 1); - error = prepend_path(&pwd, &tmp, &cwd, &buflen); + error = prepend_path(&pwd, &root, &cwd, &buflen); write_sequnlock(&rename_lock); - if (error) + if (error < 0) goto out; /* Unreachable from current root */ - if (!path_equal(&tmp, &root)) { + if (error > 0) { error = prepend_unreachable(&cwd, &buflen); if (error) goto out; @@ -2827,31 +2865,6 @@ int is_subdir(struct dentry *new_dentry, struct dentry *old_dentry) return result; } -int path_is_under(struct path *path1, struct path *path2) -{ - struct vfsmount *mnt = path1->mnt; - struct dentry *dentry = path1->dentry; - int res; - - br_read_lock(vfsmount_lock); - if (mnt != path2->mnt) { - for (;;) { - if (mnt->mnt_parent == mnt) { - br_read_unlock(vfsmount_lock); - return 0; - } - if (mnt->mnt_parent == path2->mnt) - break; - mnt = mnt->mnt_parent; - } - dentry = mnt->mnt_mountpoint; - } - res = is_subdir(dentry, path2->dentry); - br_read_unlock(vfsmount_lock); - return res; -} -EXPORT_SYMBOL(path_is_under); - void d_genocide(struct dentry *root) { struct dentry *this_parent; diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c index 90f76575c056..f65d4455c5e5 100644 --- a/fs/debugfs/file.c +++ b/fs/debugfs/file.c @@ -15,9 +15,11 @@ #include <linux/module.h> #include <linux/fs.h> +#include <linux/seq_file.h> #include <linux/pagemap.h> #include <linux/namei.h> #include <linux/debugfs.h> +#include <linux/io.h> static ssize_t default_read_file(struct file *file, char __user *buf, size_t count, loff_t *ppos) @@ -95,7 +97,7 @@ DEFINE_SIMPLE_ATTRIBUTE(fops_u8_wo, NULL, debugfs_u8_set, "%llu\n"); * %NULL or !%NULL instead as to eliminate the need for #ifdef in the calling * code. */ -struct dentry *debugfs_create_u8(const char *name, mode_t mode, +struct dentry *debugfs_create_u8(const char *name, umode_t mode, struct dentry *parent, u8 *value) { /* if there are no write bits set, make read only */ @@ -147,7 +149,7 @@ DEFINE_SIMPLE_ATTRIBUTE(fops_u16_wo, NULL, debugfs_u16_set, "%llu\n"); * %NULL or !%NULL instead as to eliminate the need for #ifdef in the calling * code. */ -struct dentry *debugfs_create_u16(const char *name, mode_t mode, +struct dentry *debugfs_create_u16(const char *name, umode_t mode, struct dentry *parent, u16 *value) { /* if there are no write bits set, make read only */ @@ -199,7 +201,7 @@ DEFINE_SIMPLE_ATTRIBUTE(fops_u32_wo, NULL, debugfs_u32_set, "%llu\n"); * %NULL or !%NULL instead as to eliminate the need for #ifdef in the calling * code. */ -struct dentry *debugfs_create_u32(const char *name, mode_t mode, +struct dentry *debugfs_create_u32(const char *name, umode_t mode, struct dentry *parent, u32 *value) { /* if there are no write bits set, make read only */ @@ -252,7 +254,7 @@ DEFINE_SIMPLE_ATTRIBUTE(fops_u64_wo, NULL, debugfs_u64_set, "%llu\n"); * %NULL or !%NULL instead as to eliminate the need for #ifdef in the calling * code. */ -struct dentry *debugfs_create_u64(const char *name, mode_t mode, +struct dentry *debugfs_create_u64(const char *name, umode_t mode, struct dentry *parent, u64 *value) { /* if there are no write bits set, make read only */ @@ -298,7 +300,7 @@ DEFINE_SIMPLE_ATTRIBUTE(fops_x64, debugfs_u64_get, debugfs_u64_set, "0x%016llx\n * @value: a pointer to the variable that the file should read to and write * from. */ -struct dentry *debugfs_create_x8(const char *name, mode_t mode, +struct dentry *debugfs_create_x8(const char *name, umode_t mode, struct dentry *parent, u8 *value) { /* if there are no write bits set, make read only */ @@ -322,7 +324,7 @@ EXPORT_SYMBOL_GPL(debugfs_create_x8); * @value: a pointer to the variable that the file should read to and write * from. */ -struct dentry *debugfs_create_x16(const char *name, mode_t mode, +struct dentry *debugfs_create_x16(const char *name, umode_t mode, struct dentry *parent, u16 *value) { /* if there are no write bits set, make read only */ @@ -346,7 +348,7 @@ EXPORT_SYMBOL_GPL(debugfs_create_x16); * @value: a pointer to the variable that the file should read to and write * from. */ -struct dentry *debugfs_create_x32(const char *name, mode_t mode, +struct dentry *debugfs_create_x32(const char *name, umode_t mode, struct dentry *parent, u32 *value) { /* if there are no write bits set, make read only */ @@ -370,7 +372,7 @@ EXPORT_SYMBOL_GPL(debugfs_create_x32); * @value: a pointer to the variable that the file should read to and write * from. */ -struct dentry *debugfs_create_x64(const char *name, mode_t mode, +struct dentry *debugfs_create_x64(const char *name, umode_t mode, struct dentry *parent, u64 *value) { return debugfs_create_file(name, mode, parent, value, &fops_x64); @@ -401,7 +403,7 @@ DEFINE_SIMPLE_ATTRIBUTE(fops_size_t, debugfs_size_t_get, debugfs_size_t_set, * @value: a pointer to the variable that the file should read to and write * from. */ -struct dentry *debugfs_create_size_t(const char *name, mode_t mode, +struct dentry *debugfs_create_size_t(const char *name, umode_t mode, struct dentry *parent, size_t *value) { return debugfs_create_file(name, mode, parent, value, &fops_size_t); @@ -473,7 +475,7 @@ static const struct file_operations fops_bool = { * %NULL or !%NULL instead as to eliminate the need for #ifdef in the calling * code. */ -struct dentry *debugfs_create_bool(const char *name, mode_t mode, +struct dentry *debugfs_create_bool(const char *name, umode_t mode, struct dentry *parent, u32 *value) { return debugfs_create_file(name, mode, parent, value, &fops_bool); @@ -518,10 +520,103 @@ static const struct file_operations fops_blob = { * %NULL or !%NULL instead as to eliminate the need for #ifdef in the calling * code. */ -struct dentry *debugfs_create_blob(const char *name, mode_t mode, +struct dentry *debugfs_create_blob(const char *name, umode_t mode, struct dentry *parent, struct debugfs_blob_wrapper *blob) { return debugfs_create_file(name, mode, parent, blob, &fops_blob); } EXPORT_SYMBOL_GPL(debugfs_create_blob); + +#ifdef CONFIG_HAS_IOMEM + +/* + * The regset32 stuff is used to print 32-bit registers using the + * seq_file utilities. We offer printing a register set in an already-opened + * sequential file or create a debugfs file that only prints a regset32. + */ + +/** + * debugfs_print_regs32 - use seq_print to describe a set of registers + * @s: the seq_file structure being used to generate output + * @regs: an array if struct debugfs_reg32 structures + * @mregs: the length of the above array + * @base: the base address to be used in reading the registers + * @prefix: a string to be prefixed to every output line + * + * This function outputs a text block describing the current values of + * some 32-bit hardware registers. It is meant to be used within debugfs + * files based on seq_file that need to show registers, intermixed with other + * information. The prefix argument may be used to specify a leading string, + * because some peripherals have several blocks of identical registers, + * for example configuration of dma channels + */ +int debugfs_print_regs32(struct seq_file *s, const struct debugfs_reg32 *regs, + int nregs, void __iomem *base, char *prefix) +{ + int i, ret = 0; + + for (i = 0; i < nregs; i++, regs++) { + if (prefix) + ret += seq_printf(s, "%s", prefix); + ret += seq_printf(s, "%s = 0x%08x\n", regs->name, + readl(base + regs->offset)); + } + return ret; +} +EXPORT_SYMBOL_GPL(debugfs_print_regs32); + +static int debugfs_show_regset32(struct seq_file *s, void *data) +{ + struct debugfs_regset32 *regset = s->private; + + debugfs_print_regs32(s, regset->regs, regset->nregs, regset->base, ""); + return 0; +} + +static int debugfs_open_regset32(struct inode *inode, struct file *file) +{ + return single_open(file, debugfs_show_regset32, inode->i_private); +} + +static const struct file_operations fops_regset32 = { + .open = debugfs_open_regset32, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +/** + * debugfs_create_regset32 - create a debugfs file that returns register values + * @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. + * @regset: a pointer to a struct debugfs_regset32, which contains a pointer + * to an array of register definitions, the array size and the base + * address where the register bank is to be found. + * + * This function creates a file in debugfs with the given name that reports + * the names and values of a set of 32-bit registers. If the @mode variable + * is so set it can be read from. Writing is not supported. + * + * This function will return a pointer to a dentry if it succeeds. This + * pointer must be passed to the debugfs_remove() function when the file is + * to be removed (no automatic cleanup happens if your module is unloaded, + * you are responsible here.) If an error occurs, %NULL will be returned. + * + * If debugfs is not enabled in the kernel, the value -%ENODEV will be + * returned. It is not wise to check for this value, but rather, check for + * %NULL or !%NULL instead as to eliminate the need for #ifdef in the calling + * code. + */ +struct dentry *debugfs_create_regset32(const char *name, mode_t mode, + struct dentry *parent, + struct debugfs_regset32 *regset) +{ + return debugfs_create_file(name, mode, parent, regset, &fops_regset32); +} +EXPORT_SYMBOL_GPL(debugfs_create_regset32); + +#endif /* CONFIG_HAS_IOMEM */ diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c index f3a257d7a985..956d5ddddf6e 100644 --- a/fs/debugfs/inode.c +++ b/fs/debugfs/inode.c @@ -30,7 +30,7 @@ static struct vfsmount *debugfs_mount; static int debugfs_mount_count; static bool debugfs_registered; -static struct inode *debugfs_get_inode(struct super_block *sb, int mode, dev_t dev, +static struct inode *debugfs_get_inode(struct super_block *sb, umode_t mode, dev_t dev, void *data, const struct file_operations *fops) { @@ -69,7 +69,7 @@ static struct inode *debugfs_get_inode(struct super_block *sb, int mode, dev_t d /* SMP-safe */ static int debugfs_mknod(struct inode *dir, struct dentry *dentry, - int mode, dev_t dev, void *data, + umode_t mode, dev_t dev, void *data, const struct file_operations *fops) { struct inode *inode; @@ -87,7 +87,7 @@ static int debugfs_mknod(struct inode *dir, struct dentry *dentry, return error; } -static int debugfs_mkdir(struct inode *dir, struct dentry *dentry, int mode, +static int debugfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode, void *data, const struct file_operations *fops) { int res; @@ -101,14 +101,14 @@ static int debugfs_mkdir(struct inode *dir, struct dentry *dentry, int mode, return res; } -static int debugfs_link(struct inode *dir, struct dentry *dentry, int mode, +static int debugfs_link(struct inode *dir, struct dentry *dentry, umode_t mode, void *data, const struct file_operations *fops) { mode = (mode & S_IALLUGO) | S_IFLNK; return debugfs_mknod(dir, dentry, mode, 0, data, fops); } -static int debugfs_create(struct inode *dir, struct dentry *dentry, int mode, +static int debugfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, void *data, const struct file_operations *fops) { int res; @@ -146,7 +146,7 @@ static struct file_system_type debug_fs_type = { .kill_sb = kill_litter_super, }; -static int debugfs_create_by_name(const char *name, mode_t mode, +static int debugfs_create_by_name(const char *name, umode_t mode, struct dentry *parent, struct dentry **dentry, void *data, @@ -160,7 +160,7 @@ static int debugfs_create_by_name(const char *name, mode_t mode, * have around. */ if (!parent) - parent = debugfs_mount->mnt_sb->s_root; + parent = debugfs_mount->mnt_root; *dentry = NULL; mutex_lock(&parent->d_inode->i_mutex); @@ -214,7 +214,7 @@ static int debugfs_create_by_name(const char *name, mode_t mode, * If debugfs is not enabled in the kernel, the value -%ENODEV will be * returned. */ -struct dentry *debugfs_create_file(const char *name, mode_t mode, +struct dentry *debugfs_create_file(const char *name, umode_t mode, struct dentry *parent, void *data, const struct file_operations *fops) { diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c index d5d5297efe97..c4e2a58a2e82 100644 --- a/fs/devpts/inode.c +++ b/fs/devpts/inode.c @@ -246,9 +246,9 @@ static int devpts_remount(struct super_block *sb, int *flags, char *data) return err; } -static int devpts_show_options(struct seq_file *seq, struct vfsmount *vfs) +static int devpts_show_options(struct seq_file *seq, struct dentry *root) { - struct pts_fs_info *fsi = DEVPTS_SB(vfs->mnt_sb); + struct pts_fs_info *fsi = DEVPTS_SB(root->d_sb); struct pts_mount_opts *opts = &fsi->mount_opts; if (opts->setuid) @@ -301,7 +301,7 @@ devpts_fill_super(struct super_block *s, void *data, int silent) inode = new_inode(s); if (!inode) - goto free_fsi; + goto fail; inode->i_ino = 1; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR; @@ -316,8 +316,6 @@ devpts_fill_super(struct super_block *s, void *data, int silent) printk(KERN_ERR "devpts: get root dentry failed\n"); iput(inode); -free_fsi: - kfree(s->s_fs_info); fail: return -ENOMEM; } diff --git a/fs/direct-io.c b/fs/direct-io.c index d740ab67ff6e..4a588dbd11bf 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -36,6 +36,7 @@ #include <linux/rwsem.h> #include <linux/uio.h> #include <linux/atomic.h> +#include <linux/prefetch.h> /* * How many user pages to map in one call to get_user_pages(). This determines @@ -580,9 +581,8 @@ static int get_more_blocks(struct dio *dio, struct dio_submit *sdio, { int ret; sector_t fs_startblk; /* Into file, in filesystem-sized blocks */ + sector_t fs_endblk; /* Into file, in filesystem-sized blocks */ unsigned long fs_count; /* Number of filesystem-sized blocks */ - unsigned long dio_count;/* Number of dio_block-sized blocks */ - unsigned long blkmask; int create; /* @@ -593,11 +593,9 @@ static int get_more_blocks(struct dio *dio, struct dio_submit *sdio, if (ret == 0) { BUG_ON(sdio->block_in_file >= sdio->final_block_in_request); fs_startblk = sdio->block_in_file >> sdio->blkfactor; - dio_count = sdio->final_block_in_request - sdio->block_in_file; - fs_count = dio_count >> sdio->blkfactor; - blkmask = (1 << sdio->blkfactor) - 1; - if (dio_count & blkmask) - fs_count++; + fs_endblk = (sdio->final_block_in_request - 1) >> + sdio->blkfactor; + fs_count = fs_endblk - fs_startblk + 1; map_bh->b_state = 0; map_bh->b_size = fs_count << dio->inode->i_blkbits; @@ -1090,8 +1088,8 @@ static inline int drop_refcount(struct dio *dio) * individual fields and will generate much worse code. This is important * for the whole file. */ -ssize_t -__blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, +static inline ssize_t +do_blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, struct block_device *bdev, const struct iovec *iov, loff_t offset, unsigned long nr_segs, get_block_t get_block, dio_iodone_t end_io, dio_submit_t submit_io, int flags) @@ -1100,7 +1098,6 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, size_t size; unsigned long addr; unsigned blkbits = inode->i_blkbits; - unsigned bdev_blkbits = 0; unsigned blocksize_mask = (1 << blkbits) - 1; ssize_t retval = -EINVAL; loff_t end = offset; @@ -1113,12 +1110,14 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, if (rw & WRITE) rw = WRITE_ODIRECT; - if (bdev) - bdev_blkbits = blksize_bits(bdev_logical_block_size(bdev)); + /* + * Avoid references to bdev if not absolutely needed to give + * the early prefetch in the caller enough time. + */ if (offset & blocksize_mask) { if (bdev) - blkbits = bdev_blkbits; + blkbits = blksize_bits(bdev_logical_block_size(bdev)); blocksize_mask = (1 << blkbits) - 1; if (offset & blocksize_mask) goto out; @@ -1129,11 +1128,13 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, addr = (unsigned long)iov[seg].iov_base; size = iov[seg].iov_len; end += size; - if ((addr & blocksize_mask) || (size & blocksize_mask)) { + if (unlikely((addr & blocksize_mask) || + (size & blocksize_mask))) { if (bdev) - blkbits = bdev_blkbits; + blkbits = blksize_bits( + bdev_logical_block_size(bdev)); blocksize_mask = (1 << blkbits) - 1; - if ((addr & blocksize_mask) || (size & blocksize_mask)) + if ((addr & blocksize_mask) || (size & blocksize_mask)) goto out; } } @@ -1316,6 +1317,30 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, out: return retval; } + +ssize_t +__blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, + struct block_device *bdev, const struct iovec *iov, loff_t offset, + unsigned long nr_segs, get_block_t get_block, dio_iodone_t end_io, + dio_submit_t submit_io, int flags) +{ + /* + * The block device state is needed in the end to finally + * submit everything. Since it's likely to be cache cold + * prefetch it here as first thing to hide some of the + * latency. + * + * Attempt to prefetch the pieces we likely need later. + */ + prefetch(&bdev->bd_disk->part_tbl); + prefetch(bdev->bd_queue); + prefetch((char *)bdev->bd_queue + SMP_CACHE_BYTES); + + return do_blockdev_direct_IO(rw, iocb, inode, bdev, iov, offset, + nr_segs, get_block, end_io, + submit_io, flags); +} + EXPORT_SYMBOL(__blockdev_direct_IO); static __init int dio_init(void) diff --git a/fs/dlm/config.c b/fs/dlm/config.c index 6cf72fcc0d0c..e7e327d43fa5 100644 --- a/fs/dlm/config.c +++ b/fs/dlm/config.c @@ -2,7 +2,7 @@ ******************************************************************************* ** ** Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. -** Copyright (C) 2004-2008 Red Hat, Inc. All rights reserved. +** Copyright (C) 2004-2011 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -17,6 +17,7 @@ #include <linux/slab.h> #include <linux/in.h> #include <linux/in6.h> +#include <linux/dlmconstants.h> #include <net/ipv6.h> #include <net/sock.h> @@ -36,6 +37,7 @@ static struct config_group *space_list; static struct config_group *comm_list; static struct dlm_comm *local_comm; +static uint32_t dlm_comm_count; struct dlm_clusters; struct dlm_cluster; @@ -103,6 +105,8 @@ struct dlm_cluster { unsigned int cl_timewarn_cs; unsigned int cl_waitwarn_us; unsigned int cl_new_rsb_count; + unsigned int cl_recover_callbacks; + char cl_cluster_name[DLM_LOCKSPACE_LEN]; }; enum { @@ -118,6 +122,8 @@ enum { CLUSTER_ATTR_TIMEWARN_CS, CLUSTER_ATTR_WAITWARN_US, CLUSTER_ATTR_NEW_RSB_COUNT, + CLUSTER_ATTR_RECOVER_CALLBACKS, + CLUSTER_ATTR_CLUSTER_NAME, }; struct cluster_attribute { @@ -126,6 +132,27 @@ struct cluster_attribute { ssize_t (*store)(struct dlm_cluster *, const char *, size_t); }; +static ssize_t cluster_cluster_name_read(struct dlm_cluster *cl, char *buf) +{ + return sprintf(buf, "%s\n", cl->cl_cluster_name); +} + +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); + return len; +} + +static struct cluster_attribute cluster_attr_cluster_name = { + .attr = { .ca_owner = THIS_MODULE, + .ca_name = "cluster_name", + .ca_mode = S_IRUGO | S_IWUSR }, + .show = cluster_cluster_name_read, + .store = cluster_cluster_name_write, +}; + static ssize_t cluster_set(struct dlm_cluster *cl, unsigned int *cl_field, int *info_field, int check_zero, const char *buf, size_t len) @@ -171,6 +198,7 @@ CLUSTER_ATTR(protocol, 0); CLUSTER_ATTR(timewarn_cs, 1); CLUSTER_ATTR(waitwarn_us, 0); CLUSTER_ATTR(new_rsb_count, 0); +CLUSTER_ATTR(recover_callbacks, 0); static struct configfs_attribute *cluster_attrs[] = { [CLUSTER_ATTR_TCP_PORT] = &cluster_attr_tcp_port.attr, @@ -185,6 +213,8 @@ static struct configfs_attribute *cluster_attrs[] = { [CLUSTER_ATTR_TIMEWARN_CS] = &cluster_attr_timewarn_cs.attr, [CLUSTER_ATTR_WAITWARN_US] = &cluster_attr_waitwarn_us.attr, [CLUSTER_ATTR_NEW_RSB_COUNT] = &cluster_attr_new_rsb_count.attr, + [CLUSTER_ATTR_RECOVER_CALLBACKS] = &cluster_attr_recover_callbacks.attr, + [CLUSTER_ATTR_CLUSTER_NAME] = &cluster_attr_cluster_name.attr, NULL, }; @@ -293,6 +323,7 @@ struct dlm_comms { struct dlm_comm { struct config_item item; + int seq; int nodeid; int local; int addr_count; @@ -309,6 +340,7 @@ struct dlm_node { int nodeid; int weight; int new; + int comm_seq; /* copy of cm->seq when nd->nodeid is set */ }; static struct configfs_group_operations clusters_ops = { @@ -455,6 +487,9 @@ static struct config_group *make_cluster(struct config_group *g, cl->cl_timewarn_cs = dlm_config.ci_timewarn_cs; cl->cl_waitwarn_us = dlm_config.ci_waitwarn_us; cl->cl_new_rsb_count = dlm_config.ci_new_rsb_count; + cl->cl_recover_callbacks = dlm_config.ci_recover_callbacks; + memcpy(cl->cl_cluster_name, dlm_config.ci_cluster_name, + DLM_LOCKSPACE_LEN); space_list = &sps->ss_group; comm_list = &cms->cs_group; @@ -558,6 +593,11 @@ static struct config_item *make_comm(struct config_group *g, const char *name) return ERR_PTR(-ENOMEM); config_item_init_type_name(&cm->item, name, &comm_type); + + cm->seq = dlm_comm_count++; + if (!cm->seq) + cm->seq = dlm_comm_count++; + cm->nodeid = -1; cm->local = 0; cm->addr_count = 0; @@ -801,7 +841,10 @@ static ssize_t node_nodeid_read(struct dlm_node *nd, char *buf) static ssize_t node_nodeid_write(struct dlm_node *nd, const char *buf, size_t len) { + uint32_t seq = 0; nd->nodeid = simple_strtol(buf, NULL, 0); + dlm_comm_seq(nd->nodeid, &seq); + nd->comm_seq = seq; return len; } @@ -908,13 +951,13 @@ static void put_comm(struct dlm_comm *cm) } /* caller must free mem */ -int dlm_nodeid_list(char *lsname, int **ids_out, int *ids_count_out, - int **new_out, int *new_count_out) +int dlm_config_nodes(char *lsname, struct dlm_config_node **nodes_out, + int *count_out) { struct dlm_space *sp; struct dlm_node *nd; - int i = 0, rv = 0, ids_count = 0, new_count = 0; - int *ids, *new; + struct dlm_config_node *nodes, *node; + int rv, count; sp = get_space(lsname); if (!sp) @@ -927,73 +970,42 @@ int dlm_nodeid_list(char *lsname, int **ids_out, int *ids_count_out, goto out; } - ids_count = sp->members_count; + count = sp->members_count; - ids = kcalloc(ids_count, sizeof(int), GFP_NOFS); - if (!ids) { + nodes = kcalloc(count, sizeof(struct dlm_config_node), GFP_NOFS); + if (!nodes) { rv = -ENOMEM; goto out; } + node = nodes; list_for_each_entry(nd, &sp->members, list) { - ids[i++] = nd->nodeid; - if (nd->new) - new_count++; - } - - if (ids_count != i) - printk(KERN_ERR "dlm: bad nodeid count %d %d\n", ids_count, i); - - if (!new_count) - goto out_ids; + node->nodeid = nd->nodeid; + node->weight = nd->weight; + node->new = nd->new; + node->comm_seq = nd->comm_seq; + node++; - new = kcalloc(new_count, sizeof(int), GFP_NOFS); - if (!new) { - kfree(ids); - rv = -ENOMEM; - goto out; + nd->new = 0; } - i = 0; - list_for_each_entry(nd, &sp->members, list) { - if (nd->new) { - new[i++] = nd->nodeid; - nd->new = 0; - } - } - *new_count_out = new_count; - *new_out = new; - - out_ids: - *ids_count_out = ids_count; - *ids_out = ids; + *count_out = count; + *nodes_out = nodes; + rv = 0; out: mutex_unlock(&sp->members_lock); put_space(sp); return rv; } -int dlm_node_weight(char *lsname, int nodeid) +int dlm_comm_seq(int nodeid, uint32_t *seq) { - struct dlm_space *sp; - struct dlm_node *nd; - int w = -EEXIST; - - sp = get_space(lsname); - if (!sp) - goto out; - - mutex_lock(&sp->members_lock); - list_for_each_entry(nd, &sp->members, list) { - if (nd->nodeid != nodeid) - continue; - w = nd->weight; - break; - } - mutex_unlock(&sp->members_lock); - put_space(sp); - out: - return w; + struct dlm_comm *cm = get_comm(nodeid, NULL); + if (!cm) + return -EEXIST; + *seq = cm->seq; + put_comm(cm); + return 0; } int dlm_nodeid_to_addr(int nodeid, struct sockaddr_storage *addr) @@ -1047,6 +1059,8 @@ int dlm_our_addr(struct sockaddr_storage *addr, int num) #define DEFAULT_TIMEWARN_CS 500 /* 5 sec = 500 centiseconds */ #define DEFAULT_WAITWARN_US 0 #define DEFAULT_NEW_RSB_COUNT 128 +#define DEFAULT_RECOVER_CALLBACKS 0 +#define DEFAULT_CLUSTER_NAME "" struct dlm_config_info dlm_config = { .ci_tcp_port = DEFAULT_TCP_PORT, @@ -1060,6 +1074,8 @@ struct dlm_config_info dlm_config = { .ci_protocol = DEFAULT_PROTOCOL, .ci_timewarn_cs = DEFAULT_TIMEWARN_CS, .ci_waitwarn_us = DEFAULT_WAITWARN_US, - .ci_new_rsb_count = DEFAULT_NEW_RSB_COUNT + .ci_new_rsb_count = DEFAULT_NEW_RSB_COUNT, + .ci_recover_callbacks = DEFAULT_RECOVER_CALLBACKS, + .ci_cluster_name = DEFAULT_CLUSTER_NAME }; diff --git a/fs/dlm/config.h b/fs/dlm/config.h index 3099d0dd26c0..9f5e3663bb0c 100644 --- a/fs/dlm/config.h +++ b/fs/dlm/config.h @@ -2,7 +2,7 @@ ******************************************************************************* ** ** Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. -** Copyright (C) 2004-2007 Red Hat, Inc. All rights reserved. +** Copyright (C) 2004-2011 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -14,6 +14,13 @@ #ifndef __CONFIG_DOT_H__ #define __CONFIG_DOT_H__ +struct dlm_config_node { + int nodeid; + int weight; + int new; + uint32_t comm_seq; +}; + #define DLM_MAX_ADDR_COUNT 3 struct dlm_config_info { @@ -29,15 +36,17 @@ struct dlm_config_info { int ci_timewarn_cs; int ci_waitwarn_us; int ci_new_rsb_count; + int ci_recover_callbacks; + char ci_cluster_name[DLM_LOCKSPACE_LEN]; }; extern struct dlm_config_info dlm_config; int dlm_config_init(void); void dlm_config_exit(void); -int dlm_node_weight(char *lsname, int nodeid); -int dlm_nodeid_list(char *lsname, int **ids_out, int *ids_count_out, - int **new_out, int *new_count_out); +int dlm_config_nodes(char *lsname, struct dlm_config_node **nodes_out, + int *count_out); +int dlm_comm_seq(int nodeid, uint32_t *seq); int dlm_nodeid_to_addr(int nodeid, struct sockaddr_storage *addr); int dlm_addr_to_nodeid(struct sockaddr_storage *addr, int *nodeid); int dlm_our_nodeid(void); diff --git a/fs/dlm/debug_fs.c b/fs/dlm/debug_fs.c index 59779237e2b4..3dca2b39e83f 100644 --- a/fs/dlm/debug_fs.c +++ b/fs/dlm/debug_fs.c @@ -393,6 +393,7 @@ static const struct seq_operations format3_seq_ops; static void *table_seq_start(struct seq_file *seq, loff_t *pos) { + struct rb_node *node; struct dlm_ls *ls = seq->private; struct rsbtbl_iter *ri; struct dlm_rsb *r; @@ -418,9 +419,10 @@ static void *table_seq_start(struct seq_file *seq, loff_t *pos) ri->format = 3; spin_lock(&ls->ls_rsbtbl[bucket].lock); - if (!list_empty(&ls->ls_rsbtbl[bucket].list)) { - list_for_each_entry(r, &ls->ls_rsbtbl[bucket].list, - res_hashchain) { + if (!RB_EMPTY_ROOT(&ls->ls_rsbtbl[bucket].keep)) { + for (node = rb_first(&ls->ls_rsbtbl[bucket].keep); node; + node = rb_next(node)) { + r = rb_entry(node, struct dlm_rsb, res_hashnode); if (!entry--) { dlm_hold_rsb(r); ri->rsb = r; @@ -449,9 +451,9 @@ static void *table_seq_start(struct seq_file *seq, loff_t *pos) } spin_lock(&ls->ls_rsbtbl[bucket].lock); - if (!list_empty(&ls->ls_rsbtbl[bucket].list)) { - r = list_first_entry(&ls->ls_rsbtbl[bucket].list, - struct dlm_rsb, res_hashchain); + if (!RB_EMPTY_ROOT(&ls->ls_rsbtbl[bucket].keep)) { + node = rb_first(&ls->ls_rsbtbl[bucket].keep); + r = rb_entry(node, struct dlm_rsb, res_hashnode); dlm_hold_rsb(r); ri->rsb = r; ri->bucket = bucket; @@ -467,7 +469,7 @@ static void *table_seq_next(struct seq_file *seq, void *iter_ptr, loff_t *pos) { struct dlm_ls *ls = seq->private; struct rsbtbl_iter *ri = iter_ptr; - struct list_head *next; + struct rb_node *next; struct dlm_rsb *r, *rp; loff_t n = *pos; unsigned bucket; @@ -480,10 +482,10 @@ static void *table_seq_next(struct seq_file *seq, void *iter_ptr, loff_t *pos) spin_lock(&ls->ls_rsbtbl[bucket].lock); rp = ri->rsb; - next = rp->res_hashchain.next; + next = rb_next(&rp->res_hashnode); - if (next != &ls->ls_rsbtbl[bucket].list) { - r = list_entry(next, struct dlm_rsb, res_hashchain); + if (next) { + r = rb_entry(next, struct dlm_rsb, res_hashnode); dlm_hold_rsb(r); ri->rsb = r; spin_unlock(&ls->ls_rsbtbl[bucket].lock); @@ -511,9 +513,9 @@ static void *table_seq_next(struct seq_file *seq, void *iter_ptr, loff_t *pos) } spin_lock(&ls->ls_rsbtbl[bucket].lock); - if (!list_empty(&ls->ls_rsbtbl[bucket].list)) { - r = list_first_entry(&ls->ls_rsbtbl[bucket].list, - struct dlm_rsb, res_hashchain); + if (!RB_EMPTY_ROOT(&ls->ls_rsbtbl[bucket].keep)) { + next = rb_first(&ls->ls_rsbtbl[bucket].keep); + r = rb_entry(next, struct dlm_rsb, res_hashnode); dlm_hold_rsb(r); ri->rsb = r; ri->bucket = bucket; diff --git a/fs/dlm/dir.c b/fs/dlm/dir.c index 7b84c1dbc82e..83641574b016 100644 --- a/fs/dlm/dir.c +++ b/fs/dlm/dir.c @@ -290,7 +290,6 @@ int dlm_recover_directory(struct dlm_ls *ls) out_status: error = 0; - dlm_set_recover_status(ls, DLM_RS_DIR); log_debug(ls, "dlm_recover_directory %d entries", count); out_free: kfree(last_name); diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index fe2860c02449..3a564d197e99 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -2,7 +2,7 @@ ******************************************************************************* ** ** Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. -** Copyright (C) 2004-2010 Red Hat, Inc. All rights reserved. +** Copyright (C) 2004-2011 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -103,8 +103,8 @@ struct dlm_dirtable { }; struct dlm_rsbtable { - struct list_head list; - struct list_head toss; + struct rb_root keep; + struct rb_root toss; spinlock_t lock; }; @@ -117,6 +117,10 @@ struct dlm_member { struct list_head list; int nodeid; int weight; + int slot; + int slot_prev; + int comm_seq; + uint32_t generation; }; /* @@ -125,10 +129,8 @@ struct dlm_member { struct dlm_recover { struct list_head list; - int *nodeids; /* nodeids of all members */ - int node_count; - int *new; /* nodeids of new members */ - int new_count; + struct dlm_config_node *nodes; + int nodes_count; uint64_t seq; }; @@ -285,7 +287,10 @@ struct dlm_rsb { unsigned long res_toss_time; uint32_t res_first_lkid; struct list_head res_lookup; /* lkbs waiting on first */ - struct list_head res_hashchain; /* rsbtbl */ + union { + struct list_head res_hashchain; + struct rb_node res_hashnode; /* rsbtbl */ + }; struct list_head res_grantqueue; struct list_head res_convertqueue; struct list_head res_waitqueue; @@ -334,7 +339,9 @@ static inline int rsb_flag(struct dlm_rsb *r, enum rsb_flags flag) /* dlm_header is first element of all structs sent between nodes */ #define DLM_HEADER_MAJOR 0x00030000 -#define DLM_HEADER_MINOR 0x00000000 +#define DLM_HEADER_MINOR 0x00000001 + +#define DLM_HEADER_SLOTS 0x00000001 #define DLM_MSG 1 #define DLM_RCOM 2 @@ -422,10 +429,34 @@ union dlm_packet { struct dlm_rcom rcom; }; +#define DLM_RSF_NEED_SLOTS 0x00000001 + +/* RCOM_STATUS data */ +struct rcom_status { + __le32 rs_flags; + __le32 rs_unused1; + __le64 rs_unused2; +}; + +/* RCOM_STATUS_REPLY data */ struct rcom_config { __le32 rf_lvblen; __le32 rf_lsflags; - __le64 rf_unused; + + /* DLM_HEADER_SLOTS adds: */ + __le32 rf_flags; + __le16 rf_our_slot; + __le16 rf_num_slots; + __le32 rf_generation; + __le32 rf_unused1; + __le64 rf_unused2; +}; + +struct rcom_slot { + __le32 ro_nodeid; + __le16 ro_slot; + __le16 ro_unused1; + __le64 ro_unused2; }; struct rcom_lock { @@ -452,6 +483,7 @@ struct dlm_ls { struct list_head ls_list; /* list of lockspaces */ dlm_lockspace_t *ls_local_handle; uint32_t ls_global_id; /* global unique lockspace ID */ + uint32_t ls_generation; uint32_t ls_exflags; int ls_lvblen; int ls_count; /* refcount of processes in @@ -490,6 +522,11 @@ struct dlm_ls { int ls_total_weight; int *ls_node_array; + int ls_slot; + int ls_num_slots; + int ls_slots_size; + struct dlm_slot *ls_slots; + struct dlm_rsb ls_stub_rsb; /* for returning errors */ struct dlm_lkb ls_stub_lkb; /* for returning errors */ struct dlm_message ls_stub_ms; /* for faking a reply */ @@ -537,6 +574,9 @@ struct dlm_ls { struct list_head ls_root_list; /* root resources */ struct rw_semaphore ls_root_sem; /* protect root_list */ + const struct dlm_lockspace_ops *ls_ops; + void *ls_ops_arg; + int ls_namelen; char ls_name[1]; }; diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 83b5e32514e1..d47183043c59 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -56,6 +56,7 @@ L: receive_xxxx_reply() <- R: send_xxxx_reply() */ #include <linux/types.h> +#include <linux/rbtree.h> #include <linux/slab.h> #include "dlm_internal.h" #include <linux/dlm_device.h> @@ -380,6 +381,8 @@ static int get_rsb_struct(struct dlm_ls *ls, char *name, int len, r = list_first_entry(&ls->ls_new_rsb, struct dlm_rsb, res_hashchain); list_del(&r->res_hashchain); + /* Convert the empty list_head to a NULL rb_node for tree usage: */ + memset(&r->res_hashnode, 0, sizeof(struct rb_node)); ls->ls_new_rsb_count--; spin_unlock(&ls->ls_new_rsb_spin); @@ -388,7 +391,6 @@ static int get_rsb_struct(struct dlm_ls *ls, char *name, int len, memcpy(r->res_name, name, len); mutex_init(&r->res_mutex); - INIT_LIST_HEAD(&r->res_hashchain); INIT_LIST_HEAD(&r->res_lookup); INIT_LIST_HEAD(&r->res_grantqueue); INIT_LIST_HEAD(&r->res_convertqueue); @@ -400,14 +402,31 @@ static int get_rsb_struct(struct dlm_ls *ls, char *name, int len, return 0; } -static int search_rsb_list(struct list_head *head, char *name, int len, +static int rsb_cmp(struct dlm_rsb *r, const char *name, int nlen) +{ + char maxname[DLM_RESNAME_MAXLEN]; + + memset(maxname, 0, DLM_RESNAME_MAXLEN); + memcpy(maxname, name, nlen); + return memcmp(r->res_name, maxname, DLM_RESNAME_MAXLEN); +} + +static int search_rsb_tree(struct rb_root *tree, char *name, int len, unsigned int flags, struct dlm_rsb **r_ret) { + struct rb_node *node = tree->rb_node; struct dlm_rsb *r; int error = 0; - - list_for_each_entry(r, head, res_hashchain) { - if (len == r->res_length && !memcmp(name, r->res_name, len)) + int rc; + + while (node) { + r = rb_entry(node, struct dlm_rsb, res_hashnode); + rc = rsb_cmp(r, name, len); + if (rc < 0) + node = node->rb_left; + else if (rc > 0) + node = node->rb_right; + else goto found; } *r_ret = NULL; @@ -420,22 +439,54 @@ static int search_rsb_list(struct list_head *head, char *name, int len, return error; } +static int rsb_insert(struct dlm_rsb *rsb, struct rb_root *tree) +{ + struct rb_node **newn = &tree->rb_node; + struct rb_node *parent = NULL; + int rc; + + while (*newn) { + struct dlm_rsb *cur = rb_entry(*newn, struct dlm_rsb, + res_hashnode); + + parent = *newn; + rc = rsb_cmp(cur, rsb->res_name, rsb->res_length); + if (rc < 0) + newn = &parent->rb_left; + else if (rc > 0) + newn = &parent->rb_right; + else { + log_print("rsb_insert match"); + dlm_dump_rsb(rsb); + dlm_dump_rsb(cur); + return -EEXIST; + } + } + + rb_link_node(&rsb->res_hashnode, parent, newn); + rb_insert_color(&rsb->res_hashnode, tree); + return 0; +} + static int _search_rsb(struct dlm_ls *ls, char *name, int len, int b, unsigned int flags, struct dlm_rsb **r_ret) { struct dlm_rsb *r; int error; - error = search_rsb_list(&ls->ls_rsbtbl[b].list, name, len, flags, &r); + error = search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, flags, &r); if (!error) { kref_get(&r->res_ref); goto out; } - error = search_rsb_list(&ls->ls_rsbtbl[b].toss, name, len, flags, &r); + error = search_rsb_tree(&ls->ls_rsbtbl[b].toss, name, len, flags, &r); if (error) goto out; - list_move(&r->res_hashchain, &ls->ls_rsbtbl[b].list); + rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].toss); + error = rsb_insert(r, &ls->ls_rsbtbl[b].keep); + if (error) + return error; if (dlm_no_directory(ls)) goto out; @@ -527,8 +578,7 @@ static int find_rsb(struct dlm_ls *ls, char *name, int namelen, nodeid = 0; r->res_nodeid = nodeid; } - list_add(&r->res_hashchain, &ls->ls_rsbtbl[bucket].list); - error = 0; + error = rsb_insert(r, &ls->ls_rsbtbl[bucket].keep); out_unlock: spin_unlock(&ls->ls_rsbtbl[bucket].lock); out: @@ -556,7 +606,8 @@ static void toss_rsb(struct kref *kref) DLM_ASSERT(list_empty(&r->res_root_list), dlm_print_rsb(r);); kref_init(&r->res_ref); - list_move(&r->res_hashchain, &ls->ls_rsbtbl[r->res_bucket].toss); + rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[r->res_bucket].keep); + rsb_insert(r, &ls->ls_rsbtbl[r->res_bucket].toss); r->res_toss_time = jiffies; if (r->res_lvbptr) { dlm_free_lvb(r->res_lvbptr); @@ -1082,19 +1133,19 @@ static void dir_remove(struct dlm_rsb *r) r->res_name, r->res_length); } -/* FIXME: shouldn't this be able to exit as soon as one non-due rsb is - found since they are in order of newest to oldest? */ +/* FIXME: make this more efficient */ static int shrink_bucket(struct dlm_ls *ls, int b) { + struct rb_node *n; struct dlm_rsb *r; int count = 0, found; for (;;) { found = 0; spin_lock(&ls->ls_rsbtbl[b].lock); - list_for_each_entry_reverse(r, &ls->ls_rsbtbl[b].toss, - res_hashchain) { + for (n = rb_first(&ls->ls_rsbtbl[b].toss); n; n = rb_next(n)) { + r = rb_entry(n, struct dlm_rsb, res_hashnode); if (!time_after_eq(jiffies, r->res_toss_time + dlm_config.ci_toss_secs * HZ)) continue; @@ -1108,7 +1159,7 @@ static int shrink_bucket(struct dlm_ls *ls, int b) } if (kref_put(&r->res_ref, kill_rsb)) { - list_del(&r->res_hashchain); + rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].toss); spin_unlock(&ls->ls_rsbtbl[b].lock); if (is_master(r)) @@ -4441,10 +4492,12 @@ int dlm_purge_locks(struct dlm_ls *ls) static struct dlm_rsb *find_purged_rsb(struct dlm_ls *ls, int bucket) { + struct rb_node *n; struct dlm_rsb *r, *r_ret = NULL; spin_lock(&ls->ls_rsbtbl[bucket].lock); - list_for_each_entry(r, &ls->ls_rsbtbl[bucket].list, res_hashchain) { + for (n = rb_first(&ls->ls_rsbtbl[bucket].keep); n; n = rb_next(n)) { + r = rb_entry(n, struct dlm_rsb, res_hashnode); if (!rsb_flag(r, RSB_LOCKS_PURGED)) continue; hold_rsb(r); diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index a1d8f1af144b..a1ea25face82 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -2,7 +2,7 @@ ******************************************************************************* ** ** Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. -** Copyright (C) 2004-2008 Red Hat, Inc. All rights reserved. +** Copyright (C) 2004-2011 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -386,12 +386,15 @@ static void threads_stop(void) dlm_lowcomms_stop(); } -static int new_lockspace(const char *name, int namelen, void **lockspace, - uint32_t flags, int lvblen) +static int new_lockspace(const char *name, const char *cluster, + uint32_t flags, int lvblen, + const struct dlm_lockspace_ops *ops, void *ops_arg, + int *ops_result, dlm_lockspace_t **lockspace) { struct dlm_ls *ls; int i, size, error; int do_unreg = 0; + int namelen = strlen(name); if (namelen > DLM_LOCKSPACE_LEN) return -EINVAL; @@ -403,8 +406,24 @@ static int new_lockspace(const char *name, int namelen, void **lockspace, return -EINVAL; if (!dlm_user_daemon_available()) { - module_put(THIS_MODULE); - return -EUNATCH; + log_print("dlm user daemon not available"); + error = -EUNATCH; + goto out; + } + + if (ops && ops_result) { + if (!dlm_config.ci_recover_callbacks) + *ops_result = -EOPNOTSUPP; + else + *ops_result = 0; + } + + if (dlm_config.ci_recover_callbacks && cluster && + strncmp(cluster, dlm_config.ci_cluster_name, DLM_LOCKSPACE_LEN)) { + log_print("dlm cluster name %s mismatch %s", + dlm_config.ci_cluster_name, cluster); + error = -EBADR; + goto out; } error = 0; @@ -442,6 +461,11 @@ static int new_lockspace(const char *name, int namelen, void **lockspace, ls->ls_flags = 0; ls->ls_scan_time = jiffies; + if (ops && dlm_config.ci_recover_callbacks) { + ls->ls_ops = ops; + ls->ls_ops_arg = ops_arg; + } + if (flags & DLM_LSFL_TIMEWARN) set_bit(LSFL_TIMEWARN, &ls->ls_flags); @@ -457,8 +481,8 @@ static int new_lockspace(const char *name, int namelen, void **lockspace, if (!ls->ls_rsbtbl) goto out_lsfree; for (i = 0; i < size; i++) { - INIT_LIST_HEAD(&ls->ls_rsbtbl[i].list); - INIT_LIST_HEAD(&ls->ls_rsbtbl[i].toss); + ls->ls_rsbtbl[i].keep.rb_node = NULL; + ls->ls_rsbtbl[i].toss.rb_node = NULL; spin_lock_init(&ls->ls_rsbtbl[i].lock); } @@ -525,6 +549,11 @@ static int new_lockspace(const char *name, int namelen, void **lockspace, if (!ls->ls_recover_buf) goto out_dirfree; + ls->ls_slot = 0; + ls->ls_num_slots = 0; + ls->ls_slots_size = 0; + ls->ls_slots = NULL; + INIT_LIST_HEAD(&ls->ls_recover_list); spin_lock_init(&ls->ls_recover_list_lock); ls->ls_recover_list_count = 0; @@ -614,8 +643,10 @@ static int new_lockspace(const char *name, int namelen, void **lockspace, return error; } -int dlm_new_lockspace(const char *name, int namelen, void **lockspace, - uint32_t flags, int lvblen) +int dlm_new_lockspace(const char *name, const char *cluster, + uint32_t flags, int lvblen, + const struct dlm_lockspace_ops *ops, void *ops_arg, + int *ops_result, dlm_lockspace_t **lockspace) { int error = 0; @@ -625,7 +656,8 @@ int dlm_new_lockspace(const char *name, int namelen, void **lockspace, if (error) goto out; - error = new_lockspace(name, namelen, lockspace, flags, lvblen); + error = new_lockspace(name, cluster, flags, lvblen, ops, ops_arg, + ops_result, lockspace); if (!error) ls_count++; if (error > 0) @@ -685,7 +717,7 @@ static int lockspace_busy(struct dlm_ls *ls, int force) static int release_lockspace(struct dlm_ls *ls, int force) { struct dlm_rsb *rsb; - struct list_head *head; + struct rb_node *n; int i, busy, rv; busy = lockspace_busy(ls, force); @@ -746,20 +778,15 @@ static int release_lockspace(struct dlm_ls *ls, int force) */ for (i = 0; i < ls->ls_rsbtbl_size; i++) { - head = &ls->ls_rsbtbl[i].list; - while (!list_empty(head)) { - rsb = list_entry(head->next, struct dlm_rsb, - res_hashchain); - - list_del(&rsb->res_hashchain); + while ((n = rb_first(&ls->ls_rsbtbl[i].keep))) { + rsb = rb_entry(n, struct dlm_rsb, res_hashnode); + rb_erase(n, &ls->ls_rsbtbl[i].keep); dlm_free_rsb(rsb); } - head = &ls->ls_rsbtbl[i].toss; - while (!list_empty(head)) { - rsb = list_entry(head->next, struct dlm_rsb, - res_hashchain); - list_del(&rsb->res_hashchain); + while ((n = rb_first(&ls->ls_rsbtbl[i].toss))) { + rsb = rb_entry(n, struct dlm_rsb, res_hashnode); + rb_erase(n, &ls->ls_rsbtbl[i].toss); dlm_free_rsb(rsb); } } diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c index 990626e7da80..0b3109ee4257 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms.c @@ -281,7 +281,7 @@ static int nodeid_to_addr(int nodeid, struct sockaddr *retaddr) } else { struct sockaddr_in6 *in6 = (struct sockaddr_in6 *) &addr; struct sockaddr_in6 *ret6 = (struct sockaddr_in6 *) retaddr; - ipv6_addr_copy(&ret6->sin6_addr, &in6->sin6_addr); + ret6->sin6_addr = in6->sin6_addr; } return 0; diff --git a/fs/dlm/member.c b/fs/dlm/member.c index b12532e553f8..862640a36d5c 100644 --- a/fs/dlm/member.c +++ b/fs/dlm/member.c @@ -1,7 +1,7 @@ /****************************************************************************** ******************************************************************************* ** -** Copyright (C) 2005-2009 Red Hat, Inc. All rights reserved. +** Copyright (C) 2005-2011 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -19,6 +19,280 @@ #include "config.h" #include "lowcomms.h" +int dlm_slots_version(struct dlm_header *h) +{ + if ((h->h_version & 0x0000FFFF) < DLM_HEADER_SLOTS) + return 0; + return 1; +} + +void dlm_slot_save(struct dlm_ls *ls, struct dlm_rcom *rc, + struct dlm_member *memb) +{ + struct rcom_config *rf = (struct rcom_config *)rc->rc_buf; + + if (!dlm_slots_version(&rc->rc_header)) + return; + + memb->slot = le16_to_cpu(rf->rf_our_slot); + memb->generation = le32_to_cpu(rf->rf_generation); +} + +void dlm_slots_copy_out(struct dlm_ls *ls, struct dlm_rcom *rc) +{ + struct dlm_slot *slot; + struct rcom_slot *ro; + int i; + + ro = (struct rcom_slot *)(rc->rc_buf + sizeof(struct rcom_config)); + + /* ls_slots array is sparse, but not rcom_slots */ + + for (i = 0; i < ls->ls_slots_size; i++) { + slot = &ls->ls_slots[i]; + if (!slot->nodeid) + continue; + ro->ro_nodeid = cpu_to_le32(slot->nodeid); + ro->ro_slot = cpu_to_le16(slot->slot); + ro++; + } +} + +#define SLOT_DEBUG_LINE 128 + +static void log_debug_slots(struct dlm_ls *ls, uint32_t gen, int num_slots, + struct rcom_slot *ro0, struct dlm_slot *array, + int array_size) +{ + char line[SLOT_DEBUG_LINE]; + int len = SLOT_DEBUG_LINE - 1; + int pos = 0; + int ret, i; + + if (!dlm_config.ci_log_debug) + return; + + memset(line, 0, sizeof(line)); + + if (array) { + for (i = 0; i < array_size; i++) { + if (!array[i].nodeid) + continue; + + ret = snprintf(line + pos, len - pos, " %d:%d", + array[i].slot, array[i].nodeid); + if (ret >= len - pos) + break; + pos += ret; + } + } else if (ro0) { + for (i = 0; i < num_slots; i++) { + ret = snprintf(line + pos, len - pos, " %d:%d", + ro0[i].ro_slot, ro0[i].ro_nodeid); + if (ret >= len - pos) + break; + pos += ret; + } + } + + log_debug(ls, "generation %u slots %d%s", gen, num_slots, line); +} + +int dlm_slots_copy_in(struct dlm_ls *ls) +{ + struct dlm_member *memb; + struct dlm_rcom *rc = ls->ls_recover_buf; + struct rcom_config *rf = (struct rcom_config *)rc->rc_buf; + struct rcom_slot *ro0, *ro; + int our_nodeid = dlm_our_nodeid(); + int i, num_slots; + uint32_t gen; + + if (!dlm_slots_version(&rc->rc_header)) + return -1; + + gen = le32_to_cpu(rf->rf_generation); + if (gen <= ls->ls_generation) { + log_error(ls, "dlm_slots_copy_in gen %u old %u", + gen, ls->ls_generation); + } + ls->ls_generation = gen; + + num_slots = le16_to_cpu(rf->rf_num_slots); + if (!num_slots) + return -1; + + ro0 = (struct rcom_slot *)(rc->rc_buf + sizeof(struct rcom_config)); + + for (i = 0, ro = ro0; i < num_slots; i++, ro++) { + ro->ro_nodeid = le32_to_cpu(ro->ro_nodeid); + ro->ro_slot = le16_to_cpu(ro->ro_slot); + } + + log_debug_slots(ls, gen, num_slots, ro0, NULL, 0); + + list_for_each_entry(memb, &ls->ls_nodes, list) { + for (i = 0, ro = ro0; i < num_slots; i++, ro++) { + if (ro->ro_nodeid != memb->nodeid) + continue; + memb->slot = ro->ro_slot; + memb->slot_prev = memb->slot; + break; + } + + if (memb->nodeid == our_nodeid) { + if (ls->ls_slot && ls->ls_slot != memb->slot) { + log_error(ls, "dlm_slots_copy_in our slot " + "changed %d %d", ls->ls_slot, + memb->slot); + return -1; + } + + if (!ls->ls_slot) + ls->ls_slot = memb->slot; + } + + if (!memb->slot) { + log_error(ls, "dlm_slots_copy_in nodeid %d no slot", + memb->nodeid); + return -1; + } + } + + return 0; +} + +/* for any nodes that do not support slots, we will not have set memb->slot + in wait_status_all(), so memb->slot will remain -1, and we will not + assign slots or set ls_num_slots here */ + +int dlm_slots_assign(struct dlm_ls *ls, int *num_slots, int *slots_size, + struct dlm_slot **slots_out, uint32_t *gen_out) +{ + struct dlm_member *memb; + struct dlm_slot *array; + int our_nodeid = dlm_our_nodeid(); + int array_size, max_slots, i; + int need = 0; + int max = 0; + int num = 0; + uint32_t gen = 0; + + /* our own memb struct will have slot -1 gen 0 */ + + list_for_each_entry(memb, &ls->ls_nodes, list) { + if (memb->nodeid == our_nodeid) { + memb->slot = ls->ls_slot; + memb->generation = ls->ls_generation; + break; + } + } + + list_for_each_entry(memb, &ls->ls_nodes, list) { + if (memb->generation > gen) + gen = memb->generation; + + /* node doesn't support slots */ + + if (memb->slot == -1) + return -1; + + /* node needs a slot assigned */ + + if (!memb->slot) + need++; + + /* node has a slot assigned */ + + num++; + + if (!max || max < memb->slot) + max = memb->slot; + + /* sanity check, once slot is assigned it shouldn't change */ + + if (memb->slot_prev && memb->slot && memb->slot_prev != memb->slot) { + log_error(ls, "nodeid %d slot changed %d %d", + memb->nodeid, memb->slot_prev, memb->slot); + return -1; + } + memb->slot_prev = memb->slot; + } + + array_size = max + need; + + array = kzalloc(array_size * sizeof(struct dlm_slot), GFP_NOFS); + if (!array) + return -ENOMEM; + + num = 0; + + /* fill in slots (offsets) that are used */ + + list_for_each_entry(memb, &ls->ls_nodes, list) { + if (!memb->slot) + continue; + + if (memb->slot > array_size) { + log_error(ls, "invalid slot number %d", memb->slot); + kfree(array); + return -1; + } + + array[memb->slot - 1].nodeid = memb->nodeid; + array[memb->slot - 1].slot = memb->slot; + num++; + } + + /* assign new slots from unused offsets */ + + list_for_each_entry(memb, &ls->ls_nodes, list) { + if (memb->slot) + continue; + + for (i = 0; i < array_size; i++) { + if (array[i].nodeid) + continue; + + memb->slot = i + 1; + memb->slot_prev = memb->slot; + array[i].nodeid = memb->nodeid; + array[i].slot = memb->slot; + num++; + + if (!ls->ls_slot && memb->nodeid == our_nodeid) + ls->ls_slot = memb->slot; + break; + } + + if (!memb->slot) { + log_error(ls, "no free slot found"); + kfree(array); + return -1; + } + } + + gen++; + + log_debug_slots(ls, gen, num, NULL, array, array_size); + + max_slots = (dlm_config.ci_buffer_size - sizeof(struct dlm_rcom) - + sizeof(struct rcom_config)) / sizeof(struct rcom_slot); + + if (num > max_slots) { + log_error(ls, "num_slots %d exceeds max_slots %d", + num, max_slots); + kfree(array); + return -1; + } + + *gen_out = gen; + *slots_out = array; + *slots_size = array_size; + *num_slots = num; + return 0; +} + static void add_ordered_member(struct dlm_ls *ls, struct dlm_member *new) { struct dlm_member *memb = NULL; @@ -43,59 +317,51 @@ static void add_ordered_member(struct dlm_ls *ls, struct dlm_member *new) } } -static int dlm_add_member(struct dlm_ls *ls, int nodeid) +static int dlm_add_member(struct dlm_ls *ls, struct dlm_config_node *node) { struct dlm_member *memb; - int w, error; + int error; memb = kzalloc(sizeof(struct dlm_member), GFP_NOFS); if (!memb) return -ENOMEM; - w = dlm_node_weight(ls->ls_name, nodeid); - if (w < 0) { - kfree(memb); - return w; - } - - error = dlm_lowcomms_connect_node(nodeid); + error = dlm_lowcomms_connect_node(node->nodeid); if (error < 0) { kfree(memb); return error; } - memb->nodeid = nodeid; - memb->weight = w; + memb->nodeid = node->nodeid; + memb->weight = node->weight; + memb->comm_seq = node->comm_seq; add_ordered_member(ls, memb); ls->ls_num_nodes++; return 0; } -static void dlm_remove_member(struct dlm_ls *ls, struct dlm_member *memb) -{ - list_move(&memb->list, &ls->ls_nodes_gone); - ls->ls_num_nodes--; -} - -int dlm_is_member(struct dlm_ls *ls, int nodeid) +static struct dlm_member *find_memb(struct list_head *head, int nodeid) { struct dlm_member *memb; - list_for_each_entry(memb, &ls->ls_nodes, list) { + list_for_each_entry(memb, head, list) { if (memb->nodeid == nodeid) - return 1; + return memb; } + return NULL; +} + +int dlm_is_member(struct dlm_ls *ls, int nodeid) +{ + if (find_memb(&ls->ls_nodes, nodeid)) + return 1; return 0; } int dlm_is_removed(struct dlm_ls *ls, int nodeid) { - struct dlm_member *memb; - - list_for_each_entry(memb, &ls->ls_nodes_gone, list) { - if (memb->nodeid == nodeid) - return 1; - } + if (find_memb(&ls->ls_nodes_gone, nodeid)) + return 1; return 0; } @@ -176,7 +442,7 @@ static int ping_members(struct dlm_ls *ls) error = dlm_recovery_stopped(ls); if (error) break; - error = dlm_rcom_status(ls, memb->nodeid); + error = dlm_rcom_status(ls, memb->nodeid, 0); if (error) break; } @@ -186,10 +452,88 @@ static int ping_members(struct dlm_ls *ls) return error; } +static void dlm_lsop_recover_prep(struct dlm_ls *ls) +{ + if (!ls->ls_ops || !ls->ls_ops->recover_prep) + return; + ls->ls_ops->recover_prep(ls->ls_ops_arg); +} + +static void dlm_lsop_recover_slot(struct dlm_ls *ls, struct dlm_member *memb) +{ + struct dlm_slot slot; + uint32_t seq; + int error; + + if (!ls->ls_ops || !ls->ls_ops->recover_slot) + return; + + /* if there is no comms connection with this node + or the present comms connection is newer + than the one when this member was added, then + we consider the node to have failed (versus + being removed due to dlm_release_lockspace) */ + + error = dlm_comm_seq(memb->nodeid, &seq); + + if (!error && seq == memb->comm_seq) + return; + + slot.nodeid = memb->nodeid; + slot.slot = memb->slot; + + ls->ls_ops->recover_slot(ls->ls_ops_arg, &slot); +} + +void dlm_lsop_recover_done(struct dlm_ls *ls) +{ + struct dlm_member *memb; + struct dlm_slot *slots; + int i, num; + + if (!ls->ls_ops || !ls->ls_ops->recover_done) + return; + + num = ls->ls_num_nodes; + + slots = kzalloc(num * sizeof(struct dlm_slot), GFP_KERNEL); + if (!slots) + return; + + i = 0; + list_for_each_entry(memb, &ls->ls_nodes, list) { + if (i == num) { + log_error(ls, "dlm_lsop_recover_done bad num %d", num); + goto out; + } + slots[i].nodeid = memb->nodeid; + slots[i].slot = memb->slot; + i++; + } + + ls->ls_ops->recover_done(ls->ls_ops_arg, slots, num, + ls->ls_slot, ls->ls_generation); + out: + kfree(slots); +} + +static struct dlm_config_node *find_config_node(struct dlm_recover *rv, + int nodeid) +{ + int i; + + for (i = 0; i < rv->nodes_count; i++) { + if (rv->nodes[i].nodeid == nodeid) + return &rv->nodes[i]; + } + return NULL; +} + int dlm_recover_members(struct dlm_ls *ls, struct dlm_recover *rv, int *neg_out) { struct dlm_member *memb, *safe; - int i, error, found, pos = 0, neg = 0, low = -1; + struct dlm_config_node *node; + int i, error, neg = 0, low = -1; /* previously removed members that we've not finished removing need to count as a negative change so the "neg" recovery steps will happen */ @@ -202,46 +546,32 @@ int dlm_recover_members(struct dlm_ls *ls, struct dlm_recover *rv, int *neg_out) /* move departed members from ls_nodes to ls_nodes_gone */ list_for_each_entry_safe(memb, safe, &ls->ls_nodes, list) { - found = 0; - for (i = 0; i < rv->node_count; i++) { - if (memb->nodeid == rv->nodeids[i]) { - found = 1; - break; - } - } + node = find_config_node(rv, memb->nodeid); + if (node && !node->new) + continue; - if (!found) { - neg++; - dlm_remove_member(ls, memb); + if (!node) { log_debug(ls, "remove member %d", memb->nodeid); + } else { + /* removed and re-added */ + log_debug(ls, "remove member %d comm_seq %u %u", + memb->nodeid, memb->comm_seq, node->comm_seq); } - } - - /* Add an entry to ls_nodes_gone for members that were removed and - then added again, so that previous state for these nodes will be - cleared during recovery. */ - - for (i = 0; i < rv->new_count; i++) { - if (!dlm_is_member(ls, rv->new[i])) - continue; - log_debug(ls, "new nodeid %d is a re-added member", rv->new[i]); - memb = kzalloc(sizeof(struct dlm_member), GFP_NOFS); - if (!memb) - return -ENOMEM; - memb->nodeid = rv->new[i]; - list_add_tail(&memb->list, &ls->ls_nodes_gone); neg++; + list_move(&memb->list, &ls->ls_nodes_gone); + ls->ls_num_nodes--; + dlm_lsop_recover_slot(ls, memb); } /* add new members to ls_nodes */ - for (i = 0; i < rv->node_count; i++) { - if (dlm_is_member(ls, rv->nodeids[i])) + for (i = 0; i < rv->nodes_count; i++) { + node = &rv->nodes[i]; + if (dlm_is_member(ls, node->nodeid)) continue; - dlm_add_member(ls, rv->nodeids[i]); - pos++; - log_debug(ls, "add member %d", rv->nodeids[i]); + dlm_add_member(ls, node); + log_debug(ls, "add member %d", node->nodeid); } list_for_each_entry(memb, &ls->ls_nodes, list) { @@ -251,7 +581,6 @@ int dlm_recover_members(struct dlm_ls *ls, struct dlm_recover *rv, int *neg_out) ls->ls_low_nodeid = low; make_member_array(ls); - dlm_set_recover_status(ls, DLM_RS_NODES); *neg_out = neg; error = ping_members(ls); @@ -261,12 +590,8 @@ int dlm_recover_members(struct dlm_ls *ls, struct dlm_recover *rv, int *neg_out) ls->ls_members_result = error; complete(&ls->ls_members_done); } - if (error) - goto out; - error = dlm_recover_members_wait(ls); - out: - log_debug(ls, "total members %d error %d", ls->ls_num_nodes, error); + log_debug(ls, "dlm_recover_members %d nodes", ls->ls_num_nodes); return error; } @@ -327,26 +652,35 @@ int dlm_ls_stop(struct dlm_ls *ls) */ dlm_recoverd_suspend(ls); + + spin_lock(&ls->ls_recover_lock); + kfree(ls->ls_slots); + ls->ls_slots = NULL; + ls->ls_num_slots = 0; + ls->ls_slots_size = 0; ls->ls_recover_status = 0; + spin_unlock(&ls->ls_recover_lock); + dlm_recoverd_resume(ls); if (!ls->ls_recover_begin) ls->ls_recover_begin = jiffies; + + dlm_lsop_recover_prep(ls); return 0; } int dlm_ls_start(struct dlm_ls *ls) { struct dlm_recover *rv = NULL, *rv_old; - int *ids = NULL, *new = NULL; - int error, ids_count = 0, new_count = 0; + struct dlm_config_node *nodes; + int error, count; rv = kzalloc(sizeof(struct dlm_recover), GFP_NOFS); if (!rv) return -ENOMEM; - error = dlm_nodeid_list(ls->ls_name, &ids, &ids_count, - &new, &new_count); + error = dlm_config_nodes(ls->ls_name, &nodes, &count); if (error < 0) goto fail; @@ -361,10 +695,8 @@ int dlm_ls_start(struct dlm_ls *ls) goto fail; } - rv->nodeids = ids; - rv->node_count = ids_count; - rv->new = new; - rv->new_count = new_count; + rv->nodes = nodes; + rv->nodes_count = count; rv->seq = ++ls->ls_recover_seq; rv_old = ls->ls_recover_args; ls->ls_recover_args = rv; @@ -372,9 +704,8 @@ int dlm_ls_start(struct dlm_ls *ls) if (rv_old) { log_error(ls, "unused recovery %llx %d", - (unsigned long long)rv_old->seq, rv_old->node_count); - kfree(rv_old->nodeids); - kfree(rv_old->new); + (unsigned long long)rv_old->seq, rv_old->nodes_count); + kfree(rv_old->nodes); kfree(rv_old); } @@ -383,8 +714,7 @@ int dlm_ls_start(struct dlm_ls *ls) fail: kfree(rv); - kfree(ids); - kfree(new); + kfree(nodes); return error; } diff --git a/fs/dlm/member.h b/fs/dlm/member.h index 7a26fca1e0b5..3deb70661c69 100644 --- a/fs/dlm/member.h +++ b/fs/dlm/member.h @@ -1,7 +1,7 @@ /****************************************************************************** ******************************************************************************* ** -** Copyright (C) 2005-2008 Red Hat, Inc. All rights reserved. +** Copyright (C) 2005-2011 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -20,6 +20,14 @@ void dlm_clear_members_gone(struct dlm_ls *ls); int dlm_recover_members(struct dlm_ls *ls, struct dlm_recover *rv,int *neg_out); int dlm_is_removed(struct dlm_ls *ls, int nodeid); int dlm_is_member(struct dlm_ls *ls, int nodeid); +int dlm_slots_version(struct dlm_header *h); +void dlm_slot_save(struct dlm_ls *ls, struct dlm_rcom *rc, + struct dlm_member *memb); +void dlm_slots_copy_out(struct dlm_ls *ls, struct dlm_rcom *rc); +int dlm_slots_copy_in(struct dlm_ls *ls); +int dlm_slots_assign(struct dlm_ls *ls, int *num_slots, int *slots_size, + struct dlm_slot **slots_out, uint32_t *gen_out); +void dlm_lsop_recover_done(struct dlm_ls *ls); #endif /* __MEMBER_DOT_H__ */ diff --git a/fs/dlm/rcom.c b/fs/dlm/rcom.c index f10a50f24e8f..ac5c616c9696 100644 --- a/fs/dlm/rcom.c +++ b/fs/dlm/rcom.c @@ -23,6 +23,7 @@ #include "memory.h" #include "lock.h" #include "util.h" +#include "member.h" static int rcom_response(struct dlm_ls *ls) @@ -72,20 +73,30 @@ static void send_rcom(struct dlm_ls *ls, struct dlm_mhandle *mh, dlm_lowcomms_commit_buffer(mh); } +static void set_rcom_status(struct dlm_ls *ls, struct rcom_status *rs, + uint32_t flags) +{ + rs->rs_flags = cpu_to_le32(flags); +} + /* When replying to a status request, a node also sends back its configuration values. The requesting node then checks that the remote node is configured the same way as itself. */ -static void make_config(struct dlm_ls *ls, struct rcom_config *rf) +static void set_rcom_config(struct dlm_ls *ls, struct rcom_config *rf, + uint32_t num_slots) { rf->rf_lvblen = cpu_to_le32(ls->ls_lvblen); rf->rf_lsflags = cpu_to_le32(ls->ls_exflags); + + rf->rf_our_slot = cpu_to_le16(ls->ls_slot); + rf->rf_num_slots = cpu_to_le16(num_slots); + rf->rf_generation = cpu_to_le32(ls->ls_generation); } -static int check_config(struct dlm_ls *ls, struct dlm_rcom *rc, int nodeid) +static int check_rcom_config(struct dlm_ls *ls, struct dlm_rcom *rc, int nodeid) { struct rcom_config *rf = (struct rcom_config *) rc->rc_buf; - size_t conf_size = sizeof(struct dlm_rcom) + sizeof(struct rcom_config); if ((rc->rc_header.h_version & 0xFFFF0000) != DLM_HEADER_MAJOR) { log_error(ls, "version mismatch: %x nodeid %d: %x", @@ -94,12 +105,6 @@ static int check_config(struct dlm_ls *ls, struct dlm_rcom *rc, int nodeid) return -EPROTO; } - if (rc->rc_header.h_length < conf_size) { - log_error(ls, "config too short: %d nodeid %d", - rc->rc_header.h_length, nodeid); - return -EPROTO; - } - if (le32_to_cpu(rf->rf_lvblen) != ls->ls_lvblen || le32_to_cpu(rf->rf_lsflags) != ls->ls_exflags) { log_error(ls, "config mismatch: %d,%x nodeid %d: %d,%x", @@ -127,7 +132,18 @@ static void disallow_sync_reply(struct dlm_ls *ls) spin_unlock(&ls->ls_rcom_spin); } -int dlm_rcom_status(struct dlm_ls *ls, int nodeid) +/* + * low nodeid gathers one slot value at a time from each node. + * it sets need_slots=0, and saves rf_our_slot returned from each + * rcom_config. + * + * other nodes gather all slot values at once from the low nodeid. + * they set need_slots=1, and ignore the rf_our_slot returned from each + * rcom_config. they use the rf_num_slots returned from the low + * node's rcom_config. + */ + +int dlm_rcom_status(struct dlm_ls *ls, int nodeid, uint32_t status_flags) { struct dlm_rcom *rc; struct dlm_mhandle *mh; @@ -141,10 +157,13 @@ int dlm_rcom_status(struct dlm_ls *ls, int nodeid) goto out; } - error = create_rcom(ls, nodeid, DLM_RCOM_STATUS, 0, &rc, &mh); + error = create_rcom(ls, nodeid, DLM_RCOM_STATUS, + sizeof(struct rcom_status), &rc, &mh); if (error) goto out; + set_rcom_status(ls, (struct rcom_status *)rc->rc_buf, status_flags); + allow_sync_reply(ls, &rc->rc_id); memset(ls->ls_recover_buf, 0, dlm_config.ci_buffer_size); @@ -161,8 +180,11 @@ int dlm_rcom_status(struct dlm_ls *ls, int nodeid) /* we pretend the remote lockspace exists with 0 status */ log_debug(ls, "remote node %d not ready", nodeid); rc->rc_result = 0; - } else - error = check_config(ls, rc, nodeid); + error = 0; + } else { + error = check_rcom_config(ls, rc, nodeid); + } + /* the caller looks at rc_result for the remote recovery status */ out: return error; @@ -172,17 +194,60 @@ static void receive_rcom_status(struct dlm_ls *ls, struct dlm_rcom *rc_in) { struct dlm_rcom *rc; struct dlm_mhandle *mh; - int error, nodeid = rc_in->rc_header.h_nodeid; + struct rcom_status *rs; + uint32_t status; + int nodeid = rc_in->rc_header.h_nodeid; + int len = sizeof(struct rcom_config); + int num_slots = 0; + int error; + + if (!dlm_slots_version(&rc_in->rc_header)) { + status = dlm_recover_status(ls); + goto do_create; + } + + rs = (struct rcom_status *)rc_in->rc_buf; + if (!(rs->rs_flags & DLM_RSF_NEED_SLOTS)) { + status = dlm_recover_status(ls); + goto do_create; + } + + spin_lock(&ls->ls_recover_lock); + status = ls->ls_recover_status; + num_slots = ls->ls_num_slots; + spin_unlock(&ls->ls_recover_lock); + len += num_slots * sizeof(struct rcom_slot); + + do_create: error = create_rcom(ls, nodeid, DLM_RCOM_STATUS_REPLY, - sizeof(struct rcom_config), &rc, &mh); + len, &rc, &mh); if (error) return; + rc->rc_id = rc_in->rc_id; rc->rc_seq_reply = rc_in->rc_seq; - rc->rc_result = dlm_recover_status(ls); - make_config(ls, (struct rcom_config *) rc->rc_buf); + rc->rc_result = status; + + set_rcom_config(ls, (struct rcom_config *)rc->rc_buf, num_slots); + + if (!num_slots) + goto do_send; + + spin_lock(&ls->ls_recover_lock); + if (ls->ls_num_slots != num_slots) { + spin_unlock(&ls->ls_recover_lock); + log_debug(ls, "receive_rcom_status num_slots %d to %d", + num_slots, ls->ls_num_slots); + rc->rc_result = 0; + set_rcom_config(ls, (struct rcom_config *)rc->rc_buf, 0); + goto do_send; + } + + dlm_slots_copy_out(ls, rc); + spin_unlock(&ls->ls_recover_lock); + do_send: send_rcom(ls, mh, rc); } diff --git a/fs/dlm/rcom.h b/fs/dlm/rcom.h index b09abd29ba38..206723ab744d 100644 --- a/fs/dlm/rcom.h +++ b/fs/dlm/rcom.h @@ -14,7 +14,7 @@ #ifndef __RCOM_DOT_H__ #define __RCOM_DOT_H__ -int dlm_rcom_status(struct dlm_ls *ls, int nodeid); +int dlm_rcom_status(struct dlm_ls *ls, int nodeid, uint32_t status_flags); int dlm_rcom_names(struct dlm_ls *ls, int nodeid, char *last_name,int last_len); int dlm_send_rcom_lookup(struct dlm_rsb *r, int dir_nodeid); int dlm_send_rcom_lock(struct dlm_rsb *r, struct dlm_lkb *lkb); diff --git a/fs/dlm/recover.c b/fs/dlm/recover.c index 14638235f7b2..34d5adf1fce7 100644 --- a/fs/dlm/recover.c +++ b/fs/dlm/recover.c @@ -85,14 +85,20 @@ uint32_t dlm_recover_status(struct dlm_ls *ls) return status; } +static void _set_recover_status(struct dlm_ls *ls, uint32_t status) +{ + ls->ls_recover_status |= status; +} + void dlm_set_recover_status(struct dlm_ls *ls, uint32_t status) { spin_lock(&ls->ls_recover_lock); - ls->ls_recover_status |= status; + _set_recover_status(ls, status); spin_unlock(&ls->ls_recover_lock); } -static int wait_status_all(struct dlm_ls *ls, uint32_t wait_status) +static int wait_status_all(struct dlm_ls *ls, uint32_t wait_status, + int save_slots) { struct dlm_rcom *rc = ls->ls_recover_buf; struct dlm_member *memb; @@ -106,10 +112,13 @@ static int wait_status_all(struct dlm_ls *ls, uint32_t wait_status) goto out; } - error = dlm_rcom_status(ls, memb->nodeid); + error = dlm_rcom_status(ls, memb->nodeid, 0); if (error) goto out; + if (save_slots) + dlm_slot_save(ls, rc, memb); + if (rc->rc_result & wait_status) break; if (delay < 1000) @@ -121,7 +130,8 @@ static int wait_status_all(struct dlm_ls *ls, uint32_t wait_status) return error; } -static int wait_status_low(struct dlm_ls *ls, uint32_t wait_status) +static int wait_status_low(struct dlm_ls *ls, uint32_t wait_status, + uint32_t status_flags) { struct dlm_rcom *rc = ls->ls_recover_buf; int error = 0, delay = 0, nodeid = ls->ls_low_nodeid; @@ -132,7 +142,7 @@ static int wait_status_low(struct dlm_ls *ls, uint32_t wait_status) goto out; } - error = dlm_rcom_status(ls, nodeid); + error = dlm_rcom_status(ls, nodeid, status_flags); if (error) break; @@ -152,18 +162,56 @@ static int wait_status(struct dlm_ls *ls, uint32_t status) int error; if (ls->ls_low_nodeid == dlm_our_nodeid()) { - error = wait_status_all(ls, status); + error = wait_status_all(ls, status, 0); if (!error) dlm_set_recover_status(ls, status_all); } else - error = wait_status_low(ls, status_all); + error = wait_status_low(ls, status_all, 0); return error; } int dlm_recover_members_wait(struct dlm_ls *ls) { - return wait_status(ls, DLM_RS_NODES); + struct dlm_member *memb; + struct dlm_slot *slots; + int num_slots, slots_size; + int error, rv; + uint32_t gen; + + list_for_each_entry(memb, &ls->ls_nodes, list) { + memb->slot = -1; + memb->generation = 0; + } + + if (ls->ls_low_nodeid == dlm_our_nodeid()) { + error = wait_status_all(ls, DLM_RS_NODES, 1); + if (error) + goto out; + + /* slots array is sparse, slots_size may be > num_slots */ + + rv = dlm_slots_assign(ls, &num_slots, &slots_size, &slots, &gen); + if (!rv) { + spin_lock(&ls->ls_recover_lock); + _set_recover_status(ls, DLM_RS_NODES_ALL); + ls->ls_num_slots = num_slots; + ls->ls_slots_size = slots_size; + ls->ls_slots = slots; + ls->ls_generation = gen; + spin_unlock(&ls->ls_recover_lock); + } else { + dlm_set_recover_status(ls, DLM_RS_NODES_ALL); + } + } else { + error = wait_status_low(ls, DLM_RS_NODES_ALL, DLM_RSF_NEED_SLOTS); + if (error) + goto out; + + dlm_slots_copy_in(ls); + } + out: + return error; } int dlm_recover_directory_wait(struct dlm_ls *ls) @@ -542,8 +590,6 @@ int dlm_recover_locks(struct dlm_ls *ls) out: if (error) recover_list_clear(ls); - else - dlm_set_recover_status(ls, DLM_RS_LOCKS); return error; } @@ -715,6 +761,7 @@ void dlm_recover_rsbs(struct dlm_ls *ls) int dlm_create_root_list(struct dlm_ls *ls) { + struct rb_node *n; struct dlm_rsb *r; int i, error = 0; @@ -727,7 +774,8 @@ int dlm_create_root_list(struct dlm_ls *ls) for (i = 0; i < ls->ls_rsbtbl_size; i++) { spin_lock(&ls->ls_rsbtbl[i].lock); - list_for_each_entry(r, &ls->ls_rsbtbl[i].list, res_hashchain) { + for (n = rb_first(&ls->ls_rsbtbl[i].keep); n; n = rb_next(n)) { + r = rb_entry(n, struct dlm_rsb, res_hashnode); list_add(&r->res_root_list, &ls->ls_root_list); dlm_hold_rsb(r); } @@ -741,7 +789,8 @@ int dlm_create_root_list(struct dlm_ls *ls) continue; } - list_for_each_entry(r, &ls->ls_rsbtbl[i].toss, res_hashchain) { + for (n = rb_first(&ls->ls_rsbtbl[i].toss); n; n = rb_next(n)) { + r = rb_entry(n, struct dlm_rsb, res_hashnode); list_add(&r->res_root_list, &ls->ls_root_list); dlm_hold_rsb(r); } @@ -771,16 +820,18 @@ void dlm_release_root_list(struct dlm_ls *ls) void dlm_clear_toss_list(struct dlm_ls *ls) { - struct dlm_rsb *r, *safe; + struct rb_node *n, *next; + struct dlm_rsb *rsb; int i; for (i = 0; i < ls->ls_rsbtbl_size; i++) { spin_lock(&ls->ls_rsbtbl[i].lock); - list_for_each_entry_safe(r, safe, &ls->ls_rsbtbl[i].toss, - res_hashchain) { - if (dlm_no_directory(ls) || !is_master(r)) { - list_del(&r->res_hashchain); - dlm_free_rsb(r); + for (n = rb_first(&ls->ls_rsbtbl[i].toss); n; n = next) { + next = rb_next(n);; + rsb = rb_entry(n, struct dlm_rsb, res_hashnode); + if (dlm_no_directory(ls) || !is_master(rsb)) { + rb_erase(n, &ls->ls_rsbtbl[i].toss); + dlm_free_rsb(rsb); } } spin_unlock(&ls->ls_rsbtbl[i].lock); diff --git a/fs/dlm/recoverd.c b/fs/dlm/recoverd.c index 774da3cf92c6..3780caf7ae0c 100644 --- a/fs/dlm/recoverd.c +++ b/fs/dlm/recoverd.c @@ -2,7 +2,7 @@ ******************************************************************************* ** ** Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. -** Copyright (C) 2004-2007 Red Hat, Inc. All rights reserved. +** Copyright (C) 2004-2011 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -54,7 +54,7 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) unsigned long start; int error, neg = 0; - log_debug(ls, "recover %llx", (unsigned long long)rv->seq); + log_debug(ls, "dlm_recover %llx", (unsigned long long)rv->seq); mutex_lock(&ls->ls_recoverd_active); @@ -76,14 +76,22 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) /* * Add or remove nodes from the lockspace's ls_nodes list. - * Also waits for all nodes to complete dlm_recover_members. */ error = dlm_recover_members(ls, rv, &neg); if (error) { - log_debug(ls, "recover_members failed %d", error); + log_debug(ls, "dlm_recover_members error %d", error); goto fail; } + + dlm_set_recover_status(ls, DLM_RS_NODES); + + error = dlm_recover_members_wait(ls); + if (error) { + log_debug(ls, "dlm_recover_members_wait error %d", error); + goto fail; + } + start = jiffies; /* @@ -93,17 +101,15 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) error = dlm_recover_directory(ls); if (error) { - log_debug(ls, "recover_directory failed %d", error); + log_debug(ls, "dlm_recover_directory error %d", error); goto fail; } - /* - * Wait for all nodes to complete directory rebuild. - */ + dlm_set_recover_status(ls, DLM_RS_DIR); error = dlm_recover_directory_wait(ls); if (error) { - log_debug(ls, "recover_directory_wait failed %d", error); + log_debug(ls, "dlm_recover_directory_wait error %d", error); goto fail; } @@ -133,7 +139,7 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) error = dlm_recover_masters(ls); if (error) { - log_debug(ls, "recover_masters failed %d", error); + log_debug(ls, "dlm_recover_masters error %d", error); goto fail; } @@ -143,13 +149,15 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) error = dlm_recover_locks(ls); if (error) { - log_debug(ls, "recover_locks failed %d", error); + log_debug(ls, "dlm_recover_locks error %d", error); goto fail; } + dlm_set_recover_status(ls, DLM_RS_LOCKS); + error = dlm_recover_locks_wait(ls); if (error) { - log_debug(ls, "recover_locks_wait failed %d", error); + log_debug(ls, "dlm_recover_locks_wait error %d", error); goto fail; } @@ -170,7 +178,7 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) error = dlm_recover_locks_wait(ls); if (error) { - log_debug(ls, "recover_locks_wait failed %d", error); + log_debug(ls, "dlm_recover_locks_wait error %d", error); goto fail; } } @@ -186,9 +194,10 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) dlm_purge_requestqueue(ls); dlm_set_recover_status(ls, DLM_RS_DONE); + error = dlm_recover_done_wait(ls); if (error) { - log_debug(ls, "recover_done_wait failed %d", error); + log_debug(ls, "dlm_recover_done_wait error %d", error); goto fail; } @@ -200,34 +209,35 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) error = enable_locking(ls, rv->seq); if (error) { - log_debug(ls, "enable_locking failed %d", error); + log_debug(ls, "enable_locking error %d", error); goto fail; } error = dlm_process_requestqueue(ls); if (error) { - log_debug(ls, "process_requestqueue failed %d", error); + log_debug(ls, "dlm_process_requestqueue error %d", error); goto fail; } error = dlm_recover_waiters_post(ls); if (error) { - log_debug(ls, "recover_waiters_post failed %d", error); + log_debug(ls, "dlm_recover_waiters_post error %d", error); goto fail; } dlm_grant_after_purge(ls); - log_debug(ls, "recover %llx done: %u ms", - (unsigned long long)rv->seq, + log_debug(ls, "dlm_recover %llx generation %u done: %u ms", + (unsigned long long)rv->seq, ls->ls_generation, jiffies_to_msecs(jiffies - start)); mutex_unlock(&ls->ls_recoverd_active); + dlm_lsop_recover_done(ls); return 0; fail: dlm_release_root_list(ls); - log_debug(ls, "recover %llx error %d", + log_debug(ls, "dlm_recover %llx error %d", (unsigned long long)rv->seq, error); mutex_unlock(&ls->ls_recoverd_active); return error; @@ -250,8 +260,7 @@ static void do_ls_recovery(struct dlm_ls *ls) if (rv) { ls_recover(ls, rv); - kfree(rv->nodeids); - kfree(rv->new); + kfree(rv->nodes); kfree(rv); } } diff --git a/fs/dlm/user.c b/fs/dlm/user.c index d8ea60756403..eb4ed9ba3098 100644 --- a/fs/dlm/user.c +++ b/fs/dlm/user.c @@ -392,8 +392,9 @@ static int device_create_lockspace(struct dlm_lspace_params *params) if (!capable(CAP_SYS_ADMIN)) return -EPERM; - error = dlm_new_lockspace(params->name, strlen(params->name), - &lockspace, params->flags, DLM_USER_LVB_LEN); + error = dlm_new_lockspace(params->name, NULL, params->flags, + DLM_USER_LVB_LEN, NULL, NULL, NULL, + &lockspace); if (error) return error; diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index 58609bde3b9f..2a834255c75d 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -967,7 +967,7 @@ static void ecryptfs_set_default_crypt_stat_vals( /** * ecryptfs_new_file_context - * @ecryptfs_dentry: The eCryptfs dentry + * @ecryptfs_inode: The eCryptfs inode * * If the crypto context for the file has not yet been established, * this is where we do that. Establishing a new crypto context @@ -984,13 +984,13 @@ static void ecryptfs_set_default_crypt_stat_vals( * * Returns zero on success; non-zero otherwise */ -int ecryptfs_new_file_context(struct dentry *ecryptfs_dentry) +int ecryptfs_new_file_context(struct inode *ecryptfs_inode) { struct ecryptfs_crypt_stat *crypt_stat = - &ecryptfs_inode_to_private(ecryptfs_dentry->d_inode)->crypt_stat; + &ecryptfs_inode_to_private(ecryptfs_inode)->crypt_stat; struct ecryptfs_mount_crypt_stat *mount_crypt_stat = &ecryptfs_superblock_to_private( - ecryptfs_dentry->d_sb)->mount_crypt_stat; + ecryptfs_inode->i_sb)->mount_crypt_stat; int cipher_name_len; int rc = 0; @@ -1299,12 +1299,12 @@ static int ecryptfs_write_headers_virt(char *page_virt, size_t max, } static int -ecryptfs_write_metadata_to_contents(struct dentry *ecryptfs_dentry, +ecryptfs_write_metadata_to_contents(struct inode *ecryptfs_inode, char *virt, size_t virt_len) { int rc; - rc = ecryptfs_write_lower(ecryptfs_dentry->d_inode, virt, + rc = ecryptfs_write_lower(ecryptfs_inode, virt, 0, virt_len); if (rc < 0) printk(KERN_ERR "%s: Error attempting to write header " @@ -1338,7 +1338,8 @@ static unsigned long ecryptfs_get_zeroed_pages(gfp_t gfp_mask, /** * ecryptfs_write_metadata - * @ecryptfs_dentry: The eCryptfs dentry + * @ecryptfs_dentry: The eCryptfs dentry, which should be negative + * @ecryptfs_inode: The newly created eCryptfs inode * * Write the file headers out. This will likely involve a userspace * callout, in which the session key is encrypted with one or more @@ -1348,10 +1349,11 @@ static unsigned long ecryptfs_get_zeroed_pages(gfp_t gfp_mask, * * Returns zero on success; non-zero on error */ -int ecryptfs_write_metadata(struct dentry *ecryptfs_dentry) +int ecryptfs_write_metadata(struct dentry *ecryptfs_dentry, + struct inode *ecryptfs_inode) { struct ecryptfs_crypt_stat *crypt_stat = - &ecryptfs_inode_to_private(ecryptfs_dentry->d_inode)->crypt_stat; + &ecryptfs_inode_to_private(ecryptfs_inode)->crypt_stat; unsigned int order; char *virt; size_t virt_len; @@ -1391,7 +1393,7 @@ int ecryptfs_write_metadata(struct dentry *ecryptfs_dentry) rc = ecryptfs_write_metadata_to_xattr(ecryptfs_dentry, virt, size); else - rc = ecryptfs_write_metadata_to_contents(ecryptfs_dentry, virt, + rc = ecryptfs_write_metadata_to_contents(ecryptfs_inode, virt, virt_len); if (rc) { printk(KERN_ERR "%s: Error writing metadata out to lower file; " @@ -1943,7 +1945,7 @@ static unsigned char *portable_filename_chars = ("-.0123456789ABCD" /* We could either offset on every reverse map or just pad some 0x00's * at the front here */ -static const unsigned char filename_rev_map[] = { +static const unsigned char filename_rev_map[256] = { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 7 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 15 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 23 */ @@ -1959,7 +1961,7 @@ static const unsigned char filename_rev_map[] = { 0x00, 0x26, 0x27, 0x28, 0x29, 0x2A, 0x2B, 0x2C, /* 103 */ 0x2D, 0x2E, 0x2F, 0x30, 0x31, 0x32, 0x33, 0x34, /* 111 */ 0x35, 0x36, 0x37, 0x38, 0x39, 0x3A, 0x3B, 0x3C, /* 119 */ - 0x3D, 0x3E, 0x3F + 0x3D, 0x3E, 0x3F /* 123 - 255 initialized to 0x00 */ }; /** diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index 54481a3b2c79..a9f29b12fbf2 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -584,9 +584,10 @@ int ecryptfs_init_crypt_ctx(struct ecryptfs_crypt_stat *crypt_stat); int ecryptfs_write_inode_size_to_metadata(struct inode *ecryptfs_inode); int ecryptfs_encrypt_page(struct page *page); int ecryptfs_decrypt_page(struct page *page); -int ecryptfs_write_metadata(struct dentry *ecryptfs_dentry); +int ecryptfs_write_metadata(struct dentry *ecryptfs_dentry, + struct inode *ecryptfs_inode); int ecryptfs_read_metadata(struct dentry *ecryptfs_dentry); -int ecryptfs_new_file_context(struct dentry *ecryptfs_dentry); +int ecryptfs_new_file_context(struct inode *ecryptfs_inode); void ecryptfs_write_crypt_stat_flags(char *page_virt, struct ecryptfs_crypt_stat *crypt_stat, size_t *written); diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index c6ac98cf9baa..d3f95f941c47 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -139,6 +139,27 @@ out: return rc; } +static void ecryptfs_vma_close(struct vm_area_struct *vma) +{ + filemap_write_and_wait(vma->vm_file->f_mapping); +} + +static const struct vm_operations_struct ecryptfs_file_vm_ops = { + .close = ecryptfs_vma_close, + .fault = filemap_fault, +}; + +static int ecryptfs_file_mmap(struct file *file, struct vm_area_struct *vma) +{ + int rc; + + rc = generic_file_mmap(file, vma); + if (!rc) + vma->vm_ops = &ecryptfs_file_vm_ops; + + return rc; +} + struct kmem_cache *ecryptfs_file_info_cache; /** @@ -349,7 +370,7 @@ const struct file_operations ecryptfs_main_fops = { #ifdef CONFIG_COMPAT .compat_ioctl = ecryptfs_compat_ioctl, #endif - .mmap = generic_file_mmap, + .mmap = ecryptfs_file_mmap, .open = ecryptfs_open, .flush = ecryptfs_flush, .release = ecryptfs_release, diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index a36d327f1521..19a8ca4ab1dd 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -144,24 +144,6 @@ static int ecryptfs_interpose(struct dentry *lower_dentry, } /** - * ecryptfs_create_underlying_file - * @lower_dir_inode: inode of the parent in the lower fs of the new file - * @dentry: New file's dentry - * @mode: The mode of the new file - * - * Creates the file in the lower file system. - * - * Returns zero on success; non-zero on error condition - */ -static int -ecryptfs_create_underlying_file(struct inode *lower_dir_inode, - struct dentry *dentry, int mode) -{ - struct dentry *lower_dentry = ecryptfs_dentry_to_lower(dentry); - return vfs_create(lower_dir_inode, lower_dentry, mode, NULL); -} - -/** * ecryptfs_do_create * @directory_inode: inode of the new file's dentry's parent in ecryptfs * @ecryptfs_dentry: New file's dentry in ecryptfs @@ -172,43 +154,42 @@ ecryptfs_create_underlying_file(struct inode *lower_dir_inode, * it. It will also update the eCryptfs directory inode to mimic the * stat of the lower directory inode. * - * Returns zero on success; non-zero on error condition + * Returns the new eCryptfs inode on success; an ERR_PTR on error condition */ -static int +static struct inode * ecryptfs_do_create(struct inode *directory_inode, - struct dentry *ecryptfs_dentry, int mode) + struct dentry *ecryptfs_dentry, umode_t mode) { int rc; struct dentry *lower_dentry; struct dentry *lower_dir_dentry; + struct inode *inode; lower_dentry = ecryptfs_dentry_to_lower(ecryptfs_dentry); lower_dir_dentry = lock_parent(lower_dentry); if (IS_ERR(lower_dir_dentry)) { ecryptfs_printk(KERN_ERR, "Error locking directory of " "dentry\n"); - rc = PTR_ERR(lower_dir_dentry); + inode = ERR_CAST(lower_dir_dentry); goto out; } - rc = ecryptfs_create_underlying_file(lower_dir_dentry->d_inode, - ecryptfs_dentry, mode); + rc = vfs_create(lower_dir_dentry->d_inode, lower_dentry, mode, NULL); if (rc) { printk(KERN_ERR "%s: Failure to create dentry in lower fs; " "rc = [%d]\n", __func__, rc); + inode = ERR_PTR(rc); goto out_lock; } - rc = ecryptfs_interpose(lower_dentry, ecryptfs_dentry, - directory_inode->i_sb); - if (rc) { - ecryptfs_printk(KERN_ERR, "Failure in ecryptfs_interpose\n"); + inode = __ecryptfs_get_inode(lower_dentry->d_inode, + directory_inode->i_sb); + if (IS_ERR(inode)) goto out_lock; - } fsstack_copy_attr_times(directory_inode, lower_dir_dentry->d_inode); fsstack_copy_inode_size(directory_inode, lower_dir_dentry->d_inode); out_lock: unlock_dir(lower_dir_dentry); out: - return rc; + return inode; } /** @@ -219,26 +200,26 @@ out: * * Returns zero on success */ -static int ecryptfs_initialize_file(struct dentry *ecryptfs_dentry) +static int ecryptfs_initialize_file(struct dentry *ecryptfs_dentry, + struct inode *ecryptfs_inode) { struct ecryptfs_crypt_stat *crypt_stat = - &ecryptfs_inode_to_private(ecryptfs_dentry->d_inode)->crypt_stat; + &ecryptfs_inode_to_private(ecryptfs_inode)->crypt_stat; int rc = 0; - if (S_ISDIR(ecryptfs_dentry->d_inode->i_mode)) { + if (S_ISDIR(ecryptfs_inode->i_mode)) { ecryptfs_printk(KERN_DEBUG, "This is a directory\n"); crypt_stat->flags &= ~(ECRYPTFS_ENCRYPTED); goto out; } ecryptfs_printk(KERN_DEBUG, "Initializing crypto context\n"); - rc = ecryptfs_new_file_context(ecryptfs_dentry); + rc = ecryptfs_new_file_context(ecryptfs_inode); if (rc) { ecryptfs_printk(KERN_ERR, "Error creating new file " "context; rc = [%d]\n", rc); goto out; } - rc = ecryptfs_get_lower_file(ecryptfs_dentry, - ecryptfs_dentry->d_inode); + rc = ecryptfs_get_lower_file(ecryptfs_dentry, ecryptfs_inode); if (rc) { printk(KERN_ERR "%s: Error attempting to initialize " "the lower file for the dentry with name " @@ -246,10 +227,10 @@ static int ecryptfs_initialize_file(struct dentry *ecryptfs_dentry) ecryptfs_dentry->d_name.name, rc); goto out; } - rc = ecryptfs_write_metadata(ecryptfs_dentry); + rc = ecryptfs_write_metadata(ecryptfs_dentry, ecryptfs_inode); if (rc) printk(KERN_ERR "Error writing headers; rc = [%d]\n", rc); - ecryptfs_put_lower_file(ecryptfs_dentry->d_inode); + ecryptfs_put_lower_file(ecryptfs_inode); out: return rc; } @@ -267,20 +248,30 @@ out: */ static int ecryptfs_create(struct inode *directory_inode, struct dentry *ecryptfs_dentry, - int mode, struct nameidata *nd) + umode_t mode, struct nameidata *nd) { + struct inode *ecryptfs_inode; int rc; - /* ecryptfs_do_create() calls ecryptfs_interpose() */ - rc = ecryptfs_do_create(directory_inode, ecryptfs_dentry, mode); - if (unlikely(rc)) { + ecryptfs_inode = ecryptfs_do_create(directory_inode, ecryptfs_dentry, + mode); + if (unlikely(IS_ERR(ecryptfs_inode))) { ecryptfs_printk(KERN_WARNING, "Failed to create file in" "lower filesystem\n"); + rc = PTR_ERR(ecryptfs_inode); goto out; } /* At this point, a file exists on "disk"; we need to make sure * that this on disk file is prepared to be an ecryptfs file */ - rc = ecryptfs_initialize_file(ecryptfs_dentry); + rc = ecryptfs_initialize_file(ecryptfs_dentry, ecryptfs_inode); + if (rc) { + drop_nlink(ecryptfs_inode); + unlock_new_inode(ecryptfs_inode); + iput(ecryptfs_inode); + goto out; + } + d_instantiate(ecryptfs_dentry, ecryptfs_inode); + unlock_new_inode(ecryptfs_inode); out: return rc; } @@ -549,7 +540,7 @@ out_lock: return rc; } -static int ecryptfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int ecryptfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { int rc; struct dentry *lower_dentry; @@ -597,7 +588,7 @@ static int ecryptfs_rmdir(struct inode *dir, struct dentry *dentry) } static int -ecryptfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) +ecryptfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev) { int rc; struct dentry *lower_dentry; diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c index dbd52d40df4c..9df7fd6e0c39 100644 --- a/fs/ecryptfs/super.c +++ b/fs/ecryptfs/super.c @@ -69,7 +69,6 @@ static void ecryptfs_i_callback(struct rcu_head *head) struct ecryptfs_inode_info *inode_info; inode_info = ecryptfs_inode_to_private(inode); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(ecryptfs_inode_info_cache, inode_info); } @@ -132,9 +131,9 @@ static void ecryptfs_evict_inode(struct inode *inode) * Prints the mount options for a given superblock. * Returns zero; does not fail. */ -static int ecryptfs_show_options(struct seq_file *m, struct vfsmount *mnt) +static int ecryptfs_show_options(struct seq_file *m, struct dentry *root) { - struct super_block *sb = mnt->mnt_sb; + struct super_block *sb = root->d_sb; struct ecryptfs_mount_crypt_stat *mount_crypt_stat = &ecryptfs_superblock_to_private(sb)->mount_crypt_stat; struct ecryptfs_global_auth_tok *walker; diff --git a/fs/efs/super.c b/fs/efs/super.c index 0f31acb0131c..981106429a9f 100644 --- a/fs/efs/super.c +++ b/fs/efs/super.c @@ -68,7 +68,6 @@ static struct inode *efs_alloc_inode(struct super_block *sb) static void efs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(efs_inode_cachep, INODE_INFO(inode)); } diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 828e750af23a..aabdfc38cf24 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -197,6 +197,12 @@ struct eventpoll { /* The user that created the eventpoll descriptor */ struct user_struct *user; + + struct file *file; + + /* used to optimize loop detection check */ + int visited; + struct list_head visited_list_link; }; /* Wait structure used by the poll hooks */ @@ -255,6 +261,15 @@ static struct kmem_cache *epi_cache __read_mostly; /* Slab cache used to allocate "struct eppoll_entry" */ static struct kmem_cache *pwq_cache __read_mostly; +/* Visited nodes during ep_loop_check(), so we can unset them when we finish */ +static LIST_HEAD(visited_list); + +/* + * List of files with newly added links, where we may need to limit the number + * of emanating paths. Protected by the epmutex. + */ +static LIST_HEAD(tfile_check_list); + #ifdef CONFIG_SYSCTL #include <linux/sysctl.h> @@ -276,6 +291,12 @@ ctl_table epoll_table[] = { }; #endif /* CONFIG_SYSCTL */ +static const struct file_operations eventpoll_fops; + +static inline int is_file_epoll(struct file *f) +{ + return f->f_op == &eventpoll_fops; +} /* Setup the structure that is used as key for the RB tree */ static inline void ep_set_ffd(struct epoll_filefd *ffd, @@ -711,12 +732,6 @@ static const struct file_operations eventpoll_fops = { .llseek = noop_llseek, }; -/* Fast test to see if the file is an eventpoll file */ -static inline int is_file_epoll(struct file *f) -{ - return f->f_op == &eventpoll_fops; -} - /* * This is called from eventpoll_release() to unlink files from the eventpoll * interface. We need to have this facility to cleanup correctly files that are @@ -926,6 +941,99 @@ static void ep_rbtree_insert(struct eventpoll *ep, struct epitem *epi) rb_insert_color(&epi->rbn, &ep->rbr); } + + +#define PATH_ARR_SIZE 5 +/* + * These are the number paths of length 1 to 5, that we are allowing to emanate + * from a single file of interest. For example, we allow 1000 paths of length + * 1, to emanate from each file of interest. This essentially represents the + * potential wakeup paths, which need to be limited in order to avoid massive + * uncontrolled wakeup storms. The common use case should be a single ep which + * is connected to n file sources. In this case each file source has 1 path + * of length 1. Thus, the numbers below should be more than sufficient. These + * path limits are enforced during an EPOLL_CTL_ADD operation, since a modify + * and delete can't add additional paths. Protected by the epmutex. + */ +static const int path_limits[PATH_ARR_SIZE] = { 1000, 500, 100, 50, 10 }; +static int path_count[PATH_ARR_SIZE]; + +static int path_count_inc(int nests) +{ + if (++path_count[nests] > path_limits[nests]) + return -1; + return 0; +} + +static void path_count_init(void) +{ + int i; + + for (i = 0; i < PATH_ARR_SIZE; i++) + path_count[i] = 0; +} + +static int reverse_path_check_proc(void *priv, void *cookie, int call_nests) +{ + int error = 0; + struct file *file = priv; + struct file *child_file; + struct epitem *epi; + + list_for_each_entry(epi, &file->f_ep_links, fllink) { + child_file = epi->ep->file; + if (is_file_epoll(child_file)) { + if (list_empty(&child_file->f_ep_links)) { + if (path_count_inc(call_nests)) { + error = -1; + break; + } + } else { + error = ep_call_nested(&poll_loop_ncalls, + EP_MAX_NESTS, + reverse_path_check_proc, + child_file, child_file, + current); + } + if (error != 0) + break; + } else { + printk(KERN_ERR "reverse_path_check_proc: " + "file is not an ep!\n"); + } + } + return error; +} + +/** + * reverse_path_check - The tfile_check_list is list of file *, which have + * links that are proposed to be newly added. We need to + * make sure that those added links don't add too many + * paths such that we will spend all our time waking up + * eventpoll objects. + * + * Returns: Returns zero if the proposed links don't create too many paths, + * -1 otherwise. + */ +static int reverse_path_check(void) +{ + int length = 0; + int error = 0; + struct file *current_file; + + /* let's call this for all tfiles */ + list_for_each_entry(current_file, &tfile_check_list, f_tfile_llink) { + length++; + path_count_init(); + error = ep_call_nested(&poll_loop_ncalls, EP_MAX_NESTS, + reverse_path_check_proc, current_file, + current_file, current); + if (error) + break; + } + return error; +} + /* * Must be called with "mtx" held. */ @@ -987,6 +1095,11 @@ static int ep_insert(struct eventpoll *ep, struct epoll_event *event, */ ep_rbtree_insert(ep, epi); + /* now check if we've created too many backpaths */ + error = -EINVAL; + if (reverse_path_check()) + goto error_remove_epi; + /* We have to drop the new item inside our item list to keep track of it */ spin_lock_irqsave(&ep->lock, flags); @@ -1011,6 +1124,14 @@ static int ep_insert(struct eventpoll *ep, struct epoll_event *event, return 0; +error_remove_epi: + spin_lock(&tfile->f_lock); + if (ep_is_linked(&epi->fllink)) + list_del_init(&epi->fllink); + spin_unlock(&tfile->f_lock); + + rb_erase(&epi->rbn, &ep->rbr); + error_unregister: ep_unregister_pollwait(ep, epi); @@ -1275,18 +1396,36 @@ static int ep_loop_check_proc(void *priv, void *cookie, int call_nests) int error = 0; struct file *file = priv; struct eventpoll *ep = file->private_data; + struct eventpoll *ep_tovisit; struct rb_node *rbp; struct epitem *epi; mutex_lock_nested(&ep->mtx, call_nests + 1); + ep->visited = 1; + list_add(&ep->visited_list_link, &visited_list); for (rbp = rb_first(&ep->rbr); rbp; rbp = rb_next(rbp)) { epi = rb_entry(rbp, struct epitem, rbn); if (unlikely(is_file_epoll(epi->ffd.file))) { + ep_tovisit = epi->ffd.file->private_data; + if (ep_tovisit->visited) + continue; error = ep_call_nested(&poll_loop_ncalls, EP_MAX_NESTS, - ep_loop_check_proc, epi->ffd.file, - epi->ffd.file->private_data, current); + ep_loop_check_proc, epi->ffd.file, + ep_tovisit, current); if (error != 0) break; + } else { + /* + * If we've reached a file that is not associated with + * an ep, then we need to check if the newly added + * links are going to add too many wakeup paths. We do + * this by adding it to the tfile_check_list, if it's + * not already there, and calling reverse_path_check() + * during ep_insert(). + */ + if (list_empty(&epi->ffd.file->f_tfile_llink)) + list_add(&epi->ffd.file->f_tfile_llink, + &tfile_check_list); } } mutex_unlock(&ep->mtx); @@ -1307,8 +1446,31 @@ static int ep_loop_check_proc(void *priv, void *cookie, int call_nests) */ static int ep_loop_check(struct eventpoll *ep, struct file *file) { - return ep_call_nested(&poll_loop_ncalls, EP_MAX_NESTS, + int ret; + struct eventpoll *ep_cur, *ep_next; + + ret = ep_call_nested(&poll_loop_ncalls, EP_MAX_NESTS, ep_loop_check_proc, file, ep, current); + /* clear visited list */ + list_for_each_entry_safe(ep_cur, ep_next, &visited_list, + visited_list_link) { + ep_cur->visited = 0; + list_del(&ep_cur->visited_list_link); + } + return ret; +} + +static void clear_tfile_check_list(void) +{ + struct file *file; + + /* first clear the tfile_check_list */ + while (!list_empty(&tfile_check_list)) { + file = list_first_entry(&tfile_check_list, struct file, + f_tfile_llink); + list_del_init(&file->f_tfile_llink); + } + INIT_LIST_HEAD(&tfile_check_list); } /* @@ -1316,8 +1478,9 @@ static int ep_loop_check(struct eventpoll *ep, struct file *file) */ SYSCALL_DEFINE1(epoll_create1, int, flags) { - int error; + int error, fd; struct eventpoll *ep = NULL; + struct file *file; /* Check the EPOLL_* constant for consistency. */ BUILD_BUG_ON(EPOLL_CLOEXEC != O_CLOEXEC); @@ -1334,11 +1497,25 @@ SYSCALL_DEFINE1(epoll_create1, int, flags) * Creates all the items needed to setup an eventpoll file. That is, * a file structure and a free file descriptor. */ - error = anon_inode_getfd("[eventpoll]", &eventpoll_fops, ep, + fd = get_unused_fd_flags(O_RDWR | (flags & O_CLOEXEC)); + if (fd < 0) { + error = fd; + goto out_free_ep; + } + file = anon_inode_getfile("[eventpoll]", &eventpoll_fops, ep, O_RDWR | (flags & O_CLOEXEC)); - if (error < 0) - ep_free(ep); - + if (IS_ERR(file)) { + error = PTR_ERR(file); + goto out_free_fd; + } + fd_install(fd, file); + ep->file = file; + return fd; + +out_free_fd: + put_unused_fd(fd); +out_free_ep: + ep_free(ep); return error; } @@ -1404,21 +1581,27 @@ SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd, /* * When we insert an epoll file descriptor, inside another epoll file * descriptor, there is the change of creating closed loops, which are - * better be handled here, than in more critical paths. + * better be handled here, than in more critical paths. While we are + * checking for loops we also determine the list of files reachable + * and hang them on the tfile_check_list, so we can check that we + * haven't created too many possible wakeup paths. * - * We hold epmutex across the loop check and the insert in this case, in - * order to prevent two separate inserts from racing and each doing the - * insert "at the same time" such that ep_loop_check passes on both - * before either one does the insert, thereby creating a cycle. + * We need to hold the epmutex across both ep_insert and ep_remove + * b/c we want to make sure we are looking at a coherent view of + * epoll network. */ - if (unlikely(is_file_epoll(tfile) && op == EPOLL_CTL_ADD)) { + if (op == EPOLL_CTL_ADD || op == EPOLL_CTL_DEL) { mutex_lock(&epmutex); did_lock_epmutex = 1; - error = -ELOOP; - if (ep_loop_check(ep, tfile) != 0) - goto error_tgt_fput; } - + if (op == EPOLL_CTL_ADD) { + if (is_file_epoll(tfile)) { + error = -ELOOP; + if (ep_loop_check(ep, tfile) != 0) + goto error_tgt_fput; + } else + list_add(&tfile->f_tfile_llink, &tfile_check_list); + } mutex_lock_nested(&ep->mtx, 0); @@ -1437,6 +1620,7 @@ SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd, error = ep_insert(ep, &epds, tfile, fd); } else error = -EEXIST; + clear_tfile_check_list(); break; case EPOLL_CTL_DEL: if (epi) @@ -1455,7 +1639,7 @@ SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd, mutex_unlock(&ep->mtx); error_tgt_fput: - if (unlikely(did_lock_epmutex)) + if (did_lock_epmutex) mutex_unlock(&epmutex); fput(tfile); diff --git a/fs/exec.c b/fs/exec.c index 36254645b7cc..aeb135c7ff5c 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -59,6 +59,8 @@ #include <asm/uaccess.h> #include <asm/mmu_context.h> #include <asm/tlb.h> + +#include <trace/events/task.h> #include "internal.h" int core_uses_pid; @@ -1054,6 +1056,8 @@ void set_task_comm(struct task_struct *tsk, char *buf) { task_lock(tsk); + trace_task_rename(tsk, buf); + /* * Threads may access current->comm without holding * the task lock, so write the string carefully. @@ -1225,7 +1229,7 @@ EXPORT_SYMBOL(install_exec_creds); * - the caller must hold ->cred_guard_mutex to protect against * PTRACE_ATTACH */ -int check_unsafe_exec(struct linux_binprm *bprm) +static int check_unsafe_exec(struct linux_binprm *bprm) { struct task_struct *p = current, *t; unsigned n_fs; diff --git a/fs/exofs/Kconfig b/fs/exofs/Kconfig index da42f32c49be..86194b2f799d 100644 --- a/fs/exofs/Kconfig +++ b/fs/exofs/Kconfig @@ -1,14 +1,3 @@ -# Note ORE needs to "select ASYNC_XOR". So Not to force multiple selects -# for every ORE user we do it like this. Any user should add itself here -# at the "depends on EXOFS_FS || ..." with an ||. The dependencies are -# selected here, and we default to "ON". So in effect it is like been -# selected by any of the users. -config ORE - tristate - depends on EXOFS_FS || PNFS_OBJLAYOUT - select ASYNC_XOR - default SCSI_OSD_ULD - config EXOFS_FS tristate "exofs: OSD based file system support" depends on SCSI_OSD_ULD diff --git a/fs/exofs/Kconfig.ore b/fs/exofs/Kconfig.ore new file mode 100644 index 000000000000..1ca7fb7b6ba8 --- /dev/null +++ b/fs/exofs/Kconfig.ore @@ -0,0 +1,12 @@ +# ORE - Objects Raid Engine (libore.ko) +# +# Note ORE needs to "select ASYNC_XOR". So Not to force multiple selects +# for every ORE user we do it like this. Any user should add itself here +# at the "depends on EXOFS_FS || ..." with an ||. The dependencies are +# selected here, and we default to "ON". So in effect it is like been +# selected by any of the users. +config ORE + tristate + depends on EXOFS_FS || PNFS_OBJLAYOUT + select ASYNC_XOR + default SCSI_OSD_ULD diff --git a/fs/exofs/dir.c b/fs/exofs/dir.c index d0941c6a1f72..80405836ba6e 100644 --- a/fs/exofs/dir.c +++ b/fs/exofs/dir.c @@ -234,7 +234,7 @@ static unsigned char exofs_type_by_mode[S_IFMT >> S_SHIFT] = { static inline void exofs_set_de_type(struct exofs_dir_entry *de, struct inode *inode) { - mode_t mode = inode->i_mode; + umode_t mode = inode->i_mode; de->file_type = exofs_type_by_mode[(mode & S_IFMT) >> S_SHIFT]; } diff --git a/fs/exofs/exofs.h b/fs/exofs/exofs.h index 51f4b4c40f09..ca9d49665ef6 100644 --- a/fs/exofs/exofs.h +++ b/fs/exofs/exofs.h @@ -154,7 +154,7 @@ int exofs_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata); extern struct inode *exofs_iget(struct super_block *, unsigned long); -struct inode *exofs_new_inode(struct inode *, int); +struct inode *exofs_new_inode(struct inode *, umode_t); extern int exofs_write_inode(struct inode *, struct writeback_control *wbc); extern void exofs_evict_inode(struct inode *); diff --git a/fs/exofs/inode.c b/fs/exofs/inode.c index f6dbf7768ce6..ea5e1f97806a 100644 --- a/fs/exofs/inode.c +++ b/fs/exofs/inode.c @@ -1276,7 +1276,7 @@ static void create_done(struct ore_io_state *ios, void *p) /* * Set up a new inode and create an object for it on the OSD */ -struct inode *exofs_new_inode(struct inode *dir, int mode) +struct inode *exofs_new_inode(struct inode *dir, umode_t mode) { struct super_block *sb = dir->i_sb; struct exofs_sb_info *sbi = sb->s_fs_info; diff --git a/fs/exofs/namei.c b/fs/exofs/namei.c index b54c43775f17..9dbf0c301030 100644 --- a/fs/exofs/namei.c +++ b/fs/exofs/namei.c @@ -59,7 +59,7 @@ static struct dentry *exofs_lookup(struct inode *dir, struct dentry *dentry, return d_splice_alias(inode, dentry); } -static int exofs_create(struct inode *dir, struct dentry *dentry, int mode, +static int exofs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { struct inode *inode = exofs_new_inode(dir, mode); @@ -74,7 +74,7 @@ static int exofs_create(struct inode *dir, struct dentry *dentry, int mode, return err; } -static int exofs_mknod(struct inode *dir, struct dentry *dentry, int mode, +static int exofs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) { struct inode *inode; @@ -153,7 +153,7 @@ static int exofs_link(struct dentry *old_dentry, struct inode *dir, return exofs_add_nondir(dentry, inode); } -static int exofs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int exofs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct inode *inode; int err = -EMLINK; diff --git a/fs/exofs/ore.c b/fs/exofs/ore.c index d271ad837202..49cf230554a2 100644 --- a/fs/exofs/ore.c +++ b/fs/exofs/ore.c @@ -266,7 +266,7 @@ int ore_get_rw_state(struct ore_layout *layout, struct ore_components *oc, /* first/last seg is split */ num_raid_units += layout->group_width; - sgs_per_dev = div_u64(num_raid_units, data_devs); + sgs_per_dev = div_u64(num_raid_units, data_devs) + 2; } else { /* For Writes add parity pages array. */ max_par_pages = num_raid_units * pages_in_unit * @@ -445,10 +445,10 @@ int ore_check_io(struct ore_io_state *ios, ore_on_dev_error on_dev_error) u64 residual = ios->reading ? or->in.residual : or->out.residual; u64 offset = (ios->offset + ios->length) - residual; - struct ore_dev *od = ios->oc->ods[ - per_dev->dev - ios->oc->first_dev]; + unsigned dev = per_dev->dev - ios->oc->first_dev; + struct ore_dev *od = ios->oc->ods[dev]; - on_dev_error(ios, od, per_dev->dev, osi.osd_err_pri, + on_dev_error(ios, od, dev, osi.osd_err_pri, offset, residual); } if (osi.osd_err_pri >= acumulated_osd_err) { diff --git a/fs/exofs/ore_raid.c b/fs/exofs/ore_raid.c index 29c47e5c4a86..d222c77cfa1b 100644 --- a/fs/exofs/ore_raid.c +++ b/fs/exofs/ore_raid.c @@ -328,8 +328,8 @@ static int _alloc_read_4_write(struct ore_io_state *ios) /* @si contains info of the to-be-inserted page. Update of @si should be * maintained by caller. Specificaly si->dev, si->obj_offset, ... */ -static int _add_to_read_4_write(struct ore_io_state *ios, - struct ore_striping_info *si, struct page *page) +static int _add_to_r4w(struct ore_io_state *ios, struct ore_striping_info *si, + struct page *page, unsigned pg_len) { struct request_queue *q; struct ore_per_dev_state *per_dev; @@ -366,17 +366,60 @@ static int _add_to_read_4_write(struct ore_io_state *ios, _ore_add_sg_seg(per_dev, gap, true); } q = osd_request_queue(ore_comp_dev(read_ios->oc, per_dev->dev)); - added_len = bio_add_pc_page(q, per_dev->bio, page, PAGE_SIZE, 0); - if (unlikely(added_len != PAGE_SIZE)) { + added_len = bio_add_pc_page(q, per_dev->bio, page, pg_len, + si->obj_offset % PAGE_SIZE); + if (unlikely(added_len != pg_len)) { ORE_DBGMSG("Failed to bio_add_pc_page bi_vcnt=%d\n", per_dev->bio->bi_vcnt); return -ENOMEM; } - per_dev->length += PAGE_SIZE; + per_dev->length += pg_len; return 0; } +/* read the beginning of an unaligned first page */ +static int _add_to_r4w_first_page(struct ore_io_state *ios, struct page *page) +{ + struct ore_striping_info si; + unsigned pg_len; + + ore_calc_stripe_info(ios->layout, ios->offset, 0, &si); + + pg_len = si.obj_offset % PAGE_SIZE; + si.obj_offset -= pg_len; + + ORE_DBGMSG("offset=0x%llx len=0x%x index=0x%lx dev=%x\n", + _LLU(si.obj_offset), pg_len, page->index, si.dev); + + return _add_to_r4w(ios, &si, page, pg_len); +} + +/* read the end of an incomplete last page */ +static int _add_to_r4w_last_page(struct ore_io_state *ios, u64 *offset) +{ + struct ore_striping_info si; + struct page *page; + unsigned pg_len, p, c; + + ore_calc_stripe_info(ios->layout, *offset, 0, &si); + + p = si.unit_off / PAGE_SIZE; + c = _dev_order(ios->layout->group_width * ios->layout->mirrors_p1, + ios->layout->mirrors_p1, si.par_dev, si.dev); + page = ios->sp2d->_1p_stripes[p].pages[c]; + + pg_len = PAGE_SIZE - (si.unit_off % PAGE_SIZE); + *offset += pg_len; + + ORE_DBGMSG("p=%d, c=%d next-offset=0x%llx len=0x%x dev=%x par_dev=%d\n", + p, c, _LLU(*offset), pg_len, si.dev, si.par_dev); + + BUG_ON(!page); + + return _add_to_r4w(ios, &si, page, pg_len); +} + static void _mark_read4write_pages_uptodate(struct ore_io_state *ios, int ret) { struct bio_vec *bv; @@ -444,9 +487,13 @@ static int _read_4_write(struct ore_io_state *ios) struct page **pp = &_1ps->pages[c]; bool uptodate; - if (*pp) + if (*pp) { + if (ios->offset % PAGE_SIZE) + /* Read the remainder of the page */ + _add_to_r4w_first_page(ios, *pp); /* to-be-written pages start here */ goto read_last_stripe; + } *pp = ios->r4w->get_page(ios->private, offset, &uptodate); @@ -454,7 +501,7 @@ static int _read_4_write(struct ore_io_state *ios) return -ENOMEM; if (!uptodate) - _add_to_read_4_write(ios, &read_si, *pp); + _add_to_r4w(ios, &read_si, *pp, PAGE_SIZE); /* Mark read-pages to be cache_released */ _1ps->page_is_read[c] = true; @@ -465,8 +512,11 @@ static int _read_4_write(struct ore_io_state *ios) } read_last_stripe: - offset = ios->offset + (ios->length + PAGE_SIZE - 1) / - PAGE_SIZE * PAGE_SIZE; + offset = ios->offset + ios->length; + if (offset % PAGE_SIZE) + _add_to_r4w_last_page(ios, &offset); + /* offset will be aligned to next page */ + last_stripe_end = div_u64(offset + bytes_in_stripe - 1, bytes_in_stripe) * bytes_in_stripe; if (offset == last_stripe_end) /* Optimize for the aligned case */ @@ -503,7 +553,7 @@ read_last_stripe: /* Mark read-pages to be cache_released */ _1ps->page_is_read[c] = true; if (!uptodate) - _add_to_read_4_write(ios, &read_si, page); + _add_to_r4w(ios, &read_si, page, PAGE_SIZE); } offset += PAGE_SIZE; @@ -551,7 +601,11 @@ int _ore_add_parity_unit(struct ore_io_state *ios, unsigned cur_len) { if (ios->reading) { - BUG_ON(per_dev->cur_sg >= ios->sgs_per_dev); + if (per_dev->cur_sg >= ios->sgs_per_dev) { + ORE_DBGMSG("cur_sg(%d) >= sgs_per_dev(%d)\n" , + per_dev->cur_sg, ios->sgs_per_dev); + return -ENOMEM; + } _ore_add_sg_seg(per_dev, cur_len, true); } else { struct __stripe_pages_2d *sp2d = ios->sp2d; @@ -612,8 +666,6 @@ int _ore_post_alloc_raid_stuff(struct ore_io_state *ios) return -ENOMEM; } - BUG_ON(ios->offset % PAGE_SIZE); - /* Round io down to last full strip */ first_stripe = div_u64(ios->offset, stripe_size); last_stripe = div_u64(ios->offset + ios->length, stripe_size); diff --git a/fs/exofs/super.c b/fs/exofs/super.c index e6085ec192d6..d22cd168c6ee 100644 --- a/fs/exofs/super.c +++ b/fs/exofs/super.c @@ -166,7 +166,6 @@ static struct inode *exofs_alloc_inode(struct super_block *sb) static void exofs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(exofs_inode_cachep, exofs_i(inode)); } @@ -839,6 +838,8 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent) ret = bdi_setup_and_register(&sbi->bdi, "exofs", BDI_CAP_MAP_COPY); if (ret) { EXOFS_DBGMSG("Failed to bdi_setup_and_register\n"); + dput(sb->s_root); + sb->s_root = NULL; goto free_sbi; } diff --git a/fs/ext2/dir.c b/fs/ext2/dir.c index 47cda410b548..d37df352d324 100644 --- a/fs/ext2/dir.c +++ b/fs/ext2/dir.c @@ -279,7 +279,7 @@ static unsigned char ext2_type_by_mode[S_IFMT >> S_SHIFT] = { static inline void ext2_set_de_type(ext2_dirent *de, struct inode *inode) { - mode_t mode = inode->i_mode; + umode_t mode = inode->i_mode; if (EXT2_HAS_INCOMPAT_FEATURE(inode->i_sb, EXT2_FEATURE_INCOMPAT_FILETYPE)) de->file_type = ext2_type_by_mode[(mode & S_IFMT)>>S_SHIFT]; else diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h index 9a4e5e206d08..75ad433c6691 100644 --- a/fs/ext2/ext2.h +++ b/fs/ext2/ext2.h @@ -110,7 +110,7 @@ extern struct ext2_dir_entry_2 * ext2_dotdot (struct inode *, struct page **); extern void ext2_set_link(struct inode *, struct ext2_dir_entry_2 *, struct page *, struct inode *, int); /* ialloc.c */ -extern struct inode * ext2_new_inode (struct inode *, int, const struct qstr *); +extern struct inode * ext2_new_inode (struct inode *, umode_t, const struct qstr *); extern void ext2_free_inode (struct inode *); extern unsigned long ext2_count_free_inodes (struct super_block *); extern void ext2_check_inodes_bitmap (struct super_block *); diff --git a/fs/ext2/ialloc.c b/fs/ext2/ialloc.c index c4e81dfb74ba..8b15cf8cef37 100644 --- a/fs/ext2/ialloc.c +++ b/fs/ext2/ialloc.c @@ -429,7 +429,7 @@ found: return group; } -struct inode *ext2_new_inode(struct inode *dir, int mode, +struct inode *ext2_new_inode(struct inode *dir, umode_t mode, const struct qstr *qstr) { struct super_block *sb; @@ -573,8 +573,11 @@ got: inode->i_generation = sbi->s_next_generation++; spin_unlock(&sbi->s_next_gen_lock); if (insert_inode_locked(inode) < 0) { - err = -EINVAL; - goto fail_drop; + ext2_error(sb, "ext2_new_inode", + "inode number already in use - inode=%lu", + (unsigned long) ino); + err = -EIO; + goto fail; } dquot_initialize(inode); diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index 91a6945af6d8..740cad8dcd8d 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -26,7 +26,6 @@ #include <linux/highuid.h> #include <linux/pagemap.h> #include <linux/quotaops.h> -#include <linux/module.h> #include <linux/writeback.h> #include <linux/buffer_head.h> #include <linux/mpage.h> @@ -36,10 +35,6 @@ #include "acl.h" #include "xip.h" -MODULE_AUTHOR("Remy Card and others"); -MODULE_DESCRIPTION("Second Extended Filesystem"); -MODULE_LICENSE("GPL"); - static int __ext2_write_inode(struct inode *inode, int do_sync); /* diff --git a/fs/ext2/ioctl.c b/fs/ext2/ioctl.c index f81e250ac5c4..1089f760c847 100644 --- a/fs/ext2/ioctl.c +++ b/fs/ext2/ioctl.c @@ -35,7 +35,7 @@ long ext2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) case EXT2_IOC_SETFLAGS: { unsigned int oldflags; - ret = mnt_want_write(filp->f_path.mnt); + ret = mnt_want_write_file(filp); if (ret) return ret; @@ -83,7 +83,7 @@ long ext2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) inode->i_ctime = CURRENT_TIME_SEC; mark_inode_dirty(inode); setflags_out: - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return ret; } case EXT2_IOC_GETVERSION: @@ -91,7 +91,7 @@ setflags_out: case EXT2_IOC_SETVERSION: if (!inode_owner_or_capable(inode)) return -EPERM; - ret = mnt_want_write(filp->f_path.mnt); + ret = mnt_want_write_file(filp); if (ret) return ret; if (get_user(inode->i_generation, (int __user *) arg)) { @@ -100,7 +100,7 @@ setflags_out: inode->i_ctime = CURRENT_TIME_SEC; mark_inode_dirty(inode); } - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return ret; case EXT2_IOC_GETRSVSZ: if (test_opt(inode->i_sb, RESERVATION) @@ -121,7 +121,7 @@ setflags_out: if (get_user(rsv_window_size, (int __user *)arg)) return -EFAULT; - ret = mnt_want_write(filp->f_path.mnt); + ret = mnt_want_write_file(filp); if (ret) return ret; @@ -145,7 +145,7 @@ setflags_out: rsv->rsv_goal_size = rsv_window_size; } mutex_unlock(&ei->truncate_mutex); - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return 0; } default: diff --git a/fs/ext2/namei.c b/fs/ext2/namei.c index 761fde807fc9..080419814bae 100644 --- a/fs/ext2/namei.c +++ b/fs/ext2/namei.c @@ -94,7 +94,7 @@ struct dentry *ext2_get_parent(struct dentry *child) * If the create succeeds, we fill in the inode information * with d_instantiate(). */ -static int ext2_create (struct inode * dir, struct dentry * dentry, int mode, struct nameidata *nd) +static int ext2_create (struct inode * dir, struct dentry * dentry, umode_t mode, struct nameidata *nd) { struct inode *inode; @@ -119,7 +119,7 @@ static int ext2_create (struct inode * dir, struct dentry * dentry, int mode, st return ext2_add_nondir(dentry, inode); } -static int ext2_mknod (struct inode * dir, struct dentry *dentry, int mode, dev_t rdev) +static int ext2_mknod (struct inode * dir, struct dentry *dentry, umode_t mode, dev_t rdev) { struct inode * inode; int err; @@ -214,7 +214,7 @@ static int ext2_link (struct dentry * old_dentry, struct inode * dir, return err; } -static int ext2_mkdir(struct inode * dir, struct dentry * dentry, int mode) +static int ext2_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode) { struct inode * inode; int err = -EMLINK; diff --git a/fs/ext2/super.c b/fs/ext2/super.c index bd8ac164a3bf..0090595beb28 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -173,7 +173,6 @@ static struct inode *ext2_alloc_inode(struct super_block *sb) static void ext2_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(ext2_inode_cachep, EXT2_I(inode)); } @@ -211,9 +210,9 @@ static void destroy_inodecache(void) kmem_cache_destroy(ext2_inode_cachep); } -static int ext2_show_options(struct seq_file *seq, struct vfsmount *vfs) +static int ext2_show_options(struct seq_file *seq, struct dentry *root) { - struct super_block *sb = vfs->mnt_sb; + struct super_block *sb = root->d_sb; struct ext2_sb_info *sbi = EXT2_SB(sb); struct ext2_super_block *es = sbi->s_es; unsigned long def_mount_opts; @@ -1521,5 +1520,8 @@ static void __exit exit_ext2_fs(void) exit_ext2_xattr(); } +MODULE_AUTHOR("Remy Card and others"); +MODULE_DESCRIPTION("Second Extended Filesystem"); +MODULE_LICENSE("GPL"); module_init(init_ext2_fs) module_exit(exit_ext2_fs) diff --git a/fs/ext2/xattr.c b/fs/ext2/xattr.c index d27b71f1d183..6dcafc7efdfd 100644 --- a/fs/ext2/xattr.c +++ b/fs/ext2/xattr.c @@ -54,7 +54,6 @@ */ #include <linux/buffer_head.h> -#include <linux/module.h> #include <linux/init.h> #include <linux/slab.h> #include <linux/mbcache.h> diff --git a/fs/ext2/xattr_security.c b/fs/ext2/xattr_security.c index c922adc8ef41..be7a8d02c9a7 100644 --- a/fs/ext2/xattr_security.c +++ b/fs/ext2/xattr_security.c @@ -3,7 +3,6 @@ * Handler for storing security labels as extended attributes. */ -#include <linux/module.h> #include <linux/slab.h> #include <linux/string.h> #include <linux/fs.h> diff --git a/fs/ext2/xattr_trusted.c b/fs/ext2/xattr_trusted.c index 667e46a8d62d..2989467d3595 100644 --- a/fs/ext2/xattr_trusted.c +++ b/fs/ext2/xattr_trusted.c @@ -5,7 +5,6 @@ * Copyright (C) 2003 by Andreas Gruenbacher, <a.gruenbacher@computer.org> */ -#include <linux/module.h> #include <linux/string.h> #include <linux/capability.h> #include <linux/fs.h> diff --git a/fs/ext2/xattr_user.c b/fs/ext2/xattr_user.c index 099d20f47163..f470e44c4b8d 100644 --- a/fs/ext2/xattr_user.c +++ b/fs/ext2/xattr_user.c @@ -6,7 +6,6 @@ */ #include <linux/init.h> -#include <linux/module.h> #include <linux/string.h> #include "ext2.h" #include "xattr.h" diff --git a/fs/ext3/ialloc.c b/fs/ext3/ialloc.c index 5c866e06e7ab..1cde28438014 100644 --- a/fs/ext3/ialloc.c +++ b/fs/ext3/ialloc.c @@ -371,7 +371,7 @@ static int find_group_other(struct super_block *sb, struct inode *parent) * group to find a free inode. */ struct inode *ext3_new_inode(handle_t *handle, struct inode * dir, - const struct qstr *qstr, int mode) + const struct qstr *qstr, umode_t mode) { struct super_block *sb; struct buffer_head *bitmap_bh = NULL; @@ -525,8 +525,12 @@ got: if (IS_DIRSYNC(inode)) handle->h_sync = 1; if (insert_inode_locked(inode) < 0) { - err = -EINVAL; - goto fail_drop; + /* + * Likely a bitmap corruption causing inode to be allocated + * twice. + */ + err = -EIO; + goto fail; } spin_lock(&sbi->s_next_gen_lock); inode->i_generation = sbi->s_next_generation++; diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index 85fe655fe3e0..2d0afeca0b47 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -22,7 +22,6 @@ * Assorted race fixes, rewrite of ext3_get_block() by Al Viro, 2000 */ -#include <linux/module.h> #include <linux/fs.h> #include <linux/time.h> #include <linux/ext3_jbd.h> @@ -223,8 +222,12 @@ void ext3_evict_inode (struct inode *inode) * * Note that directories do not have this problem because they don't * use page cache. + * + * The s_journal check handles the case when ext3_get_journal() fails + * and puts the journal inode. */ if (inode->i_nlink && ext3_should_journal_data(inode) && + EXT3_SB(inode->i_sb)->s_journal && (S_ISLNK(inode->i_mode) || S_ISREG(inode->i_mode))) { tid_t commit_tid = atomic_read(&ei->i_datasync_tid); journal_t *journal = EXT3_SB(inode->i_sb)->s_journal; @@ -1132,9 +1135,11 @@ struct buffer_head *ext3_bread(handle_t *handle, struct inode *inode, bh = ext3_getblk(handle, inode, block, create, err); if (!bh) return bh; - if (buffer_uptodate(bh)) + if (bh_uptodate_or_lock(bh)) return bh; - ll_rw_block(READ | REQ_META | REQ_PRIO, 1, &bh); + get_bh(bh); + bh->b_end_io = end_buffer_read_sync; + submit_bh(READ | REQ_META | REQ_PRIO, bh); wait_on_buffer(bh); if (buffer_uptodate(bh)) return bh; @@ -1617,7 +1622,13 @@ static int ext3_ordered_writepage(struct page *page, int err; J_ASSERT(PageLocked(page)); - WARN_ON_ONCE(IS_RDONLY(inode)); + /* + * We don't want to warn for emergency remount. The condition is + * ordered to avoid dereferencing inode->i_sb in non-error case to + * avoid slow-downs. + */ + WARN_ON_ONCE(IS_RDONLY(inode) && + !(EXT3_SB(inode->i_sb)->s_mount_state & EXT3_ERROR_FS)); /* * We give up here if we're reentered, because it might be for a @@ -1692,7 +1703,13 @@ static int ext3_writeback_writepage(struct page *page, int err; J_ASSERT(PageLocked(page)); - WARN_ON_ONCE(IS_RDONLY(inode)); + /* + * We don't want to warn for emergency remount. The condition is + * ordered to avoid dereferencing inode->i_sb in non-error case to + * avoid slow-downs. + */ + WARN_ON_ONCE(IS_RDONLY(inode) && + !(EXT3_SB(inode->i_sb)->s_mount_state & EXT3_ERROR_FS)); if (ext3_journal_current_handle()) goto out_fail; @@ -1735,7 +1752,13 @@ static int ext3_journalled_writepage(struct page *page, int err; J_ASSERT(PageLocked(page)); - WARN_ON_ONCE(IS_RDONLY(inode)); + /* + * We don't want to warn for emergency remount. The condition is + * ordered to avoid dereferencing inode->i_sb in non-error case to + * avoid slow-downs. + */ + WARN_ON_ONCE(IS_RDONLY(inode) && + !(EXT3_SB(inode->i_sb)->s_mount_state & EXT3_ERROR_FS)); if (ext3_journal_current_handle()) goto no_write; @@ -2064,12 +2087,10 @@ static int ext3_block_truncate_page(struct inode *inode, loff_t from) if (PageUptodate(page)) set_buffer_uptodate(bh); - if (!buffer_uptodate(bh)) { - err = -EIO; - ll_rw_block(READ, 1, &bh); - wait_on_buffer(bh); + if (!bh_uptodate_or_lock(bh)) { + err = bh_submit_read(bh); /* Uhhuh. Read error. Complain and punt. */ - if (!buffer_uptodate(bh)) + if (err) goto unlock; } @@ -2490,7 +2511,7 @@ int ext3_can_truncate(struct inode *inode) * transaction, and VFS/VM ensures that ext3_truncate() cannot run * simultaneously on behalf of the same inode. * - * As we work through the truncate and commmit bits of it to the journal there + * As we work through the truncate and commit bits of it to the journal there * is one core, guiding principle: the file's tree must always be consistent on * disk. We must be able to restart the truncate after a crash. * diff --git a/fs/ext3/ioctl.c b/fs/ext3/ioctl.c index ba1b54e23cae..4af574ce4a46 100644 --- a/fs/ext3/ioctl.c +++ b/fs/ext3/ioctl.c @@ -44,7 +44,7 @@ long ext3_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) if (get_user(flags, (int __user *) arg)) return -EFAULT; - err = mnt_want_write(filp->f_path.mnt); + err = mnt_want_write_file(filp); if (err) return err; @@ -110,7 +110,7 @@ flags_err: err = ext3_change_inode_journal_flag(inode, jflag); flags_out: mutex_unlock(&inode->i_mutex); - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return err; } case EXT3_IOC_GETVERSION: @@ -126,7 +126,7 @@ flags_out: if (!inode_owner_or_capable(inode)) return -EPERM; - err = mnt_want_write(filp->f_path.mnt); + err = mnt_want_write_file(filp); if (err) return err; if (get_user(generation, (int __user *) arg)) { @@ -134,10 +134,11 @@ flags_out: goto setversion_out; } + mutex_lock(&inode->i_mutex); handle = ext3_journal_start(inode, 1); if (IS_ERR(handle)) { err = PTR_ERR(handle); - goto setversion_out; + goto unlock_out; } err = ext3_reserve_inode_write(handle, inode, &iloc); if (err == 0) { @@ -146,8 +147,11 @@ flags_out: err = ext3_mark_iloc_dirty(handle, inode, &iloc); } ext3_journal_stop(handle); + +unlock_out: + mutex_unlock(&inode->i_mutex); setversion_out: - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return err; } case EXT3_IOC_GETRSVSZ: @@ -164,7 +168,7 @@ setversion_out: if (!test_opt(inode->i_sb, RESERVATION) ||!S_ISREG(inode->i_mode)) return -ENOTTY; - err = mnt_want_write(filp->f_path.mnt); + err = mnt_want_write_file(filp); if (err) return err; @@ -195,7 +199,7 @@ setversion_out: } mutex_unlock(&ei->truncate_mutex); setrsvsz_out: - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return err; } case EXT3_IOC_GROUP_EXTEND: { @@ -206,7 +210,7 @@ setrsvsz_out: if (!capable(CAP_SYS_RESOURCE)) return -EPERM; - err = mnt_want_write(filp->f_path.mnt); + err = mnt_want_write_file(filp); if (err) return err; @@ -221,7 +225,7 @@ setrsvsz_out: if (err == 0) err = err2; group_extend_out: - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return err; } case EXT3_IOC_GROUP_ADD: { @@ -232,7 +236,7 @@ group_extend_out: if (!capable(CAP_SYS_RESOURCE)) return -EPERM; - err = mnt_want_write(filp->f_path.mnt); + err = mnt_want_write_file(filp); if (err) return err; @@ -249,7 +253,7 @@ group_extend_out: if (err == 0) err = err2; group_add_out: - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return err; } case FITRIM: { diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index 642dc6d66dfd..e8e211795e9f 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -921,9 +921,12 @@ restart: num++; bh = ext3_getblk(NULL, dir, b++, 0, &err); bh_use[ra_max] = bh; - if (bh) - ll_rw_block(READ | REQ_META | REQ_PRIO, - 1, &bh); + if (bh && !bh_uptodate_or_lock(bh)) { + get_bh(bh); + bh->b_end_io = end_buffer_read_sync; + submit_bh(READ | REQ_META | REQ_PRIO, + bh); + } } } if ((bh = bh_use[ra_ptr++]) == NULL) @@ -1698,7 +1701,7 @@ static int ext3_add_nondir(handle_t *handle, * If the create succeeds, we fill in the inode information * with d_instantiate(). */ -static int ext3_create (struct inode * dir, struct dentry * dentry, int mode, +static int ext3_create (struct inode * dir, struct dentry * dentry, umode_t mode, struct nameidata *nd) { handle_t *handle; @@ -1732,7 +1735,7 @@ retry: } static int ext3_mknod (struct inode * dir, struct dentry *dentry, - int mode, dev_t rdev) + umode_t mode, dev_t rdev) { handle_t *handle; struct inode *inode; @@ -1768,7 +1771,7 @@ retry: return err; } -static int ext3_mkdir(struct inode * dir, struct dentry * dentry, int mode) +static int ext3_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode) { handle_t *handle; struct inode * inode; @@ -2272,7 +2275,7 @@ retry: err = PTR_ERR(handle); goto err_drop_inode; } - inc_nlink(inode); + set_nlink(inode, 1); err = ext3_orphan_del(handle, inode); if (err) { ext3_journal_stop(handle); diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 922d289aeeb3..726c7ef6cdf1 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -511,7 +511,6 @@ static int ext3_drop_inode(struct inode *inode) static void ext3_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(ext3_inode_cachep, EXT3_I(inode)); } @@ -611,9 +610,9 @@ static char *data_mode_string(unsigned long mode) * - it's set to a non-default value OR * - if the per-sb default is different from the global default */ -static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs) +static int ext3_show_options(struct seq_file *seq, struct dentry *root) { - struct super_block *sb = vfs->mnt_sb; + struct super_block *sb = root->d_sb; struct ext3_sb_info *sbi = EXT3_SB(sb); struct ext3_super_block *es = sbi->s_es; unsigned long def_mount_opts; @@ -2060,9 +2059,10 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) EXT3_SB(sb)->s_mount_state |= EXT3_ORPHAN_FS; ext3_orphan_cleanup(sb, es); EXT3_SB(sb)->s_mount_state &= ~EXT3_ORPHAN_FS; - if (needs_recovery) + if (needs_recovery) { + ext3_mark_recovery_complete(sb, es); ext3_msg(sb, KERN_INFO, "recovery complete"); - ext3_mark_recovery_complete(sb, es); + } ext3_msg(sb, KERN_INFO, "mounted filesystem with %s data mode", test_opt(sb,DATA_FLAGS) == EXT3_MOUNT_JOURNAL_DATA ? "journal": test_opt(sb,DATA_FLAGS) == EXT3_MOUNT_ORDERED_DATA ? "ordered": @@ -2230,11 +2230,11 @@ static journal_t *ext3_get_dev_journal(struct super_block *sb, goto out_bdev; } journal->j_private = sb; - ll_rw_block(READ, 1, &journal->j_sb_buffer); - wait_on_buffer(journal->j_sb_buffer); - if (!buffer_uptodate(journal->j_sb_buffer)) { - ext3_msg(sb, KERN_ERR, "I/O error on journal device"); - goto out_journal; + if (!bh_uptodate_or_lock(journal->j_sb_buffer)) { + if (bh_submit_read(journal->j_sb_buffer)) { + ext3_msg(sb, KERN_ERR, "I/O error on journal device"); + goto out_journal; + } } if (be32_to_cpu(journal->j_superblock->s_nr_users) != 1) { ext3_msg(sb, KERN_ERR, @@ -2910,7 +2910,7 @@ static int ext3_quota_on(struct super_block *sb, int type, int format_id, return -EINVAL; /* Quotafile not on the same filesystem? */ - if (path->mnt->mnt_sb != sb) + if (path->dentry->d_sb != sb) return -EXDEV; /* Journaling quota? */ if (EXT3_SB(sb)->s_qf_names[type]) { diff --git a/fs/ext3/xattr_security.c b/fs/ext3/xattr_security.c index 3c218b8a51d4..ea26f2acab94 100644 --- a/fs/ext3/xattr_security.c +++ b/fs/ext3/xattr_security.c @@ -3,7 +3,6 @@ * Handler for storing security labels as extended attributes. */ -#include <linux/module.h> #include <linux/slab.h> #include <linux/string.h> #include <linux/fs.h> diff --git a/fs/ext3/xattr_trusted.c b/fs/ext3/xattr_trusted.c index dc8edda9ffe0..2526a8829de8 100644 --- a/fs/ext3/xattr_trusted.c +++ b/fs/ext3/xattr_trusted.c @@ -5,7 +5,6 @@ * Copyright (C) 2003 by Andreas Gruenbacher, <a.gruenbacher@computer.org> */ -#include <linux/module.h> #include <linux/string.h> #include <linux/capability.h> #include <linux/fs.h> diff --git a/fs/ext3/xattr_user.c b/fs/ext3/xattr_user.c index 7a321974d584..b32e473a1e33 100644 --- a/fs/ext3/xattr_user.c +++ b/fs/ext3/xattr_user.c @@ -5,7 +5,6 @@ * Copyright (C) 2001 by Andreas Gruenbacher, <a.gruenbacher@computer.org> */ -#include <linux/module.h> #include <linux/string.h> #include <linux/fs.h> #include <linux/ext3_jbd.h> diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index f6dba4505f1c..f9e2cd8cf711 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -23,6 +23,8 @@ #include <trace/events/ext4.h> +static unsigned ext4_num_base_meta_clusters(struct super_block *sb, + ext4_group_t block_group); /* * balloc.c contains the blocks allocation and deallocation routines */ @@ -565,7 +567,7 @@ ext4_fsblk_t ext4_count_free_clusters(struct super_block *sb) brelse(bitmap_bh); printk(KERN_DEBUG "ext4_count_free_clusters: stored = %llu" ", computed = %llu, %llu\n", - EXT4_B2C(sbi, ext4_free_blocks_count(es)), + EXT4_B2C(EXT4_SB(sb), ext4_free_blocks_count(es)), desc_count, bitmap_count); return bitmap_count; #else @@ -668,7 +670,7 @@ unsigned long ext4_bg_num_gdb(struct super_block *sb, ext4_group_t group) * This function returns the number of file system metadata clusters at * the beginning of a block group, including the reserved gdt blocks. */ -unsigned ext4_num_base_meta_clusters(struct super_block *sb, +static unsigned ext4_num_base_meta_clusters(struct super_block *sb, ext4_group_t block_group) { struct ext4_sb_info *sbi = EXT4_SB(sb); diff --git a/fs/ext4/block_validity.c b/fs/ext4/block_validity.c index 8efb2f0a3447..3f11656bd72e 100644 --- a/fs/ext4/block_validity.c +++ b/fs/ext4/block_validity.c @@ -13,7 +13,6 @@ #include <linux/namei.h> #include <linux/quotaops.h> #include <linux/buffer_head.h> -#include <linux/module.h> #include <linux/swap.h> #include <linux/pagemap.h> #include <linux/blkdev.h> diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 5b0e26a1272d..513004fc3d84 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -511,6 +511,14 @@ struct ext4_new_group_data { __u32 free_blocks_count; }; +/* Indexes used to index group tables in ext4_new_group_data */ +enum { + BLOCK_BITMAP = 0, /* block bitmap */ + INODE_BITMAP, /* inode bitmap */ + INODE_TABLE, /* inode tables */ + GROUP_TABLE_COUNT, +}; + /* * Flags used by ext4_map_blocks() */ @@ -575,6 +583,7 @@ struct ext4_new_group_data { /* note ioctl 11 reserved for filesystem-independent FIEMAP ioctl */ #define EXT4_IOC_ALLOC_DA_BLKS _IO('f', 12) #define EXT4_IOC_MOVE_EXT _IOWR('f', 15, struct move_extent) +#define EXT4_IOC_RESIZE_FS _IOW('f', 16, __u64) #if defined(__KERNEL__) && defined(CONFIG_COMPAT) /* @@ -957,12 +966,13 @@ struct ext4_inode_info { #define test_opt2(sb, opt) (EXT4_SB(sb)->s_mount_opt2 & \ EXT4_MOUNT2_##opt) -#define ext4_set_bit __test_and_set_bit_le +#define ext4_test_and_set_bit __test_and_set_bit_le +#define ext4_set_bit __set_bit_le #define ext4_set_bit_atomic ext2_set_bit_atomic -#define ext4_clear_bit __test_and_clear_bit_le +#define ext4_test_and_clear_bit __test_and_clear_bit_le +#define ext4_clear_bit __clear_bit_le #define ext4_clear_bit_atomic ext2_clear_bit_atomic #define ext4_test_bit test_bit_le -#define ext4_find_first_zero_bit find_first_zero_bit_le #define ext4_find_next_zero_bit find_next_zero_bit_le #define ext4_find_next_bit find_next_bit_le @@ -1397,6 +1407,7 @@ static inline void ext4_clear_state_flags(struct ext4_inode_info *ei) #define EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE 0x0040 #define EXT4_FEATURE_RO_COMPAT_QUOTA 0x0100 #define EXT4_FEATURE_RO_COMPAT_BIGALLOC 0x0200 +#define EXT4_FEATURE_RO_COMPAT_METADATA_CSUM 0x0400 #define EXT4_FEATURE_INCOMPAT_COMPRESSION 0x0001 #define EXT4_FEATURE_INCOMPAT_FILETYPE 0x0002 @@ -1409,6 +1420,8 @@ static inline void ext4_clear_state_flags(struct ext4_inode_info *ei) #define EXT4_FEATURE_INCOMPAT_FLEX_BG 0x0200 #define EXT4_FEATURE_INCOMPAT_EA_INODE 0x0400 /* EA in inode */ #define EXT4_FEATURE_INCOMPAT_DIRDATA 0x1000 /* data in dirent */ +#define EXT4_FEATURE_INCOMPAT_INLINEDATA 0x2000 /* data in inode */ +#define EXT4_FEATURE_INCOMPAT_LARGEDIR 0x4000 /* >2GB or 3-lvl htree */ #define EXT2_FEATURE_COMPAT_SUPP EXT4_FEATURE_COMPAT_EXT_ATTR #define EXT2_FEATURE_INCOMPAT_SUPP (EXT4_FEATURE_INCOMPAT_FILETYPE| \ @@ -1790,8 +1803,6 @@ extern void ext4_init_block_bitmap(struct super_block *sb, extern unsigned ext4_free_clusters_after_init(struct super_block *sb, ext4_group_t block_group, struct ext4_group_desc *gdp); -extern unsigned ext4_num_base_meta_clusters(struct super_block *sb, - ext4_group_t block_group); extern unsigned ext4_num_overhead_clusters(struct super_block *sb, ext4_group_t block_group, struct ext4_group_desc *gdp); @@ -1819,7 +1830,7 @@ extern int ext4fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo); /* ialloc.c */ -extern struct inode *ext4_new_inode(handle_t *, struct inode *, int, +extern struct inode *ext4_new_inode(handle_t *, struct inode *, umode_t, const struct qstr *qstr, __u32 goal, uid_t *owner); extern void ext4_free_inode(handle_t *, struct inode *); @@ -1880,16 +1891,9 @@ 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_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_discard_partial_page_buffers(handle_t *handle, struct address_space *mapping, loff_t from, loff_t length, int flags); -extern 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); 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, @@ -1924,6 +1928,7 @@ extern int ext4_group_add(struct super_block *sb, extern int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, ext4_fsblk_t n_blocks_count); +extern int ext4_resize_fs(struct super_block *sb, ext4_fsblk_t n_blocks_count); /* super.c */ extern void *ext4_kvmalloc(size_t size, gfp_t flags); diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 61fa9e1614af..74f23c292e1b 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -29,7 +29,6 @@ * - smart tree reduction */ -#include <linux/module.h> #include <linux/fs.h> #include <linux/time.h> #include <linux/jbd2.h> @@ -1095,7 +1094,7 @@ static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode, le32_to_cpu(EXT_FIRST_INDEX(neh)->ei_block), ext4_idx_pblock(EXT_FIRST_INDEX(neh))); - neh->eh_depth = cpu_to_le16(neh->eh_depth + 1); + neh->eh_depth = cpu_to_le16(le16_to_cpu(neh->eh_depth) + 1); ext4_mark_inode_dirty(handle, inode); out: brelse(bh); @@ -2955,7 +2954,6 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, /* Pre-conditions */ BUG_ON(!ext4_ext_is_uninitialized(ex)); BUG_ON(!in_range(map->m_lblk, ee_block, ee_len)); - BUG_ON(map->m_lblk + map->m_len > ee_block + ee_len); /* * Attempt to transfer newly initialized blocks from the currently @@ -3282,6 +3280,9 @@ static int ext4_find_delalloc_range(struct inode *inode, ext4_lblk_t i, pg_lblk; pgoff_t index; + if (!test_opt(inode->i_sb, DELALLOC)) + return 0; + /* reverse search wont work if fs block size is less than page size */ if (inode->i_blkbits < PAGE_CACHE_SHIFT) search_hint_reverse = 0; @@ -3454,8 +3455,8 @@ ext4_ext_handle_uninitialized_extents(handle_t *handle, struct inode *inode, int err = 0; ext4_io_end_t *io = EXT4_I(inode)->cur_aio_dio; - ext_debug("ext4_ext_handle_uninitialized_extents: inode %lu, logical" - "block %llu, max_blocks %u, flags %d, allocated %u", + ext_debug("ext4_ext_handle_uninitialized_extents: inode %lu, logical " + "block %llu, max_blocks %u, flags %x, allocated %u\n", inode->i_ino, (unsigned long long)map->m_lblk, map->m_len, flags, allocated); ext4_ext_show_leaf(inode, path); @@ -3626,7 +3627,7 @@ static int get_implied_cluster_alloc(struct super_block *sb, struct ext4_sb_info *sbi = EXT4_SB(sb); ext4_lblk_t c_offset = map->m_lblk & (sbi->s_cluster_ratio-1); ext4_lblk_t ex_cluster_start, ex_cluster_end; - ext4_lblk_t rr_cluster_start, rr_cluster_end; + ext4_lblk_t rr_cluster_start; ext4_lblk_t ee_block = le32_to_cpu(ex->ee_block); ext4_fsblk_t ee_start = ext4_ext_pblock(ex); unsigned short ee_len = ext4_ext_get_actual_len(ex); @@ -3637,7 +3638,6 @@ static int get_implied_cluster_alloc(struct super_block *sb, /* The requested region passed into ext4_map_blocks() */ rr_cluster_start = EXT4_B2C(sbi, map->m_lblk); - rr_cluster_end = EXT4_B2C(sbi, map->m_lblk + map->m_len - 1); if ((rr_cluster_start == ex_cluster_end) || (rr_cluster_start == ex_cluster_start)) { diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index 00beb4f9cc4f..25d8c9781ad9 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -252,7 +252,7 @@ void ext4_free_inode(handle_t *handle, struct inode *inode) fatal = ext4_journal_get_write_access(handle, bh2); } ext4_lock_group(sb, block_group); - cleared = ext4_clear_bit(bit, bitmap_bh->b_data); + cleared = ext4_test_and_clear_bit(bit, bitmap_bh->b_data); if (fatal || !cleared) { ext4_unlock_group(sb, block_group); goto out; @@ -351,14 +351,14 @@ static void get_orlov_stats(struct super_block *sb, ext4_group_t g, */ static int find_group_orlov(struct super_block *sb, struct inode *parent, - ext4_group_t *group, int mode, + ext4_group_t *group, umode_t mode, const struct qstr *qstr) { ext4_group_t parent_group = EXT4_I(parent)->i_block_group; struct ext4_sb_info *sbi = EXT4_SB(sb); ext4_group_t real_ngroups = ext4_get_groups_count(sb); int inodes_per_group = EXT4_INODES_PER_GROUP(sb); - unsigned int freei, avefreei; + unsigned int freei, avefreei, grp_free; ext4_fsblk_t freeb, avefreec; unsigned int ndirs; int max_dirs, min_inodes; @@ -477,8 +477,8 @@ fallback_retry: for (i = 0; i < ngroups; i++) { grp = (parent_group + i) % ngroups; desc = ext4_get_group_desc(sb, grp, NULL); - if (desc && ext4_free_inodes_count(sb, desc) && - ext4_free_inodes_count(sb, desc) >= avefreei) { + grp_free = ext4_free_inodes_count(sb, desc); + if (desc && grp_free && grp_free >= avefreei) { *group = grp; return 0; } @@ -497,7 +497,7 @@ fallback_retry: } static int find_group_other(struct super_block *sb, struct inode *parent, - ext4_group_t *group, int mode) + ext4_group_t *group, umode_t mode) { ext4_group_t parent_group = EXT4_I(parent)->i_block_group; ext4_group_t i, last, ngroups = ext4_get_groups_count(sb); @@ -602,7 +602,7 @@ static int find_group_other(struct super_block *sb, struct inode *parent, */ static int ext4_claim_inode(struct super_block *sb, struct buffer_head *inode_bitmap_bh, - unsigned long ino, ext4_group_t group, int mode) + unsigned long ino, ext4_group_t group, umode_t mode) { int free = 0, retval = 0, count; struct ext4_sb_info *sbi = EXT4_SB(sb); @@ -618,7 +618,7 @@ static int ext4_claim_inode(struct super_block *sb, */ down_read(&grp->alloc_sem); ext4_lock_group(sb, group); - if (ext4_set_bit(ino, inode_bitmap_bh->b_data)) { + if (ext4_test_and_set_bit(ino, inode_bitmap_bh->b_data)) { /* not a free inode */ retval = 1; goto err_ret; @@ -690,7 +690,7 @@ err_ret: * For other inodes, search forward from the parent directory's block * group to find a free inode. */ -struct inode *ext4_new_inode(handle_t *handle, struct inode *dir, int mode, +struct inode *ext4_new_inode(handle_t *handle, struct inode *dir, umode_t mode, const struct qstr *qstr, __u32 goal, uid_t *owner) { struct super_block *sb; @@ -885,8 +885,12 @@ got: if (IS_DIRSYNC(inode)) ext4_handle_sync(handle); if (insert_inode_locked(inode) < 0) { - err = -EINVAL; - goto fail_drop; + /* + * Likely a bitmap corruption causing inode to be allocated + * twice. + */ + err = -EIO; + goto fail; } spin_lock(&sbi->s_next_gen_lock); inode->i_generation = sbi->s_next_generation++; diff --git a/fs/ext4/indirect.c b/fs/ext4/indirect.c index 3cfc73fbca8e..830e1b2bf145 100644 --- a/fs/ext4/indirect.c +++ b/fs/ext4/indirect.c @@ -20,7 +20,6 @@ * (sct@redhat.com), 1993, 1998 */ -#include <linux/module.h> #include "ext4_jbd2.h" #include "truncate.h" diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 240f6e2dc7ee..feaa82fe629d 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -18,7 +18,6 @@ * Assorted race fixes, rewrite of ext4_get_block() by Al Viro, 2000 */ -#include <linux/module.h> #include <linux/fs.h> #include <linux/time.h> #include <linux/jbd2.h> @@ -72,6 +71,9 @@ static int ext4_set_bh_endio(struct buffer_head *bh, struct inode *inode); static void ext4_end_io_buffer_write(struct buffer_head *bh, int uptodate); 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); /* * Test whether an inode is a fast symlink. @@ -1339,8 +1341,11 @@ static int mpage_da_submit_io(struct mpage_da_data *mpd, clear_buffer_unwritten(bh); } - /* skip page if block allocation undone */ - if (buffer_delay(bh) || 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; @@ -1878,7 +1883,7 @@ static void ext4_end_io_buffer_write(struct buffer_head *bh, int uptodate); * a[0] = 'a'; * truncate(f, 4096); * we have in the page first buffer_head mapped via page_mkwrite call back - * but other bufer_heads would be unmapped but dirty(dirty done via the + * but other buffer_heads would be unmapped but dirty (dirty done via the * do_wp_page). So writepage should write the first block. If we modify * the mmap area beyond 1024 we will again get a page_fault and the * page_mkwrite callback will do the block allocation and mark the @@ -2270,6 +2275,7 @@ retry: 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; } @@ -2386,7 +2392,6 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping, pgoff_t index; struct inode *inode = mapping->host; handle_t *handle; - loff_t page_len; index = pos >> PAGE_CACHE_SHIFT; @@ -2433,13 +2438,6 @@ retry: */ if (pos + len > inode->i_size) ext4_truncate_failed_write(inode); - } else { - page_len = pos & (PAGE_CACHE_SIZE - 1); - if (page_len > 0) { - ret = ext4_discard_partial_page_buffers_no_lock(handle, - inode, page, pos - page_len, page_len, - EXT4_DISCARD_PARTIAL_PG_ZERO_UNMAPPED); - } } if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) @@ -2482,7 +2480,6 @@ static int ext4_da_write_end(struct file *file, loff_t new_i_size; unsigned long start, end; int write_mode = (int)(unsigned long)fsdata; - loff_t page_len; if (write_mode == FALL_BACK_TO_NONDELALLOC) { if (ext4_should_order_data(inode)) { @@ -2507,7 +2504,7 @@ static int ext4_da_write_end(struct file *file, */ new_i_size = pos + copied; - if (new_i_size > EXT4_I(inode)->i_disksize) { + if (copied && new_i_size > EXT4_I(inode)->i_disksize) { if (ext4_da_should_update_i_disksize(page, end)) { down_write(&EXT4_I(inode)->i_data_sem); if (new_i_size > EXT4_I(inode)->i_disksize) { @@ -2531,16 +2528,6 @@ static int ext4_da_write_end(struct file *file, } ret2 = generic_write_end(file, mapping, pos, len, copied, page, fsdata); - - page_len = PAGE_CACHE_SIZE - - ((pos + copied - 1) & (PAGE_CACHE_SIZE - 1)); - - if (page_len > 0) { - ret = ext4_discard_partial_page_buffers_no_lock(handle, - inode, page, pos + copied - 1, page_len, - EXT4_DISCARD_PARTIAL_PG_ZERO_UNMAPPED); - } - copied = ret2; if (ret2 < 0) ret = ret2; @@ -2775,15 +2762,16 @@ static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset, if (!io_end || !size) goto out; - ext_debug("ext4_end_io_dio(): io_end 0x%p" + ext_debug("ext4_end_io_dio(): io_end 0x%p " "for inode %lu, iocb 0x%p, offset %llu, size %llu\n", iocb->private, io_end->inode->i_ino, iocb, 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); - iocb->private = NULL; out: if (is_async) aio_complete(iocb, ret, 0); @@ -2807,7 +2795,6 @@ out: /* queue the work to convert unwritten extents to written */ queue_work(wq, &io_end->work); - iocb->private = NULL; /* XXX: probably should move into the real I/O completion handler */ inode_dio_done(inode); @@ -3176,7 +3163,7 @@ int ext4_discard_partial_page_buffers(handle_t *handle, * * Returns zero on sucess or negative on failure. */ -int ext4_discard_partial_page_buffers_no_lock(handle_t *handle, +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) { @@ -3202,26 +3189,8 @@ int ext4_discard_partial_page_buffers_no_lock(handle_t *handle, iblock = index << (PAGE_CACHE_SHIFT - inode->i_sb->s_blocksize_bits); - if (!page_has_buffers(page)) { - /* - * If the range to be discarded covers a partial block - * we need to get the page buffers. This is because - * partial blocks cannot be released and the page needs - * to be updated with the contents of the block before - * we write the zeros on top of it. - */ - if ((from & (blocksize - 1)) || - ((from + length) & (blocksize - 1))) { - create_empty_buffers(page, blocksize, 0); - } else { - /* - * If there are no partial blocks, - * there is nothing to update, - * so we can return now - */ - return 0; - } - } + if (!page_has_buffers(page)) + create_empty_buffers(page, blocksize, 0); /* Find the buffer that contains "offset" */ bh = page_buffers(page); @@ -3334,126 +3303,6 @@ next: return err; } -/* - * 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_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; - - blocksize = inode->i_sb->s_blocksize; - length = blocksize - (offset & (blocksize - 1)); - - return ext4_block_zero_page_range(handle, mapping, from, length); -} - -/* - * 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' - */ -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 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; - - page = find_or_create_page(mapping, from >> PAGE_CACHE_SHIFT, - mapping_gfp_mask(mapping) & ~__GFP_FS); - if (!page) - return -ENOMEM; - - 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 block - */ - if (length > max || length < 0) - length = max; - - iblock = index << (PAGE_CACHE_SHIFT - inode->i_sb->s_blocksize_bits); - - if (!page_has_buffers(page)) - create_empty_buffers(page, blocksize, 0); - - /* Find the buffer that contains "offset" */ - bh = page_buffers(page); - pos = blocksize; - while (offset >= pos) { - bh = bh->b_this_page; - iblock++; - pos += blocksize; - } - - err = 0; - 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_TRACE(bh, "still unmapped"); - goto unlock; - } - } - - /* 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 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"); - - err = 0; - if (ext4_should_journal_data(inode)) { - err = ext4_handle_dirty_metadata(handle, inode, bh); - } else - mark_buffer_dirty(bh); - -unlock: - unlock_page(page); - page_cache_release(page); - return err; -} - int ext4_can_truncate(struct inode *inode) { if (S_ISREG(inode->i_mode)) @@ -3502,7 +3351,7 @@ int ext4_punch_hole(struct file *file, loff_t offset, loff_t length) * transaction, and VFS/VM ensures that ext4_truncate() cannot run * simultaneously on behalf of the same inode. * - * As we work through the truncate and commmit bits of it to the journal there + * As we work through the truncate and commit bits of it to the journal there * is one core, guiding principle: the file's tree must always be consistent on * disk. We must be able to restart the truncate after a crash. * @@ -4680,9 +4529,19 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val) return 0; if (is_journal_aborted(journal)) return -EROFS; + /* We have to allocate physical blocks for delalloc blocks + * before flushing journal. otherwise delalloc blocks can not + * be allocated any more. even more truncate on delalloc blocks + * could trigger BUG by flushing delalloc blocks in journal. + * There is no delalloc block in non-journal data mode. + */ + if (val && test_opt(inode->i_sb, DELALLOC)) { + err = ext4_alloc_da_blocks(inode); + if (err < 0) + return err; + } jbd2_journal_lock_updates(journal); - jbd2_journal_flush(journal); /* * OK, there are no updates running now, and all cached data is @@ -4694,8 +4553,10 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val) if (val) ext4_set_inode_flag(inode, EXT4_INODE_JOURNAL_DATA); - else + else { + jbd2_journal_flush(journal); ext4_clear_inode_flag(inode, EXT4_INODE_JOURNAL_DATA); + } ext4_set_aops(inode); jbd2_journal_unlock_updates(journal); diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index a56796814d6a..6eee25591b81 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -18,6 +18,8 @@ #include "ext4_jbd2.h" #include "ext4.h" +#define MAX_32_NUM ((((unsigned long long) 1) << 32) - 1) + long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { struct inode *inode = filp->f_dentry->d_inode; @@ -45,7 +47,7 @@ long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) if (get_user(flags, (int __user *) arg)) return -EFAULT; - err = mnt_want_write(filp->f_path.mnt); + err = mnt_want_write_file(filp); if (err) return err; @@ -134,7 +136,7 @@ flags_err: err = ext4_ext_migrate(inode); flags_out: mutex_unlock(&inode->i_mutex); - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return err; } case EXT4_IOC_GETVERSION: @@ -150,7 +152,7 @@ flags_out: if (!inode_owner_or_capable(inode)) return -EPERM; - err = mnt_want_write(filp->f_path.mnt); + err = mnt_want_write_file(filp); if (err) return err; if (get_user(generation, (int __user *) arg)) { @@ -158,10 +160,11 @@ flags_out: goto setversion_out; } + mutex_lock(&inode->i_mutex); handle = ext4_journal_start(inode, 1); if (IS_ERR(handle)) { err = PTR_ERR(handle); - goto setversion_out; + goto unlock_out; } err = ext4_reserve_inode_write(handle, inode, &iloc); if (err == 0) { @@ -170,8 +173,11 @@ flags_out: err = ext4_mark_iloc_dirty(handle, inode, &iloc); } ext4_journal_stop(handle); + +unlock_out: + mutex_unlock(&inode->i_mutex); setversion_out: - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return err; } case EXT4_IOC_GROUP_EXTEND: { @@ -182,19 +188,22 @@ setversion_out: if (err) return err; - if (get_user(n_blocks_count, (__u32 __user *)arg)) - return -EFAULT; + if (get_user(n_blocks_count, (__u32 __user *)arg)) { + err = -EFAULT; + goto group_extend_out; + } if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_BIGALLOC)) { ext4_msg(sb, KERN_ERR, "Online resizing not supported with bigalloc"); - return -EOPNOTSUPP; + err = -EOPNOTSUPP; + goto group_extend_out; } - err = mnt_want_write(filp->f_path.mnt); + err = mnt_want_write_file(filp); if (err) - return err; + goto group_extend_out; err = ext4_group_extend(sb, EXT4_SB(sb)->s_es, n_blocks_count); if (EXT4_SB(sb)->s_journal) { @@ -204,9 +213,9 @@ setversion_out: } if (err == 0) err = err2; - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); +group_extend_out: ext4_resize_end(sb); - return err; } @@ -240,15 +249,14 @@ setversion_out: return -EOPNOTSUPP; } - err = mnt_want_write(filp->f_path.mnt); + err = mnt_want_write_file(filp); if (err) goto mext_out; err = ext4_move_extents(filp, donor_filp, me.orig_start, me.donor_start, me.len, &me.moved_len); + mnt_drop_write_file(filp); mnt_drop_write(filp->f_path.mnt); - if (me.moved_len > 0) - file_remove_suid(donor_filp); if (copy_to_user((struct move_extent __user *)arg, &me, sizeof(me))) @@ -267,19 +275,22 @@ mext_out: return err; if (copy_from_user(&input, (struct ext4_new_group_input __user *)arg, - sizeof(input))) - return -EFAULT; + sizeof(input))) { + err = -EFAULT; + goto group_add_out; + } if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_BIGALLOC)) { ext4_msg(sb, KERN_ERR, "Online resizing not supported with bigalloc"); - return -EOPNOTSUPP; + err = -EOPNOTSUPP; + goto group_add_out; } - err = mnt_want_write(filp->f_path.mnt); + err = mnt_want_write_file(filp); if (err) - return err; + goto group_add_out; err = ext4_group_add(sb, &input); if (EXT4_SB(sb)->s_journal) { @@ -289,9 +300,9 @@ mext_out: } if (err == 0) err = err2; - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); +group_add_out: ext4_resize_end(sb); - return err; } @@ -301,7 +312,7 @@ mext_out: if (!inode_owner_or_capable(inode)) return -EACCES; - err = mnt_want_write(filp->f_path.mnt); + err = mnt_want_write_file(filp); if (err) return err; /* @@ -313,7 +324,7 @@ mext_out: mutex_lock(&(inode->i_mutex)); err = ext4_ext_migrate(inode); mutex_unlock(&(inode->i_mutex)); - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return err; } @@ -323,11 +334,65 @@ mext_out: if (!inode_owner_or_capable(inode)) return -EACCES; - err = mnt_want_write(filp->f_path.mnt); + err = mnt_want_write_file(filp); if (err) return err; err = ext4_alloc_da_blocks(inode); + mnt_drop_write_file(filp); + return err; + } + + case EXT4_IOC_RESIZE_FS: { + ext4_fsblk_t n_blocks_count; + struct super_block *sb = inode->i_sb; + int err = 0, err2 = 0; + + if (EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_BIGALLOC)) { + ext4_msg(sb, KERN_ERR, + "Online resizing not (yet) supported with bigalloc"); + return -EOPNOTSUPP; + } + + if (EXT4_HAS_INCOMPAT_FEATURE(sb, + EXT4_FEATURE_INCOMPAT_META_BG)) { + ext4_msg(sb, KERN_ERR, + "Online resizing not (yet) supported with meta_bg"); + return -EOPNOTSUPP; + } + + if (copy_from_user(&n_blocks_count, (__u64 __user *)arg, + sizeof(__u64))) { + return -EFAULT; + } + + if (n_blocks_count > MAX_32_NUM && + !EXT4_HAS_INCOMPAT_FEATURE(sb, + EXT4_FEATURE_INCOMPAT_64BIT)) { + ext4_msg(sb, KERN_ERR, + "File system only supports 32-bit block numbers"); + return -EOPNOTSUPP; + } + + err = ext4_resize_begin(sb); + if (err) + return err; + + err = mnt_want_write(filp->f_path.mnt); + if (err) + goto resizefs_out; + + err = ext4_resize_fs(sb, n_blocks_count); + if (EXT4_SB(sb)->s_journal) { + jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal); + err2 = jbd2_journal_flush(EXT4_SB(sb)->s_journal); + jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal); + } + if (err == 0) + err = err2; mnt_drop_write(filp->f_path.mnt); +resizefs_out: + ext4_resize_end(sb); return err; } @@ -429,6 +494,7 @@ long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) } case EXT4_IOC_MOVE_EXT: case FITRIM: + case EXT4_IOC_RESIZE_FS: break; default: return -ENOIOCTLCMD; diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index e2d8be8f28bf..cb990b21c698 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -3671,7 +3671,7 @@ ext4_mb_release_group_pa(struct ext4_buddy *e4b, ext4_group_t group; ext4_grpblk_t bit; - trace_ext4_mb_release_group_pa(pa); + trace_ext4_mb_release_group_pa(sb, pa); BUG_ON(pa->pa_deleted == 0); ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, &bit); BUG_ON(group != e4b->bd_group && pa->pa_len != 0); diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c index 16ac228dbec6..e7d6bb0acfa6 100644 --- a/fs/ext4/migrate.c +++ b/fs/ext4/migrate.c @@ -12,7 +12,6 @@ * */ -#include <linux/module.h> #include <linux/slab.h> #include "ext4_jbd2.h" diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index aa4c782c9dd7..2043f482375d 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -1736,7 +1736,7 @@ static int ext4_add_nondir(handle_t *handle, * If the create succeeds, we fill in the inode information * with d_instantiate(). */ -static int ext4_create(struct inode *dir, struct dentry *dentry, int mode, +static int ext4_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { handle_t *handle; @@ -1770,7 +1770,7 @@ retry: } static int ext4_mknod(struct inode *dir, struct dentry *dentry, - int mode, dev_t rdev) + umode_t mode, dev_t rdev) { handle_t *handle; struct inode *inode; @@ -1806,7 +1806,7 @@ retry: return err; } -static int ext4_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int ext4_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { handle_t *handle; struct inode *inode; @@ -2315,7 +2315,7 @@ retry: err = PTR_ERR(handle); goto err_drop_inode; } - inc_nlink(inode); + set_nlink(inode, 1); err = ext4_orphan_del(handle, inode); if (err) { ext4_journal_stop(handle); diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c index 7ce1d0b19c94..475851896518 100644 --- a/fs/ext4/page-io.c +++ b/fs/ext4/page-io.c @@ -6,7 +6,6 @@ * Written by Theodore Ts'o, 2010. */ -#include <linux/module.h> #include <linux/fs.h> #include <linux/time.h> #include <linux/jbd2.h> @@ -385,6 +384,18 @@ int ext4_bio_write_page(struct ext4_io_submit *io, block_end = block_start + blocksize; if (block_start >= len) { + /* + * Comments copied from block_write_full_page_endio: + * + * The page straddles i_size. It must be zeroed out on + * each and every writepage invocation because it may + * be mmapped. "A file is mapped in multiples of the + * page size. For a file that is not a multiple of + * the page size, the remaining memory is zeroed when + * mapped, and writes to that region are not written + * out to the file." + */ + zero_user_segment(page, block_start, block_end); clear_buffer_dirty(bh); set_buffer_uptodate(bh); continue; diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index 996780ab4f4e..f9d948f0eb86 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -134,6 +134,172 @@ static int verify_group_input(struct super_block *sb, return err; } +/* + * ext4_new_flex_group_data is used by 64bit-resize interface to add a flex + * group each time. + */ +struct ext4_new_flex_group_data { + struct ext4_new_group_data *groups; /* new_group_data for groups + in the flex group */ + __u16 *bg_flags; /* block group flags of groups + in @groups */ + ext4_group_t count; /* number of groups in @groups + */ +}; + +/* + * alloc_flex_gd() allocates a ext4_new_flex_group_data with size of + * @flexbg_size. + * + * Returns NULL on failure otherwise address of the allocated structure. + */ +static struct ext4_new_flex_group_data *alloc_flex_gd(unsigned long flexbg_size) +{ + struct ext4_new_flex_group_data *flex_gd; + + flex_gd = kmalloc(sizeof(*flex_gd), GFP_NOFS); + if (flex_gd == NULL) + goto out3; + + flex_gd->count = flexbg_size; + + flex_gd->groups = kmalloc(sizeof(struct ext4_new_group_data) * + flexbg_size, GFP_NOFS); + if (flex_gd->groups == NULL) + goto out2; + + flex_gd->bg_flags = kmalloc(flexbg_size * sizeof(__u16), GFP_NOFS); + if (flex_gd->bg_flags == NULL) + goto out1; + + return flex_gd; + +out1: + kfree(flex_gd->groups); +out2: + kfree(flex_gd); +out3: + return NULL; +} + +static void free_flex_gd(struct ext4_new_flex_group_data *flex_gd) +{ + kfree(flex_gd->bg_flags); + kfree(flex_gd->groups); + kfree(flex_gd); +} + +/* + * ext4_alloc_group_tables() allocates block bitmaps, inode bitmaps + * and inode tables for a flex group. + * + * This function is used by 64bit-resize. Note that this function allocates + * group tables from the 1st group of groups contained by @flexgd, which may + * be a partial of a flex group. + * + * @sb: super block of fs to which the groups belongs + */ +static void ext4_alloc_group_tables(struct super_block *sb, + struct ext4_new_flex_group_data *flex_gd, + int flexbg_size) +{ + struct ext4_new_group_data *group_data = flex_gd->groups; + struct ext4_super_block *es = EXT4_SB(sb)->s_es; + ext4_fsblk_t start_blk; + ext4_fsblk_t last_blk; + ext4_group_t src_group; + ext4_group_t bb_index = 0; + ext4_group_t ib_index = 0; + ext4_group_t it_index = 0; + ext4_group_t group; + ext4_group_t last_group; + unsigned overhead; + + BUG_ON(flex_gd->count == 0 || group_data == NULL); + + src_group = group_data[0].group; + last_group = src_group + flex_gd->count - 1; + + BUG_ON((flexbg_size > 1) && ((src_group & ~(flexbg_size - 1)) != + (last_group & ~(flexbg_size - 1)))); +next_group: + group = group_data[0].group; + start_blk = ext4_group_first_block_no(sb, src_group); + last_blk = start_blk + group_data[src_group - group].blocks_count; + + overhead = ext4_bg_has_super(sb, src_group) ? + (1 + ext4_bg_num_gdb(sb, src_group) + + le16_to_cpu(es->s_reserved_gdt_blocks)) : 0; + + start_blk += overhead; + + BUG_ON(src_group >= group_data[0].group + flex_gd->count); + /* We collect contiguous blocks as much as possible. */ + src_group++; + for (; src_group <= last_group; src_group++) + if (!ext4_bg_has_super(sb, src_group)) + last_blk += group_data[src_group - group].blocks_count; + else + break; + + /* Allocate block bitmaps */ + for (; bb_index < flex_gd->count; bb_index++) { + if (start_blk >= last_blk) + goto next_group; + group_data[bb_index].block_bitmap = start_blk++; + ext4_get_group_no_and_offset(sb, start_blk - 1, &group, NULL); + group -= group_data[0].group; + group_data[group].free_blocks_count--; + if (flexbg_size > 1) + flex_gd->bg_flags[group] &= ~EXT4_BG_BLOCK_UNINIT; + } + + /* Allocate inode bitmaps */ + for (; ib_index < flex_gd->count; ib_index++) { + if (start_blk >= last_blk) + goto next_group; + group_data[ib_index].inode_bitmap = start_blk++; + ext4_get_group_no_and_offset(sb, start_blk - 1, &group, NULL); + group -= group_data[0].group; + group_data[group].free_blocks_count--; + if (flexbg_size > 1) + flex_gd->bg_flags[group] &= ~EXT4_BG_BLOCK_UNINIT; + } + + /* Allocate inode tables */ + for (; it_index < flex_gd->count; it_index++) { + if (start_blk + EXT4_SB(sb)->s_itb_per_group > last_blk) + goto next_group; + group_data[it_index].inode_table = start_blk; + ext4_get_group_no_and_offset(sb, start_blk, &group, NULL); + group -= group_data[0].group; + group_data[group].free_blocks_count -= + EXT4_SB(sb)->s_itb_per_group; + if (flexbg_size > 1) + flex_gd->bg_flags[group] &= ~EXT4_BG_BLOCK_UNINIT; + + start_blk += EXT4_SB(sb)->s_itb_per_group; + } + + if (test_opt(sb, DEBUG)) { + int i; + group = group_data[0].group; + + printk(KERN_DEBUG "EXT4-fs: adding a flex group with " + "%d groups, flexbg size is %d:\n", flex_gd->count, + flexbg_size); + + for (i = 0; i < flex_gd->count; i++) { + printk(KERN_DEBUG "adding %s group %u: %u " + "blocks (%d free)\n", + ext4_bg_has_super(sb, group + i) ? "normal" : + "no-super", group + i, + group_data[i].blocks_count, + group_data[i].free_blocks_count); + } + } +} + static struct buffer_head *bclean(handle_t *handle, struct super_block *sb, ext4_fsblk_t blk) { @@ -179,131 +345,250 @@ static int extend_or_restart_transaction(handle_t *handle, int thresh) } /* - * Set up the block and inode bitmaps, and the inode table for the new group. + * set_flexbg_block_bitmap() mark @count blocks starting from @block used. + * + * Helper function for ext4_setup_new_group_blocks() which set . + * + * @sb: super block + * @handle: journal handle + * @flex_gd: flex group data + */ +static int set_flexbg_block_bitmap(struct super_block *sb, handle_t *handle, + struct ext4_new_flex_group_data *flex_gd, + ext4_fsblk_t block, ext4_group_t count) +{ + ext4_group_t count2; + + ext4_debug("mark blocks [%llu/%u] used\n", block, count); + for (count2 = count; count > 0; count -= count2, block += count2) { + ext4_fsblk_t start; + struct buffer_head *bh; + ext4_group_t group; + int err; + + ext4_get_group_no_and_offset(sb, block, &group, NULL); + start = ext4_group_first_block_no(sb, group); + group -= flex_gd->groups[0].group; + + count2 = sb->s_blocksize * 8 - (block - start); + if (count2 > count) + count2 = count; + + if (flex_gd->bg_flags[group] & EXT4_BG_BLOCK_UNINIT) { + BUG_ON(flex_gd->count > 1); + continue; + } + + err = extend_or_restart_transaction(handle, 1); + if (err) + return err; + + bh = sb_getblk(sb, flex_gd->groups[group].block_bitmap); + if (!bh) + return -EIO; + + err = ext4_journal_get_write_access(handle, bh); + if (err) + return err; + ext4_debug("mark block bitmap %#04llx (+%llu/%u)\n", block, + block - start, count2); + ext4_set_bits(bh->b_data, block - start, count2); + + err = ext4_handle_dirty_metadata(handle, NULL, bh); + if (unlikely(err)) + return err; + brelse(bh); + } + + return 0; +} + +/* + * Set up the block and inode bitmaps, and the inode table for the new groups. * This doesn't need to be part of the main transaction, since we are only * changing blocks outside the actual filesystem. We still do journaling to * ensure the recovery is correct in case of a failure just after resize. * If any part of this fails, we simply abort the resize. + * + * setup_new_flex_group_blocks handles a flex group as follow: + * 1. copy super block and GDT, and initialize group tables if necessary. + * In this step, we only set bits in blocks bitmaps for blocks taken by + * super block and GDT. + * 2. allocate group tables in block bitmaps, that is, set bits in block + * bitmap for blocks taken by group tables. */ -static int setup_new_group_blocks(struct super_block *sb, - struct ext4_new_group_data *input) +static int setup_new_flex_group_blocks(struct super_block *sb, + struct ext4_new_flex_group_data *flex_gd) { + int group_table_count[] = {1, 1, EXT4_SB(sb)->s_itb_per_group}; + ext4_fsblk_t start; + ext4_fsblk_t block; struct ext4_sb_info *sbi = EXT4_SB(sb); - ext4_fsblk_t start = ext4_group_first_block_no(sb, input->group); - int reserved_gdb = ext4_bg_has_super(sb, input->group) ? - le16_to_cpu(sbi->s_es->s_reserved_gdt_blocks) : 0; - unsigned long gdblocks = ext4_bg_num_gdb(sb, input->group); - struct buffer_head *bh; + struct ext4_super_block *es = sbi->s_es; + struct ext4_new_group_data *group_data = flex_gd->groups; + __u16 *bg_flags = flex_gd->bg_flags; handle_t *handle; - ext4_fsblk_t block; - ext4_grpblk_t bit; - int i; - int err = 0, err2; + ext4_group_t group, count; + struct buffer_head *bh = NULL; + int reserved_gdb, i, j, err = 0, err2; + + BUG_ON(!flex_gd->count || !group_data || + group_data[0].group != sbi->s_groups_count); + + reserved_gdb = le16_to_cpu(es->s_reserved_gdt_blocks); /* This transaction may be extended/restarted along the way */ handle = ext4_journal_start_sb(sb, EXT4_MAX_TRANS_DATA); - if (IS_ERR(handle)) return PTR_ERR(handle); - BUG_ON(input->group != sbi->s_groups_count); + group = group_data[0].group; + for (i = 0; i < flex_gd->count; i++, group++) { + unsigned long gdblocks; - /* Copy all of the GDT blocks into the backup in this group */ - for (i = 0, bit = 1, block = start + 1; - i < gdblocks; i++, block++, bit++) { - struct buffer_head *gdb; + gdblocks = ext4_bg_num_gdb(sb, group); + start = ext4_group_first_block_no(sb, group); - ext4_debug("update backup group %#04llx (+%d)\n", block, bit); - err = extend_or_restart_transaction(handle, 1); - if (err) - goto exit_journal; + /* Copy all of the GDT blocks into the backup in this group */ + for (j = 0, block = start + 1; j < gdblocks; j++, block++) { + struct buffer_head *gdb; - gdb = sb_getblk(sb, block); - if (!gdb) { - err = -EIO; - goto exit_journal; - } - if ((err = ext4_journal_get_write_access(handle, gdb))) { + ext4_debug("update backup group %#04llx\n", block); + err = extend_or_restart_transaction(handle, 1); + if (err) + goto out; + + gdb = sb_getblk(sb, block); + if (!gdb) { + err = -EIO; + goto out; + } + + err = ext4_journal_get_write_access(handle, gdb); + if (err) { + brelse(gdb); + goto out; + } + memcpy(gdb->b_data, sbi->s_group_desc[j]->b_data, + gdb->b_size); + set_buffer_uptodate(gdb); + + err = ext4_handle_dirty_metadata(handle, NULL, gdb); + if (unlikely(err)) { + brelse(gdb); + goto out; + } brelse(gdb); - goto exit_journal; } - memcpy(gdb->b_data, sbi->s_group_desc[i]->b_data, gdb->b_size); - set_buffer_uptodate(gdb); - err = ext4_handle_dirty_metadata(handle, NULL, gdb); - if (unlikely(err)) { - brelse(gdb); - goto exit_journal; + + /* Zero out all of the reserved backup group descriptor + * table blocks + */ + if (ext4_bg_has_super(sb, group)) { + err = sb_issue_zeroout(sb, gdblocks + start + 1, + reserved_gdb, GFP_NOFS); + if (err) + goto out; } - brelse(gdb); - } - /* Zero out all of the reserved backup group descriptor table blocks */ - ext4_debug("clear inode table blocks %#04llx -> %#04lx\n", - block, sbi->s_itb_per_group); - err = sb_issue_zeroout(sb, gdblocks + start + 1, reserved_gdb, - GFP_NOFS); - if (err) - goto exit_journal; + /* Initialize group tables of the grop @group */ + if (!(bg_flags[i] & EXT4_BG_INODE_ZEROED)) + goto handle_bb; - err = extend_or_restart_transaction(handle, 2); - if (err) - goto exit_journal; + /* Zero out all of the inode table blocks */ + block = group_data[i].inode_table; + ext4_debug("clear inode table blocks %#04llx -> %#04lx\n", + block, sbi->s_itb_per_group); + err = sb_issue_zeroout(sb, block, sbi->s_itb_per_group, + GFP_NOFS); + if (err) + goto out; - bh = bclean(handle, sb, input->block_bitmap); - if (IS_ERR(bh)) { - err = PTR_ERR(bh); - goto exit_journal; - } +handle_bb: + if (bg_flags[i] & EXT4_BG_BLOCK_UNINIT) + goto handle_ib; - if (ext4_bg_has_super(sb, input->group)) { - ext4_debug("mark backup group tables %#04llx (+0)\n", start); - ext4_set_bits(bh->b_data, 0, gdblocks + reserved_gdb + 1); - } + /* Initialize block bitmap of the @group */ + block = group_data[i].block_bitmap; + err = extend_or_restart_transaction(handle, 1); + if (err) + goto out; - ext4_debug("mark block bitmap %#04llx (+%llu)\n", input->block_bitmap, - input->block_bitmap - start); - ext4_set_bit(input->block_bitmap - start, bh->b_data); - ext4_debug("mark inode bitmap %#04llx (+%llu)\n", input->inode_bitmap, - input->inode_bitmap - start); - ext4_set_bit(input->inode_bitmap - start, bh->b_data); - - /* Zero out all of the inode table blocks */ - block = input->inode_table; - ext4_debug("clear inode table blocks %#04llx -> %#04lx\n", - block, sbi->s_itb_per_group); - err = sb_issue_zeroout(sb, block, sbi->s_itb_per_group, GFP_NOFS); - if (err) - goto exit_bh; - ext4_set_bits(bh->b_data, input->inode_table - start, - sbi->s_itb_per_group); + bh = bclean(handle, sb, block); + if (IS_ERR(bh)) { + err = PTR_ERR(bh); + goto out; + } + if (ext4_bg_has_super(sb, group)) { + ext4_debug("mark backup superblock %#04llx (+0)\n", + start); + ext4_set_bits(bh->b_data, 0, gdblocks + reserved_gdb + + 1); + } + ext4_mark_bitmap_end(group_data[i].blocks_count, + sb->s_blocksize * 8, bh->b_data); + err = ext4_handle_dirty_metadata(handle, NULL, bh); + if (err) + goto out; + brelse(bh); +handle_ib: + if (bg_flags[i] & EXT4_BG_INODE_UNINIT) + continue; - ext4_mark_bitmap_end(input->blocks_count, sb->s_blocksize * 8, - bh->b_data); - err = ext4_handle_dirty_metadata(handle, NULL, bh); - if (unlikely(err)) { - ext4_std_error(sb, err); - goto exit_bh; + /* Initialize inode bitmap of the @group */ + block = group_data[i].inode_bitmap; + err = extend_or_restart_transaction(handle, 1); + if (err) + goto out; + /* Mark unused entries in inode bitmap used */ + bh = bclean(handle, sb, block); + if (IS_ERR(bh)) { + err = PTR_ERR(bh); + goto out; + } + + ext4_mark_bitmap_end(EXT4_INODES_PER_GROUP(sb), + sb->s_blocksize * 8, bh->b_data); + err = ext4_handle_dirty_metadata(handle, NULL, bh); + if (err) + goto out; + brelse(bh); } - brelse(bh); - /* Mark unused entries in inode bitmap used */ - ext4_debug("clear inode bitmap %#04llx (+%llu)\n", - input->inode_bitmap, input->inode_bitmap - start); - if (IS_ERR(bh = bclean(handle, sb, input->inode_bitmap))) { - err = PTR_ERR(bh); - goto exit_journal; + bh = NULL; + + /* Mark group tables in block bitmap */ + for (j = 0; j < GROUP_TABLE_COUNT; j++) { + count = group_table_count[j]; + start = (&group_data[0].block_bitmap)[j]; + block = start; + for (i = 1; i < flex_gd->count; i++) { + block += group_table_count[j]; + if (block == (&group_data[i].block_bitmap)[j]) { + count += group_table_count[j]; + continue; + } + err = set_flexbg_block_bitmap(sb, handle, + flex_gd, start, count); + if (err) + goto out; + count = group_table_count[j]; + start = group_data[i].block_bitmap; + block = start; + } + + if (count) { + err = set_flexbg_block_bitmap(sb, handle, + flex_gd, start, count); + if (err) + goto out; + } } - ext4_mark_bitmap_end(EXT4_INODES_PER_GROUP(sb), sb->s_blocksize * 8, - bh->b_data); - err = ext4_handle_dirty_metadata(handle, NULL, bh); - if (unlikely(err)) - ext4_std_error(sb, err); -exit_bh: +out: brelse(bh); - -exit_journal: - if ((err2 = ext4_journal_stop(handle)) && !err) + err2 = ext4_journal_stop(handle); + if (err2 && !err) err = err2; return err; @@ -351,10 +636,10 @@ static unsigned ext4_list_backups(struct super_block *sb, unsigned *three, * groups in current filesystem that have BACKUPS, or -ve error code. */ static int verify_reserved_gdb(struct super_block *sb, + ext4_group_t end, struct buffer_head *primary) { const ext4_fsblk_t blk = primary->b_blocknr; - const ext4_group_t end = EXT4_SB(sb)->s_groups_count; unsigned three = 1; unsigned five = 5; unsigned seven = 7; @@ -429,7 +714,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, if (!gdb_bh) return -EIO; - gdbackups = verify_reserved_gdb(sb, gdb_bh); + gdbackups = verify_reserved_gdb(sb, group, gdb_bh); if (gdbackups < 0) { err = gdbackups; goto exit_bh; @@ -592,7 +877,8 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode, err = -EIO; goto exit_bh; } - if ((gdbackups = verify_reserved_gdb(sb, primary[res])) < 0) { + gdbackups = verify_reserved_gdb(sb, group, primary[res]); + if (gdbackups < 0) { brelse(primary[res]); err = gdbackups; goto exit_bh; @@ -735,6 +1021,348 @@ exit_err: } } +/* + * ext4_add_new_descs() adds @count group descriptor of groups + * starting at @group + * + * @handle: journal handle + * @sb: super block + * @group: the group no. of the first group desc to be added + * @resize_inode: the resize inode + * @count: number of group descriptors to be added + */ +static int ext4_add_new_descs(handle_t *handle, struct super_block *sb, + ext4_group_t group, struct inode *resize_inode, + ext4_group_t count) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_super_block *es = sbi->s_es; + struct buffer_head *gdb_bh; + int i, gdb_off, gdb_num, err = 0; + + for (i = 0; i < count; i++, group++) { + int reserved_gdb = ext4_bg_has_super(sb, group) ? + le16_to_cpu(es->s_reserved_gdt_blocks) : 0; + + gdb_off = group % EXT4_DESC_PER_BLOCK(sb); + gdb_num = group / EXT4_DESC_PER_BLOCK(sb); + + /* + * We will only either add reserved group blocks to a backup group + * or remove reserved blocks for the first group in a new group block. + * Doing both would be mean more complex code, and sane people don't + * use non-sparse filesystems anymore. This is already checked above. + */ + if (gdb_off) { + gdb_bh = sbi->s_group_desc[gdb_num]; + err = ext4_journal_get_write_access(handle, gdb_bh); + + if (!err && reserved_gdb && ext4_bg_num_gdb(sb, group)) + err = reserve_backup_gdb(handle, resize_inode, group); + } else + err = add_new_gdb(handle, resize_inode, group); + if (err) + break; + } + return err; +} + +/* + * ext4_setup_new_descs() will set up the group descriptor descriptors of a flex bg + */ +static int ext4_setup_new_descs(handle_t *handle, struct super_block *sb, + struct ext4_new_flex_group_data *flex_gd) +{ + struct ext4_new_group_data *group_data = flex_gd->groups; + struct ext4_group_desc *gdp; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct buffer_head *gdb_bh; + ext4_group_t group; + __u16 *bg_flags = flex_gd->bg_flags; + int i, gdb_off, gdb_num, err = 0; + + + for (i = 0; i < flex_gd->count; i++, group_data++, bg_flags++) { + group = group_data->group; + + gdb_off = group % EXT4_DESC_PER_BLOCK(sb); + gdb_num = group / EXT4_DESC_PER_BLOCK(sb); + + /* + * get_write_access() has been called on gdb_bh by ext4_add_new_desc(). + */ + gdb_bh = sbi->s_group_desc[gdb_num]; + /* Update group descriptor block for new group */ + gdp = (struct ext4_group_desc *)((char *)gdb_bh->b_data + + gdb_off * EXT4_DESC_SIZE(sb)); + + memset(gdp, 0, EXT4_DESC_SIZE(sb)); + ext4_block_bitmap_set(sb, gdp, group_data->block_bitmap); + ext4_inode_bitmap_set(sb, gdp, group_data->inode_bitmap); + ext4_inode_table_set(sb, gdp, group_data->inode_table); + ext4_free_group_clusters_set(sb, gdp, + EXT4_B2C(sbi, group_data->free_blocks_count)); + ext4_free_inodes_set(sb, gdp, EXT4_INODES_PER_GROUP(sb)); + gdp->bg_flags = cpu_to_le16(*bg_flags); + gdp->bg_checksum = ext4_group_desc_csum(sbi, group, gdp); + + err = ext4_handle_dirty_metadata(handle, NULL, gdb_bh); + if (unlikely(err)) { + ext4_std_error(sb, err); + break; + } + + /* + * We can allocate memory for mb_alloc based on the new group + * descriptor + */ + err = ext4_mb_add_groupinfo(sb, group, gdp); + if (err) + break; + } + return err; +} + +/* + * ext4_update_super() updates the super block so that the newly added + * groups can be seen by the filesystem. + * + * @sb: super block + * @flex_gd: new added groups + */ +static void ext4_update_super(struct super_block *sb, + struct ext4_new_flex_group_data *flex_gd) +{ + ext4_fsblk_t blocks_count = 0; + ext4_fsblk_t free_blocks = 0; + ext4_fsblk_t reserved_blocks = 0; + struct ext4_new_group_data *group_data = flex_gd->groups; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_super_block *es = sbi->s_es; + int i; + + BUG_ON(flex_gd->count == 0 || group_data == NULL); + /* + * Make the new blocks and inodes valid next. We do this before + * increasing the group count so that once the group is enabled, + * all of its blocks and inodes are already valid. + * + * We always allocate group-by-group, then block-by-block or + * inode-by-inode within a group, so enabling these + * blocks/inodes before the group is live won't actually let us + * allocate the new space yet. + */ + for (i = 0; i < flex_gd->count; i++) { + blocks_count += group_data[i].blocks_count; + free_blocks += group_data[i].free_blocks_count; + } + + reserved_blocks = ext4_r_blocks_count(es) * 100; + do_div(reserved_blocks, ext4_blocks_count(es)); + reserved_blocks *= blocks_count; + do_div(reserved_blocks, 100); + + ext4_blocks_count_set(es, ext4_blocks_count(es) + blocks_count); + le32_add_cpu(&es->s_inodes_count, EXT4_INODES_PER_GROUP(sb) * + flex_gd->count); + + /* + * We need to protect s_groups_count against other CPUs seeing + * inconsistent state in the superblock. + * + * The precise rules we use are: + * + * * Writers must perform a smp_wmb() after updating all + * dependent data and before modifying the groups count + * + * * Readers must perform an smp_rmb() after reading the groups + * count and before reading any dependent data. + * + * NB. These rules can be relaxed when checking the group count + * while freeing data, as we can only allocate from a block + * group after serialising against the group count, and we can + * only then free after serialising in turn against that + * allocation. + */ + smp_wmb(); + + /* Update the global fs size fields */ + sbi->s_groups_count += flex_gd->count; + + /* Update the reserved block counts only once the new group is + * active. */ + ext4_r_blocks_count_set(es, ext4_r_blocks_count(es) + + reserved_blocks); + + /* Update the free space counts */ + percpu_counter_add(&sbi->s_freeclusters_counter, + EXT4_B2C(sbi, free_blocks)); + percpu_counter_add(&sbi->s_freeinodes_counter, + EXT4_INODES_PER_GROUP(sb) * flex_gd->count); + + if (EXT4_HAS_INCOMPAT_FEATURE(sb, + EXT4_FEATURE_INCOMPAT_FLEX_BG) && + sbi->s_log_groups_per_flex) { + ext4_group_t flex_group; + flex_group = ext4_flex_group(sbi, group_data[0].group); + atomic_add(EXT4_B2C(sbi, free_blocks), + &sbi->s_flex_groups[flex_group].free_clusters); + atomic_add(EXT4_INODES_PER_GROUP(sb) * flex_gd->count, + &sbi->s_flex_groups[flex_group].free_inodes); + } + + if (test_opt(sb, DEBUG)) + printk(KERN_DEBUG "EXT4-fs: added group %u:" + "%llu blocks(%llu free %llu reserved)\n", flex_gd->count, + blocks_count, free_blocks, reserved_blocks); +} + +/* Add a flex group to an fs. Ensure we handle all possible error conditions + * _before_ we start modifying the filesystem, because we cannot abort the + * transaction and not have it write the data to disk. + */ +static int ext4_flex_group_add(struct super_block *sb, + struct inode *resize_inode, + struct ext4_new_flex_group_data *flex_gd) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_super_block *es = sbi->s_es; + ext4_fsblk_t o_blocks_count; + ext4_grpblk_t last; + ext4_group_t group; + handle_t *handle; + unsigned reserved_gdb; + int err = 0, err2 = 0, credit; + + BUG_ON(!flex_gd->count || !flex_gd->groups || !flex_gd->bg_flags); + + reserved_gdb = le16_to_cpu(es->s_reserved_gdt_blocks); + o_blocks_count = ext4_blocks_count(es); + ext4_get_group_no_and_offset(sb, o_blocks_count, &group, &last); + BUG_ON(last); + + err = setup_new_flex_group_blocks(sb, flex_gd); + if (err) + goto exit; + /* + * We will always be modifying at least the superblock and GDT + * block. If we are adding a group past the last current GDT block, + * we will also modify the inode and the dindirect block. If we + * are adding a group with superblock/GDT backups we will also + * modify each of the reserved GDT dindirect blocks. + */ + credit = flex_gd->count * 4 + reserved_gdb; + handle = ext4_journal_start_sb(sb, credit); + if (IS_ERR(handle)) { + err = PTR_ERR(handle); + goto exit; + } + + err = ext4_journal_get_write_access(handle, sbi->s_sbh); + if (err) + goto exit_journal; + + group = flex_gd->groups[0].group; + BUG_ON(group != EXT4_SB(sb)->s_groups_count); + err = ext4_add_new_descs(handle, sb, group, + resize_inode, flex_gd->count); + if (err) + goto exit_journal; + + err = ext4_setup_new_descs(handle, sb, flex_gd); + if (err) + goto exit_journal; + + ext4_update_super(sb, flex_gd); + + err = ext4_handle_dirty_super(handle, sb); + +exit_journal: + err2 = ext4_journal_stop(handle); + if (!err) + err = err2; + + if (!err) { + int i; + update_backups(sb, sbi->s_sbh->b_blocknr, (char *)es, + sizeof(struct ext4_super_block)); + for (i = 0; i < flex_gd->count; i++, group++) { + struct buffer_head *gdb_bh; + int gdb_num; + gdb_num = group / EXT4_BLOCKS_PER_GROUP(sb); + gdb_bh = sbi->s_group_desc[gdb_num]; + update_backups(sb, gdb_bh->b_blocknr, gdb_bh->b_data, + gdb_bh->b_size); + } + } +exit: + return err; +} + +static int ext4_setup_next_flex_gd(struct super_block *sb, + struct ext4_new_flex_group_data *flex_gd, + ext4_fsblk_t n_blocks_count, + unsigned long flexbg_size) +{ + struct ext4_super_block *es = EXT4_SB(sb)->s_es; + struct ext4_new_group_data *group_data = flex_gd->groups; + ext4_fsblk_t o_blocks_count; + ext4_group_t n_group; + ext4_group_t group; + ext4_group_t last_group; + ext4_grpblk_t last; + ext4_grpblk_t blocks_per_group; + unsigned long i; + + blocks_per_group = EXT4_BLOCKS_PER_GROUP(sb); + + o_blocks_count = ext4_blocks_count(es); + + if (o_blocks_count == n_blocks_count) + return 0; + + ext4_get_group_no_and_offset(sb, o_blocks_count, &group, &last); + BUG_ON(last); + ext4_get_group_no_and_offset(sb, n_blocks_count - 1, &n_group, &last); + + last_group = group | (flexbg_size - 1); + if (last_group > n_group) + last_group = n_group; + + flex_gd->count = last_group - group + 1; + + for (i = 0; i < flex_gd->count; i++) { + int overhead; + + group_data[i].group = group + i; + group_data[i].blocks_count = blocks_per_group; + overhead = ext4_bg_has_super(sb, group + i) ? + (1 + ext4_bg_num_gdb(sb, group + i) + + le16_to_cpu(es->s_reserved_gdt_blocks)) : 0; + group_data[i].free_blocks_count = blocks_per_group - overhead; + if (EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_GDT_CSUM)) + flex_gd->bg_flags[i] = EXT4_BG_BLOCK_UNINIT | + EXT4_BG_INODE_UNINIT; + else + flex_gd->bg_flags[i] = EXT4_BG_INODE_ZEROED; + } + + if (last_group == n_group && + EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_GDT_CSUM)) + /* We need to initialize block bitmap of last group. */ + flex_gd->bg_flags[i - 1] &= ~EXT4_BG_BLOCK_UNINIT; + + if ((last_group == n_group) && (last != blocks_per_group - 1)) { + group_data[i - 1].blocks_count = last + 1; + group_data[i - 1].free_blocks_count -= blocks_per_group- + last - 1; + } + + return 1; +} + /* Add group descriptor data to an existing or new group descriptor block. * Ensure we handle all possible error conditions _before_ we start modifying * the filesystem, because we cannot abort the transaction and not have it @@ -750,16 +1378,15 @@ exit_err: */ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) { + struct ext4_new_flex_group_data flex_gd; struct ext4_sb_info *sbi = EXT4_SB(sb); struct ext4_super_block *es = sbi->s_es; int reserved_gdb = ext4_bg_has_super(sb, input->group) ? le16_to_cpu(es->s_reserved_gdt_blocks) : 0; - struct buffer_head *primary = NULL; - struct ext4_group_desc *gdp; struct inode *inode = NULL; - handle_t *handle; int gdb_off, gdb_num; - int err, err2; + int err; + __u16 bg_flags = 0; gdb_num = input->group / EXT4_DESC_PER_BLOCK(sb); gdb_off = input->group % EXT4_DESC_PER_BLOCK(sb); @@ -798,175 +1425,69 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) } - if ((err = verify_group_input(sb, input))) - goto exit_put; + err = verify_group_input(sb, input); + if (err) + goto out; - if ((err = setup_new_group_blocks(sb, input))) - goto exit_put; + flex_gd.count = 1; + flex_gd.groups = input; + flex_gd.bg_flags = &bg_flags; + err = ext4_flex_group_add(sb, inode, &flex_gd); +out: + iput(inode); + return err; +} /* ext4_group_add */ - /* - * We will always be modifying at least the superblock and a GDT - * block. If we are adding a group past the last current GDT block, - * we will also modify the inode and the dindirect block. If we - * are adding a group with superblock/GDT backups we will also - * modify each of the reserved GDT dindirect blocks. +/* + * extend a group without checking assuming that checking has been done. + */ +static int ext4_group_extend_no_check(struct super_block *sb, + ext4_fsblk_t o_blocks_count, ext4_grpblk_t add) +{ + struct ext4_super_block *es = EXT4_SB(sb)->s_es; + handle_t *handle; + int err = 0, err2; + + /* We will update the superblock, one block bitmap, and + * one group descriptor via ext4_group_add_blocks(). */ - handle = ext4_journal_start_sb(sb, - ext4_bg_has_super(sb, input->group) ? - 3 + reserved_gdb : 4); + handle = ext4_journal_start_sb(sb, 3); if (IS_ERR(handle)) { err = PTR_ERR(handle); - goto exit_put; + ext4_warning(sb, "error %d on journal start", err); + return err; } - if ((err = ext4_journal_get_write_access(handle, sbi->s_sbh))) - goto exit_journal; - - /* - * We will only either add reserved group blocks to a backup group - * or remove reserved blocks for the first group in a new group block. - * Doing both would be mean more complex code, and sane people don't - * use non-sparse filesystems anymore. This is already checked above. - */ - if (gdb_off) { - primary = sbi->s_group_desc[gdb_num]; - if ((err = ext4_journal_get_write_access(handle, primary))) - goto exit_journal; - - if (reserved_gdb && ext4_bg_num_gdb(sb, input->group)) { - err = reserve_backup_gdb(handle, inode, input->group); - if (err) - goto exit_journal; - } - } else { - /* - * Note that we can access new group descriptor block safely - * only if add_new_gdb() succeeds. - */ - err = add_new_gdb(handle, inode, input->group); - if (err) - goto exit_journal; - primary = sbi->s_group_desc[gdb_num]; + err = ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh); + if (err) { + ext4_warning(sb, "error %d on journal write access", err); + goto errout; } - /* - * OK, now we've set up the new group. Time to make it active. - * - * so we have to be safe wrt. concurrent accesses the group - * data. So we need to be careful to set all of the relevant - * group descriptor data etc. *before* we enable the group. - * - * The key field here is sbi->s_groups_count: as long as - * that retains its old value, nobody is going to access the new - * group. - * - * So first we update all the descriptor metadata for the new - * group; then we update the total disk blocks count; then we - * update the groups count to enable the group; then finally we - * update the free space counts so that the system can start - * using the new disk blocks. - */ - - /* Update group descriptor block for new group */ - gdp = (struct ext4_group_desc *)((char *)primary->b_data + - gdb_off * EXT4_DESC_SIZE(sb)); - - memset(gdp, 0, EXT4_DESC_SIZE(sb)); - ext4_block_bitmap_set(sb, gdp, input->block_bitmap); /* LV FIXME */ - ext4_inode_bitmap_set(sb, gdp, input->inode_bitmap); /* LV FIXME */ - ext4_inode_table_set(sb, gdp, input->inode_table); /* LV FIXME */ - ext4_free_group_clusters_set(sb, gdp, input->free_blocks_count); - ext4_free_inodes_set(sb, gdp, EXT4_INODES_PER_GROUP(sb)); - gdp->bg_flags = cpu_to_le16(EXT4_BG_INODE_ZEROED); - gdp->bg_checksum = ext4_group_desc_csum(sbi, input->group, gdp); - - /* - * We can allocate memory for mb_alloc based on the new group - * descriptor - */ - err = ext4_mb_add_groupinfo(sb, input->group, gdp); + ext4_blocks_count_set(es, o_blocks_count + add); + ext4_debug("freeing blocks %llu through %llu\n", o_blocks_count, + o_blocks_count + add); + /* We add the blocks to the bitmap and set the group need init bit */ + err = ext4_group_add_blocks(handle, sb, o_blocks_count, add); if (err) - goto exit_journal; - - /* - * Make the new blocks and inodes valid next. We do this before - * increasing the group count so that once the group is enabled, - * all of its blocks and inodes are already valid. - * - * We always allocate group-by-group, then block-by-block or - * inode-by-inode within a group, so enabling these - * blocks/inodes before the group is live won't actually let us - * allocate the new space yet. - */ - ext4_blocks_count_set(es, ext4_blocks_count(es) + - input->blocks_count); - le32_add_cpu(&es->s_inodes_count, EXT4_INODES_PER_GROUP(sb)); - - /* - * We need to protect s_groups_count against other CPUs seeing - * inconsistent state in the superblock. - * - * The precise rules we use are: - * - * * Writers must perform a smp_wmb() after updating all dependent - * data and before modifying the groups count - * - * * Readers must perform an smp_rmb() after reading the groups count - * and before reading any dependent data. - * - * NB. These rules can be relaxed when checking the group count - * while freeing data, as we can only allocate from a block - * group after serialising against the group count, and we can - * only then free after serialising in turn against that - * allocation. - */ - smp_wmb(); - - /* Update the global fs size fields */ - sbi->s_groups_count++; - - err = ext4_handle_dirty_metadata(handle, NULL, primary); - if (unlikely(err)) { - ext4_std_error(sb, err); - goto exit_journal; - } - - /* Update the reserved block counts only once the new group is - * active. */ - ext4_r_blocks_count_set(es, ext4_r_blocks_count(es) + - input->reserved_blocks); - - /* Update the free space counts */ - percpu_counter_add(&sbi->s_freeclusters_counter, - EXT4_B2C(sbi, input->free_blocks_count)); - percpu_counter_add(&sbi->s_freeinodes_counter, - EXT4_INODES_PER_GROUP(sb)); - - if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG) && - sbi->s_log_groups_per_flex) { - ext4_group_t flex_group; - flex_group = ext4_flex_group(sbi, input->group); - atomic_add(EXT4_B2C(sbi, input->free_blocks_count), - &sbi->s_flex_groups[flex_group].free_clusters); - atomic_add(EXT4_INODES_PER_GROUP(sb), - &sbi->s_flex_groups[flex_group].free_inodes); - } - + goto errout; ext4_handle_dirty_super(handle, sb); - -exit_journal: - if ((err2 = ext4_journal_stop(handle)) && !err) + ext4_debug("freed blocks %llu through %llu\n", o_blocks_count, + o_blocks_count + add); +errout: + err2 = ext4_journal_stop(handle); + if (err2 && !err) err = err2; - if (!err && primary) { - update_backups(sb, sbi->s_sbh->b_blocknr, (char *)es, + + if (!err) { + 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, (char *)es, sizeof(struct ext4_super_block)); - update_backups(sb, primary->b_blocknr, primary->b_data, - primary->b_size); } -exit_put: - iput(inode); return err; -} /* ext4_group_add */ +} /* * Extend the filesystem to the new number of blocks specified. This entry @@ -985,8 +1506,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, ext4_grpblk_t last; ext4_grpblk_t add; struct buffer_head *bh; - handle_t *handle; - int err, err2; + int err; ext4_group_t group; o_blocks_count = ext4_blocks_count(es); @@ -1042,42 +1562,119 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, } brelse(bh); - /* We will update the superblock, one block bitmap, and - * one group descriptor via ext4_free_blocks(). - */ - handle = ext4_journal_start_sb(sb, 3); - if (IS_ERR(handle)) { - err = PTR_ERR(handle); - ext4_warning(sb, "error %d on journal start", err); - goto exit_put; + err = ext4_group_extend_no_check(sb, o_blocks_count, add); + return err; +} /* ext4_group_extend */ + +/* + * ext4_resize_fs() resizes a fs to new size specified by @n_blocks_count + * + * @sb: super block of the fs to be resized + * @n_blocks_count: the number of blocks resides in the resized fs + */ +int ext4_resize_fs(struct super_block *sb, ext4_fsblk_t n_blocks_count) +{ + struct ext4_new_flex_group_data *flex_gd = NULL; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_super_block *es = sbi->s_es; + struct buffer_head *bh; + struct inode *resize_inode; + ext4_fsblk_t o_blocks_count; + ext4_group_t o_group; + ext4_group_t n_group; + ext4_grpblk_t offset; + unsigned long n_desc_blocks; + unsigned long o_desc_blocks; + unsigned long desc_blocks; + int err = 0, flexbg_size = 1; + + o_blocks_count = ext4_blocks_count(es); + + if (test_opt(sb, DEBUG)) + printk(KERN_DEBUG "EXT4-fs: resizing filesystem from %llu " + "upto %llu blocks\n", o_blocks_count, n_blocks_count); + + if (n_blocks_count < o_blocks_count) { + /* On-line shrinking not supported */ + ext4_warning(sb, "can't shrink FS - resize aborted"); + return -EINVAL; } - if ((err = ext4_journal_get_write_access(handle, - EXT4_SB(sb)->s_sbh))) { - ext4_warning(sb, "error %d on journal write access", err); - ext4_journal_stop(handle); - goto exit_put; + if (n_blocks_count == o_blocks_count) + /* Nothing need to do */ + return 0; + + ext4_get_group_no_and_offset(sb, n_blocks_count - 1, &n_group, &offset); + ext4_get_group_no_and_offset(sb, o_blocks_count, &o_group, &offset); + + n_desc_blocks = (n_group + EXT4_DESC_PER_BLOCK(sb)) / + EXT4_DESC_PER_BLOCK(sb); + o_desc_blocks = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) / + EXT4_DESC_PER_BLOCK(sb); + desc_blocks = n_desc_blocks - o_desc_blocks; + + if (desc_blocks && + (!EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_RESIZE_INODE) || + le16_to_cpu(es->s_reserved_gdt_blocks) < desc_blocks)) { + ext4_warning(sb, "No reserved GDT blocks, can't resize"); + return -EPERM; } - ext4_blocks_count_set(es, o_blocks_count + add); - ext4_debug("freeing blocks %llu through %llu\n", o_blocks_count, - o_blocks_count + add); - /* We add the blocks to the bitmap and set the group need init bit */ - err = ext4_group_add_blocks(handle, sb, o_blocks_count, add); - ext4_handle_dirty_super(handle, sb); - ext4_debug("freed blocks %llu through %llu\n", o_blocks_count, - o_blocks_count + add); - err2 = ext4_journal_stop(handle); - if (!err && err2) - err = err2; - if (err) - goto exit_put; + resize_inode = ext4_iget(sb, EXT4_RESIZE_INO); + if (IS_ERR(resize_inode)) { + ext4_warning(sb, "Error opening resize inode"); + return PTR_ERR(resize_inode); + } + /* See if the device is actually as big as what was requested */ + bh = sb_bread(sb, n_blocks_count - 1); + if (!bh) { + ext4_warning(sb, "can't read last block, resize aborted"); + return -ENOSPC; + } + brelse(bh); + + if (offset != 0) { + /* extend the last group */ + ext4_grpblk_t add; + add = EXT4_BLOCKS_PER_GROUP(sb) - offset; + err = ext4_group_extend_no_check(sb, o_blocks_count, add); + if (err) + goto out; + } + + if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG) && + es->s_log_groups_per_flex) + flexbg_size = 1 << es->s_log_groups_per_flex; + + o_blocks_count = ext4_blocks_count(es); + if (o_blocks_count == n_blocks_count) + goto out; + + flex_gd = alloc_flex_gd(flexbg_size); + if (flex_gd == NULL) { + err = -ENOMEM; + goto out; + } + + /* Add flex groups. Note that a regular group is a + * flex group with 1 group. + */ + while (ext4_setup_next_flex_gd(sb, flex_gd, n_blocks_count, + flexbg_size)) { + ext4_alloc_group_tables(sb, flex_gd, flexbg_size); + err = ext4_flex_group_add(sb, resize_inode, flex_gd); + if (unlikely(err)) + break; + } + +out: + if (flex_gd) + free_flex_gd(flex_gd); + + iput(resize_inode); 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, (char *)es, - sizeof(struct ext4_super_block)); -exit_put: + printk(KERN_DEBUG "EXT4-fs: resized filesystem from %llu " + "upto %llu blocks\n", o_blocks_count, n_blocks_count); return err; -} /* ext4_group_extend */ +} diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 9953d80145ad..502c61fd7392 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -930,7 +930,6 @@ static int ext4_drop_inode(struct inode *inode) static void ext4_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(ext4_inode_cachep, EXT4_I(inode)); } @@ -1033,11 +1032,11 @@ static inline void ext4_show_quota_options(struct seq_file *seq, * - it's set to a non-default value OR * - if the per-sb default is different from the global default */ -static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) +static int ext4_show_options(struct seq_file *seq, struct dentry *root) { int def_errors; unsigned long def_mount_opts; - struct super_block *sb = vfs->mnt_sb; + struct super_block *sb = root->d_sb; struct ext4_sb_info *sbi = EXT4_SB(sb); struct ext4_super_block *es = sbi->s_es; @@ -1096,7 +1095,7 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) } if (sbi->s_max_batch_time != EXT4_DEF_MAX_BATCH_TIME) { seq_printf(seq, ",max_batch_time=%u", - (unsigned) sbi->s_min_batch_time); + (unsigned) sbi->s_max_batch_time); } /* @@ -1155,9 +1154,9 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) seq_puts(seq, ",block_validity"); if (!test_opt(sb, INIT_INODE_TABLE)) - seq_puts(seq, ",noinit_inode_table"); + seq_puts(seq, ",noinit_itable"); else if (sbi->s_li_wait_mult != EXT4_DEF_LI_WAIT_MULT) - seq_printf(seq, ",init_inode_table=%u", + seq_printf(seq, ",init_itable=%u", (unsigned) sbi->s_li_wait_mult); ext4_show_quota_options(seq, sb); @@ -1333,8 +1332,7 @@ enum { Opt_nomblk_io_submit, Opt_block_validity, Opt_noblock_validity, Opt_inode_readahead_blks, Opt_journal_ioprio, Opt_dioread_nolock, Opt_dioread_lock, - Opt_discard, Opt_nodiscard, - Opt_init_inode_table, Opt_noinit_inode_table, + Opt_discard, Opt_nodiscard, Opt_init_itable, Opt_noinit_itable, }; static const match_table_t tokens = { @@ -1407,9 +1405,9 @@ static const match_table_t tokens = { {Opt_dioread_lock, "dioread_lock"}, {Opt_discard, "discard"}, {Opt_nodiscard, "nodiscard"}, - {Opt_init_inode_table, "init_itable=%u"}, - {Opt_init_inode_table, "init_itable"}, - {Opt_noinit_inode_table, "noinit_itable"}, + {Opt_init_itable, "init_itable=%u"}, + {Opt_init_itable, "init_itable"}, + {Opt_noinit_itable, "noinit_itable"}, {Opt_err, NULL}, }; @@ -1683,7 +1681,9 @@ static int parse_options(char *options, struct super_block *sb, data_opt = EXT4_MOUNT_WRITEBACK_DATA; datacheck: if (is_remount) { - if (test_opt(sb, DATA_FLAGS) != data_opt) { + if (!sbi->s_journal) + ext4_msg(sb, KERN_WARNING, "Remounting file system with no journal so ignoring journalled data option"); + else if (test_opt(sb, DATA_FLAGS) != data_opt) { ext4_msg(sb, KERN_ERR, "Cannot change data mode on remount"); return 0; @@ -1890,7 +1890,7 @@ set_qf_format: case Opt_dioread_lock: clear_opt(sb, DIOREAD_NOLOCK); break; - case Opt_init_inode_table: + case Opt_init_itable: set_opt(sb, INIT_INODE_TABLE); if (args[0].from) { if (match_int(&args[0], &option)) @@ -1901,7 +1901,7 @@ set_qf_format: return 0; sbi->s_li_wait_mult = option; break; - case Opt_noinit_inode_table: + case Opt_noinit_itable: clear_opt(sb, INIT_INODE_TABLE); break; default: @@ -2005,17 +2005,16 @@ static int ext4_fill_flex_info(struct super_block *sb) struct ext4_group_desc *gdp = NULL; ext4_group_t flex_group_count; ext4_group_t flex_group; - int groups_per_flex = 0; + unsigned int groups_per_flex = 0; size_t size; int i; sbi->s_log_groups_per_flex = sbi->s_es->s_log_groups_per_flex; - groups_per_flex = 1 << sbi->s_log_groups_per_flex; - - if (groups_per_flex < 2) { + if (sbi->s_log_groups_per_flex < 1 || sbi->s_log_groups_per_flex > 31) { sbi->s_log_groups_per_flex = 0; return 1; } + groups_per_flex = 1 << sbi->s_log_groups_per_flex; /* We allocate both existing and potentially added groups */ flex_group_count = ((sbi->s_groups_count + groups_per_flex - 1) + @@ -2882,8 +2881,7 @@ cont_thread: } mutex_unlock(&eli->li_list_mtx); - if (freezing(current)) - refrigerator(); + try_to_freeze(); cur = jiffies; if ((time_after_eq(cur, next_wakeup)) || @@ -3099,8 +3097,6 @@ static void ext4_destroy_lazyinit_thread(void) } static int ext4_fill_super(struct super_block *sb, void *data, int silent) - __releases(kernel_lock) - __acquires(kernel_lock) { char *orig_data = kstrdup(data, GFP_KERNEL); struct buffer_head *bh; @@ -3509,7 +3505,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) * of the filesystem. */ if (le32_to_cpu(es->s_first_data_block) >= ext4_blocks_count(es)) { - ext4_msg(sb, KERN_WARNING, "bad geometry: first data" + ext4_msg(sb, KERN_WARNING, "bad geometry: first data " "block %u is beyond end of filesystem (%llu)", le32_to_cpu(es->s_first_data_block), ext4_blocks_count(es)); @@ -3736,10 +3732,12 @@ no_journal: } if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) { ext4_msg(sb, KERN_ERR, "corrupt root inode, run e2fsck"); + iput(root); goto failed_mount4; } sb->s_root = d_alloc_root(root); if (!sb->s_root) { + iput(root); ext4_msg(sb, KERN_ERR, "get root dentry failed"); ret = -ENOMEM; goto failed_mount4; @@ -3776,7 +3774,7 @@ no_journal: if (err) { ext4_msg(sb, KERN_ERR, "failed to initialize system " "zone (%d)", err); - goto failed_mount4; + goto failed_mount4a; } ext4_ext_init(sb); @@ -3833,13 +3831,14 @@ cantfind_ext4: failed_mount7: ext4_unregister_li_request(sb); failed_mount6: - ext4_ext_release(sb); -failed_mount5: ext4_mb_release(sb); +failed_mount5: + ext4_ext_release(sb); ext4_release_system_zone(sb); -failed_mount4: - iput(root); +failed_mount4a: + dput(sb->s_root); sb->s_root = NULL; +failed_mount4: ext4_msg(sb, KERN_ERR, "mount failed"); destroy_workqueue(EXT4_SB(sb)->dio_unwritten_wq); failed_mount_wq: @@ -4783,7 +4782,7 @@ static int ext4_quota_on(struct super_block *sb, int type, int format_id, return -EINVAL; /* Quotafile not on the same filesystem? */ - if (path->mnt->mnt_sb != sb) + if (path->dentry->d_sb != sb) return -EXDEV; /* Journaling quota? */ if (EXT4_SB(sb)->s_qf_names[type]) { diff --git a/fs/ext4/xattr_security.c b/fs/ext4/xattr_security.c index 34e4350dd4d9..d2a200624af5 100644 --- a/fs/ext4/xattr_security.c +++ b/fs/ext4/xattr_security.c @@ -3,7 +3,6 @@ * Handler for storing security labels as extended attributes. */ -#include <linux/module.h> #include <linux/string.h> #include <linux/fs.h> #include <linux/security.h> @@ -48,8 +47,9 @@ ext4_xattr_security_set(struct dentry *dentry, const char *name, name, value, size, flags); } -int ext4_initxattrs(struct inode *inode, const struct xattr *xattr_array, - void *fs_info) +static int +ext4_initxattrs(struct inode *inode, const struct xattr *xattr_array, + void *fs_info) { const struct xattr *xattr; handle_t *handle = fs_info; diff --git a/fs/ext4/xattr_trusted.c b/fs/ext4/xattr_trusted.c index 37e6ebca2cc3..95f1f4ab59a4 100644 --- a/fs/ext4/xattr_trusted.c +++ b/fs/ext4/xattr_trusted.c @@ -5,7 +5,6 @@ * Copyright (C) 2003 by Andreas Gruenbacher, <a.gruenbacher@computer.org> */ -#include <linux/module.h> #include <linux/string.h> #include <linux/capability.h> #include <linux/fs.h> diff --git a/fs/ext4/xattr_user.c b/fs/ext4/xattr_user.c index 98c375352d0e..0edb7611ffbe 100644 --- a/fs/ext4/xattr_user.c +++ b/fs/ext4/xattr_user.c @@ -5,7 +5,6 @@ * Copyright (C) 2001 by Andreas Gruenbacher, <a.gruenbacher@computer.org> */ -#include <linux/module.h> #include <linux/string.h> #include <linux/fs.h> #include "ext4_jbd2.h" diff --git a/fs/fat/fat.h b/fs/fat/fat.h index 1510a4d51990..66994f316e18 100644 --- a/fs/fat/fat.h +++ b/fs/fat/fat.h @@ -141,7 +141,7 @@ static inline struct msdos_inode_info *MSDOS_I(struct inode *inode) static inline int fat_mode_can_hold_ro(struct inode *inode) { struct msdos_sb_info *sbi = MSDOS_SB(inode->i_sb); - mode_t mask; + umode_t mask; if (S_ISDIR(inode->i_mode)) { if (!sbi->options.rodir) @@ -156,8 +156,8 @@ static inline int fat_mode_can_hold_ro(struct inode *inode) } /* Convert attribute bits and a mask to the UNIX mode. */ -static inline mode_t fat_make_mode(struct msdos_sb_info *sbi, - u8 attrs, mode_t mode) +static inline umode_t fat_make_mode(struct msdos_sb_info *sbi, + u8 attrs, umode_t mode) { if (attrs & ATTR_RO && !((attrs & ATTR_DIR) && !sbi->options.rodir)) mode &= ~S_IWUGO; diff --git a/fs/fat/file.c b/fs/fat/file.c index c118acf16e43..a71fe3715ee8 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -44,7 +44,7 @@ static int fat_ioctl_set_attributes(struct file *file, u32 __user *user_attr) goto out; mutex_lock(&inode->i_mutex); - err = mnt_want_write(file->f_path.mnt); + err = mnt_want_write_file(file); if (err) goto out_unlock_inode; @@ -108,7 +108,7 @@ static int fat_ioctl_set_attributes(struct file *file, u32 __user *user_attr) fat_save_attrs(inode, attr); mark_inode_dirty(inode); out_drop_write: - mnt_drop_write(file->f_path.mnt); + mnt_drop_write_file(file); out_unlock_inode: mutex_unlock(&inode->i_mutex); out: @@ -314,7 +314,7 @@ EXPORT_SYMBOL_GPL(fat_getattr); static int fat_sanitize_mode(const struct msdos_sb_info *sbi, struct inode *inode, umode_t *mode_ptr) { - mode_t mask, perm; + umode_t mask, perm; /* * Note, the basic check is already done by a caller of @@ -351,7 +351,7 @@ static int fat_sanitize_mode(const struct msdos_sb_info *sbi, static int fat_allow_set_time(struct msdos_sb_info *sbi, struct inode *inode) { - mode_t allow_utime = sbi->options.allow_utime; + umode_t allow_utime = sbi->options.allow_utime; if (current_fsuid() != inode->i_uid) { if (in_group_p(inode->i_gid)) diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 808cac7edcfb..3ab841054d53 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -518,7 +518,6 @@ static struct inode *fat_alloc_inode(struct super_block *sb) static void fat_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(fat_inode_cachep, MSDOS_I(inode)); } @@ -672,7 +671,7 @@ int fat_sync_inode(struct inode *inode) EXPORT_SYMBOL_GPL(fat_sync_inode); -static int fat_show_options(struct seq_file *m, struct vfsmount *mnt); +static int fat_show_options(struct seq_file *m, struct dentry *root); static const struct super_operations fat_sops = { .alloc_inode = fat_alloc_inode, .destroy_inode = fat_destroy_inode, @@ -811,9 +810,9 @@ static const struct export_operations fat_export_ops = { .get_parent = fat_get_parent, }; -static int fat_show_options(struct seq_file *m, struct vfsmount *mnt) +static int fat_show_options(struct seq_file *m, struct dentry *root) { - struct msdos_sb_info *sbi = MSDOS_SB(mnt->mnt_sb); + struct msdos_sb_info *sbi = MSDOS_SB(root->d_sb); struct fat_mount_options *opts = &sbi->options; int isvfat = opts->isvfat; @@ -898,7 +897,7 @@ enum { Opt_charset, Opt_shortname_lower, Opt_shortname_win95, Opt_shortname_winnt, Opt_shortname_mixed, Opt_utf8_no, Opt_utf8_yes, Opt_uni_xl_no, Opt_uni_xl_yes, Opt_nonumtail_no, Opt_nonumtail_yes, - Opt_obsolate, Opt_flush, Opt_tz_utc, Opt_rodir, Opt_err_cont, + Opt_obsolete, Opt_flush, Opt_tz_utc, Opt_rodir, Opt_err_cont, Opt_err_panic, Opt_err_ro, Opt_discard, Opt_err, }; @@ -928,17 +927,17 @@ static const match_table_t fat_tokens = { {Opt_err_panic, "errors=panic"}, {Opt_err_ro, "errors=remount-ro"}, {Opt_discard, "discard"}, - {Opt_obsolate, "conv=binary"}, - {Opt_obsolate, "conv=text"}, - {Opt_obsolate, "conv=auto"}, - {Opt_obsolate, "conv=b"}, - {Opt_obsolate, "conv=t"}, - {Opt_obsolate, "conv=a"}, - {Opt_obsolate, "fat=%u"}, - {Opt_obsolate, "blocksize=%u"}, - {Opt_obsolate, "cvf_format=%20s"}, - {Opt_obsolate, "cvf_options=%100s"}, - {Opt_obsolate, "posix"}, + {Opt_obsolete, "conv=binary"}, + {Opt_obsolete, "conv=text"}, + {Opt_obsolete, "conv=auto"}, + {Opt_obsolete, "conv=b"}, + {Opt_obsolete, "conv=t"}, + {Opt_obsolete, "conv=a"}, + {Opt_obsolete, "fat=%u"}, + {Opt_obsolete, "blocksize=%u"}, + {Opt_obsolete, "cvf_format=%20s"}, + {Opt_obsolete, "cvf_options=%100s"}, + {Opt_obsolete, "posix"}, {Opt_err, NULL}, }; static const match_table_t msdos_tokens = { @@ -1170,7 +1169,7 @@ static int parse_options(struct super_block *sb, char *options, int is_vfat, break; /* obsolete mount options */ - case Opt_obsolate: + case Opt_obsolete: fat_msg(sb, KERN_INFO, "\"%s\" option is obsolete, " "not supported now", p); break; diff --git a/fs/fat/namei_msdos.c b/fs/fat/namei_msdos.c index 216b419f30e2..c5938c9084b9 100644 --- a/fs/fat/namei_msdos.c +++ b/fs/fat/namei_msdos.c @@ -264,7 +264,7 @@ static int msdos_add_entry(struct inode *dir, const unsigned char *name, } /***** Create a file */ -static int msdos_create(struct inode *dir, struct dentry *dentry, int mode, +static int msdos_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { struct super_block *sb = dir->i_sb; @@ -346,7 +346,7 @@ out: } /***** Make a directory */ -static int msdos_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int msdos_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct super_block *sb = dir->i_sb; struct fat_slot_info sinfo; diff --git a/fs/fat/namei_vfat.c b/fs/fat/namei_vfat.c index a87a65663c25..a81eb2367d39 100644 --- a/fs/fat/namei_vfat.c +++ b/fs/fat/namei_vfat.c @@ -512,7 +512,8 @@ xlate_to_uni(const unsigned char *name, int len, unsigned char *outname, int charlen; if (utf8) { - *outlen = utf8s_to_utf16s(name, len, (wchar_t *)outname); + *outlen = utf8s_to_utf16s(name, len, UTF16_HOST_ENDIAN, + (wchar_t *) outname, FAT_LFN_LEN + 2); if (*outlen < 0) return *outlen; else if (*outlen > FAT_LFN_LEN) @@ -781,7 +782,7 @@ error: return ERR_PTR(err); } -static int vfat_create(struct inode *dir, struct dentry *dentry, int mode, +static int vfat_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { struct super_block *sb = dir->i_sb; @@ -870,7 +871,7 @@ out: return err; } -static int vfat_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int vfat_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct super_block *sb = dir->i_sb; struct inode *inode; diff --git a/fs/fhandle.c b/fs/fhandle.c index 6b088641f5bf..a48e4a139be1 100644 --- a/fs/fhandle.c +++ b/fs/fhandle.c @@ -10,6 +10,7 @@ #include <linux/personality.h> #include <asm/uaccess.h> #include "internal.h" +#include "mount.h" static long do_sys_name_to_handle(struct path *path, struct file_handle __user *ufh, @@ -24,8 +25,8 @@ static long do_sys_name_to_handle(struct path *path, * We need t make sure wether the file system * support decoding of the file handle */ - if (!path->mnt->mnt_sb->s_export_op || - !path->mnt->mnt_sb->s_export_op->fh_to_dentry) + if (!path->dentry->d_sb->s_export_op || + !path->dentry->d_sb->s_export_op->fh_to_dentry) return -EOPNOTSUPP; if (copy_from_user(&f_handle, ufh, sizeof(struct file_handle))) @@ -66,7 +67,8 @@ static long do_sys_name_to_handle(struct path *path, } else retval = 0; /* copy the mount id */ - if (copy_to_user(mnt_id, &path->mnt->mnt_id, sizeof(*mnt_id)) || + if (copy_to_user(mnt_id, &real_mount(path->mnt)->mnt_id, + sizeof(*mnt_id)) || copy_to_user(ufh, handle, sizeof(struct file_handle) + handle_bytes)) retval = -EFAULT; diff --git a/fs/file_table.c b/fs/file_table.c index c322794f7360..20002e39754d 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -474,29 +474,6 @@ void file_sb_list_del(struct file *file) #endif -int fs_may_remount_ro(struct super_block *sb) -{ - struct file *file; - /* Check that no files are currently opened for writing. */ - lg_global_lock(files_lglock); - do_file_list_for_each_entry(sb, file) { - struct inode *inode = file->f_path.dentry->d_inode; - - /* File with pending delete? */ - if (inode->i_nlink == 0) - goto too_bad; - - /* Writeable file? */ - if (S_ISREG(inode->i_mode) && (file->f_mode & FMODE_WRITE)) - goto too_bad; - } while_file_list_for_each_entry; - lg_global_unlock(files_lglock); - return 1; /* Tis' cool bro. */ -too_bad: - lg_global_unlock(files_lglock); - return 0; -} - /** * mark_files_ro - mark all files read-only * @sb: superblock in question diff --git a/fs/filesystems.c b/fs/filesystems.c index 0845f84f2a5f..96f24286667a 100644 --- a/fs/filesystems.c +++ b/fs/filesystems.c @@ -74,7 +74,6 @@ int register_filesystem(struct file_system_type * fs) BUG_ON(strchr(fs->name, '.')); if (fs->next) return -EBUSY; - INIT_LIST_HEAD(&fs->fs_supers); write_lock(&file_systems_lock); p = find_filesystem(fs->name, strlen(fs->name)); if (*p) diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c index 7b2af5abe2fa..cf9ef918a2a9 100644 --- a/fs/freevxfs/vxfs_inode.c +++ b/fs/freevxfs/vxfs_inode.c @@ -187,10 +187,10 @@ vxfs_stiget(struct super_block *sbp, ino_t ino) * vxfs_transmod returns a Linux mode_t for a given * VxFS inode structure. */ -static __inline__ mode_t +static __inline__ umode_t vxfs_transmod(struct vxfs_inode_info *vip) { - mode_t ret = vip->vii_mode & ~VXFS_TYPE_MASK; + umode_t ret = vip->vii_mode & ~VXFS_TYPE_MASK; if (VXFS_ISFIFO(vip)) ret |= S_IFIFO; @@ -340,7 +340,6 @@ vxfs_iget(struct super_block *sbp, ino_t ino) static void vxfs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(vxfs_inode_cachep, inode->i_private); } diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 73c3992b2bb4..f855916657ba 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -20,16 +20,21 @@ #include <linux/sched.h> #include <linux/fs.h> #include <linux/mm.h> +#include <linux/pagemap.h> #include <linux/kthread.h> #include <linux/freezer.h> #include <linux/writeback.h> #include <linux/blkdev.h> #include <linux/backing-dev.h> -#include <linux/buffer_head.h> #include <linux/tracepoint.h> #include "internal.h" /* + * 4MB minimal write chunk size + */ +#define MIN_WRITEBACK_PAGES (4096UL >> (PAGE_CACHE_SHIFT - 10)) + +/* * Passed into wb_writeback(), essentially a subset of writeback_control */ struct wb_writeback_work { @@ -47,17 +52,6 @@ struct wb_writeback_work { struct completion *done; /* set if the caller waits */ }; -const char *wb_reason_name[] = { - [WB_REASON_BACKGROUND] = "background", - [WB_REASON_TRY_TO_FREE_PAGES] = "try_to_free_pages", - [WB_REASON_SYNC] = "sync", - [WB_REASON_PERIODIC] = "periodic", - [WB_REASON_LAPTOP_TIMER] = "laptop_timer", - [WB_REASON_FREE_MORE_MEM] = "free_more_memory", - [WB_REASON_FS_FREE_SPACE] = "fs_free_space", - [WB_REASON_FORKER_THREAD] = "forker_thread" -}; - /* * Include the creation of the trace points after defining the * wb_writeback_work structure so that the definition remains local to this @@ -156,6 +150,7 @@ __bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages, * bdi_start_writeback - start writeback * @bdi: the backing device to write from * @nr_pages: the number of pages to write + * @reason: reason why some writeback work was initiated * * Description: * This does WB_SYNC_NONE opportunistic writeback. The IO is only @@ -753,11 +748,17 @@ static long wb_writeback(struct bdi_writeback *wb, if (work->for_background && !over_bground_thresh(wb->bdi)) break; + /* + * Kupdate and background works are special and we want to + * include all inodes that need writing. Livelock avoidance is + * handled by these works yielding to any other work so we are + * safe. + */ if (work->for_kupdate) { oldest_jif = jiffies - msecs_to_jiffies(dirty_expire_interval * 10); - work->older_than_this = &oldest_jif; - } + } else if (work->for_background) + oldest_jif = jiffies; trace_writeback_start(wb->bdi, work); if (list_empty(&wb->b_io)) @@ -947,7 +948,7 @@ int bdi_writeback_thread(void *data) trace_writeback_thread_start(bdi); - while (!kthread_should_stop()) { + while (!kthread_freezable_should_stop(NULL)) { /* * Remove own delayed wake-up timer, since we are already awake * and we'll take care of the preriodic write-back. @@ -977,8 +978,6 @@ int bdi_writeback_thread(void *data) */ schedule(); } - - try_to_freeze(); } /* Flush any work that raced with us exiting */ @@ -1223,6 +1222,7 @@ static void wait_sb_inodes(struct super_block *sb) * writeback_inodes_sb_nr - writeback dirty inodes from given super_block * @sb: the superblock * @nr: the number of pages to write + * @reason: reason why some writeback work initiated * * Start writeback on some inodes on this super_block. No guarantees are made * on how many (if any) will be written, and this function does not wait @@ -1251,6 +1251,7 @@ EXPORT_SYMBOL(writeback_inodes_sb_nr); /** * writeback_inodes_sb - writeback dirty inodes from given super_block * @sb: the superblock + * @reason: reason why some writeback work was initiated * * Start writeback on some inodes on this super_block. No guarantees are made * on how many (if any) will be written, and this function does not wait @@ -1265,6 +1266,7 @@ EXPORT_SYMBOL(writeback_inodes_sb); /** * writeback_inodes_sb_if_idle - start writeback if none underway * @sb: the superblock + * @reason: reason why some writeback work was initiated * * Invoke writeback_inodes_sb if no writeback is currently underway. * Returns 1 if writeback was started, 0 if not. @@ -1285,6 +1287,7 @@ EXPORT_SYMBOL(writeback_inodes_sb_if_idle); * writeback_inodes_sb_if_idle - start writeback if none underway * @sb: the superblock * @nr: the number of pages to write + * @reason: reason why some writeback work was initiated * * Invoke writeback_inodes_sb if no writeback is currently underway. * Returns 1 if writeback was started, 0 if not. diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 5cb8614508c3..5f3368ab0fa9 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -1378,7 +1378,59 @@ static int fuse_notify_inval_entry(struct fuse_conn *fc, unsigned int size, down_read(&fc->killsb); err = -ENOENT; if (fc->sb) - err = fuse_reverse_inval_entry(fc->sb, outarg.parent, &name); + err = fuse_reverse_inval_entry(fc->sb, outarg.parent, 0, &name); + up_read(&fc->killsb); + kfree(buf); + return err; + +err: + kfree(buf); + fuse_copy_finish(cs); + return err; +} + +static int fuse_notify_delete(struct fuse_conn *fc, unsigned int size, + struct fuse_copy_state *cs) +{ + struct fuse_notify_delete_out outarg; + int err = -ENOMEM; + char *buf; + struct qstr name; + + buf = kzalloc(FUSE_NAME_MAX + 1, GFP_KERNEL); + if (!buf) + goto err; + + err = -EINVAL; + if (size < sizeof(outarg)) + goto err; + + err = fuse_copy_one(cs, &outarg, sizeof(outarg)); + if (err) + goto err; + + err = -ENAMETOOLONG; + if (outarg.namelen > FUSE_NAME_MAX) + goto err; + + err = -EINVAL; + if (size != sizeof(outarg) + outarg.namelen + 1) + goto err; + + name.name = buf; + name.len = outarg.namelen; + err = fuse_copy_one(cs, buf, outarg.namelen + 1); + if (err) + goto err; + fuse_copy_finish(cs); + buf[outarg.namelen] = 0; + name.hash = full_name_hash(name.name, name.len); + + down_read(&fc->killsb); + err = -ENOENT; + if (fc->sb) + err = fuse_reverse_inval_entry(fc->sb, outarg.parent, + outarg.child, &name); up_read(&fc->killsb); kfree(buf); return err; @@ -1512,7 +1564,7 @@ static int fuse_retrieve(struct fuse_conn *fc, struct inode *inode, else if (outarg->offset + num > file_size) num = file_size - outarg->offset; - while (num) { + while (num && req->num_pages < FUSE_MAX_PAGES_PER_REQ) { struct page *page; unsigned int this_num; @@ -1526,6 +1578,7 @@ static int fuse_retrieve(struct fuse_conn *fc, struct inode *inode, num -= this_num; total_len += this_num; + index++; } req->misc.retrieve_in.offset = outarg->offset; req->misc.retrieve_in.size = total_len; @@ -1596,6 +1649,9 @@ static int fuse_notify(struct fuse_conn *fc, enum fuse_notify_code code, case FUSE_NOTIFY_RETRIEVE: return fuse_notify_retrieve(fc, size, cs); + case FUSE_NOTIFY_DELETE: + return fuse_notify_delete(fc, size, cs); + default: fuse_copy_finish(cs); return -EINVAL; diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 9f63e493a9b6..206632887bb4 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -369,8 +369,8 @@ static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry, * If the filesystem doesn't support this, then fall back to separate * 'mknod' + 'open' requests. */ -static int fuse_create_open(struct inode *dir, struct dentry *entry, int mode, - struct nameidata *nd) +static int fuse_create_open(struct inode *dir, struct dentry *entry, + umode_t mode, struct nameidata *nd) { int err; struct inode *inode; @@ -480,7 +480,7 @@ static int fuse_create_open(struct inode *dir, struct dentry *entry, int mode, */ static int create_new_entry(struct fuse_conn *fc, struct fuse_req *req, struct inode *dir, struct dentry *entry, - int mode) + umode_t mode) { struct fuse_entry_out outarg; struct inode *inode; @@ -547,7 +547,7 @@ static int create_new_entry(struct fuse_conn *fc, struct fuse_req *req, return err; } -static int fuse_mknod(struct inode *dir, struct dentry *entry, int mode, +static int fuse_mknod(struct inode *dir, struct dentry *entry, umode_t mode, dev_t rdev) { struct fuse_mknod_in inarg; @@ -573,7 +573,7 @@ static int fuse_mknod(struct inode *dir, struct dentry *entry, int mode, return create_new_entry(fc, req, dir, entry, mode); } -static int fuse_create(struct inode *dir, struct dentry *entry, int mode, +static int fuse_create(struct inode *dir, struct dentry *entry, umode_t mode, struct nameidata *nd) { if (nd) { @@ -585,7 +585,7 @@ static int fuse_create(struct inode *dir, struct dentry *entry, int mode, return fuse_mknod(dir, entry, mode, 0); } -static int fuse_mkdir(struct inode *dir, struct dentry *entry, int mode) +static int fuse_mkdir(struct inode *dir, struct dentry *entry, umode_t mode) { struct fuse_mkdir_in inarg; struct fuse_conn *fc = get_fuse_conn(dir); @@ -868,7 +868,7 @@ int fuse_update_attributes(struct inode *inode, struct kstat *stat, } int fuse_reverse_inval_entry(struct super_block *sb, u64 parent_nodeid, - struct qstr *name) + u64 child_nodeid, struct qstr *name) { int err = -ENOTDIR; struct inode *parent; @@ -895,8 +895,36 @@ int fuse_reverse_inval_entry(struct super_block *sb, u64 parent_nodeid, fuse_invalidate_attr(parent); fuse_invalidate_entry(entry); + + if (child_nodeid != 0 && entry->d_inode) { + mutex_lock(&entry->d_inode->i_mutex); + if (get_node_id(entry->d_inode) != child_nodeid) { + err = -ENOENT; + goto badentry; + } + if (d_mountpoint(entry)) { + err = -EBUSY; + goto badentry; + } + if (S_ISDIR(entry->d_inode->i_mode)) { + shrink_dcache_parent(entry); + if (!simple_empty(entry)) { + err = -ENOTEMPTY; + goto badentry; + } + entry->d_inode->i_flags |= S_DEAD; + } + dont_mount(entry); + clear_nlink(entry->d_inode); + err = 0; + badentry: + mutex_unlock(&entry->d_inode->i_mutex); + if (!err) + d_delete(entry); + } else { + err = 0; + } dput(entry); - err = 0; unlock: mutex_unlock(&parent->i_mutex); @@ -1182,6 +1210,30 @@ static int fuse_dir_fsync(struct file *file, loff_t start, loff_t end, return fuse_fsync_common(file, start, end, datasync, 1); } +static long fuse_dir_ioctl(struct file *file, unsigned int cmd, + unsigned long arg) +{ + struct fuse_conn *fc = get_fuse_conn(file->f_mapping->host); + + /* FUSE_IOCTL_DIR only supported for API version >= 7.18 */ + if (fc->minor < 18) + return -ENOTTY; + + return fuse_ioctl_common(file, cmd, arg, FUSE_IOCTL_DIR); +} + +static long fuse_dir_compat_ioctl(struct file *file, unsigned int cmd, + unsigned long arg) +{ + struct fuse_conn *fc = get_fuse_conn(file->f_mapping->host); + + if (fc->minor < 18) + return -ENOTTY; + + return fuse_ioctl_common(file, cmd, arg, + FUSE_IOCTL_COMPAT | FUSE_IOCTL_DIR); +} + static bool update_mtime(unsigned ivalid) { /* Always update if mtime is explicitly set */ @@ -1596,6 +1648,8 @@ static const struct file_operations fuse_dir_operations = { .open = fuse_dir_open, .release = fuse_dir_release, .fsync = fuse_dir_fsync, + .unlocked_ioctl = fuse_dir_ioctl, + .compat_ioctl = fuse_dir_compat_ioctl, }; static const struct inode_operations fuse_common_inode_operations = { diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 594f07a81c28..4a199fd93fbd 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -1555,44 +1555,16 @@ static loff_t fuse_file_llseek(struct file *file, loff_t offset, int origin) loff_t retval; struct inode *inode = file->f_path.dentry->d_inode; - mutex_lock(&inode->i_mutex); - if (origin != SEEK_CUR || origin != SEEK_SET) { - retval = fuse_update_attributes(inode, NULL, file, NULL); - if (retval) - goto exit; - } + /* No i_mutex protection necessary for SEEK_CUR and SEEK_SET */ + if (origin == SEEK_CUR || origin == SEEK_SET) + return generic_file_llseek(file, offset, origin); - switch (origin) { - case SEEK_END: - offset += i_size_read(inode); - break; - case SEEK_CUR: - offset += file->f_pos; - break; - case SEEK_DATA: - if (offset >= i_size_read(inode)) { - retval = -ENXIO; - goto exit; - } - break; - case SEEK_HOLE: - if (offset >= i_size_read(inode)) { - retval = -ENXIO; - goto exit; - } - offset = i_size_read(inode); - break; - } - retval = -EINVAL; - if (offset >= 0 && offset <= inode->i_sb->s_maxbytes) { - if (offset != file->f_pos) { - file->f_pos = offset; - file->f_version = 0; - } - retval = offset; - } -exit: + mutex_lock(&inode->i_mutex); + retval = fuse_update_attributes(inode, NULL, file, NULL); + if (!retval) + retval = generic_file_llseek(file, offset, origin); mutex_unlock(&inode->i_mutex); + return retval; } @@ -1804,7 +1776,7 @@ long fuse_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg, BUILD_BUG_ON(sizeof(struct fuse_ioctl_iovec) * FUSE_IOCTL_MAX_IOV > PAGE_SIZE); err = -ENOMEM; - pages = kzalloc(sizeof(pages[0]) * FUSE_MAX_PAGES_PER_REQ, GFP_KERNEL); + pages = kcalloc(FUSE_MAX_PAGES_PER_REQ, sizeof(pages[0]), GFP_KERNEL); iov_page = (struct iovec *) __get_free_page(GFP_KERNEL); if (!pages || !iov_page) goto out; @@ -1954,8 +1926,8 @@ long fuse_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg, } EXPORT_SYMBOL_GPL(fuse_do_ioctl); -static long fuse_file_ioctl_common(struct file *file, unsigned int cmd, - unsigned long arg, unsigned int flags) +long fuse_ioctl_common(struct file *file, unsigned int cmd, + unsigned long arg, unsigned int flags) { struct inode *inode = file->f_dentry->d_inode; struct fuse_conn *fc = get_fuse_conn(inode); @@ -1972,13 +1944,13 @@ static long fuse_file_ioctl_common(struct file *file, unsigned int cmd, static long fuse_file_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { - return fuse_file_ioctl_common(file, cmd, arg, 0); + return fuse_ioctl_common(file, cmd, arg, 0); } static long fuse_file_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { - return fuse_file_ioctl_common(file, cmd, arg, FUSE_IOCTL_COMPAT); + return fuse_ioctl_common(file, cmd, arg, FUSE_IOCTL_COMPAT); } /* diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index cf6db0a93219..572cefc78012 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -80,7 +80,7 @@ struct fuse_inode { /** The sticky bit in inode->i_mode may have been removed, so preserve the original mode */ - mode_t orig_i_mode; + umode_t orig_i_mode; /** Version of last attribute change */ u64 attr_version; @@ -755,9 +755,15 @@ int fuse_reverse_inval_inode(struct super_block *sb, u64 nodeid, /** * File-system tells the kernel to invalidate parent attributes and * the dentry matching parent/name. + * + * If the child_nodeid is non-zero and: + * - matches the inode number for the dentry matching parent/name, + * - is not a mount point + * - is a file or oan empty directory + * then the dentry is unhashed (d_delete()). */ int fuse_reverse_inval_entry(struct super_block *sb, u64 parent_nodeid, - struct qstr *name); + u64 child_nodeid, struct qstr *name); int fuse_do_open(struct fuse_conn *fc, u64 nodeid, struct file *file, bool isdir); @@ -765,6 +771,8 @@ ssize_t fuse_direct_io(struct file *file, const char __user *buf, size_t count, loff_t *ppos, int write); long fuse_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg, unsigned int flags); +long fuse_ioctl_common(struct file *file, unsigned int cmd, + unsigned long arg, unsigned int flags); unsigned fuse_file_poll(struct file *file, poll_table *wait); int fuse_dev_release(struct inode *inode, struct file *file); diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 3e6d72756479..64cf8d07393e 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -107,7 +107,6 @@ static struct inode *fuse_alloc_inode(struct super_block *sb) static void fuse_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(fuse_inode_cachep, inode); } @@ -498,9 +497,10 @@ static int parse_fuse_opt(char *opt, struct fuse_mount_data *d, int is_bdev) return 1; } -static int fuse_show_options(struct seq_file *m, struct vfsmount *mnt) +static int fuse_show_options(struct seq_file *m, struct dentry *root) { - struct fuse_conn *fc = get_fuse_conn_super(mnt->mnt_sb); + struct super_block *sb = root->d_sb; + struct fuse_conn *fc = get_fuse_conn_super(sb); seq_printf(m, ",user_id=%u", fc->user_id); seq_printf(m, ",group_id=%u", fc->group_id); @@ -510,9 +510,8 @@ static int fuse_show_options(struct seq_file *m, struct vfsmount *mnt) seq_puts(m, ",allow_other"); if (fc->max_read != ~0) seq_printf(m, ",max_read=%u", fc->max_read); - if (mnt->mnt_sb->s_bdev && - mnt->mnt_sb->s_blocksize != FUSE_DEFAULT_BLKSIZE) - seq_printf(m, ",blksize=%lu", mnt->mnt_sb->s_blocksize); + if (sb->s_bdev && sb->s_blocksize != FUSE_DEFAULT_BLKSIZE) + seq_printf(m, ",blksize=%lu", sb->s_blocksize); return 0; } @@ -1138,28 +1137,28 @@ static int __init fuse_fs_init(void) { int err; - err = register_filesystem(&fuse_fs_type); - if (err) - goto out; - - err = register_fuseblk(); - if (err) - goto out_unreg; - fuse_inode_cachep = kmem_cache_create("fuse_inode", sizeof(struct fuse_inode), 0, SLAB_HWCACHE_ALIGN, fuse_inode_init_once); err = -ENOMEM; if (!fuse_inode_cachep) - goto out_unreg2; + goto out; + + err = register_fuseblk(); + if (err) + goto out2; + + err = register_filesystem(&fuse_fs_type); + if (err) + goto out3; return 0; - out_unreg2: + out3: unregister_fuseblk(); - out_unreg: - unregister_filesystem(&fuse_fs_type); + out2: + kmem_cache_destroy(fuse_inode_cachep); out: return err; } diff --git a/fs/gfs2/acl.c b/fs/gfs2/acl.c index 65978d7885c8..230eb0f005b6 100644 --- a/fs/gfs2/acl.c +++ b/fs/gfs2/acl.c @@ -38,8 +38,9 @@ static const char *gfs2_acl_name(int type) return NULL; } -static struct posix_acl *gfs2_acl_get(struct gfs2_inode *ip, int type) +struct posix_acl *gfs2_get_acl(struct inode *inode, int type) { + struct gfs2_inode *ip = GFS2_I(inode); struct posix_acl *acl; const char *name; char *data; @@ -67,11 +68,6 @@ static struct posix_acl *gfs2_acl_get(struct gfs2_inode *ip, int type) return acl; } -struct posix_acl *gfs2_get_acl(struct inode *inode, int type) -{ - return gfs2_acl_get(GFS2_I(inode), type); -} - static int gfs2_set_mode(struct inode *inode, umode_t mode) { int error = 0; @@ -125,7 +121,7 @@ int gfs2_acl_create(struct gfs2_inode *dip, struct inode *inode) if (S_ISLNK(inode->i_mode)) return 0; - acl = gfs2_acl_get(dip, ACL_TYPE_DEFAULT); + acl = gfs2_get_acl(&dip->i_inode, ACL_TYPE_DEFAULT); if (IS_ERR(acl)) return PTR_ERR(acl); if (!acl) { @@ -166,7 +162,7 @@ int gfs2_acl_chmod(struct gfs2_inode *ip, struct iattr *attr) unsigned int len; int error; - acl = gfs2_acl_get(ip, ACL_TYPE_ACCESS); + acl = gfs2_get_acl(&ip->i_inode, ACL_TYPE_ACCESS); if (IS_ERR(acl)) return PTR_ERR(acl); if (!acl) @@ -216,7 +212,7 @@ static int gfs2_xattr_system_get(struct dentry *dentry, const char *name, if (type < 0) return type; - acl = gfs2_acl_get(GFS2_I(inode), type); + acl = gfs2_get_acl(inode, type); if (IS_ERR(acl)) return PTR_ERR(acl); if (acl == NULL) diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c index 4858e1fed8b1..501e5cba09b3 100644 --- a/fs/gfs2/aops.c +++ b/fs/gfs2/aops.c @@ -615,7 +615,7 @@ static int gfs2_write_begin(struct file *file, struct address_space *mapping, unsigned int data_blocks = 0, ind_blocks = 0, rblocks; int alloc_required; int error = 0; - struct gfs2_alloc *al = NULL; + struct gfs2_qadata *qa = NULL; pgoff_t index = pos >> PAGE_CACHE_SHIFT; unsigned from = pos & (PAGE_CACHE_SIZE - 1); struct page *page; @@ -639,8 +639,8 @@ static int gfs2_write_begin(struct file *file, struct address_space *mapping, gfs2_write_calc_reserv(ip, len, &data_blocks, &ind_blocks); if (alloc_required) { - al = gfs2_alloc_get(ip); - if (!al) { + qa = gfs2_qadata_get(ip); + if (!qa) { error = -ENOMEM; goto out_unlock; } @@ -649,8 +649,7 @@ static int gfs2_write_begin(struct file *file, struct address_space *mapping, if (error) goto out_alloc_put; - al->al_requested = data_blocks + ind_blocks; - error = gfs2_inplace_reserve(ip); + error = gfs2_inplace_reserve(ip, data_blocks + ind_blocks); if (error) goto out_qunlock; } @@ -711,7 +710,7 @@ out_trans_fail: out_qunlock: gfs2_quota_unlock(ip); out_alloc_put: - gfs2_alloc_put(ip); + gfs2_qadata_put(ip); } out_unlock: if (&ip->i_inode == sdp->sd_rindex) { @@ -848,7 +847,7 @@ static int gfs2_write_end(struct file *file, struct address_space *mapping, struct gfs2_sbd *sdp = GFS2_SB(inode); struct gfs2_inode *m_ip = GFS2_I(sdp->sd_statfs_inode); struct buffer_head *dibh; - struct gfs2_alloc *al = ip->i_alloc; + struct gfs2_qadata *qa = ip->i_qadata; unsigned int from = pos & (PAGE_CACHE_SIZE - 1); unsigned int to = from + len; int ret; @@ -880,10 +879,11 @@ static int gfs2_write_end(struct file *file, struct address_space *mapping, brelse(dibh); failed: gfs2_trans_end(sdp); - if (al) { + if (ip->i_res) gfs2_inplace_release(ip); + if (qa) { gfs2_quota_unlock(ip); - gfs2_alloc_put(ip); + gfs2_qadata_put(ip); } if (inode == sdp->sd_rindex) { gfs2_glock_dq(&m_ip->i_gh); diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index 41d494d79709..14a704015970 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -133,7 +133,7 @@ int gfs2_unstuff_dinode(struct gfs2_inode *ip, struct page *page) and write it out to disk */ unsigned int n = 1; - error = gfs2_alloc_block(ip, &block, &n); + error = gfs2_alloc_blocks(ip, &block, &n, 0, NULL); if (error) goto out_brelse; if (isdir) { @@ -503,7 +503,7 @@ static int gfs2_bmap_alloc(struct inode *inode, const sector_t lblock, do { int error; n = blks - alloced; - error = gfs2_alloc_block(ip, &bn, &n); + error = gfs2_alloc_blocks(ip, &bn, &n, 0, NULL); if (error) return error; alloced += n; @@ -743,9 +743,6 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, else if (ip->i_depth) revokes = sdp->sd_inptrs; - if (error) - return error; - memset(&rlist, 0, sizeof(struct gfs2_rgrp_list)); bstart = 0; blen = 0; @@ -1044,7 +1041,7 @@ static int trunc_dealloc(struct gfs2_inode *ip, u64 size) lblock = (size - 1) >> sdp->sd_sb.sb_bsize_shift; find_metapath(sdp, lblock, &mp, ip->i_height); - if (!gfs2_alloc_get(ip)) + if (!gfs2_qadata_get(ip)) return -ENOMEM; error = gfs2_quota_hold(ip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE); @@ -1064,7 +1061,7 @@ static int trunc_dealloc(struct gfs2_inode *ip, u64 size) gfs2_quota_unhold(ip); out: - gfs2_alloc_put(ip); + gfs2_qadata_put(ip); return error; } @@ -1166,21 +1163,20 @@ static int do_grow(struct inode *inode, u64 size) struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); struct buffer_head *dibh; - struct gfs2_alloc *al = NULL; + struct gfs2_qadata *qa = NULL; int error; if (gfs2_is_stuffed(ip) && (size > (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode)))) { - al = gfs2_alloc_get(ip); - if (al == NULL) + qa = gfs2_qadata_get(ip); + if (qa == NULL) return -ENOMEM; error = gfs2_quota_lock_check(ip); if (error) goto do_grow_alloc_put; - al->al_requested = 1; - error = gfs2_inplace_reserve(ip); + error = gfs2_inplace_reserve(ip, 1); if (error) goto do_grow_qunlock; } @@ -1189,7 +1185,7 @@ static int do_grow(struct inode *inode, u64 size) if (error) goto do_grow_release; - if (al) { + if (qa) { error = gfs2_unstuff_dinode(ip, NULL); if (error) goto do_end_trans; @@ -1208,12 +1204,12 @@ static int do_grow(struct inode *inode, u64 size) do_end_trans: gfs2_trans_end(sdp); do_grow_release: - if (al) { + if (qa) { gfs2_inplace_release(ip); do_grow_qunlock: gfs2_quota_unlock(ip); do_grow_alloc_put: - gfs2_alloc_put(ip); + gfs2_qadata_put(ip); } return error; } diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index 8ccad2467cb6..c35573abd371 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -76,6 +76,8 @@ #define IS_LEAF 1 /* Hashed (leaf) directory */ #define IS_DINODE 2 /* Linear (stuffed dinode block) directory */ +#define MAX_RA_BLOCKS 32 /* max read-ahead blocks */ + #define gfs2_disk_hash2offset(h) (((u64)(h)) >> 1) #define gfs2_dir_offset2hash(p) ((u32)(((u64)(p)) << 1)) @@ -821,7 +823,7 @@ static struct gfs2_leaf *new_leaf(struct inode *inode, struct buffer_head **pbh, struct gfs2_dirent *dent; struct qstr name = { .name = "", .len = 0, .hash = 0 }; - error = gfs2_alloc_block(ip, &bn, &n); + error = gfs2_alloc_blocks(ip, &bn, &n, 0, NULL); if (error) return NULL; bh = gfs2_meta_new(ip->i_gl, bn); @@ -1376,6 +1378,52 @@ out: return error; } +/** + * gfs2_dir_readahead - Issue read-ahead requests for leaf blocks. + * + * Note: we can't calculate each index like dir_e_read can because we don't + * have the leaf, and therefore we don't have the depth, and therefore we + * don't have the length. So we have to just read enough ahead to make up + * for the loss of information. + */ +static void gfs2_dir_readahead(struct inode *inode, unsigned hsize, u32 index, + struct file_ra_state *f_ra) +{ + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_glock *gl = ip->i_gl; + struct buffer_head *bh; + u64 blocknr = 0, last; + unsigned count; + + /* First check if we've already read-ahead for the whole range. */ + if (index + MAX_RA_BLOCKS < f_ra->start) + return; + + f_ra->start = max((pgoff_t)index, f_ra->start); + for (count = 0; count < MAX_RA_BLOCKS; count++) { + if (f_ra->start >= hsize) /* if exceeded the hash table */ + break; + + last = blocknr; + blocknr = be64_to_cpu(ip->i_hash_cache[f_ra->start]); + f_ra->start++; + if (blocknr == last) + continue; + + bh = gfs2_getbuf(gl, blocknr, 1); + if (trylock_buffer(bh)) { + if (buffer_uptodate(bh)) { + unlock_buffer(bh); + brelse(bh); + continue; + } + bh->b_end_io = end_buffer_read_sync; + submit_bh(READA | REQ_META, bh); + continue; + } + brelse(bh); + } +} /** * dir_e_read - Reads the entries from a directory into a filldir buffer @@ -1388,7 +1436,7 @@ out: */ static int dir_e_read(struct inode *inode, u64 *offset, void *opaque, - filldir_t filldir) + filldir_t filldir, struct file_ra_state *f_ra) { struct gfs2_inode *dip = GFS2_I(inode); u32 hsize, len = 0; @@ -1402,10 +1450,14 @@ static int dir_e_read(struct inode *inode, u64 *offset, void *opaque, hash = gfs2_dir_offset2hash(*offset); index = hash >> (32 - dip->i_depth); + if (dip->i_hash_cache == NULL) + f_ra->start = 0; lp = gfs2_dir_get_hash_table(dip); if (IS_ERR(lp)) return PTR_ERR(lp); + gfs2_dir_readahead(inode, hsize, index, f_ra); + while (index < hsize) { error = gfs2_dir_read_leaf(inode, offset, opaque, filldir, &copied, &depth, @@ -1423,7 +1475,7 @@ static int dir_e_read(struct inode *inode, u64 *offset, void *opaque, } int gfs2_dir_read(struct inode *inode, u64 *offset, void *opaque, - filldir_t filldir) + filldir_t filldir, struct file_ra_state *f_ra) { struct gfs2_inode *dip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); @@ -1437,7 +1489,7 @@ int gfs2_dir_read(struct inode *inode, u64 *offset, void *opaque, return 0; if (dip->i_diskflags & GFS2_DIF_EXHASH) - return dir_e_read(inode, offset, opaque, filldir); + return dir_e_read(inode, offset, opaque, filldir, f_ra); if (!gfs2_is_stuffed(dip)) { gfs2_consist_inode(dip); @@ -1798,7 +1850,7 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len, if (!ht) return -ENOMEM; - if (!gfs2_alloc_get(dip)) { + if (!gfs2_qadata_get(dip)) { error = -ENOMEM; goto out; } @@ -1887,7 +1939,7 @@ out_rlist: gfs2_rlist_free(&rlist); gfs2_quota_unhold(dip); out_put: - gfs2_alloc_put(dip); + gfs2_qadata_put(dip); out: kfree(ht); return error; diff --git a/fs/gfs2/dir.h b/fs/gfs2/dir.h index ff5772fbf024..98c960beab35 100644 --- a/fs/gfs2/dir.h +++ b/fs/gfs2/dir.h @@ -25,7 +25,7 @@ extern int gfs2_dir_add(struct inode *inode, const struct qstr *filename, const struct gfs2_inode *ip); extern int gfs2_dir_del(struct gfs2_inode *dip, const struct dentry *dentry); extern int gfs2_dir_read(struct inode *inode, u64 *offset, void *opaque, - filldir_t filldir); + filldir_t filldir, struct file_ra_state *f_ra); extern int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename, const struct gfs2_inode *nip, unsigned int new_type); diff --git a/fs/gfs2/export.c b/fs/gfs2/export.c index fe9945f2ff72..70ba891654f8 100644 --- a/fs/gfs2/export.c +++ b/fs/gfs2/export.c @@ -99,6 +99,7 @@ static int gfs2_get_name(struct dentry *parent, char *name, struct gfs2_holder gh; u64 offset = 0; int error; + struct file_ra_state f_ra = { .start = 0 }; if (!dir) return -EINVAL; @@ -118,7 +119,7 @@ static int gfs2_get_name(struct dentry *parent, char *name, if (error) return error; - error = gfs2_dir_read(dir, &offset, &gnfd, get_name_filldir); + error = gfs2_dir_read(dir, &offset, &gnfd, get_name_filldir, &f_ra); gfs2_glock_dq_uninit(&gh); diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index ce36a56dfeac..c5fb3597f696 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -105,7 +105,7 @@ static int gfs2_readdir(struct file *file, void *dirent, filldir_t filldir) return error; } - error = gfs2_dir_read(dir, &offset, dirent, filldir); + error = gfs2_dir_read(dir, &offset, dirent, filldir, &file->f_ra); gfs2_glock_dq_uninit(&d_gh); @@ -223,7 +223,7 @@ static int do_gfs2_set_flags(struct file *filp, u32 reqflags, u32 mask) int error; u32 new_flags, flags; - error = mnt_want_write(filp->f_path.mnt); + error = mnt_want_write_file(filp); if (error) return error; @@ -285,7 +285,7 @@ out_trans_end: out: gfs2_glock_dq_uninit(&gh); out_drop_write: - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return error; } @@ -365,7 +365,7 @@ static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) u64 pos = page->index << PAGE_CACHE_SHIFT; unsigned int data_blocks, ind_blocks, rblocks; struct gfs2_holder gh; - struct gfs2_alloc *al; + struct gfs2_qadata *qa; loff_t size; int ret; @@ -393,16 +393,15 @@ static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) } ret = -ENOMEM; - al = gfs2_alloc_get(ip); - if (al == NULL) + qa = gfs2_qadata_get(ip); + if (qa == NULL) goto out_unlock; ret = gfs2_quota_lock_check(ip); if (ret) goto out_alloc_put; gfs2_write_calc_reserv(ip, PAGE_CACHE_SIZE, &data_blocks, &ind_blocks); - al->al_requested = data_blocks + ind_blocks; - ret = gfs2_inplace_reserve(ip); + ret = gfs2_inplace_reserve(ip, data_blocks + ind_blocks); if (ret) goto out_quota_unlock; @@ -448,7 +447,7 @@ out_trans_fail: out_quota_unlock: gfs2_quota_unlock(ip); out_alloc_put: - gfs2_alloc_put(ip); + gfs2_qadata_put(ip); out_unlock: gfs2_glock_dq(&gh); out: @@ -609,7 +608,7 @@ static int gfs2_fsync(struct file *file, loff_t start, loff_t end, struct inode *inode = mapping->host; int sync_state = inode->i_state & (I_DIRTY_SYNC|I_DIRTY_DATASYNC); struct gfs2_inode *ip = GFS2_I(inode); - int ret, ret1 = 0; + int ret = 0, ret1 = 0; if (mapping->nrpages) { ret1 = filemap_fdatawrite_range(mapping, start, end); @@ -750,8 +749,10 @@ static long gfs2_fallocate(struct file *file, int mode, loff_t offset, struct gfs2_inode *ip = GFS2_I(inode); unsigned int data_blocks = 0, ind_blocks = 0, rblocks; loff_t bytes, max_bytes; - struct gfs2_alloc *al; + struct gfs2_qadata *qa; int error; + const loff_t pos = offset; + const loff_t count = len; loff_t bsize_mask = ~((loff_t)sdp->sd_sb.sb_bsize - 1); loff_t next = (offset + len - 1) >> sdp->sd_sb.sb_bsize_shift; loff_t max_chunk_size = UINT_MAX & bsize_mask; @@ -782,8 +783,8 @@ static long gfs2_fallocate(struct file *file, int mode, loff_t offset, while (len > 0) { if (len < bytes) bytes = len; - al = gfs2_alloc_get(ip); - if (!al) { + qa = gfs2_qadata_get(ip); + if (!qa) { error = -ENOMEM; goto out_unlock; } @@ -795,8 +796,7 @@ static long gfs2_fallocate(struct file *file, int mode, loff_t offset, retry: gfs2_write_calc_reserv(ip, bytes, &data_blocks, &ind_blocks); - al->al_requested = data_blocks + ind_blocks; - error = gfs2_inplace_reserve(ip); + error = gfs2_inplace_reserve(ip, data_blocks + ind_blocks); if (error) { if (error == -ENOSPC && bytes > sdp->sd_sb.sb_bsize) { bytes >>= 1; @@ -810,7 +810,6 @@ retry: max_bytes = bytes; calc_max_reserv(ip, (len > max_chunk_size)? max_chunk_size: len, &max_bytes, &data_blocks, &ind_blocks); - al->al_requested = data_blocks + ind_blocks; rblocks = RES_DINODE + ind_blocks + RES_STATFS + RES_QUOTA + RES_RG_HDR + gfs2_rg_blocks(ip); @@ -832,8 +831,11 @@ retry: offset += max_bytes; gfs2_inplace_release(ip); gfs2_quota_unlock(ip); - gfs2_alloc_put(ip); + gfs2_qadata_put(ip); } + + if (error == 0) + error = generic_write_sync(file, pos, count); goto out_unlock; out_trans_fail: @@ -841,7 +843,7 @@ out_trans_fail: out_qunlock: gfs2_quota_unlock(ip); out_alloc_put: - gfs2_alloc_put(ip); + gfs2_qadata_put(ip); out_unlock: gfs2_glock_dq(&ip->i_gh); out_uninit: diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 88e8a23d0026..376816fcd040 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -1353,7 +1353,7 @@ void gfs2_glock_complete(struct gfs2_glock *gl, int ret) spin_lock(&gl->gl_spin); gl->gl_reply = ret; - if (unlikely(test_bit(DFL_BLOCK_LOCKS, &ls->ls_flags))) { + if (unlikely(test_bit(DFL_BLOCK_LOCKS, &ls->ls_recover_flags))) { if (gfs2_should_freeze(gl)) { set_bit(GLF_FROZEN, &gl->gl_flags); spin_unlock(&gl->gl_spin); diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h index 2553b858a72e..307ac31df781 100644 --- a/fs/gfs2/glock.h +++ b/fs/gfs2/glock.h @@ -121,8 +121,11 @@ enum { struct lm_lockops { const char *lm_proto_name; - int (*lm_mount) (struct gfs2_sbd *sdp, const char *fsname); - void (*lm_unmount) (struct gfs2_sbd *sdp); + int (*lm_mount) (struct gfs2_sbd *sdp, const char *table); + void (*lm_first_done) (struct gfs2_sbd *sdp); + void (*lm_recovery_result) (struct gfs2_sbd *sdp, unsigned int jid, + unsigned int result); + void (*lm_unmount) (struct gfs2_sbd *sdp); void (*lm_withdraw) (struct gfs2_sbd *sdp); void (*lm_put_lock) (struct gfs2_glock *gl); int (*lm_lock) (struct gfs2_glock *gl, unsigned int req_state, diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index 7389dfdcc9ef..97742a7ea9cc 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -139,8 +139,45 @@ struct gfs2_bufdata { #define GDLM_STRNAME_BYTES 25 #define GDLM_LVB_SIZE 32 +/* + * ls_recover_flags: + * + * DFL_BLOCK_LOCKS: dlm is in recovery and will grant locks that had been + * held by failed nodes whose journals need recovery. Those locks should + * only be used for journal recovery until the journal recovery is done. + * This is set by the dlm recover_prep callback and cleared by the + * gfs2_control thread when journal recovery is complete. To avoid + * races between recover_prep setting and gfs2_control clearing, recover_spin + * is held while changing this bit and reading/writing recover_block + * and recover_start. + * + * DFL_NO_DLM_OPS: dlm lockspace ops/callbacks are not being used. + * + * DFL_FIRST_MOUNT: this node is the first to mount this fs and is doing + * recovery of all journals before allowing other nodes to mount the fs. + * This is cleared when FIRST_MOUNT_DONE is set. + * + * DFL_FIRST_MOUNT_DONE: this node was the first mounter, and has finished + * recovery of all journals, and now allows other nodes to mount the fs. + * + * DFL_MOUNT_DONE: gdlm_mount has completed successfully and cleared + * BLOCK_LOCKS for the first time. The gfs2_control thread should now + * control clearing BLOCK_LOCKS for further recoveries. + * + * DFL_UNMOUNT: gdlm_unmount sets to keep sdp off gfs2_control_wq. + * + * DFL_DLM_RECOVERY: set while dlm is in recovery, between recover_prep() + * and recover_done(), i.e. set while recover_block == recover_start. + */ + enum { DFL_BLOCK_LOCKS = 0, + DFL_NO_DLM_OPS = 1, + DFL_FIRST_MOUNT = 2, + DFL_FIRST_MOUNT_DONE = 3, + DFL_MOUNT_DONE = 4, + DFL_UNMOUNT = 5, + DFL_DLM_RECOVERY = 6, }; struct lm_lockname { @@ -244,17 +281,16 @@ struct gfs2_glock { #define GFS2_MIN_LVB_SIZE 32 /* Min size of LVB that gfs2 supports */ -struct gfs2_alloc { +struct gfs2_qadata { /* quota allocation data */ /* Quota stuff */ - struct gfs2_quota_data *al_qd[2*MAXQUOTAS]; - struct gfs2_holder al_qd_ghs[2*MAXQUOTAS]; - unsigned int al_qd_num; - - u32 al_requested; /* Filled in by caller of gfs2_inplace_reserve() */ - u32 al_alloced; /* Filled in by gfs2_alloc_*() */ + struct gfs2_quota_data *qa_qd[2*MAXQUOTAS]; + struct gfs2_holder qa_qd_ghs[2*MAXQUOTAS]; + unsigned int qa_qd_num; +}; - /* Filled in by gfs2_inplace_reserve() */ - struct gfs2_holder al_rgd_gh; +struct gfs2_blkreserv { + u32 rs_requested; /* Filled in by caller of gfs2_inplace_reserve() */ + struct gfs2_holder rs_rgd_gh; /* Filled in by gfs2_inplace_reserve() */ }; enum { @@ -275,7 +311,8 @@ struct gfs2_inode { struct gfs2_glock *i_gl; /* Move into i_gh? */ struct gfs2_holder i_iopen_gh; struct gfs2_holder i_gh; /* for prepare/commit_write only */ - struct gfs2_alloc *i_alloc; + struct gfs2_qadata *i_qadata; /* quota allocation data */ + struct gfs2_blkreserv *i_res; /* resource group block reservation */ struct gfs2_rgrpd *i_rgd; u64 i_goal; /* goal block for allocations */ struct rw_semaphore i_rw_mutex; @@ -392,6 +429,7 @@ struct gfs2_jdesc { #define JDF_RECOVERY 1 unsigned int jd_jid; unsigned int jd_blocks; + int jd_recover_error; }; struct gfs2_statfs_change_host { @@ -461,6 +499,7 @@ enum { SDF_NORECOVERY = 4, SDF_DEMOTE = 5, SDF_NOJOURNALID = 6, + SDF_RORECOVERY = 7, /* read only recovery */ }; #define GFS2_FSNAME_LEN 256 @@ -499,14 +538,26 @@ struct gfs2_sb_host { struct lm_lockstruct { int ls_jid; unsigned int ls_first; - unsigned int ls_first_done; unsigned int ls_nodir; const struct lm_lockops *ls_ops; - unsigned long ls_flags; dlm_lockspace_t *ls_dlm; - int ls_recover_jid_done; - int ls_recover_jid_status; + int ls_recover_jid_done; /* These two are deprecated, */ + int ls_recover_jid_status; /* used previously by gfs_controld */ + + struct dlm_lksb ls_mounted_lksb; /* mounted_lock */ + struct dlm_lksb ls_control_lksb; /* control_lock */ + char ls_control_lvb[GDLM_LVB_SIZE]; /* control_lock lvb */ + struct completion ls_sync_wait; /* {control,mounted}_{lock,unlock} */ + + spinlock_t ls_recover_spin; /* protects following fields */ + unsigned long ls_recover_flags; /* DFL_ */ + uint32_t ls_recover_mount; /* gen in first recover_done cb */ + uint32_t ls_recover_start; /* gen in last recover_done cb */ + uint32_t ls_recover_block; /* copy recover_start in last recover_prep */ + uint32_t ls_recover_size; /* size of recover_submit, recover_result */ + uint32_t *ls_recover_submit; /* gen in last recover_slot cb per jid */ + uint32_t *ls_recover_result; /* result of last jid recovery */ }; struct gfs2_sbd { @@ -544,6 +595,7 @@ struct gfs2_sbd { wait_queue_head_t sd_glock_wait; atomic_t sd_glock_disposal; struct completion sd_locking_init; + struct delayed_work sd_control_work; /* Inode Stuff */ diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index cfd4959b218c..a7d611b93f0f 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -333,7 +333,7 @@ out: */ static int create_ok(struct gfs2_inode *dip, const struct qstr *name, - unsigned int mode) + umode_t mode) { int error; @@ -364,7 +364,7 @@ static int create_ok(struct gfs2_inode *dip, const struct qstr *name, return 0; } -static void munge_mode_uid_gid(struct gfs2_inode *dip, unsigned int *mode, +static void munge_mode_uid_gid(struct gfs2_inode *dip, umode_t *mode, unsigned int *uid, unsigned int *gid) { if (GFS2_SB(&dip->i_inode)->sd_args.ar_suiddir && @@ -389,12 +389,13 @@ static int alloc_dinode(struct gfs2_inode *dip, u64 *no_addr, u64 *generation) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); int error; + int dblocks = 1; - if (gfs2_alloc_get(dip) == NULL) - return -ENOMEM; + error = gfs2_rindex_update(sdp); + if (error) + fs_warn(sdp, "rindex update returns %d\n", error); - dip->i_alloc->al_requested = RES_DINODE; - error = gfs2_inplace_reserve(dip); + error = gfs2_inplace_reserve(dip, RES_DINODE); if (error) goto out; @@ -402,14 +403,13 @@ static int alloc_dinode(struct gfs2_inode *dip, u64 *no_addr, u64 *generation) if (error) goto out_ipreserv; - error = gfs2_alloc_di(dip, no_addr, generation); + error = gfs2_alloc_blocks(dip, no_addr, &dblocks, 1, generation); gfs2_trans_end(sdp); out_ipreserv: gfs2_inplace_release(dip); out: - gfs2_alloc_put(dip); return error; } @@ -447,7 +447,7 @@ static void gfs2_init_dir(struct buffer_head *dibh, */ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, - const struct gfs2_inum_host *inum, unsigned int mode, + const struct gfs2_inum_host *inum, umode_t mode, unsigned int uid, unsigned int gid, const u64 *generation, dev_t dev, const char *symname, unsigned size, struct buffer_head **bhp) @@ -516,7 +516,7 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, } static int make_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, - unsigned int mode, const struct gfs2_inum_host *inum, + umode_t mode, const struct gfs2_inum_host *inum, const u64 *generation, dev_t dev, const char *symname, unsigned int size, struct buffer_head **bhp) { @@ -525,7 +525,7 @@ static int make_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, int error; munge_mode_uid_gid(dip, &mode, &uid, &gid); - if (!gfs2_alloc_get(dip)) + if (!gfs2_qadata_get(dip)) return -ENOMEM; error = gfs2_quota_lock(dip, uid, gid); @@ -547,7 +547,7 @@ static int make_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, out_quota: gfs2_quota_unlock(dip); out: - gfs2_alloc_put(dip); + gfs2_qadata_put(dip); return error; } @@ -555,13 +555,13 @@ static int link_dinode(struct gfs2_inode *dip, const struct qstr *name, struct gfs2_inode *ip) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); - struct gfs2_alloc *al; + struct gfs2_qadata *qa; int alloc_required; struct buffer_head *dibh; int error; - al = gfs2_alloc_get(dip); - if (!al) + qa = gfs2_qadata_get(dip); + if (!qa) return -ENOMEM; error = gfs2_quota_lock(dip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE); @@ -576,9 +576,7 @@ static int link_dinode(struct gfs2_inode *dip, const struct qstr *name, if (error) goto fail_quota_locks; - al->al_requested = sdp->sd_max_dirres; - - error = gfs2_inplace_reserve(dip); + error = gfs2_inplace_reserve(dip, sdp->sd_max_dirres); if (error) goto fail_quota_locks; @@ -601,9 +599,7 @@ static int link_dinode(struct gfs2_inode *dip, const struct qstr *name, error = gfs2_meta_inode_buffer(ip, &dibh); if (error) goto fail_end_trans; - inc_nlink(&ip->i_inode); - if (S_ISDIR(ip->i_inode.i_mode)) - inc_nlink(&ip->i_inode); + set_nlink(&ip->i_inode, S_ISDIR(ip->i_inode.i_mode) ? 2 : 1); gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); @@ -619,11 +615,11 @@ fail_quota_locks: gfs2_quota_unlock(dip); fail: - gfs2_alloc_put(dip); + gfs2_qadata_put(dip); return error; } -int gfs2_initxattrs(struct inode *inode, const struct xattr *xattr_array, +static int gfs2_initxattrs(struct inode *inode, const struct xattr *xattr_array, void *fs_info) { const struct xattr *xattr; @@ -659,7 +655,7 @@ static int gfs2_security_init(struct gfs2_inode *dip, struct gfs2_inode *ip, */ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry, - unsigned int mode, dev_t dev, const char *symname, + umode_t mode, dev_t dev, const char *symname, unsigned int size, int excl) { const struct qstr *name = &dentry->d_name; @@ -728,9 +724,12 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry, brelse(bh); gfs2_trans_end(sdp); - gfs2_inplace_release(dip); + /* Check if we reserved space in the rgrp. Function link_dinode may + not, depending on whether alloc is required. */ + if (dip->i_res) + gfs2_inplace_release(dip); gfs2_quota_unlock(dip); - gfs2_alloc_put(dip); + gfs2_qadata_put(dip); mark_inode_dirty(inode); gfs2_glock_dq_uninit_m(2, ghs); d_instantiate(dentry, inode); @@ -760,7 +759,7 @@ fail: */ static int gfs2_create(struct inode *dir, struct dentry *dentry, - int mode, struct nameidata *nd) + umode_t mode, struct nameidata *nd) { int excl = 0; if (nd && (nd->flags & LOOKUP_EXCL)) @@ -875,8 +874,9 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir, error = 0; if (alloc_required) { - struct gfs2_alloc *al = gfs2_alloc_get(dip); - if (!al) { + struct gfs2_qadata *qa = gfs2_qadata_get(dip); + + if (!qa) { error = -ENOMEM; goto out_gunlock; } @@ -885,9 +885,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir, if (error) goto out_alloc; - al->al_requested = sdp->sd_max_dirres; - - error = gfs2_inplace_reserve(dip); + error = gfs2_inplace_reserve(dip, sdp->sd_max_dirres); if (error) goto out_gunlock_q; @@ -930,7 +928,7 @@ out_gunlock_q: gfs2_quota_unlock(dip); out_alloc: if (alloc_required) - gfs2_alloc_put(dip); + gfs2_qadata_put(dip); out_gunlock: gfs2_glock_dq(ghs + 1); out_child: @@ -1037,12 +1035,14 @@ static int gfs2_unlink(struct inode *dir, struct dentry *dentry) struct buffer_head *bh; struct gfs2_holder ghs[3]; struct gfs2_rgrpd *rgd; - int error; + int error = -EROFS; gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + 1); rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr); + if (!rgd) + goto out_inodes; gfs2_holder_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + 2); @@ -1088,12 +1088,13 @@ out_end_trans: out_gunlock: gfs2_glock_dq(ghs + 2); out_rgrp: - gfs2_holder_uninit(ghs + 2); gfs2_glock_dq(ghs + 1); out_child: - gfs2_holder_uninit(ghs + 1); gfs2_glock_dq(ghs); out_parent: + gfs2_holder_uninit(ghs + 2); +out_inodes: + gfs2_holder_uninit(ghs + 1); gfs2_holder_uninit(ghs); return error; } @@ -1129,7 +1130,7 @@ static int gfs2_symlink(struct inode *dir, struct dentry *dentry, * Returns: errno */ -static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { return gfs2_create_inode(dir, dentry, S_IFDIR | mode, 0, NULL, 0, 0); } @@ -1143,7 +1144,7 @@ static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode) * */ -static int gfs2_mknod(struct inode *dir, struct dentry *dentry, int 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); @@ -1350,8 +1351,9 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, error = 0; if (alloc_required) { - struct gfs2_alloc *al = gfs2_alloc_get(ndip); - if (!al) { + struct gfs2_qadata *qa = gfs2_qadata_get(ndip); + + if (!qa) { error = -ENOMEM; goto out_gunlock; } @@ -1360,9 +1362,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, if (error) goto out_alloc; - al->al_requested = sdp->sd_max_dirres; - - error = gfs2_inplace_reserve(ndip); + error = gfs2_inplace_reserve(ndip, sdp->sd_max_dirres); if (error) goto out_gunlock_q; @@ -1423,7 +1423,7 @@ out_gunlock_q: gfs2_quota_unlock(ndip); out_alloc: if (alloc_required) - gfs2_alloc_put(ndip); + gfs2_qadata_put(ndip); out_gunlock: while (x--) { gfs2_glock_dq(ghs + x); @@ -1584,7 +1584,7 @@ static int setattr_chown(struct inode *inode, struct iattr *attr) if (!(attr->ia_valid & ATTR_GID) || ogid == ngid) ogid = ngid = NO_QUOTA_CHANGE; - if (!gfs2_alloc_get(ip)) + if (!gfs2_qadata_get(ip)) return -ENOMEM; error = gfs2_quota_lock(ip, nuid, ngid); @@ -1616,7 +1616,7 @@ out_end_trans: out_gunlock_q: gfs2_quota_unlock(ip); out_alloc: - gfs2_alloc_put(ip); + gfs2_qadata_put(ip); return error; } diff --git a/fs/gfs2/lock_dlm.c b/fs/gfs2/lock_dlm.c index 98c80d8c2a62..8944d1e32ab5 100644 --- a/fs/gfs2/lock_dlm.c +++ b/fs/gfs2/lock_dlm.c @@ -1,6 +1,6 @@ /* * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. - * Copyright (C) 2004-2009 Red Hat, Inc. All rights reserved. + * Copyright 2004-2011 Red Hat, Inc. * * This copyrighted material is made available to anyone wishing to use, * modify, copy, or redistribute it subject to the terms and conditions @@ -11,12 +11,15 @@ #include <linux/dlm.h> #include <linux/slab.h> #include <linux/types.h> +#include <linux/delay.h> #include <linux/gfs2_ondisk.h> #include "incore.h" #include "glock.h" #include "util.h" +#include "sys.h" +extern struct workqueue_struct *gfs2_control_wq; static void gdlm_ast(void *arg) { @@ -185,34 +188,1002 @@ static void gdlm_cancel(struct gfs2_glock *gl) dlm_unlock(ls->ls_dlm, gl->gl_lksb.sb_lkid, DLM_LKF_CANCEL, NULL, gl); } -static int gdlm_mount(struct gfs2_sbd *sdp, const char *fsname) +/* + * dlm/gfs2 recovery coordination using dlm_recover callbacks + * + * 1. dlm_controld sees lockspace members change + * 2. dlm_controld blocks dlm-kernel locking activity + * 3. dlm_controld within dlm-kernel notifies gfs2 (recover_prep) + * 4. dlm_controld starts and finishes its own user level recovery + * 5. dlm_controld starts dlm-kernel dlm_recoverd to do kernel recovery + * 6. dlm_recoverd notifies gfs2 of failed nodes (recover_slot) + * 7. dlm_recoverd does its own lock recovery + * 8. dlm_recoverd unblocks dlm-kernel locking activity + * 9. dlm_recoverd notifies gfs2 when done (recover_done with new generation) + * 10. gfs2_control updates control_lock lvb with new generation and jid bits + * 11. gfs2_control enqueues journals for gfs2_recover to recover (maybe none) + * 12. gfs2_recover dequeues and recovers journals of failed nodes + * 13. gfs2_recover provides recovery results to gfs2_control (recovery_result) + * 14. gfs2_control updates control_lock lvb jid bits for recovered journals + * 15. gfs2_control unblocks normal locking when all journals are recovered + * + * - failures during recovery + * + * recover_prep() may set BLOCK_LOCKS (step 3) again before gfs2_control + * clears BLOCK_LOCKS (step 15), e.g. another node fails while still + * recovering for a prior failure. gfs2_control needs a way to detect + * this so it can leave BLOCK_LOCKS set in step 15. This is managed using + * the recover_block and recover_start values. + * + * recover_done() provides a new lockspace generation number each time it + * is called (step 9). This generation number is saved as recover_start. + * When recover_prep() is called, it sets BLOCK_LOCKS and sets + * recover_block = recover_start. So, while recover_block is equal to + * recover_start, BLOCK_LOCKS should remain set. (recover_spin must + * be held around the BLOCK_LOCKS/recover_block/recover_start logic.) + * + * - more specific gfs2 steps in sequence above + * + * 3. recover_prep sets BLOCK_LOCKS and sets recover_block = recover_start + * 6. recover_slot records any failed jids (maybe none) + * 9. recover_done sets recover_start = new generation number + * 10. gfs2_control sets control_lock lvb = new gen + bits for failed jids + * 12. gfs2_recover does journal recoveries for failed jids identified above + * 14. gfs2_control clears control_lock lvb bits for recovered jids + * 15. gfs2_control checks if recover_block == recover_start (step 3 occured + * again) then do nothing, otherwise if recover_start > recover_block + * then clear BLOCK_LOCKS. + * + * - parallel recovery steps across all nodes + * + * All nodes attempt to update the control_lock lvb with the new generation + * number and jid bits, but only the first to get the control_lock EX will + * do so; others will see that it's already done (lvb already contains new + * generation number.) + * + * . All nodes get the same recover_prep/recover_slot/recover_done callbacks + * . All nodes attempt to set control_lock lvb gen + bits for the new gen + * . One node gets control_lock first and writes the lvb, others see it's done + * . All nodes attempt to recover jids for which they see control_lock bits set + * . One node succeeds for a jid, and that one clears the jid bit in the lvb + * . All nodes will eventually see all lvb bits clear and unblock locks + * + * - is there a problem with clearing an lvb bit that should be set + * and missing a journal recovery? + * + * 1. jid fails + * 2. lvb bit set for step 1 + * 3. jid recovered for step 1 + * 4. jid taken again (new mount) + * 5. jid fails (for step 4) + * 6. lvb bit set for step 5 (will already be set) + * 7. lvb bit cleared for step 3 + * + * This is not a problem because the failure in step 5 does not + * require recovery, because the mount in step 4 could not have + * progressed far enough to unblock locks and access the fs. The + * control_mount() function waits for all recoveries to be complete + * for the latest lockspace generation before ever unblocking locks + * and returning. The mount in step 4 waits until the recovery in + * step 1 is done. + * + * - special case of first mounter: first node to mount the fs + * + * The first node to mount a gfs2 fs needs to check all the journals + * and recover any that need recovery before other nodes are allowed + * to mount the fs. (Others may begin mounting, but they must wait + * for the first mounter to be done before taking locks on the fs + * or accessing the fs.) This has two parts: + * + * 1. The mounted_lock tells a node it's the first to mount the fs. + * Each node holds the mounted_lock in PR while it's mounted. + * Each node tries to acquire the mounted_lock in EX when it mounts. + * If a node is granted the mounted_lock EX it means there are no + * other mounted nodes (no PR locks exist), and it is the first mounter. + * The mounted_lock is demoted to PR when first recovery is done, so + * others will fail to get an EX lock, but will get a PR lock. + * + * 2. The control_lock blocks others in control_mount() while the first + * mounter is doing first mount recovery of all journals. + * A mounting node needs to acquire control_lock in EX mode before + * it can proceed. The first mounter holds control_lock in EX while doing + * the first mount recovery, blocking mounts from other nodes, then demotes + * control_lock to NL when it's done (others_may_mount/first_done), + * allowing other nodes to continue mounting. + * + * first mounter: + * control_lock EX/NOQUEUE success + * mounted_lock EX/NOQUEUE success (no other PR, so no other mounters) + * set first=1 + * do first mounter recovery + * mounted_lock EX->PR + * control_lock EX->NL, write lvb generation + * + * other mounter: + * control_lock EX/NOQUEUE success (if fail -EAGAIN, retry) + * mounted_lock EX/NOQUEUE fail -EAGAIN (expected due to other mounters PR) + * mounted_lock PR/NOQUEUE success + * read lvb generation + * control_lock EX->NL + * set first=0 + * + * - mount during recovery + * + * If a node mounts while others are doing recovery (not first mounter), + * the mounting node will get its initial recover_done() callback without + * having seen any previous failures/callbacks. + * + * It must wait for all recoveries preceding its mount to be finished + * before it unblocks locks. It does this by repeating the "other mounter" + * steps above until the lvb generation number is >= its mount generation + * number (from initial recover_done) and all lvb bits are clear. + * + * - control_lock lvb format + * + * 4 bytes generation number: the latest dlm lockspace generation number + * from recover_done callback. Indicates the jid bitmap has been updated + * to reflect all slot failures through that generation. + * 4 bytes unused. + * GDLM_LVB_SIZE-8 bytes of jid bit map. If bit N is set, it indicates + * that jid N needs recovery. + */ + +#define JID_BITMAP_OFFSET 8 /* 4 byte generation number + 4 byte unused */ + +static void control_lvb_read(struct lm_lockstruct *ls, uint32_t *lvb_gen, + char *lvb_bits) +{ + uint32_t gen; + memcpy(lvb_bits, ls->ls_control_lvb, GDLM_LVB_SIZE); + memcpy(&gen, lvb_bits, sizeof(uint32_t)); + *lvb_gen = le32_to_cpu(gen); +} + +static void control_lvb_write(struct lm_lockstruct *ls, uint32_t lvb_gen, + char *lvb_bits) +{ + uint32_t gen; + memcpy(ls->ls_control_lvb, lvb_bits, GDLM_LVB_SIZE); + gen = cpu_to_le32(lvb_gen); + memcpy(ls->ls_control_lvb, &gen, sizeof(uint32_t)); +} + +static int all_jid_bits_clear(char *lvb) +{ + int i; + for (i = JID_BITMAP_OFFSET; i < GDLM_LVB_SIZE; i++) { + if (lvb[i]) + return 0; + } + return 1; +} + +static void sync_wait_cb(void *arg) +{ + struct lm_lockstruct *ls = arg; + complete(&ls->ls_sync_wait); +} + +static int sync_unlock(struct gfs2_sbd *sdp, struct dlm_lksb *lksb, char *name) { struct lm_lockstruct *ls = &sdp->sd_lockstruct; int error; - if (fsname == NULL) { - fs_info(sdp, "no fsname found\n"); - return -EINVAL; + error = dlm_unlock(ls->ls_dlm, lksb->sb_lkid, 0, lksb, ls); + if (error) { + fs_err(sdp, "%s lkid %x error %d\n", + name, lksb->sb_lkid, error); + return error; + } + + wait_for_completion(&ls->ls_sync_wait); + + if (lksb->sb_status != -DLM_EUNLOCK) { + fs_err(sdp, "%s lkid %x status %d\n", + name, lksb->sb_lkid, lksb->sb_status); + return -1; + } + return 0; +} + +static int sync_lock(struct gfs2_sbd *sdp, int mode, uint32_t flags, + unsigned int num, struct dlm_lksb *lksb, char *name) +{ + struct lm_lockstruct *ls = &sdp->sd_lockstruct; + char strname[GDLM_STRNAME_BYTES]; + int error, status; + + memset(strname, 0, GDLM_STRNAME_BYTES); + snprintf(strname, GDLM_STRNAME_BYTES, "%8x%16x", LM_TYPE_NONDISK, num); + + error = dlm_lock(ls->ls_dlm, mode, lksb, flags, + strname, GDLM_STRNAME_BYTES - 1, + 0, sync_wait_cb, ls, NULL); + if (error) { + fs_err(sdp, "%s lkid %x flags %x mode %d error %d\n", + name, lksb->sb_lkid, flags, mode, error); + return error; + } + + wait_for_completion(&ls->ls_sync_wait); + + status = lksb->sb_status; + + if (status && status != -EAGAIN) { + fs_err(sdp, "%s lkid %x flags %x mode %d status %d\n", + name, lksb->sb_lkid, flags, mode, status); + } + + return status; +} + +static int mounted_unlock(struct gfs2_sbd *sdp) +{ + struct lm_lockstruct *ls = &sdp->sd_lockstruct; + return sync_unlock(sdp, &ls->ls_mounted_lksb, "mounted_lock"); +} + +static int mounted_lock(struct gfs2_sbd *sdp, int mode, uint32_t flags) +{ + struct lm_lockstruct *ls = &sdp->sd_lockstruct; + return sync_lock(sdp, mode, flags, GFS2_MOUNTED_LOCK, + &ls->ls_mounted_lksb, "mounted_lock"); +} + +static int control_unlock(struct gfs2_sbd *sdp) +{ + struct lm_lockstruct *ls = &sdp->sd_lockstruct; + return sync_unlock(sdp, &ls->ls_control_lksb, "control_lock"); +} + +static int control_lock(struct gfs2_sbd *sdp, int mode, uint32_t flags) +{ + struct lm_lockstruct *ls = &sdp->sd_lockstruct; + return sync_lock(sdp, mode, flags, GFS2_CONTROL_LOCK, + &ls->ls_control_lksb, "control_lock"); +} + +static void gfs2_control_func(struct work_struct *work) +{ + struct gfs2_sbd *sdp = container_of(work, struct gfs2_sbd, sd_control_work.work); + struct lm_lockstruct *ls = &sdp->sd_lockstruct; + char lvb_bits[GDLM_LVB_SIZE]; + uint32_t block_gen, start_gen, lvb_gen, flags; + int recover_set = 0; + int write_lvb = 0; + int recover_size; + int i, error; + + spin_lock(&ls->ls_recover_spin); + /* + * No MOUNT_DONE means we're still mounting; control_mount() + * will set this flag, after which this thread will take over + * all further clearing of BLOCK_LOCKS. + * + * FIRST_MOUNT means this node is doing first mounter recovery, + * for which recovery control is handled by + * control_mount()/control_first_done(), not this thread. + */ + if (!test_bit(DFL_MOUNT_DONE, &ls->ls_recover_flags) || + test_bit(DFL_FIRST_MOUNT, &ls->ls_recover_flags)) { + spin_unlock(&ls->ls_recover_spin); + return; + } + block_gen = ls->ls_recover_block; + start_gen = ls->ls_recover_start; + spin_unlock(&ls->ls_recover_spin); + + /* + * Equal block_gen and start_gen implies we are between + * recover_prep and recover_done callbacks, which means + * dlm recovery is in progress and dlm locking is blocked. + * There's no point trying to do any work until recover_done. + */ + + if (block_gen == start_gen) + return; + + /* + * Propagate recover_submit[] and recover_result[] to lvb: + * dlm_recoverd adds to recover_submit[] jids needing recovery + * gfs2_recover adds to recover_result[] journal recovery results + * + * set lvb bit for jids in recover_submit[] if the lvb has not + * yet been updated for the generation of the failure + * + * clear lvb bit for jids in recover_result[] if the result of + * the journal recovery is SUCCESS + */ + + error = control_lock(sdp, DLM_LOCK_EX, DLM_LKF_CONVERT|DLM_LKF_VALBLK); + if (error) { + fs_err(sdp, "control lock EX error %d\n", error); + return; + } + + control_lvb_read(ls, &lvb_gen, lvb_bits); + + spin_lock(&ls->ls_recover_spin); + if (block_gen != ls->ls_recover_block || + start_gen != ls->ls_recover_start) { + fs_info(sdp, "recover generation %u block1 %u %u\n", + start_gen, block_gen, ls->ls_recover_block); + spin_unlock(&ls->ls_recover_spin); + control_lock(sdp, DLM_LOCK_NL, DLM_LKF_CONVERT); + return; + } + + recover_size = ls->ls_recover_size; + + if (lvb_gen <= start_gen) { + /* + * Clear lvb bits for jids we've successfully recovered. + * Because all nodes attempt to recover failed journals, + * a journal can be recovered multiple times successfully + * in succession. Only the first will really do recovery, + * the others find it clean, but still report a successful + * recovery. So, another node may have already recovered + * the jid and cleared the lvb bit for it. + */ + for (i = 0; i < recover_size; i++) { + if (ls->ls_recover_result[i] != LM_RD_SUCCESS) + continue; + + ls->ls_recover_result[i] = 0; + + if (!test_bit_le(i, lvb_bits + JID_BITMAP_OFFSET)) + continue; + + __clear_bit_le(i, lvb_bits + JID_BITMAP_OFFSET); + write_lvb = 1; + } + } + + if (lvb_gen == start_gen) { + /* + * Failed slots before start_gen are already set in lvb. + */ + for (i = 0; i < recover_size; i++) { + if (!ls->ls_recover_submit[i]) + continue; + if (ls->ls_recover_submit[i] < lvb_gen) + ls->ls_recover_submit[i] = 0; + } + } else if (lvb_gen < start_gen) { + /* + * Failed slots before start_gen are not yet set in lvb. + */ + for (i = 0; i < recover_size; i++) { + if (!ls->ls_recover_submit[i]) + continue; + if (ls->ls_recover_submit[i] < start_gen) { + ls->ls_recover_submit[i] = 0; + __set_bit_le(i, lvb_bits + JID_BITMAP_OFFSET); + } + } + /* even if there are no bits to set, we need to write the + latest generation to the lvb */ + write_lvb = 1; + } else { + /* + * we should be getting a recover_done() for lvb_gen soon + */ + } + spin_unlock(&ls->ls_recover_spin); + + if (write_lvb) { + control_lvb_write(ls, start_gen, lvb_bits); + flags = DLM_LKF_CONVERT | DLM_LKF_VALBLK; + } else { + flags = DLM_LKF_CONVERT; + } + + error = control_lock(sdp, DLM_LOCK_NL, flags); + if (error) { + fs_err(sdp, "control lock NL error %d\n", error); + return; + } + + /* + * Everyone will see jid bits set in the lvb, run gfs2_recover_set(), + * and clear a jid bit in the lvb if the recovery is a success. + * Eventually all journals will be recovered, all jid bits will + * be cleared in the lvb, and everyone will clear BLOCK_LOCKS. + */ + + for (i = 0; i < recover_size; i++) { + if (test_bit_le(i, lvb_bits + JID_BITMAP_OFFSET)) { + fs_info(sdp, "recover generation %u jid %d\n", + start_gen, i); + gfs2_recover_set(sdp, i); + recover_set++; + } + } + if (recover_set) + return; + + /* + * No more jid bits set in lvb, all recovery is done, unblock locks + * (unless a new recover_prep callback has occured blocking locks + * again while working above) + */ + + spin_lock(&ls->ls_recover_spin); + if (ls->ls_recover_block == block_gen && + ls->ls_recover_start == start_gen) { + clear_bit(DFL_BLOCK_LOCKS, &ls->ls_recover_flags); + spin_unlock(&ls->ls_recover_spin); + fs_info(sdp, "recover generation %u done\n", start_gen); + gfs2_glock_thaw(sdp); + } else { + fs_info(sdp, "recover generation %u block2 %u %u\n", + start_gen, block_gen, ls->ls_recover_block); + spin_unlock(&ls->ls_recover_spin); + } +} + +static int control_mount(struct gfs2_sbd *sdp) +{ + struct lm_lockstruct *ls = &sdp->sd_lockstruct; + char lvb_bits[GDLM_LVB_SIZE]; + uint32_t start_gen, block_gen, mount_gen, lvb_gen; + int mounted_mode; + int retries = 0; + int error; + + memset(&ls->ls_mounted_lksb, 0, sizeof(struct dlm_lksb)); + memset(&ls->ls_control_lksb, 0, sizeof(struct dlm_lksb)); + memset(&ls->ls_control_lvb, 0, GDLM_LVB_SIZE); + ls->ls_control_lksb.sb_lvbptr = ls->ls_control_lvb; + init_completion(&ls->ls_sync_wait); + + set_bit(DFL_BLOCK_LOCKS, &ls->ls_recover_flags); + + error = control_lock(sdp, DLM_LOCK_NL, DLM_LKF_VALBLK); + if (error) { + fs_err(sdp, "control_mount control_lock NL error %d\n", error); + return error; + } + + error = mounted_lock(sdp, DLM_LOCK_NL, 0); + if (error) { + fs_err(sdp, "control_mount mounted_lock NL error %d\n", error); + control_unlock(sdp); + return error; + } + mounted_mode = DLM_LOCK_NL; + +restart: + if (retries++ && signal_pending(current)) { + error = -EINTR; + goto fail; + } + + /* + * We always start with both locks in NL. control_lock is + * demoted to NL below so we don't need to do it here. + */ + + if (mounted_mode != DLM_LOCK_NL) { + error = mounted_lock(sdp, DLM_LOCK_NL, DLM_LKF_CONVERT); + if (error) + goto fail; + mounted_mode = DLM_LOCK_NL; + } + + /* + * Other nodes need to do some work in dlm recovery and gfs2_control + * before the recover_done and control_lock will be ready for us below. + * A delay here is not required but often avoids having to retry. + */ + + msleep_interruptible(500); + + /* + * Acquire control_lock in EX and mounted_lock in either EX or PR. + * control_lock lvb keeps track of any pending journal recoveries. + * mounted_lock indicates if any other nodes have the fs mounted. + */ + + error = control_lock(sdp, DLM_LOCK_EX, DLM_LKF_CONVERT|DLM_LKF_NOQUEUE|DLM_LKF_VALBLK); + if (error == -EAGAIN) { + goto restart; + } else if (error) { + fs_err(sdp, "control_mount control_lock EX error %d\n", error); + goto fail; + } + + error = mounted_lock(sdp, DLM_LOCK_EX, DLM_LKF_CONVERT|DLM_LKF_NOQUEUE); + if (!error) { + mounted_mode = DLM_LOCK_EX; + goto locks_done; + } else if (error != -EAGAIN) { + fs_err(sdp, "control_mount mounted_lock EX error %d\n", error); + goto fail; + } + + error = mounted_lock(sdp, DLM_LOCK_PR, DLM_LKF_CONVERT|DLM_LKF_NOQUEUE); + if (!error) { + mounted_mode = DLM_LOCK_PR; + goto locks_done; + } else { + /* not even -EAGAIN should happen here */ + fs_err(sdp, "control_mount mounted_lock PR error %d\n", error); + goto fail; + } + +locks_done: + /* + * If we got both locks above in EX, then we're the first mounter. + * If not, then we need to wait for the control_lock lvb to be + * updated by other mounted nodes to reflect our mount generation. + * + * In simple first mounter cases, first mounter will see zero lvb_gen, + * but in cases where all existing nodes leave/fail before mounting + * nodes finish control_mount, then all nodes will be mounting and + * lvb_gen will be non-zero. + */ + + control_lvb_read(ls, &lvb_gen, lvb_bits); + + if (lvb_gen == 0xFFFFFFFF) { + /* special value to force mount attempts to fail */ + fs_err(sdp, "control_mount control_lock disabled\n"); + error = -EINVAL; + goto fail; + } + + if (mounted_mode == DLM_LOCK_EX) { + /* first mounter, keep both EX while doing first recovery */ + spin_lock(&ls->ls_recover_spin); + clear_bit(DFL_BLOCK_LOCKS, &ls->ls_recover_flags); + set_bit(DFL_MOUNT_DONE, &ls->ls_recover_flags); + set_bit(DFL_FIRST_MOUNT, &ls->ls_recover_flags); + spin_unlock(&ls->ls_recover_spin); + fs_info(sdp, "first mounter control generation %u\n", lvb_gen); + return 0; + } + + error = control_lock(sdp, DLM_LOCK_NL, DLM_LKF_CONVERT); + if (error) + goto fail; + + /* + * We are not first mounter, now we need to wait for the control_lock + * lvb generation to be >= the generation from our first recover_done + * and all lvb bits to be clear (no pending journal recoveries.) + */ + + if (!all_jid_bits_clear(lvb_bits)) { + /* journals need recovery, wait until all are clear */ + fs_info(sdp, "control_mount wait for journal recovery\n"); + goto restart; + } + + spin_lock(&ls->ls_recover_spin); + block_gen = ls->ls_recover_block; + start_gen = ls->ls_recover_start; + mount_gen = ls->ls_recover_mount; + + if (lvb_gen < mount_gen) { + /* wait for mounted nodes to update control_lock lvb to our + generation, which might include new recovery bits set */ + fs_info(sdp, "control_mount wait1 block %u start %u mount %u " + "lvb %u flags %lx\n", block_gen, start_gen, mount_gen, + lvb_gen, ls->ls_recover_flags); + spin_unlock(&ls->ls_recover_spin); + goto restart; + } + + if (lvb_gen != start_gen) { + /* wait for mounted nodes to update control_lock lvb to the + latest recovery generation */ + fs_info(sdp, "control_mount wait2 block %u start %u mount %u " + "lvb %u flags %lx\n", block_gen, start_gen, mount_gen, + lvb_gen, ls->ls_recover_flags); + spin_unlock(&ls->ls_recover_spin); + goto restart; + } + + if (block_gen == start_gen) { + /* dlm recovery in progress, wait for it to finish */ + fs_info(sdp, "control_mount wait3 block %u start %u mount %u " + "lvb %u flags %lx\n", block_gen, start_gen, mount_gen, + lvb_gen, ls->ls_recover_flags); + spin_unlock(&ls->ls_recover_spin); + goto restart; } - error = dlm_new_lockspace(fsname, strlen(fsname), &ls->ls_dlm, - DLM_LSFL_FS | DLM_LSFL_NEWEXCL | - (ls->ls_nodir ? DLM_LSFL_NODIR : 0), - GDLM_LVB_SIZE); + clear_bit(DFL_BLOCK_LOCKS, &ls->ls_recover_flags); + set_bit(DFL_MOUNT_DONE, &ls->ls_recover_flags); + memset(ls->ls_recover_submit, 0, ls->ls_recover_size*sizeof(uint32_t)); + memset(ls->ls_recover_result, 0, ls->ls_recover_size*sizeof(uint32_t)); + spin_unlock(&ls->ls_recover_spin); + return 0; + +fail: + mounted_unlock(sdp); + control_unlock(sdp); + return error; +} + +static int dlm_recovery_wait(void *word) +{ + schedule(); + return 0; +} + +static int control_first_done(struct gfs2_sbd *sdp) +{ + struct lm_lockstruct *ls = &sdp->sd_lockstruct; + char lvb_bits[GDLM_LVB_SIZE]; + uint32_t start_gen, block_gen; + int error; + +restart: + spin_lock(&ls->ls_recover_spin); + start_gen = ls->ls_recover_start; + block_gen = ls->ls_recover_block; + + if (test_bit(DFL_BLOCK_LOCKS, &ls->ls_recover_flags) || + !test_bit(DFL_MOUNT_DONE, &ls->ls_recover_flags) || + !test_bit(DFL_FIRST_MOUNT, &ls->ls_recover_flags)) { + /* sanity check, should not happen */ + fs_err(sdp, "control_first_done start %u block %u flags %lx\n", + start_gen, block_gen, ls->ls_recover_flags); + spin_unlock(&ls->ls_recover_spin); + control_unlock(sdp); + return -1; + } + + if (start_gen == block_gen) { + /* + * Wait for the end of a dlm recovery cycle to switch from + * first mounter recovery. We can ignore any recover_slot + * callbacks between the recover_prep and next recover_done + * because we are still the first mounter and any failed nodes + * have not fully mounted, so they don't need recovery. + */ + spin_unlock(&ls->ls_recover_spin); + fs_info(sdp, "control_first_done wait gen %u\n", start_gen); + + wait_on_bit(&ls->ls_recover_flags, DFL_DLM_RECOVERY, + dlm_recovery_wait, TASK_UNINTERRUPTIBLE); + goto restart; + } + + clear_bit(DFL_FIRST_MOUNT, &ls->ls_recover_flags); + set_bit(DFL_FIRST_MOUNT_DONE, &ls->ls_recover_flags); + memset(ls->ls_recover_submit, 0, ls->ls_recover_size*sizeof(uint32_t)); + memset(ls->ls_recover_result, 0, ls->ls_recover_size*sizeof(uint32_t)); + spin_unlock(&ls->ls_recover_spin); + + memset(lvb_bits, 0, sizeof(lvb_bits)); + control_lvb_write(ls, start_gen, lvb_bits); + + error = mounted_lock(sdp, DLM_LOCK_PR, DLM_LKF_CONVERT); + if (error) + fs_err(sdp, "control_first_done mounted PR error %d\n", error); + + error = control_lock(sdp, DLM_LOCK_NL, DLM_LKF_CONVERT|DLM_LKF_VALBLK); if (error) - printk(KERN_ERR "dlm_new_lockspace error %d", error); + fs_err(sdp, "control_first_done control NL error %d\n", error); return error; } +/* + * Expand static jid arrays if necessary (by increments of RECOVER_SIZE_INC) + * to accomodate the largest slot number. (NB dlm slot numbers start at 1, + * gfs2 jids start at 0, so jid = slot - 1) + */ + +#define RECOVER_SIZE_INC 16 + +static int set_recover_size(struct gfs2_sbd *sdp, struct dlm_slot *slots, + int num_slots) +{ + struct lm_lockstruct *ls = &sdp->sd_lockstruct; + uint32_t *submit = NULL; + uint32_t *result = NULL; + uint32_t old_size, new_size; + int i, max_jid; + + max_jid = 0; + for (i = 0; i < num_slots; i++) { + if (max_jid < slots[i].slot - 1) + max_jid = slots[i].slot - 1; + } + + old_size = ls->ls_recover_size; + + if (old_size >= max_jid + 1) + return 0; + + new_size = old_size + RECOVER_SIZE_INC; + + submit = kzalloc(new_size * sizeof(uint32_t), GFP_NOFS); + result = kzalloc(new_size * sizeof(uint32_t), GFP_NOFS); + if (!submit || !result) { + kfree(submit); + kfree(result); + return -ENOMEM; + } + + spin_lock(&ls->ls_recover_spin); + memcpy(submit, ls->ls_recover_submit, old_size * sizeof(uint32_t)); + memcpy(result, ls->ls_recover_result, old_size * sizeof(uint32_t)); + kfree(ls->ls_recover_submit); + kfree(ls->ls_recover_result); + ls->ls_recover_submit = submit; + ls->ls_recover_result = result; + ls->ls_recover_size = new_size; + spin_unlock(&ls->ls_recover_spin); + return 0; +} + +static void free_recover_size(struct lm_lockstruct *ls) +{ + kfree(ls->ls_recover_submit); + kfree(ls->ls_recover_result); + ls->ls_recover_submit = NULL; + ls->ls_recover_result = NULL; + ls->ls_recover_size = 0; +} + +/* dlm calls before it does lock recovery */ + +static void gdlm_recover_prep(void *arg) +{ + struct gfs2_sbd *sdp = arg; + struct lm_lockstruct *ls = &sdp->sd_lockstruct; + + spin_lock(&ls->ls_recover_spin); + ls->ls_recover_block = ls->ls_recover_start; + set_bit(DFL_DLM_RECOVERY, &ls->ls_recover_flags); + + if (!test_bit(DFL_MOUNT_DONE, &ls->ls_recover_flags) || + test_bit(DFL_FIRST_MOUNT, &ls->ls_recover_flags)) { + spin_unlock(&ls->ls_recover_spin); + return; + } + set_bit(DFL_BLOCK_LOCKS, &ls->ls_recover_flags); + spin_unlock(&ls->ls_recover_spin); +} + +/* dlm calls after recover_prep has been completed on all lockspace members; + identifies slot/jid of failed member */ + +static void gdlm_recover_slot(void *arg, struct dlm_slot *slot) +{ + struct gfs2_sbd *sdp = arg; + struct lm_lockstruct *ls = &sdp->sd_lockstruct; + int jid = slot->slot - 1; + + spin_lock(&ls->ls_recover_spin); + if (ls->ls_recover_size < jid + 1) { + fs_err(sdp, "recover_slot jid %d gen %u short size %d", + jid, ls->ls_recover_block, ls->ls_recover_size); + spin_unlock(&ls->ls_recover_spin); + return; + } + + if (ls->ls_recover_submit[jid]) { + fs_info(sdp, "recover_slot jid %d gen %u prev %u", + jid, ls->ls_recover_block, ls->ls_recover_submit[jid]); + } + ls->ls_recover_submit[jid] = ls->ls_recover_block; + spin_unlock(&ls->ls_recover_spin); +} + +/* dlm calls after recover_slot and after it completes lock recovery */ + +static void gdlm_recover_done(void *arg, struct dlm_slot *slots, int num_slots, + int our_slot, uint32_t generation) +{ + struct gfs2_sbd *sdp = arg; + struct lm_lockstruct *ls = &sdp->sd_lockstruct; + + /* ensure the ls jid arrays are large enough */ + set_recover_size(sdp, slots, num_slots); + + spin_lock(&ls->ls_recover_spin); + ls->ls_recover_start = generation; + + if (!ls->ls_recover_mount) { + ls->ls_recover_mount = generation; + ls->ls_jid = our_slot - 1; + } + + if (!test_bit(DFL_UNMOUNT, &ls->ls_recover_flags)) + queue_delayed_work(gfs2_control_wq, &sdp->sd_control_work, 0); + + clear_bit(DFL_DLM_RECOVERY, &ls->ls_recover_flags); + smp_mb__after_clear_bit(); + wake_up_bit(&ls->ls_recover_flags, DFL_DLM_RECOVERY); + spin_unlock(&ls->ls_recover_spin); +} + +/* gfs2_recover thread has a journal recovery result */ + +static void gdlm_recovery_result(struct gfs2_sbd *sdp, unsigned int jid, + unsigned int result) +{ + struct lm_lockstruct *ls = &sdp->sd_lockstruct; + + if (test_bit(DFL_NO_DLM_OPS, &ls->ls_recover_flags)) + return; + + /* don't care about the recovery of own journal during mount */ + if (jid == ls->ls_jid) + return; + + spin_lock(&ls->ls_recover_spin); + if (test_bit(DFL_FIRST_MOUNT, &ls->ls_recover_flags)) { + spin_unlock(&ls->ls_recover_spin); + return; + } + if (ls->ls_recover_size < jid + 1) { + fs_err(sdp, "recovery_result jid %d short size %d", + jid, ls->ls_recover_size); + spin_unlock(&ls->ls_recover_spin); + return; + } + + fs_info(sdp, "recover jid %d result %s\n", jid, + result == LM_RD_GAVEUP ? "busy" : "success"); + + ls->ls_recover_result[jid] = result; + + /* GAVEUP means another node is recovering the journal; delay our + next attempt to recover it, to give the other node a chance to + finish before trying again */ + + if (!test_bit(DFL_UNMOUNT, &ls->ls_recover_flags)) + queue_delayed_work(gfs2_control_wq, &sdp->sd_control_work, + result == LM_RD_GAVEUP ? HZ : 0); + spin_unlock(&ls->ls_recover_spin); +} + +const struct dlm_lockspace_ops gdlm_lockspace_ops = { + .recover_prep = gdlm_recover_prep, + .recover_slot = gdlm_recover_slot, + .recover_done = gdlm_recover_done, +}; + +static int gdlm_mount(struct gfs2_sbd *sdp, const char *table) +{ + struct lm_lockstruct *ls = &sdp->sd_lockstruct; + char cluster[GFS2_LOCKNAME_LEN]; + const char *fsname; + uint32_t flags; + int error, ops_result; + + /* + * initialize everything + */ + + INIT_DELAYED_WORK(&sdp->sd_control_work, gfs2_control_func); + spin_lock_init(&ls->ls_recover_spin); + ls->ls_recover_flags = 0; + ls->ls_recover_mount = 0; + ls->ls_recover_start = 0; + ls->ls_recover_block = 0; + ls->ls_recover_size = 0; + ls->ls_recover_submit = NULL; + ls->ls_recover_result = NULL; + + error = set_recover_size(sdp, NULL, 0); + if (error) + goto fail; + + /* + * prepare dlm_new_lockspace args + */ + + fsname = strchr(table, ':'); + if (!fsname) { + fs_info(sdp, "no fsname found\n"); + error = -EINVAL; + goto fail_free; + } + memset(cluster, 0, sizeof(cluster)); + memcpy(cluster, table, strlen(table) - strlen(fsname)); + fsname++; + + flags = DLM_LSFL_FS | DLM_LSFL_NEWEXCL; + if (ls->ls_nodir) + flags |= DLM_LSFL_NODIR; + + /* + * create/join lockspace + */ + + error = dlm_new_lockspace(fsname, cluster, flags, GDLM_LVB_SIZE, + &gdlm_lockspace_ops, sdp, &ops_result, + &ls->ls_dlm); + if (error) { + fs_err(sdp, "dlm_new_lockspace error %d\n", error); + goto fail_free; + } + + if (ops_result < 0) { + /* + * dlm does not support ops callbacks, + * old dlm_controld/gfs_controld are used, try without ops. + */ + fs_info(sdp, "dlm lockspace ops not used\n"); + free_recover_size(ls); + set_bit(DFL_NO_DLM_OPS, &ls->ls_recover_flags); + return 0; + } + + if (!test_bit(SDF_NOJOURNALID, &sdp->sd_flags)) { + fs_err(sdp, "dlm lockspace ops disallow jid preset\n"); + error = -EINVAL; + goto fail_release; + } + + /* + * control_mount() uses control_lock to determine first mounter, + * and for later mounts, waits for any recoveries to be cleared. + */ + + error = control_mount(sdp); + if (error) { + fs_err(sdp, "mount control error %d\n", error); + goto fail_release; + } + + ls->ls_first = !!test_bit(DFL_FIRST_MOUNT, &ls->ls_recover_flags); + clear_bit(SDF_NOJOURNALID, &sdp->sd_flags); + smp_mb__after_clear_bit(); + wake_up_bit(&sdp->sd_flags, SDF_NOJOURNALID); + return 0; + +fail_release: + dlm_release_lockspace(ls->ls_dlm, 2); +fail_free: + free_recover_size(ls); +fail: + return error; +} + +static void gdlm_first_done(struct gfs2_sbd *sdp) +{ + struct lm_lockstruct *ls = &sdp->sd_lockstruct; + int error; + + if (test_bit(DFL_NO_DLM_OPS, &ls->ls_recover_flags)) + return; + + error = control_first_done(sdp); + if (error) + fs_err(sdp, "mount first_done error %d\n", error); +} + static void gdlm_unmount(struct gfs2_sbd *sdp) { struct lm_lockstruct *ls = &sdp->sd_lockstruct; + if (test_bit(DFL_NO_DLM_OPS, &ls->ls_recover_flags)) + goto release; + + /* wait for gfs2_control_wq to be done with this mount */ + + spin_lock(&ls->ls_recover_spin); + set_bit(DFL_UNMOUNT, &ls->ls_recover_flags); + spin_unlock(&ls->ls_recover_spin); + flush_delayed_work_sync(&sdp->sd_control_work); + + /* mounted_lock and control_lock will be purged in dlm recovery */ +release: if (ls->ls_dlm) { dlm_release_lockspace(ls->ls_dlm, 2); ls->ls_dlm = NULL; } + + free_recover_size(ls); } static const match_table_t dlm_tokens = { @@ -226,6 +1197,8 @@ static const match_table_t dlm_tokens = { const struct lm_lockops gfs2_dlm_ops = { .lm_proto_name = "lock_dlm", .lm_mount = gdlm_mount, + .lm_first_done = gdlm_first_done, + .lm_recovery_result = gdlm_recovery_result, .lm_unmount = gdlm_unmount, .lm_put_lock = gdlm_put_lock, .lm_lock = gdlm_lock, diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index 598646434362..756fae9eaf8f 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -626,7 +626,7 @@ static void log_write_header(struct gfs2_sbd *sdp, u32 flags, int pull) if (test_bit(SDF_NOBARRIERS, &sdp->sd_flags)) submit_bh(WRITE_SYNC | REQ_META | REQ_PRIO, bh); else - submit_bh(WRITE_FLUSH_FUA | REQ_META | REQ_PRIO, bh); + submit_bh(WRITE_FLUSH_FUA | REQ_META, bh); wait_on_buffer(bh); if (!buffer_uptodate(bh)) @@ -951,8 +951,8 @@ int gfs2_logd(void *data) wake_up(&sdp->sd_log_waitq); t = gfs2_tune_get(sdp, gt_logd_secs) * HZ; - if (freezing(current)) - refrigerator(); + + try_to_freeze(); do { prepare_to_wait(&sdp->sd_logd_waitq, &wait, diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c index 8a139ff1919f..a8d9bcd0e19c 100644 --- a/fs/gfs2/main.c +++ b/fs/gfs2/main.c @@ -28,6 +28,8 @@ #include "recovery.h" #include "dir.h" +struct workqueue_struct *gfs2_control_wq; + static struct shrinker qd_shrinker = { .shrink = gfs2_shrink_qd_memory, .seeks = DEFAULT_SEEKS, @@ -40,7 +42,8 @@ static void gfs2_init_inode_once(void *foo) inode_init_once(&ip->i_inode); init_rwsem(&ip->i_rw_mutex); INIT_LIST_HEAD(&ip->i_trunc_list); - ip->i_alloc = NULL; + ip->i_qadata = NULL; + ip->i_res = NULL; ip->i_hash_cache = NULL; } @@ -145,12 +148,19 @@ static int __init init_gfs2_fs(void) if (!gfs_recovery_wq) goto fail_wq; + gfs2_control_wq = alloc_workqueue("gfs2_control", + WQ_NON_REENTRANT | WQ_UNBOUND | WQ_FREEZABLE, 0); + if (!gfs2_control_wq) + goto fail_control; + gfs2_register_debugfs(); printk("GFS2 installed\n"); return 0; +fail_control: + destroy_workqueue(gfs_recovery_wq); fail_wq: unregister_filesystem(&gfs2meta_fs_type); fail_unregister: @@ -194,6 +204,7 @@ static void __exit exit_gfs2_fs(void) unregister_filesystem(&gfs2_fs_type); unregister_filesystem(&gfs2meta_fs_type); destroy_workqueue(gfs_recovery_wq); + destroy_workqueue(gfs2_control_wq); rcu_barrier(); diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c index be29858900f6..181586e673f9 100644 --- a/fs/gfs2/meta_io.c +++ b/fs/gfs2/meta_io.c @@ -435,7 +435,7 @@ struct buffer_head *gfs2_meta_ra(struct gfs2_glock *gl, u64 dblock, u32 extlen) if (buffer_uptodate(first_bh)) goto out; if (!buffer_locked(first_bh)) - ll_rw_block(READ_SYNC | REQ_META | REQ_PRIO, 1, &first_bh); + ll_rw_block(READ_SYNC | REQ_META, 1, &first_bh); dblock++; extlen--; @@ -444,7 +444,7 @@ struct buffer_head *gfs2_meta_ra(struct gfs2_glock *gl, u64 dblock, u32 extlen) bh = gfs2_getbuf(gl, dblock, CREATE); if (!buffer_uptodate(bh) && !buffer_locked(bh)) - ll_rw_block(READA, 1, &bh); + ll_rw_block(READA | REQ_META, 1, &bh); brelse(bh); dblock++; extlen--; diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index cb23c2be731a..6aacf3f230a2 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -224,7 +224,7 @@ static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector, int silent) bio->bi_end_io = end_bio_io_page; bio->bi_private = page; - submit_bio(READ_SYNC | REQ_META | REQ_PRIO, bio); + submit_bio(READ_SYNC | REQ_META, bio); wait_on_page_locked(page); bio_put(bio); if (!PageUptodate(page)) { @@ -562,8 +562,12 @@ static void gfs2_others_may_mount(struct gfs2_sbd *sdp) { char *message = "FIRSTMOUNT=Done"; char *envp[] = { message, NULL }; - struct lm_lockstruct *ls = &sdp->sd_lockstruct; - ls->ls_first_done = 1; + + fs_info(sdp, "first mount done, others may mount\n"); + + if (sdp->sd_lockstruct.ls_ops->lm_first_done) + sdp->sd_lockstruct.ls_ops->lm_first_done(sdp); + kobject_uevent_env(&sdp->sd_kobj, KOBJ_CHANGE, envp); } @@ -944,7 +948,6 @@ static int gfs2_lm_mount(struct gfs2_sbd *sdp, int silent) struct gfs2_args *args = &sdp->sd_args; const char *proto = sdp->sd_proto_name; const char *table = sdp->sd_table_name; - const char *fsname; char *o, *options; int ret; @@ -1004,21 +1007,12 @@ hostdata_error: } } - if (sdp->sd_args.ar_spectator) - snprintf(sdp->sd_fsname, GFS2_FSNAME_LEN, "%s.s", table); - else - snprintf(sdp->sd_fsname, GFS2_FSNAME_LEN, "%s.%u", table, - sdp->sd_lockstruct.ls_jid); - - fsname = strchr(table, ':'); - if (fsname) - fsname++; if (lm->lm_mount == NULL) { fs_info(sdp, "Now mounting FS...\n"); complete_all(&sdp->sd_locking_init); return 0; } - ret = lm->lm_mount(sdp, fsname); + ret = lm->lm_mount(sdp, table); if (ret == 0) fs_info(sdp, "Joined cluster. Now mounting FS...\n"); complete_all(&sdp->sd_locking_init); @@ -1084,7 +1078,7 @@ static int fill_super(struct super_block *sb, struct gfs2_args *args, int silent if (sdp->sd_args.ar_spectator) { sb->s_flags |= MS_RDONLY; - set_bit(SDF_NORECOVERY, &sdp->sd_flags); + set_bit(SDF_RORECOVERY, &sdp->sd_flags); } if (sdp->sd_args.ar_posix_acl) sb->s_flags |= MS_POSIXACL; @@ -1124,6 +1118,8 @@ static int fill_super(struct super_block *sb, struct gfs2_args *args, int silent if (error) goto fail; + snprintf(sdp->sd_fsname, GFS2_FSNAME_LEN, "%s", sdp->sd_table_name); + gfs2_create_debugfs_file(sdp); error = gfs2_sys_fs_add(sdp); @@ -1160,6 +1156,13 @@ static int fill_super(struct super_block *sb, struct gfs2_args *args, int silent goto fail_sb; } + if (sdp->sd_args.ar_spectator) + snprintf(sdp->sd_fsname, GFS2_FSNAME_LEN, "%s.s", + sdp->sd_table_name); + else + snprintf(sdp->sd_fsname, GFS2_FSNAME_LEN, "%s.%u", + sdp->sd_table_name, sdp->sd_lockstruct.ls_jid); + error = init_inodes(sdp, DO); if (error) goto fail_sb; diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index 7e528dc14f85..a45b21b03915 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -494,11 +494,11 @@ static void qdsb_put(struct gfs2_quota_data *qd) int gfs2_quota_hold(struct gfs2_inode *ip, u32 uid, u32 gid) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al = ip->i_alloc; - struct gfs2_quota_data **qd = al->al_qd; + struct gfs2_qadata *qa = ip->i_qadata; + struct gfs2_quota_data **qd = qa->qa_qd; int error; - if (gfs2_assert_warn(sdp, !al->al_qd_num) || + if (gfs2_assert_warn(sdp, !qa->qa_qd_num) || gfs2_assert_warn(sdp, !test_bit(GIF_QD_LOCKED, &ip->i_flags))) return -EIO; @@ -508,20 +508,20 @@ int gfs2_quota_hold(struct gfs2_inode *ip, u32 uid, u32 gid) error = qdsb_get(sdp, QUOTA_USER, ip->i_inode.i_uid, qd); if (error) goto out; - al->al_qd_num++; + qa->qa_qd_num++; qd++; error = qdsb_get(sdp, QUOTA_GROUP, ip->i_inode.i_gid, qd); if (error) goto out; - al->al_qd_num++; + qa->qa_qd_num++; qd++; if (uid != NO_QUOTA_CHANGE && uid != ip->i_inode.i_uid) { error = qdsb_get(sdp, QUOTA_USER, uid, qd); if (error) goto out; - al->al_qd_num++; + qa->qa_qd_num++; qd++; } @@ -529,7 +529,7 @@ int gfs2_quota_hold(struct gfs2_inode *ip, u32 uid, u32 gid) error = qdsb_get(sdp, QUOTA_GROUP, gid, qd); if (error) goto out; - al->al_qd_num++; + qa->qa_qd_num++; qd++; } @@ -542,16 +542,16 @@ out: void gfs2_quota_unhold(struct gfs2_inode *ip) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al = ip->i_alloc; + struct gfs2_qadata *qa = ip->i_qadata; unsigned int x; gfs2_assert_warn(sdp, !test_bit(GIF_QD_LOCKED, &ip->i_flags)); - for (x = 0; x < al->al_qd_num; x++) { - qdsb_put(al->al_qd[x]); - al->al_qd[x] = NULL; + for (x = 0; x < qa->qa_qd_num; x++) { + qdsb_put(qa->qa_qd[x]); + qa->qa_qd[x] = NULL; } - al->al_qd_num = 0; + qa->qa_qd_num = 0; } static int sort_qd(const void *a, const void *b) @@ -712,7 +712,7 @@ get_a_page: set_buffer_uptodate(bh); if (!buffer_uptodate(bh)) { - ll_rw_block(READ | REQ_META | REQ_PRIO, 1, &bh); + ll_rw_block(READ | REQ_META, 1, &bh); wait_on_buffer(bh); if (!buffer_uptodate(bh)) goto unlock_out; @@ -762,7 +762,6 @@ static int do_sync(unsigned int num_qd, struct gfs2_quota_data **qda) struct gfs2_quota_data *qd; loff_t offset; unsigned int nalloc = 0, blocks; - struct gfs2_alloc *al = NULL; int error; gfs2_write_calc_reserv(ip, sizeof(struct gfs2_quota), @@ -792,26 +791,19 @@ static int do_sync(unsigned int num_qd, struct gfs2_quota_data **qda) nalloc++; } - al = gfs2_alloc_get(ip); - if (!al) { - error = -ENOMEM; - goto out_gunlock; - } /* * 1 blk for unstuffing inode if stuffed. We add this extra * block to the reservation unconditionally. If the inode * doesn't need unstuffing, the block will be released to the * rgrp since it won't be allocated during the transaction */ - al->al_requested = 1; /* +3 in the end for unstuffing block, inode size update block * and another block in case quota straddles page boundary and * two blocks need to be updated instead of 1 */ blocks = num_qd * data_blocks + RES_DINODE + num_qd + 3; - if (nalloc) - al->al_requested += nalloc * (data_blocks + ind_blocks); - error = gfs2_inplace_reserve(ip); + error = gfs2_inplace_reserve(ip, 1 + + (nalloc * (data_blocks + ind_blocks))); if (error) goto out_alloc; @@ -840,8 +832,6 @@ out_end_trans: out_ipres: gfs2_inplace_release(ip); out_alloc: - gfs2_alloc_put(ip); -out_gunlock: gfs2_glock_dq_uninit(&i_gh); out: while (qx--) @@ -925,7 +915,7 @@ fail: int gfs2_quota_lock(struct gfs2_inode *ip, u32 uid, u32 gid) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al = ip->i_alloc; + struct gfs2_qadata *qa = ip->i_qadata; struct gfs2_quota_data *qd; unsigned int x; int error = 0; @@ -938,15 +928,15 @@ int gfs2_quota_lock(struct gfs2_inode *ip, u32 uid, u32 gid) sdp->sd_args.ar_quota != GFS2_QUOTA_ON) return 0; - sort(al->al_qd, al->al_qd_num, sizeof(struct gfs2_quota_data *), + sort(qa->qa_qd, qa->qa_qd_num, sizeof(struct gfs2_quota_data *), sort_qd, NULL); - for (x = 0; x < al->al_qd_num; x++) { + for (x = 0; x < qa->qa_qd_num; x++) { int force = NO_FORCE; - qd = al->al_qd[x]; + qd = qa->qa_qd[x]; if (test_and_clear_bit(QDF_REFRESH, &qd->qd_flags)) force = FORCE; - error = do_glock(qd, force, &al->al_qd_ghs[x]); + error = do_glock(qd, force, &qa->qa_qd_ghs[x]); if (error) break; } @@ -955,7 +945,7 @@ int gfs2_quota_lock(struct gfs2_inode *ip, u32 uid, u32 gid) set_bit(GIF_QD_LOCKED, &ip->i_flags); else { while (x--) - gfs2_glock_dq_uninit(&al->al_qd_ghs[x]); + gfs2_glock_dq_uninit(&qa->qa_qd_ghs[x]); gfs2_quota_unhold(ip); } @@ -1000,7 +990,7 @@ static int need_sync(struct gfs2_quota_data *qd) void gfs2_quota_unlock(struct gfs2_inode *ip) { - struct gfs2_alloc *al = ip->i_alloc; + struct gfs2_qadata *qa = ip->i_qadata; struct gfs2_quota_data *qda[4]; unsigned int count = 0; unsigned int x; @@ -1008,14 +998,14 @@ void gfs2_quota_unlock(struct gfs2_inode *ip) if (!test_and_clear_bit(GIF_QD_LOCKED, &ip->i_flags)) goto out; - for (x = 0; x < al->al_qd_num; x++) { + for (x = 0; x < qa->qa_qd_num; x++) { struct gfs2_quota_data *qd; int sync; - qd = al->al_qd[x]; + qd = qa->qa_qd[x]; sync = need_sync(qd); - gfs2_glock_dq_uninit(&al->al_qd_ghs[x]); + gfs2_glock_dq_uninit(&qa->qa_qd_ghs[x]); if (sync && qd_trylock(qd)) qda[count++] = qd; @@ -1048,7 +1038,7 @@ static int print_message(struct gfs2_quota_data *qd, char *type) int gfs2_quota_check(struct gfs2_inode *ip, u32 uid, u32 gid) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al = ip->i_alloc; + struct gfs2_qadata *qa = ip->i_qadata; struct gfs2_quota_data *qd; s64 value; unsigned int x; @@ -1060,8 +1050,8 @@ int gfs2_quota_check(struct gfs2_inode *ip, u32 uid, u32 gid) if (sdp->sd_args.ar_quota != GFS2_QUOTA_ON) return 0; - for (x = 0; x < al->al_qd_num; x++) { - qd = al->al_qd[x]; + for (x = 0; x < qa->qa_qd_num; x++) { + qd = qa->qa_qd[x]; if (!((qd->qd_id == uid && test_bit(QDF_USER, &qd->qd_flags)) || (qd->qd_id == gid && !test_bit(QDF_USER, &qd->qd_flags)))) @@ -1099,7 +1089,7 @@ int gfs2_quota_check(struct gfs2_inode *ip, u32 uid, u32 gid) void gfs2_quota_change(struct gfs2_inode *ip, s64 change, u32 uid, u32 gid) { - struct gfs2_alloc *al = ip->i_alloc; + struct gfs2_qadata *qa = ip->i_qadata; struct gfs2_quota_data *qd; unsigned int x; @@ -1108,8 +1098,8 @@ void gfs2_quota_change(struct gfs2_inode *ip, s64 change, if (ip->i_diskflags & GFS2_DIF_SYSTEM) return; - for (x = 0; x < al->al_qd_num; x++) { - qd = al->al_qd[x]; + for (x = 0; x < qa->qa_qd_num; x++) { + qd = qa->qa_qd[x]; if ((qd->qd_id == uid && test_bit(QDF_USER, &qd->qd_flags)) || (qd->qd_id == gid && !test_bit(QDF_USER, &qd->qd_flags))) { @@ -1427,8 +1417,8 @@ int gfs2_quotad(void *data) /* Check for & recover partially truncated inodes */ quotad_check_trunc_list(sdp); - if (freezing(current)) - refrigerator(); + try_to_freeze(); + t = min(quotad_timeo, statfs_timeo); prepare_to_wait(&sdp->sd_quota_wait, &wait, TASK_INTERRUPTIBLE); @@ -1529,7 +1519,6 @@ static int gfs2_set_dqblk(struct super_block *sb, int type, qid_t id, unsigned int data_blocks, ind_blocks; unsigned int blocks = 0; int alloc_required; - struct gfs2_alloc *al; loff_t offset; int error; @@ -1594,15 +1583,12 @@ static int gfs2_set_dqblk(struct super_block *sb, int type, qid_t id, if (gfs2_is_stuffed(ip)) alloc_required = 1; if (alloc_required) { - al = gfs2_alloc_get(ip); - if (al == NULL) - goto out_i; gfs2_write_calc_reserv(ip, sizeof(struct gfs2_quota), &data_blocks, &ind_blocks); - blocks = al->al_requested = 1 + data_blocks + ind_blocks; - error = gfs2_inplace_reserve(ip); + blocks = 1 + data_blocks + ind_blocks; + error = gfs2_inplace_reserve(ip, blocks); if (error) - goto out_alloc; + goto out_i; blocks += gfs2_rg_blocks(ip); } @@ -1617,11 +1603,8 @@ static int gfs2_set_dqblk(struct super_block *sb, int type, qid_t id, gfs2_trans_end(sdp); out_release: - if (alloc_required) { + if (alloc_required) gfs2_inplace_release(ip); -out_alloc: - gfs2_alloc_put(ip); - } out_i: gfs2_glock_dq_uninit(&i_gh); out_q: diff --git a/fs/gfs2/recovery.c b/fs/gfs2/recovery.c index f2a02edcac8f..963b2d75200c 100644 --- a/fs/gfs2/recovery.c +++ b/fs/gfs2/recovery.c @@ -436,12 +436,16 @@ static void gfs2_recovery_done(struct gfs2_sbd *sdp, unsigned int jid, char env_status[20]; char *envp[] = { env_jid, env_status, NULL }; struct lm_lockstruct *ls = &sdp->sd_lockstruct; + ls->ls_recover_jid_done = jid; ls->ls_recover_jid_status = message; sprintf(env_jid, "JID=%d", jid); sprintf(env_status, "RECOVERY=%s", message == LM_RD_SUCCESS ? "Done" : "Failed"); kobject_uevent_env(&sdp->sd_kobj, KOBJ_CHANGE, envp); + + if (sdp->sd_lockstruct.ls_ops->lm_recovery_result) + sdp->sd_lockstruct.ls_ops->lm_recovery_result(sdp, jid, message); } void gfs2_recover_func(struct work_struct *work) @@ -512,7 +516,9 @@ void gfs2_recover_func(struct work_struct *work) if (error) goto fail_gunlock_ji; - if (test_bit(SDF_JOURNAL_CHECKED, &sdp->sd_flags)) { + if (test_bit(SDF_RORECOVERY, &sdp->sd_flags)) { + ro = 1; + } else if (test_bit(SDF_JOURNAL_CHECKED, &sdp->sd_flags)) { if (!test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags)) ro = 1; } else { @@ -577,6 +583,7 @@ fail_gunlock_j: fs_info(sdp, "jid=%u: %s\n", jd->jd_jid, (error) ? "Failed" : "Done"); fail: + jd->jd_recover_error = error; gfs2_recovery_done(sdp, jd->jd_jid, LM_RD_GAVEUP); done: clear_bit(JDF_RECOVERY, &jd->jd_flags); @@ -605,6 +612,6 @@ int gfs2_recover_journal(struct gfs2_jdesc *jd, bool wait) wait_on_bit(&jd->jd_flags, JDF_RECOVERY, gfs2_recovery_wait, TASK_UNINTERRUPTIBLE); - return 0; + return wait ? jd->jd_recover_error : 0; } diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index 96bd6d759f29..981bfa32121a 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -65,8 +65,8 @@ static const char valid_change[16] = { }; static u32 rgblk_search(struct gfs2_rgrpd *rgd, u32 goal, - unsigned char old_state, unsigned char new_state, - unsigned int *n); + unsigned char old_state, + struct gfs2_bitmap **rbi); /** * gfs2_setbit - Set a bit in the bitmaps @@ -860,22 +860,36 @@ fail: } /** - * gfs2_alloc_get - get the struct gfs2_alloc structure for an inode + * gfs2_qadata_get - get the struct gfs2_qadata structure for an inode * @ip: the incore GFS2 inode structure * - * Returns: the struct gfs2_alloc + * Returns: the struct gfs2_qadata */ -struct gfs2_alloc *gfs2_alloc_get(struct gfs2_inode *ip) +struct gfs2_qadata *gfs2_qadata_get(struct gfs2_inode *ip) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); int error; - BUG_ON(ip->i_alloc != NULL); - ip->i_alloc = kzalloc(sizeof(struct gfs2_alloc), GFP_NOFS); + BUG_ON(ip->i_qadata != NULL); + ip->i_qadata = kzalloc(sizeof(struct gfs2_qadata), GFP_NOFS); error = gfs2_rindex_update(sdp); if (error) fs_warn(sdp, "rindex update returns %d\n", error); - return ip->i_alloc; + return ip->i_qadata; +} + +/** + * gfs2_blkrsv_get - get the struct gfs2_blkreserv structure for an inode + * @ip: the incore GFS2 inode structure + * + * Returns: the struct gfs2_qadata + */ + +static struct gfs2_blkreserv *gfs2_blkrsv_get(struct gfs2_inode *ip) +{ + BUG_ON(ip->i_res != NULL); + ip->i_res = kzalloc(sizeof(struct gfs2_blkreserv), GFP_NOFS); + return ip->i_res; } /** @@ -890,15 +904,20 @@ struct gfs2_alloc *gfs2_alloc_get(struct gfs2_inode *ip) static int try_rgrp_fit(const struct gfs2_rgrpd *rgd, const struct gfs2_inode *ip) { - const struct gfs2_alloc *al = ip->i_alloc; + const struct gfs2_blkreserv *rs = ip->i_res; if (rgd->rd_flags & (GFS2_RGF_NOALLOC | GFS2_RDF_ERROR)) return 0; - if (rgd->rd_free_clone >= al->al_requested) + if (rgd->rd_free_clone >= rs->rs_requested) return 1; return 0; } +static inline u32 gfs2_bi2rgd_blk(struct gfs2_bitmap *bi, u32 blk) +{ + return (bi->bi_start * GFS2_NBBY) + blk; +} + /** * try_rgrp_unlink - Look for any unlinked, allocated, but unused inodes * @rgd: The rgrp @@ -912,20 +931,20 @@ static void try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, u64 skip u32 goal = 0, block; u64 no_addr; struct gfs2_sbd *sdp = rgd->rd_sbd; - unsigned int n; struct gfs2_glock *gl; struct gfs2_inode *ip; int error; int found = 0; + struct gfs2_bitmap *bi; while (goal < rgd->rd_data) { down_write(&sdp->sd_log_flush_lock); - n = 1; - block = rgblk_search(rgd, goal, GFS2_BLKST_UNLINKED, - GFS2_BLKST_UNLINKED, &n); + block = rgblk_search(rgd, goal, GFS2_BLKST_UNLINKED, &bi); up_write(&sdp->sd_log_flush_lock); if (block == BFITNOENT) break; + + block = gfs2_bi2rgd_blk(bi, block); /* rgblk_search can return a block < goal, so we need to keep it marching forward. */ no_addr = block + rgd->rd_data0; @@ -977,8 +996,8 @@ static int get_local_rgrp(struct gfs2_inode *ip, u64 *last_unlinked) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct gfs2_rgrpd *rgd, *begin = NULL; - struct gfs2_alloc *al = ip->i_alloc; - int error, rg_locked; + struct gfs2_blkreserv *rs = ip->i_res; + int error, rg_locked, flags = LM_FLAG_TRY; int loops = 0; if (ip->i_rgd && rgrp_contains_block(ip->i_rgd, ip->i_goal)) @@ -997,7 +1016,7 @@ static int get_local_rgrp(struct gfs2_inode *ip, u64 *last_unlinked) error = 0; } else { error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, - LM_FLAG_TRY, &al->al_rgd_gh); + flags, &rs->rs_rgd_gh); } switch (error) { case 0: @@ -1008,12 +1027,14 @@ static int get_local_rgrp(struct gfs2_inode *ip, u64 *last_unlinked) if (rgd->rd_flags & GFS2_RDF_CHECK) try_rgrp_unlink(rgd, last_unlinked, ip->i_no_addr); if (!rg_locked) - gfs2_glock_dq_uninit(&al->al_rgd_gh); + gfs2_glock_dq_uninit(&rs->rs_rgd_gh); /* fall through */ case GLR_TRYFAILED: rgd = gfs2_rgrpd_get_next(rgd); - if (rgd == begin) + if (rgd == begin) { + flags = 0; loops++; + } break; default: return error; @@ -1023,6 +1044,13 @@ static int get_local_rgrp(struct gfs2_inode *ip, u64 *last_unlinked) return -ENOSPC; } +static void gfs2_blkrsv_put(struct gfs2_inode *ip) +{ + BUG_ON(ip->i_res == NULL); + kfree(ip->i_res); + ip->i_res = NULL; +} + /** * gfs2_inplace_reserve - Reserve space in the filesystem * @ip: the inode to reserve space for @@ -1030,16 +1058,23 @@ static int get_local_rgrp(struct gfs2_inode *ip, u64 *last_unlinked) * Returns: errno */ -int gfs2_inplace_reserve(struct gfs2_inode *ip) +int gfs2_inplace_reserve(struct gfs2_inode *ip, u32 requested) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al = ip->i_alloc; + struct gfs2_blkreserv *rs; int error = 0; u64 last_unlinked = NO_BLOCK; int tries = 0; - if (gfs2_assert_warn(sdp, al->al_requested)) - return -EINVAL; + rs = gfs2_blkrsv_get(ip); + if (!rs) + return -ENOMEM; + + rs->rs_requested = requested; + if (gfs2_assert_warn(sdp, requested)) { + error = -EINVAL; + goto out; + } do { error = get_local_rgrp(ip, &last_unlinked); @@ -1056,6 +1091,9 @@ int gfs2_inplace_reserve(struct gfs2_inode *ip) gfs2_log_flush(sdp, NULL); } while (tries++ < 3); +out: + if (error) + gfs2_blkrsv_put(ip); return error; } @@ -1068,10 +1106,11 @@ int gfs2_inplace_reserve(struct gfs2_inode *ip) void gfs2_inplace_release(struct gfs2_inode *ip) { - struct gfs2_alloc *al = ip->i_alloc; + struct gfs2_blkreserv *rs = ip->i_res; - if (al->al_rgd_gh.gh_gl) - gfs2_glock_dq_uninit(&al->al_rgd_gh); + if (rs->rs_rgd_gh.gh_gl) + gfs2_glock_dq_uninit(&rs->rs_rgd_gh); + gfs2_blkrsv_put(ip); } /** @@ -1108,39 +1147,35 @@ static unsigned char gfs2_get_block_type(struct gfs2_rgrpd *rgd, u64 block) } /** - * rgblk_search - find a block in @old_state, change allocation - * state to @new_state + * rgblk_search - find a block in @state * @rgd: the resource group descriptor * @goal: the goal block within the RG (start here to search for avail block) - * @old_state: GFS2_BLKST_XXX the before-allocation state to find - * @new_state: GFS2_BLKST_XXX the after-allocation block state - * @n: The extent length + * @state: GFS2_BLKST_XXX the before-allocation state to find + * @dinode: TRUE if the first block we allocate is for a dinode + * @rbi: address of the pointer to the bitmap containing the block found * - * Walk rgrp's bitmap to find bits that represent a block in @old_state. - * Add the found bitmap buffer to the transaction. - * Set the found bits to @new_state to change block's allocation state. + * Walk rgrp's bitmap to find bits that represent a block in @state. * * This function never fails, because we wouldn't call it unless we * know (from reservation results, etc.) that a block is available. * - * Scope of @goal and returned block is just within rgrp, not the whole - * filesystem. + * Scope of @goal is just within rgrp, not the whole filesystem. + * Scope of @returned block is just within bitmap, not the whole filesystem. * - * Returns: the block number allocated + * Returns: the block number found relative to the bitmap rbi */ static u32 rgblk_search(struct gfs2_rgrpd *rgd, u32 goal, - unsigned char old_state, unsigned char new_state, - unsigned int *n) + unsigned char state, + struct gfs2_bitmap **rbi) { struct gfs2_bitmap *bi = NULL; const u32 length = rgd->rd_length; u32 blk = BFITNOENT; unsigned int buf, x; - const unsigned int elen = *n; const u8 *buffer = NULL; - *n = 0; + *rbi = NULL; /* Find bitmap block that contains bits for goal block */ for (buf = 0; buf < length; buf++) { bi = rgd->rd_bits + buf; @@ -1163,21 +1198,21 @@ do_search: bi = rgd->rd_bits + buf; if (test_bit(GBF_FULL, &bi->bi_flags) && - (old_state == GFS2_BLKST_FREE)) + (state == GFS2_BLKST_FREE)) goto skip; /* The GFS2_BLKST_UNLINKED state doesn't apply to the clone bitmaps, so we must search the originals for that. */ buffer = bi->bi_bh->b_data + bi->bi_offset; WARN_ON(!buffer_uptodate(bi->bi_bh)); - if (old_state != GFS2_BLKST_UNLINKED && bi->bi_clone) + if (state != GFS2_BLKST_UNLINKED && bi->bi_clone) buffer = bi->bi_clone + bi->bi_offset; - blk = gfs2_bitfit(buffer, bi->bi_len, goal, old_state); + blk = gfs2_bitfit(buffer, bi->bi_len, goal, state); if (blk != BFITNOENT) break; - if ((goal == 0) && (old_state == GFS2_BLKST_FREE)) + if ((goal == 0) && (state == GFS2_BLKST_FREE)) set_bit(GBF_FULL, &bi->bi_flags); /* Try next bitmap block (wrap back to rgrp header if at end) */ @@ -1187,16 +1222,37 @@ skip: goal = 0; } - if (blk == BFITNOENT) - return blk; + if (blk != BFITNOENT) + *rbi = bi; - *n = 1; - if (old_state == new_state) - goto out; + return blk; +} + +/** + * gfs2_alloc_extent - allocate an extent from a given bitmap + * @rgd: the resource group descriptor + * @bi: the bitmap within the rgrp + * @blk: the block within the bitmap + * @dinode: TRUE if the first block we allocate is for a dinode + * @n: The extent length + * + * Add the found bitmap buffer to the transaction. + * Set the found bits to @new_state to change block's allocation state. + * Returns: starting block number of the extent (fs scope) + */ +static u64 gfs2_alloc_extent(struct gfs2_rgrpd *rgd, struct gfs2_bitmap *bi, + u32 blk, bool dinode, unsigned int *n) +{ + const unsigned int elen = *n; + u32 goal; + const u8 *buffer = NULL; + *n = 0; + buffer = bi->bi_bh->b_data + bi->bi_offset; gfs2_trans_add_bh(rgd->rd_gl, bi->bi_bh, 1); gfs2_setbit(rgd, bi->bi_bh->b_data, bi->bi_clone, bi->bi_offset, - bi, blk, new_state); + bi, blk, dinode ? GFS2_BLKST_DINODE : GFS2_BLKST_USED); + (*n)++; goal = blk; while (*n < elen) { goal++; @@ -1206,11 +1262,12 @@ skip: GFS2_BLKST_FREE) break; gfs2_setbit(rgd, bi->bi_bh->b_data, bi->bi_clone, bi->bi_offset, - bi, goal, new_state); + bi, goal, GFS2_BLKST_USED); (*n)++; } -out: - return (bi->bi_start * GFS2_NBBY) + blk; + blk = gfs2_bi2rgd_blk(bi, blk); + rgd->rd_last_alloc = blk + *n - 1; + return rgd->rd_data0 + blk; } /** @@ -1298,121 +1355,93 @@ static void gfs2_rgrp_error(struct gfs2_rgrpd *rgd) } /** - * gfs2_alloc_block - Allocate one or more blocks + * gfs2_alloc_blocks - Allocate one or more blocks of data and/or a dinode * @ip: the inode to allocate the block for * @bn: Used to return the starting block number - * @n: requested number of blocks/extent length (value/result) + * @ndata: requested number of blocks/extent length (value/result) + * @dinode: 1 if we're allocating a dinode block, else 0 + * @generation: the generation number of the inode * * Returns: 0 or error */ -int gfs2_alloc_block(struct gfs2_inode *ip, u64 *bn, unsigned int *n) +int gfs2_alloc_blocks(struct gfs2_inode *ip, u64 *bn, unsigned int *nblocks, + bool dinode, u64 *generation) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct buffer_head *dibh; - struct gfs2_alloc *al = ip->i_alloc; struct gfs2_rgrpd *rgd; - u32 goal, blk; - u64 block; + unsigned int ndata; + u32 goal, blk; /* block, within the rgrp scope */ + u64 block; /* block, within the file system scope */ int error; + struct gfs2_bitmap *bi; /* Only happens if there is a bug in gfs2, return something distinctive * to ensure that it is noticed. */ - if (al == NULL) + if (ip->i_res == NULL) return -ECANCELED; rgd = ip->i_rgd; - if (rgrp_contains_block(rgd, ip->i_goal)) + if (!dinode && rgrp_contains_block(rgd, ip->i_goal)) goal = ip->i_goal - rgd->rd_data0; else goal = rgd->rd_last_alloc; - blk = rgblk_search(rgd, goal, GFS2_BLKST_FREE, GFS2_BLKST_USED, n); + blk = rgblk_search(rgd, goal, GFS2_BLKST_FREE, &bi); /* Since all blocks are reserved in advance, this shouldn't happen */ if (blk == BFITNOENT) goto rgrp_error; - rgd->rd_last_alloc = blk; - block = rgd->rd_data0 + blk; - ip->i_goal = block + *n - 1; - error = gfs2_meta_inode_buffer(ip, &dibh); - if (error == 0) { - struct gfs2_dinode *di = (struct gfs2_dinode *)dibh->b_data; - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - di->di_goal_meta = di->di_goal_data = cpu_to_be64(ip->i_goal); - brelse(dibh); + block = gfs2_alloc_extent(rgd, bi, blk, dinode, nblocks); + ndata = *nblocks; + if (dinode) + ndata--; + + if (!dinode) { + ip->i_goal = block + ndata - 1; + error = gfs2_meta_inode_buffer(ip, &dibh); + if (error == 0) { + struct gfs2_dinode *di = + (struct gfs2_dinode *)dibh->b_data; + gfs2_trans_add_bh(ip->i_gl, dibh, 1); + di->di_goal_meta = di->di_goal_data = + cpu_to_be64(ip->i_goal); + brelse(dibh); + } } - if (rgd->rd_free < *n) + if (rgd->rd_free < *nblocks) goto rgrp_error; - rgd->rd_free -= *n; - - gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1); - gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data); - - al->al_alloced += *n; - - gfs2_statfs_change(sdp, 0, -(s64)*n, 0); - gfs2_quota_change(ip, *n, ip->i_inode.i_uid, ip->i_inode.i_gid); - - rgd->rd_free_clone -= *n; - trace_gfs2_block_alloc(ip, block, *n, GFS2_BLKST_USED); - *bn = block; - return 0; - -rgrp_error: - gfs2_rgrp_error(rgd); - return -EIO; -} - -/** - * gfs2_alloc_di - Allocate a dinode - * @dip: the directory that the inode is going in - * @bn: the block number which is allocated - * @generation: the generation number of the inode - * - * Returns: 0 on success or error - */ - -int gfs2_alloc_di(struct gfs2_inode *dip, u64 *bn, u64 *generation) -{ - struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); - struct gfs2_alloc *al = dip->i_alloc; - struct gfs2_rgrpd *rgd = dip->i_rgd; - u32 blk; - u64 block; - unsigned int n = 1; - - blk = rgblk_search(rgd, rgd->rd_last_alloc, - GFS2_BLKST_FREE, GFS2_BLKST_DINODE, &n); - - /* Since all blocks are reserved in advance, this shouldn't happen */ - if (blk == BFITNOENT) - goto rgrp_error; - - rgd->rd_last_alloc = blk; - block = rgd->rd_data0 + blk; - if (rgd->rd_free == 0) - goto rgrp_error; - - rgd->rd_free--; - rgd->rd_dinodes++; - *generation = rgd->rd_igeneration++; - if (*generation == 0) + rgd->rd_free -= *nblocks; + if (dinode) { + rgd->rd_dinodes++; *generation = rgd->rd_igeneration++; + if (*generation == 0) + *generation = rgd->rd_igeneration++; + } + gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1); gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data); - al->al_alloced++; + gfs2_statfs_change(sdp, 0, -(s64)*nblocks, dinode ? 1 : 0); + if (dinode) + gfs2_trans_add_unrevoke(sdp, block, 1); - gfs2_statfs_change(sdp, 0, -1, +1); - gfs2_trans_add_unrevoke(sdp, block, 1); + /* + * This needs reviewing to see why we cannot do the quota change + * at this point in the dinode case. + */ + if (ndata) + gfs2_quota_change(ip, ndata, ip->i_inode.i_uid, + ip->i_inode.i_gid); - rgd->rd_free_clone--; - trace_gfs2_block_alloc(dip, block, 1, GFS2_BLKST_DINODE); + rgd->rd_free_clone -= *nblocks; + trace_gfs2_block_alloc(ip, block, *nblocks, + dinode ? GFS2_BLKST_DINODE : GFS2_BLKST_USED); *bn = block; return 0; diff --git a/fs/gfs2/rgrp.h b/fs/gfs2/rgrp.h index cf5c50180192..ceec9106cdf4 100644 --- a/fs/gfs2/rgrp.h +++ b/fs/gfs2/rgrp.h @@ -28,19 +28,19 @@ extern void gfs2_free_clones(struct gfs2_rgrpd *rgd); extern int gfs2_rgrp_go_lock(struct gfs2_holder *gh); extern void gfs2_rgrp_go_unlock(struct gfs2_holder *gh); -extern struct gfs2_alloc *gfs2_alloc_get(struct gfs2_inode *ip); -static inline void gfs2_alloc_put(struct gfs2_inode *ip) +extern struct gfs2_qadata *gfs2_qadata_get(struct gfs2_inode *ip); +static inline void gfs2_qadata_put(struct gfs2_inode *ip) { - BUG_ON(ip->i_alloc == NULL); - kfree(ip->i_alloc); - ip->i_alloc = NULL; + BUG_ON(ip->i_qadata == NULL); + kfree(ip->i_qadata); + ip->i_qadata = NULL; } -extern int gfs2_inplace_reserve(struct gfs2_inode *ip); +extern int gfs2_inplace_reserve(struct gfs2_inode *ip, u32 requested); extern void gfs2_inplace_release(struct gfs2_inode *ip); -extern int gfs2_alloc_block(struct gfs2_inode *ip, u64 *bn, unsigned int *n); -extern int gfs2_alloc_di(struct gfs2_inode *ip, u64 *bn, u64 *generation); +extern int gfs2_alloc_blocks(struct gfs2_inode *ip, u64 *bn, unsigned int *n, + bool dinode, u64 *generation); extern void __gfs2_free_blocks(struct gfs2_inode *ip, u64 bstart, u32 blen, int meta); extern void gfs2_free_meta(struct gfs2_inode *ip, u64 bstart, u32 blen); diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 71e420989f77..4553ce515f62 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -1284,18 +1284,18 @@ static int is_ancestor(const struct dentry *d1, const struct dentry *d2) /** * gfs2_show_options - Show mount options for /proc/mounts * @s: seq_file structure - * @mnt: vfsmount + * @root: root of this (sub)tree * * Returns: 0 on success or error code */ -static int gfs2_show_options(struct seq_file *s, struct vfsmount *mnt) +static int gfs2_show_options(struct seq_file *s, struct dentry *root) { - struct gfs2_sbd *sdp = mnt->mnt_sb->s_fs_info; + struct gfs2_sbd *sdp = root->d_sb->s_fs_info; struct gfs2_args *args = &sdp->sd_args; int val; - if (is_ancestor(mnt->mnt_root, sdp->sd_master_dir)) + if (is_ancestor(root, sdp->sd_master_dir)) seq_printf(s, ",meta"); if (args->ar_lockproto[0]) seq_printf(s, ",lockproto=%s", args->ar_lockproto); @@ -1399,8 +1399,9 @@ static void gfs2_final_release_pages(struct gfs2_inode *ip) static int gfs2_dinode_dealloc(struct gfs2_inode *ip) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al; + struct gfs2_qadata *qa; struct gfs2_rgrpd *rgd; + struct gfs2_holder gh; int error; if (gfs2_get_inode_blocks(&ip->i_inode) != 1) { @@ -1408,8 +1409,8 @@ static int gfs2_dinode_dealloc(struct gfs2_inode *ip) return -EIO; } - al = gfs2_alloc_get(ip); - if (!al) + qa = gfs2_qadata_get(ip); + if (!qa) return -ENOMEM; error = gfs2_quota_hold(ip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE); @@ -1423,8 +1424,7 @@ static int gfs2_dinode_dealloc(struct gfs2_inode *ip) goto out_qs; } - error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, - &al->al_rgd_gh); + error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, &gh); if (error) goto out_qs; @@ -1440,11 +1440,11 @@ static int gfs2_dinode_dealloc(struct gfs2_inode *ip) gfs2_trans_end(sdp); out_rg_gunlock: - gfs2_glock_dq_uninit(&al->al_rgd_gh); + gfs2_glock_dq_uninit(&gh); out_qs: gfs2_quota_unhold(ip); out: - gfs2_alloc_put(ip); + gfs2_qadata_put(ip); return error; } @@ -1582,7 +1582,6 @@ static struct inode *gfs2_alloc_inode(struct super_block *sb) static void gfs2_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(gfs2_inode_cachep, inode); } diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c index 443cabcfcd23..d33172c291ba 100644 --- a/fs/gfs2/sys.c +++ b/fs/gfs2/sys.c @@ -298,7 +298,7 @@ static ssize_t block_show(struct gfs2_sbd *sdp, char *buf) ssize_t ret; int val = 0; - if (test_bit(DFL_BLOCK_LOCKS, &ls->ls_flags)) + if (test_bit(DFL_BLOCK_LOCKS, &ls->ls_recover_flags)) val = 1; ret = sprintf(buf, "%d\n", val); return ret; @@ -313,9 +313,9 @@ static ssize_t block_store(struct gfs2_sbd *sdp, const char *buf, size_t len) val = simple_strtol(buf, NULL, 0); if (val == 1) - set_bit(DFL_BLOCK_LOCKS, &ls->ls_flags); + set_bit(DFL_BLOCK_LOCKS, &ls->ls_recover_flags); else if (val == 0) { - clear_bit(DFL_BLOCK_LOCKS, &ls->ls_flags); + clear_bit(DFL_BLOCK_LOCKS, &ls->ls_recover_flags); smp_mb__after_clear_bit(); gfs2_glock_thaw(sdp); } else { @@ -350,8 +350,8 @@ static ssize_t lkfirst_store(struct gfs2_sbd *sdp, const char *buf, size_t len) goto out; if (sdp->sd_lockstruct.ls_ops->lm_mount == NULL) goto out; - sdp->sd_lockstruct.ls_first = first; - rv = 0; + sdp->sd_lockstruct.ls_first = first; + rv = 0; out: spin_unlock(&sdp->sd_jindex_spin); return rv ? rv : len; @@ -360,19 +360,14 @@ out: static ssize_t first_done_show(struct gfs2_sbd *sdp, char *buf) { struct lm_lockstruct *ls = &sdp->sd_lockstruct; - return sprintf(buf, "%d\n", ls->ls_first_done); + return sprintf(buf, "%d\n", !!test_bit(DFL_FIRST_MOUNT_DONE, &ls->ls_recover_flags)); } -static ssize_t recover_store(struct gfs2_sbd *sdp, const char *buf, size_t len) +int gfs2_recover_set(struct gfs2_sbd *sdp, unsigned jid) { - unsigned jid; struct gfs2_jdesc *jd; int rv; - rv = sscanf(buf, "%u", &jid); - if (rv != 1) - return -EINVAL; - rv = -ESHUTDOWN; spin_lock(&sdp->sd_jindex_spin); if (test_bit(SDF_NORECOVERY, &sdp->sd_flags)) @@ -389,6 +384,20 @@ static ssize_t recover_store(struct gfs2_sbd *sdp, const char *buf, size_t len) } out: spin_unlock(&sdp->sd_jindex_spin); + return rv; +} + +static ssize_t recover_store(struct gfs2_sbd *sdp, const char *buf, size_t len) +{ + unsigned jid; + int rv; + + rv = sscanf(buf, "%u", &jid); + if (rv != 1) + return -EINVAL; + + rv = gfs2_recover_set(sdp, jid); + return rv ? rv : len; } diff --git a/fs/gfs2/sys.h b/fs/gfs2/sys.h index e94560e836d7..79182d6ad6ac 100644 --- a/fs/gfs2/sys.h +++ b/fs/gfs2/sys.h @@ -19,5 +19,7 @@ void gfs2_sys_fs_del(struct gfs2_sbd *sdp); int gfs2_sys_init(void); void gfs2_sys_uninit(void); +int gfs2_recover_set(struct gfs2_sbd *sdp, unsigned jid); + #endif /* __SYS_DOT_H__ */ diff --git a/fs/gfs2/trans.h b/fs/gfs2/trans.h index f8f101ef600c..125d4572e1c0 100644 --- a/fs/gfs2/trans.h +++ b/fs/gfs2/trans.h @@ -30,9 +30,9 @@ struct gfs2_glock; * block, or all of the blocks in the rg, whichever is smaller */ static inline unsigned int gfs2_rg_blocks(const struct gfs2_inode *ip) { - const struct gfs2_alloc *al = ip->i_alloc; - if (al->al_requested < ip->i_rgd->rd_length) - return al->al_requested + 1; + const struct gfs2_blkreserv *rs = ip->i_res; + if (rs->rs_requested < ip->i_rgd->rd_length) + return rs->rs_requested + 1; return ip->i_rgd->rd_length; } diff --git a/fs/gfs2/xattr.c b/fs/gfs2/xattr.c index 71d7bf830c09..e9636591b5d5 100644 --- a/fs/gfs2/xattr.c +++ b/fs/gfs2/xattr.c @@ -321,11 +321,11 @@ static int ea_remove_unstuffed(struct gfs2_inode *ip, struct buffer_head *bh, struct gfs2_ea_header *ea, struct gfs2_ea_header *prev, int leave) { - struct gfs2_alloc *al; + struct gfs2_qadata *qa; int error; - al = gfs2_alloc_get(ip); - if (!al) + qa = gfs2_qadata_get(ip); + if (!qa) return -ENOMEM; error = gfs2_quota_hold(ip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE); @@ -336,7 +336,7 @@ static int ea_remove_unstuffed(struct gfs2_inode *ip, struct buffer_head *bh, gfs2_quota_unhold(ip); out_alloc: - gfs2_alloc_put(ip); + gfs2_qadata_put(ip); return error; } @@ -549,9 +549,10 @@ int gfs2_xattr_acl_get(struct gfs2_inode *ip, const char *name, char **ppdata) goto out; error = gfs2_ea_get_copy(ip, &el, data, len); - if (error == 0) - error = len; - *ppdata = data; + if (error < 0) + kfree(data); + else + *ppdata = data; out: brelse(el.el_bh); return error; @@ -609,7 +610,7 @@ static int ea_alloc_blk(struct gfs2_inode *ip, struct buffer_head **bhp) u64 block; int error; - error = gfs2_alloc_block(ip, &block, &n); + error = gfs2_alloc_blocks(ip, &block, &n, 0, NULL); if (error) return error; gfs2_trans_add_unrevoke(sdp, block, 1); @@ -671,7 +672,7 @@ static int ea_write(struct gfs2_inode *ip, struct gfs2_ea_header *ea, int mh_size = sizeof(struct gfs2_meta_header); unsigned int n = 1; - error = gfs2_alloc_block(ip, &block, &n); + error = gfs2_alloc_blocks(ip, &block, &n, 0, NULL); if (error) return error; gfs2_trans_add_unrevoke(sdp, block, 1); @@ -708,21 +709,19 @@ static int ea_alloc_skeleton(struct gfs2_inode *ip, struct gfs2_ea_request *er, unsigned int blks, ea_skeleton_call_t skeleton_call, void *private) { - struct gfs2_alloc *al; + struct gfs2_qadata *qa; struct buffer_head *dibh; int error; - al = gfs2_alloc_get(ip); - if (!al) + qa = gfs2_qadata_get(ip); + if (!qa) return -ENOMEM; error = gfs2_quota_lock_check(ip); if (error) goto out; - al->al_requested = blks; - - error = gfs2_inplace_reserve(ip); + error = gfs2_inplace_reserve(ip, blks); if (error) goto out_gunlock_q; @@ -751,7 +750,7 @@ out_ipres: out_gunlock_q: gfs2_quota_unlock(ip); out: - gfs2_alloc_put(ip); + gfs2_qadata_put(ip); return error; } @@ -991,7 +990,7 @@ static int ea_set_block(struct gfs2_inode *ip, struct gfs2_ea_request *er, } else { u64 blk; unsigned int n = 1; - error = gfs2_alloc_block(ip, &blk, &n); + error = gfs2_alloc_blocks(ip, &blk, &n, 0, NULL); if (error) return error; gfs2_trans_add_unrevoke(sdp, blk, 1); @@ -1435,9 +1434,9 @@ out: static int ea_dealloc_block(struct gfs2_inode *ip) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al = ip->i_alloc; struct gfs2_rgrpd *rgd; struct buffer_head *dibh; + struct gfs2_holder gh; int error; rgd = gfs2_blk2rgrpd(sdp, ip->i_eattr); @@ -1446,8 +1445,7 @@ static int ea_dealloc_block(struct gfs2_inode *ip) return -EIO; } - error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, - &al->al_rgd_gh); + error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, &gh); if (error) return error; @@ -1471,7 +1469,7 @@ static int ea_dealloc_block(struct gfs2_inode *ip) gfs2_trans_end(sdp); out_gunlock: - gfs2_glock_dq_uninit(&al->al_rgd_gh); + gfs2_glock_dq_uninit(&gh); return error; } @@ -1484,11 +1482,11 @@ out_gunlock: int gfs2_ea_dealloc(struct gfs2_inode *ip) { - struct gfs2_alloc *al; + struct gfs2_qadata *qa; int error; - al = gfs2_alloc_get(ip); - if (!al) + qa = gfs2_qadata_get(ip); + if (!qa) return -ENOMEM; error = gfs2_quota_hold(ip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE); @@ -1510,7 +1508,7 @@ int gfs2_ea_dealloc(struct gfs2_inode *ip) out_quota: gfs2_quota_unhold(ip); out_alloc: - gfs2_alloc_put(ip); + gfs2_qadata_put(ip); return error; } diff --git a/fs/hfs/dir.c b/fs/hfs/dir.c index bce4eef91a06..62fc14ea4b73 100644 --- a/fs/hfs/dir.c +++ b/fs/hfs/dir.c @@ -186,7 +186,7 @@ static int hfs_dir_release(struct inode *inode, struct file *file) * a directory and return a corresponding inode, given the inode for * the directory and the name (and its length) of the new file. */ -static int hfs_create(struct inode *dir, struct dentry *dentry, int mode, +static int hfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { struct inode *inode; @@ -216,7 +216,7 @@ static int hfs_create(struct inode *dir, struct dentry *dentry, int mode, * in a directory, given the inode for the parent directory and the * name (and its length) of the new directory. */ -static int hfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int hfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct inode *inode; int res; diff --git a/fs/hfs/hfs_fs.h b/fs/hfs/hfs_fs.h index ad97c2d58287..1bf967c6bfdc 100644 --- a/fs/hfs/hfs_fs.h +++ b/fs/hfs/hfs_fs.h @@ -184,7 +184,7 @@ extern int hfs_get_block(struct inode *, sector_t, struct buffer_head *, int); extern const struct address_space_operations hfs_aops; extern const struct address_space_operations hfs_btree_aops; -extern struct inode *hfs_new_inode(struct inode *, struct qstr *, int); +extern struct inode *hfs_new_inode(struct inode *, struct qstr *, umode_t); extern void hfs_inode_write_fork(struct inode *, struct hfs_extent *, __be32 *, __be32 *); extern int hfs_write_inode(struct inode *, struct writeback_control *); extern int hfs_inode_setattr(struct dentry *, struct iattr *); diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c index a1a9fdcd2a00..737dbeb64320 100644 --- a/fs/hfs/inode.c +++ b/fs/hfs/inode.c @@ -169,7 +169,7 @@ const struct address_space_operations hfs_aops = { /* * hfs_new_inode */ -struct inode *hfs_new_inode(struct inode *dir, struct qstr *name, int mode) +struct inode *hfs_new_inode(struct inode *dir, struct qstr *name, umode_t mode) { struct super_block *sb = dir->i_sb; struct inode *inode = new_inode(sb); diff --git a/fs/hfs/super.c b/fs/hfs/super.c index 1b55f704fb22..8137fb3e6780 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -133,9 +133,9 @@ static int hfs_remount(struct super_block *sb, int *flags, char *data) return 0; } -static int hfs_show_options(struct seq_file *seq, struct vfsmount *mnt) +static int hfs_show_options(struct seq_file *seq, struct dentry *root) { - struct hfs_sb_info *sbi = HFS_SB(mnt->mnt_sb); + struct hfs_sb_info *sbi = HFS_SB(root->d_sb); if (sbi->s_creator != cpu_to_be32(0x3f3f3f3f)) seq_printf(seq, ",creator=%.4s", (char *)&sbi->s_creator); @@ -170,7 +170,6 @@ static struct inode *hfs_alloc_inode(struct super_block *sb) static void hfs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(hfs_inode_cachep, HFS_I(inode)); } diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c index 4536cd3f15ae..88e155f895c6 100644 --- a/fs/hfsplus/dir.c +++ b/fs/hfsplus/dir.c @@ -424,7 +424,7 @@ out: } static int hfsplus_mknod(struct inode *dir, struct dentry *dentry, - int mode, dev_t rdev) + umode_t mode, dev_t rdev) { struct hfsplus_sb_info *sbi = HFSPLUS_SB(dir->i_sb); struct inode *inode; @@ -453,13 +453,13 @@ out: return res; } -static int hfsplus_create(struct inode *dir, struct dentry *dentry, int mode, +static int hfsplus_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { return hfsplus_mknod(dir, dentry, mode, 0); } -static int hfsplus_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int hfsplus_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { return hfsplus_mknod(dir, dentry, mode | S_IFDIR, 0); } diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h index d7674d051f52..21a5b7fc6db4 100644 --- a/fs/hfsplus/hfsplus_fs.h +++ b/fs/hfsplus/hfsplus_fs.h @@ -402,7 +402,7 @@ void hfsplus_inode_read_fork(struct inode *, struct hfsplus_fork_raw *); void hfsplus_inode_write_fork(struct inode *, struct hfsplus_fork_raw *); int hfsplus_cat_read_inode(struct inode *, struct hfs_find_data *); int hfsplus_cat_write_inode(struct inode *); -struct inode *hfsplus_new_inode(struct super_block *, int); +struct inode *hfsplus_new_inode(struct super_block *, umode_t); void hfsplus_delete_inode(struct inode *); int hfsplus_file_fsync(struct file *file, loff_t start, loff_t end, int datasync); @@ -419,7 +419,7 @@ ssize_t hfsplus_listxattr(struct dentry *dentry, char *buffer, size_t size); int hfsplus_parse_options(char *, struct hfsplus_sb_info *); int hfsplus_parse_options_remount(char *input, int *force); void hfsplus_fill_defaults(struct hfsplus_sb_info *); -int hfsplus_show_options(struct seq_file *, struct vfsmount *); +int hfsplus_show_options(struct seq_file *, struct dentry *); /* super.c */ struct inode *hfsplus_iget(struct super_block *, unsigned long); diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c index 40e1413be4cf..6643b242bdd7 100644 --- a/fs/hfsplus/inode.c +++ b/fs/hfsplus/inode.c @@ -378,7 +378,7 @@ static const struct file_operations hfsplus_file_operations = { .unlocked_ioctl = hfsplus_ioctl, }; -struct inode *hfsplus_new_inode(struct super_block *sb, int mode) +struct inode *hfsplus_new_inode(struct super_block *sb, umode_t mode) { struct hfsplus_sb_info *sbi = HFSPLUS_SB(sb); struct inode *inode = new_inode(sb); diff --git a/fs/hfsplus/ioctl.c b/fs/hfsplus/ioctl.c index fbaa6690c8e0..f66c7655b3f7 100644 --- a/fs/hfsplus/ioctl.c +++ b/fs/hfsplus/ioctl.c @@ -43,7 +43,7 @@ static int hfsplus_ioctl_setflags(struct file *file, int __user *user_flags) unsigned int flags; int err = 0; - err = mnt_want_write(file->f_path.mnt); + err = mnt_want_write_file(file); if (err) goto out; @@ -94,7 +94,7 @@ static int hfsplus_ioctl_setflags(struct file *file, int __user *user_flags) out_unlock_inode: mutex_unlock(&inode->i_mutex); out_drop_write: - mnt_drop_write(file->f_path.mnt); + mnt_drop_write_file(file); out: return err; } diff --git a/fs/hfsplus/options.c b/fs/hfsplus/options.c index bb62a5882147..06fa5618600c 100644 --- a/fs/hfsplus/options.c +++ b/fs/hfsplus/options.c @@ -206,9 +206,9 @@ done: return 1; } -int hfsplus_show_options(struct seq_file *seq, struct vfsmount *mnt) +int hfsplus_show_options(struct seq_file *seq, struct dentry *root) { - struct hfsplus_sb_info *sbi = HFSPLUS_SB(mnt->mnt_sb); + struct hfsplus_sb_info *sbi = HFSPLUS_SB(root->d_sb); if (sbi->creator != HFSPLUS_DEF_CR_TYPE) seq_printf(seq, ",creator=%.4s", (char *)&sbi->creator); diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index d24a9b666a23..427682ca9e48 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -499,9 +499,16 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) if (!sbi->hidden_dir) { mutex_lock(&sbi->vh_mutex); sbi->hidden_dir = hfsplus_new_inode(sb, S_IFDIR); - hfsplus_create_cat(sbi->hidden_dir->i_ino, root, &str, - sbi->hidden_dir); + if (!sbi->hidden_dir) { + mutex_unlock(&sbi->vh_mutex); + err = -ENOMEM; + goto out_put_root; + } + err = hfsplus_create_cat(sbi->hidden_dir->i_ino, root, + &str, sbi->hidden_dir); mutex_unlock(&sbi->vh_mutex); + if (err) + goto out_put_hidden_dir; hfsplus_mark_inode_dirty(sbi->hidden_dir, HFSPLUS_I_CAT_DIRTY); @@ -558,7 +565,6 @@ static void hfsplus_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(hfsplus_inode_cachep, HFSPLUS_I(inode)); } diff --git a/fs/hostfs/hostfs.h b/fs/hostfs/hostfs.h index bf15a43016b9..3cbfa93cd782 100644 --- a/fs/hostfs/hostfs.h +++ b/fs/hostfs/hostfs.h @@ -39,7 +39,7 @@ struct hostfs_iattr { unsigned int ia_valid; - mode_t ia_mode; + unsigned short ia_mode; uid_t ia_uid; gid_t ia_gid; loff_t ia_size; diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index 2f72da5ae686..e130bd46d671 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -250,7 +250,6 @@ static void hostfs_evict_inode(struct inode *inode) static void hostfs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kfree(HOSTFS_I(inode)); } @@ -259,9 +258,9 @@ static void hostfs_destroy_inode(struct inode *inode) call_rcu(&inode->i_rcu, hostfs_i_callback); } -static int hostfs_show_options(struct seq_file *seq, struct vfsmount *vfs) +static int hostfs_show_options(struct seq_file *seq, struct dentry *root) { - const char *root_path = vfs->mnt_sb->s_fs_info; + const char *root_path = root->d_sb->s_fs_info; size_t offset = strlen(root_ino) + 1; if (strlen(root_path) > offset) @@ -552,7 +551,7 @@ static int read_name(struct inode *ino, char *name) return 0; } -int hostfs_create(struct inode *dir, struct dentry *dentry, int mode, +int hostfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { struct inode *inode; @@ -677,7 +676,7 @@ int hostfs_symlink(struct inode *ino, struct dentry *dentry, const char *to) return err; } -int hostfs_mkdir(struct inode *ino, struct dentry *dentry, int mode) +int hostfs_mkdir(struct inode *ino, struct dentry *dentry, umode_t mode) { char *file; int err; @@ -701,7 +700,7 @@ int hostfs_rmdir(struct inode *ino, struct dentry *dentry) return err; } -int hostfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) +static int hostfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev) { struct inode *inode; char *name; diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c index ea91fcb0ef9b..30dd7b10b507 100644 --- a/fs/hpfs/namei.c +++ b/fs/hpfs/namei.c @@ -8,7 +8,7 @@ #include <linux/sched.h> #include "hpfs_fn.h" -static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { const unsigned char *name = dentry->d_name.name; unsigned len = dentry->d_name.len; @@ -115,7 +115,7 @@ bail: return err; } -static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struct nameidata *nd) +static int hpfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { const unsigned char *name = dentry->d_name.name; unsigned len = dentry->d_name.len; @@ -201,7 +201,7 @@ bail: return err; } -static int hpfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev) +static int hpfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) { const unsigned char *name = dentry->d_name.name; unsigned len = dentry->d_name.len; diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c index 98580a3b5005..3690467c944e 100644 --- a/fs/hpfs/super.c +++ b/fs/hpfs/super.c @@ -181,7 +181,6 @@ static struct inode *hpfs_alloc_inode(struct super_block *sb) static void hpfs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(hpfs_inode_cachep, hpfs_i(inode)); } diff --git a/fs/hppfs/hppfs.c b/fs/hppfs/hppfs.c index f590b1160c6c..d92f4ce80925 100644 --- a/fs/hppfs/hppfs.c +++ b/fs/hppfs/hppfs.c @@ -622,7 +622,6 @@ void hppfs_evict_inode(struct inode *ino) static void hppfs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kfree(HPPFS_I(inode)); } @@ -726,7 +725,7 @@ static int hppfs_fill_super(struct super_block *sb, void *d, int silent) sb->s_fs_info = proc_mnt; err = -ENOMEM; - root_inode = get_inode(sb, dget(proc_mnt->mnt_sb->s_root)); + root_inode = get_inode(sb, dget(proc_mnt->mnt_root)); if (!root_inode) goto out_mntput; diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 0be5a78598d0..1e85a7ac0217 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -447,8 +447,8 @@ static int hugetlbfs_setattr(struct dentry *dentry, struct iattr *attr) return 0; } -static struct inode *hugetlbfs_get_inode(struct super_block *sb, uid_t uid, - gid_t gid, int mode, dev_t dev) +static struct inode *hugetlbfs_get_root(struct super_block *sb, + struct hugetlbfs_config *config) { struct inode *inode; @@ -456,9 +456,31 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb, uid_t uid, if (inode) { struct hugetlbfs_inode_info *info; inode->i_ino = get_next_ino(); - inode->i_mode = mode; - inode->i_uid = uid; - inode->i_gid = gid; + inode->i_mode = S_IFDIR | config->mode; + inode->i_uid = config->uid; + inode->i_gid = config->gid; + inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; + info = HUGETLBFS_I(inode); + mpol_shared_policy_init(&info->policy, NULL); + inode->i_op = &hugetlbfs_dir_inode_operations; + inode->i_fop = &simple_dir_operations; + /* directory inodes start off with i_nlink == 2 (for "." entry) */ + inc_nlink(inode); + } + return inode; +} + +static struct inode *hugetlbfs_get_inode(struct super_block *sb, + struct inode *dir, + umode_t mode, dev_t dev) +{ + struct inode *inode; + + inode = new_inode(sb); + if (inode) { + struct hugetlbfs_inode_info *info; + inode->i_ino = get_next_ino(); + inode_init_owner(inode, dir, mode); inode->i_mapping->a_ops = &hugetlbfs_aops; inode->i_mapping->backing_dev_info =&hugetlbfs_backing_dev_info; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; @@ -500,20 +522,12 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb, uid_t uid, * File creation. Allocate an inode, and we're done.. */ static int hugetlbfs_mknod(struct inode *dir, - struct dentry *dentry, int mode, dev_t dev) + struct dentry *dentry, umode_t mode, dev_t dev) { struct inode *inode; int error = -ENOSPC; - gid_t gid; - - if (dir->i_mode & S_ISGID) { - gid = dir->i_gid; - if (S_ISDIR(mode)) - mode |= S_ISGID; - } else { - gid = current_fsgid(); - } - inode = hugetlbfs_get_inode(dir->i_sb, current_fsuid(), gid, mode, dev); + + inode = hugetlbfs_get_inode(dir->i_sb, dir, mode, dev); if (inode) { dir->i_ctime = dir->i_mtime = CURRENT_TIME; d_instantiate(dentry, inode); @@ -523,7 +537,7 @@ static int hugetlbfs_mknod(struct inode *dir, return error; } -static int hugetlbfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int hugetlbfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { int retval = hugetlbfs_mknod(dir, dentry, mode | S_IFDIR, 0); if (!retval) @@ -531,7 +545,7 @@ static int hugetlbfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) return retval; } -static int hugetlbfs_create(struct inode *dir, struct dentry *dentry, int mode, struct nameidata *nd) +static int hugetlbfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { return hugetlbfs_mknod(dir, dentry, mode | S_IFREG, 0); } @@ -541,15 +555,8 @@ static int hugetlbfs_symlink(struct inode *dir, { struct inode *inode; int error = -ENOSPC; - gid_t gid; - - if (dir->i_mode & S_ISGID) - gid = dir->i_gid; - else - gid = current_fsgid(); - inode = hugetlbfs_get_inode(dir->i_sb, current_fsuid(), - gid, S_IFLNK|S_IRWXUGO, 0); + inode = hugetlbfs_get_inode(dir->i_sb, dir, S_IFLNK|S_IRWXUGO, 0); if (inode) { int l = strlen(symname)+1; error = page_symlink(inode, symname, l); @@ -576,7 +583,8 @@ static int hugetlbfs_set_page_dirty(struct page *page) } static int hugetlbfs_migrate_page(struct address_space *mapping, - struct page *newpage, struct page *page) + struct page *newpage, struct page *page, + enum migrate_mode mode) { int rc; @@ -666,7 +674,6 @@ static struct inode *hugetlbfs_alloc_inode(struct super_block *sb) static void hugetlbfs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(hugetlbfs_inode_cachep, HUGETLBFS_I(inode)); } @@ -858,8 +865,7 @@ hugetlbfs_fill_super(struct super_block *sb, void *data, int silent) sb->s_magic = HUGETLBFS_MAGIC; sb->s_op = &hugetlbfs_ops; sb->s_time_gran = 1; - inode = hugetlbfs_get_inode(sb, config.uid, config.gid, - S_IFDIR | config.mode, 0); + inode = hugetlbfs_get_root(sb, &config); if (!inode) goto out_free; @@ -957,8 +963,7 @@ struct file *hugetlb_file_setup(const char *name, size_t size, path.mnt = mntget(hugetlbfs_vfsmount); error = -ENOSPC; - inode = hugetlbfs_get_inode(root->d_sb, current_fsuid(), - current_fsgid(), S_IFREG | S_IRWXUGO, 0); + inode = hugetlbfs_get_inode(root->d_sb, NULL, S_IFREG | S_IRWXUGO, 0); if (!inode) goto out_dentry; diff --git a/fs/inode.c b/fs/inode.c index ee4e66b998f4..4fa4f0916af9 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -26,6 +26,7 @@ #include <linux/ima.h> #include <linux/cred.h> #include <linux/buffer_head.h> /* for inode_has_buffers */ +#include <linux/ratelimit.h> #include "internal.h" /* @@ -191,6 +192,7 @@ int inode_init_always(struct super_block *sb, struct inode *inode) } inode->i_private = NULL; inode->i_mapping = mapping; + INIT_LIST_HEAD(&inode->i_dentry); /* buggered by rcu freeing */ #ifdef CONFIG_FS_POSIX_ACL inode->i_acl = inode->i_default_acl = ACL_NOT_CACHED; #endif @@ -241,6 +243,11 @@ void __destroy_inode(struct inode *inode) BUG_ON(inode_has_buffers(inode)); security_inode_free(inode); fsnotify_inode_delete(inode); + if (!inode->i_nlink) { + WARN_ON(atomic_long_read(&inode->i_sb->s_remove_count) == 0); + atomic_long_dec(&inode->i_sb->s_remove_count); + } + #ifdef CONFIG_FS_POSIX_ACL if (inode->i_acl && inode->i_acl != ACL_NOT_CACHED) posix_acl_release(inode->i_acl); @@ -254,7 +261,6 @@ EXPORT_SYMBOL(__destroy_inode); static void i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(inode_cachep, inode); } @@ -268,6 +274,85 @@ static void destroy_inode(struct inode *inode) call_rcu(&inode->i_rcu, i_callback); } +/** + * drop_nlink - directly drop an inode's link count + * @inode: inode + * + * This is a low-level filesystem helper to replace any + * direct filesystem manipulation of i_nlink. In cases + * where we are attempting to track writes to the + * filesystem, a decrement to zero means an imminent + * write when the file is truncated and actually unlinked + * on the filesystem. + */ +void drop_nlink(struct inode *inode) +{ + WARN_ON(inode->i_nlink == 0); + inode->__i_nlink--; + if (!inode->i_nlink) + atomic_long_inc(&inode->i_sb->s_remove_count); +} +EXPORT_SYMBOL(drop_nlink); + +/** + * clear_nlink - directly zero an inode's link count + * @inode: inode + * + * This is a low-level filesystem helper to replace any + * direct filesystem manipulation of i_nlink. See + * drop_nlink() for why we care about i_nlink hitting zero. + */ +void clear_nlink(struct inode *inode) +{ + if (inode->i_nlink) { + inode->__i_nlink = 0; + atomic_long_inc(&inode->i_sb->s_remove_count); + } +} +EXPORT_SYMBOL(clear_nlink); + +/** + * set_nlink - directly set an inode's link count + * @inode: inode + * @nlink: new nlink (should be non-zero) + * + * This is a low-level filesystem helper to replace any + * direct filesystem manipulation of i_nlink. + */ +void set_nlink(struct inode *inode, unsigned int nlink) +{ + if (!nlink) { + printk_ratelimited(KERN_INFO + "set_nlink() clearing i_nlink on %s inode %li\n", + inode->i_sb->s_type->name, inode->i_ino); + clear_nlink(inode); + } else { + /* Yes, some filesystems do change nlink from zero to one */ + if (inode->i_nlink == 0) + atomic_long_dec(&inode->i_sb->s_remove_count); + + inode->__i_nlink = nlink; + } +} +EXPORT_SYMBOL(set_nlink); + +/** + * inc_nlink - directly increment an inode's link count + * @inode: inode + * + * This is a low-level filesystem helper to replace any + * direct filesystem manipulation of i_nlink. Currently, + * it is only here for parity with dec_nlink(). + */ +void inc_nlink(struct inode *inode) +{ + if (WARN_ON(inode->i_nlink == 0)) + atomic_long_dec(&inode->i_sb->s_remove_count); + + inode->__i_nlink++; +} +EXPORT_SYMBOL(inc_nlink); + void address_space_init_once(struct address_space *mapping) { memset(mapping, 0, sizeof(*mapping)); @@ -290,7 +375,6 @@ void inode_init_once(struct inode *inode) { memset(inode, 0, sizeof(*inode)); INIT_HLIST_NODE(&inode->i_hash); - INIT_LIST_HEAD(&inode->i_dentry); INIT_LIST_HEAD(&inode->i_devices); INIT_LIST_HEAD(&inode->i_wb_list); INIT_LIST_HEAD(&inode->i_lru); @@ -692,6 +776,8 @@ void prune_icache_sb(struct super_block *sb, int nr_to_scan) else __count_vm_events(PGINODESTEAL, reap); spin_unlock(&sb->s_inode_lru_lock); + if (current->reclaim_state) + current->reclaim_state->reclaimed_slab += reap; dispose_list(&freeable); } @@ -1508,7 +1594,7 @@ void file_update_time(struct file *file) if (sync_it & S_MTIME) inode->i_mtime = now; mark_inode_dirty_sync(inode); - mnt_drop_write(file->f_path.mnt); + mnt_drop_write_file(file); } EXPORT_SYMBOL(file_update_time); @@ -1647,7 +1733,7 @@ EXPORT_SYMBOL(init_special_inode); * @mode: mode of the new inode */ void inode_init_owner(struct inode *inode, const struct inode *dir, - mode_t mode) + umode_t mode) { inode->i_uid = current_fsuid(); if (dir && dir->i_mode & S_ISGID) { diff --git a/fs/internal.h b/fs/internal.h index fe327c20af83..9962c59ba280 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -15,19 +15,14 @@ struct super_block; struct file_system_type; struct linux_binprm; struct path; +struct mount; /* * block_dev.c */ #ifdef CONFIG_BLOCK -extern struct super_block *blockdev_superblock; extern void __init bdev_cache_init(void); -static inline int sb_is_blkdev_sb(struct super_block *sb) -{ - return sb == blockdev_superblock; -} - extern int __sync_blockdev(struct block_device *bdev, int wait); #else @@ -35,11 +30,6 @@ static inline void bdev_cache_init(void) { } -static inline int sb_is_blkdev_sb(struct super_block *sb) -{ - return 0; -} - static inline int __sync_blockdev(struct block_device *bdev, int wait) { return 0; @@ -52,28 +42,17 @@ static inline int __sync_blockdev(struct block_device *bdev, int wait) extern void __init chrdev_init(void); /* - * exec.c - */ -extern int check_unsafe_exec(struct linux_binprm *); - -/* * namespace.c */ extern int copy_mount_options(const void __user *, unsigned long *); extern int copy_mount_string(const void __user *, char **); -extern unsigned int mnt_get_count(struct vfsmount *mnt); -extern struct vfsmount *__lookup_mnt(struct vfsmount *, struct dentry *, int); extern struct vfsmount *lookup_mnt(struct path *); -extern void mnt_set_mountpoint(struct vfsmount *, struct dentry *, - struct vfsmount *); -extern void release_mounts(struct list_head *); -extern void umount_tree(struct vfsmount *, int, struct list_head *); -extern struct vfsmount *copy_tree(struct vfsmount *, struct dentry *, int); extern int finish_automount(struct vfsmount *, struct path *); extern void mnt_make_longterm(struct vfsmount *); extern void mnt_make_shortterm(struct vfsmount *); +extern int sb_prepare_remount_readonly(struct super_block *); extern void __init mnt_init(void); @@ -98,10 +77,9 @@ extern struct file *get_empty_filp(void); */ extern int do_remount_sb(struct super_block *, int, void *, int); extern bool grab_super_passive(struct super_block *sb); -extern void __put_super(struct super_block *sb); -extern void put_super(struct super_block *sb); extern struct dentry *mount_fs(struct file_system_type *, int, const char *, void *); +extern struct super_block *user_get_super(dev_t); /* * open.c @@ -111,7 +89,7 @@ extern struct file *nameidata_to_filp(struct nameidata *); extern void release_open_intent(struct nameidata *); struct open_flags { int open_flag; - int mode; + umode_t mode; int acc_mode; int intent; }; diff --git a/fs/ioctl.c b/fs/ioctl.c index 1d9b9fcb2db4..066836e81848 100644 --- a/fs/ioctl.c +++ b/fs/ioctl.c @@ -42,7 +42,7 @@ static long vfs_ioctl(struct file *filp, unsigned int cmd, error = filp->f_op->unlocked_ioctl(filp, cmd, arg); if (error == -ENOIOCTLCMD) - error = -EINVAL; + error = -ENOTTY; out: return error; } diff --git a/fs/ioprio.c b/fs/ioprio.c index f79dab83e17b..f84b380d65e5 100644 --- a/fs/ioprio.c +++ b/fs/ioprio.c @@ -48,28 +48,12 @@ int set_task_ioprio(struct task_struct *task, int ioprio) if (err) return err; - task_lock(task); - do { - ioc = task->io_context; - /* see wmb() in current_io_context() */ - smp_read_barrier_depends(); - if (ioc) - break; - - ioc = alloc_io_context(GFP_ATOMIC, -1); - if (!ioc) { - err = -ENOMEM; - break; - } - task->io_context = ioc; - } while (1); - - if (!err) { - ioc->ioprio = ioprio; - ioc->ioprio_changed = 1; + ioc = get_task_io_context(task, GFP_ATOMIC, NUMA_NO_NODE); + if (ioc) { + ioc_ioprio_changed(ioc, ioprio); + put_io_context(ioc, NULL); } - task_unlock(task); return err; } EXPORT_SYMBOL_GPL(set_task_ioprio); diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index f950059525fc..bd62c76fb5df 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -85,7 +85,6 @@ static struct inode *isofs_alloc_inode(struct super_block *sb) static void isofs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(isofs_inode_cachep, ISOFS_I(inode)); } @@ -170,8 +169,8 @@ struct iso9660_options{ unsigned char map; unsigned char check; unsigned int blocksize; - mode_t fmode; - mode_t dmode; + umode_t fmode; + umode_t dmode; gid_t gid; uid_t uid; char *iocharset; @@ -949,8 +948,11 @@ root_found: /* get the root dentry */ s->s_root = d_alloc_root(inode); - if (!(s->s_root)) - goto out_no_root; + if (!(s->s_root)) { + iput(inode); + error = -ENOMEM; + goto out_no_inode; + } kfree(opt.iocharset); diff --git a/fs/isofs/isofs.h b/fs/isofs/isofs.h index 7d33de84f52a..0e73f63d9274 100644 --- a/fs/isofs/isofs.h +++ b/fs/isofs/isofs.h @@ -50,14 +50,14 @@ struct isofs_sb_info { unsigned int s_uid_set:1; unsigned int s_gid_set:1; - mode_t s_fmode; - mode_t s_dmode; + umode_t s_fmode; + umode_t s_dmode; gid_t s_gid; uid_t s_uid; struct nls_table *s_nls_iocharset; /* Native language support table */ }; -#define ISOFS_INVALID_MODE ((mode_t) -1) +#define ISOFS_INVALID_MODE ((umode_t) -1) static inline struct isofs_sb_info *ISOFS_SB(struct super_block *sb) { diff --git a/fs/jbd/checkpoint.c b/fs/jbd/checkpoint.c index f94fc48ff3a0..5d1a00a5041b 100644 --- a/fs/jbd/checkpoint.c +++ b/fs/jbd/checkpoint.c @@ -537,7 +537,7 @@ int cleanup_journal_tail(journal_t *journal) * them. * * Called with j_list_lock held. - * Returns number of bufers reaped (for debug) + * Returns number of buffers reaped (for debug) */ static int journal_clean_one_cp_list(struct journal_head *jh, int *released) diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c index 8799207df058..f2b9a571f4cf 100644 --- a/fs/jbd/commit.c +++ b/fs/jbd/commit.c @@ -392,6 +392,12 @@ void journal_commit_transaction(journal_t *journal) jbd_debug (3, "JBD: commit phase 1\n"); /* + * Clear revoked flag to reflect there is no revoked buffers + * in the next transaction which is going to be started. + */ + journal_clear_buffer_revoked_flags(journal); + + /* * Switch to a new revoke table. */ journal_switch_revoke_table(journal); diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index fea8dd661d2b..59c09f9541b5 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -166,7 +166,7 @@ loop: */ jbd_debug(1, "Now suspending kjournald\n"); spin_unlock(&journal->j_state_lock); - refrigerator(); + try_to_freeze(); spin_lock(&journal->j_state_lock); } else { /* @@ -721,7 +721,6 @@ static journal_t * journal_init_common (void) init_waitqueue_head(&journal->j_wait_checkpoint); init_waitqueue_head(&journal->j_wait_commit); init_waitqueue_head(&journal->j_wait_updates); - mutex_init(&journal->j_barrier); mutex_init(&journal->j_checkpoint_mutex); spin_lock_init(&journal->j_revoke_lock); spin_lock_init(&journal->j_list_lock); diff --git a/fs/jbd/revoke.c b/fs/jbd/revoke.c index 305a90763154..25c713e7071c 100644 --- a/fs/jbd/revoke.c +++ b/fs/jbd/revoke.c @@ -47,6 +47,10 @@ * overwriting the new data. We don't even need to clear the revoke * bit here. * + * We cache revoke status of a buffer in the current transaction in b_states + * bits. As the name says, revokevalid flag indicates that the cached revoke + * status of a buffer is valid and we can rely on the cached status. + * * Revoke information on buffers is a tri-state value: * * RevokeValid clear: no cached revoke status, need to look it up @@ -479,6 +483,36 @@ int journal_cancel_revoke(handle_t *handle, struct journal_head *jh) return did_revoke; } +/* + * journal_clear_revoked_flags clears revoked flag of buffers in + * revoke table to reflect there is no revoked buffer in the next + * transaction which is going to be started. + */ +void journal_clear_buffer_revoked_flags(journal_t *journal) +{ + struct jbd_revoke_table_s *revoke = journal->j_revoke; + int i = 0; + + for (i = 0; i < revoke->hash_size; i++) { + struct list_head *hash_list; + struct list_head *list_entry; + hash_list = &revoke->hash_table[i]; + + list_for_each(list_entry, hash_list) { + struct jbd_revoke_record_s *record; + struct buffer_head *bh; + record = (struct jbd_revoke_record_s *)list_entry; + bh = __find_get_block(journal->j_fs_dev, + record->blocknr, + journal->j_blocksize); + if (bh) { + clear_buffer_revoked(bh); + __brelse(bh); + } + } + } +} + /* journal_switch_revoke table select j_revoke for next transaction * we do not want to suspend any processing until all revokes are * written -bzzz diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index 7e59c6e66f9b..7fce94b04bc3 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -426,17 +426,34 @@ int journal_restart(handle_t *handle, int nblocks) * void journal_lock_updates () - establish a transaction barrier. * @journal: Journal to establish a barrier on. * - * This locks out any further updates from being started, and blocks - * until all existing updates have completed, returning only once the - * journal is in a quiescent state with no updates running. - * - * The journal lock should not be held on entry. + * This locks out any further updates from being started, and blocks until all + * existing updates have completed, returning only once the journal is in a + * quiescent state with no updates running. + * + * We do not use simple mutex for synchronization as there are syscalls which + * want to return with filesystem locked and that trips up lockdep. Also + * hibernate needs to lock filesystem but locked mutex then blocks hibernation. + * Since locking filesystem is rare operation, we use simple counter and + * waitqueue for locking. */ void journal_lock_updates(journal_t *journal) { DEFINE_WAIT(wait); +wait: + /* Wait for previous locked operation to finish */ + wait_event(journal->j_wait_transaction_locked, + journal->j_barrier_count == 0); + spin_lock(&journal->j_state_lock); + /* + * Check reliably under the lock whether we are the ones winning the race + * and locking the journal + */ + if (journal->j_barrier_count > 0) { + spin_unlock(&journal->j_state_lock); + goto wait; + } ++journal->j_barrier_count; /* Wait until there are no running updates */ @@ -460,14 +477,6 @@ void journal_lock_updates(journal_t *journal) spin_lock(&journal->j_state_lock); } spin_unlock(&journal->j_state_lock); - - /* - * We have now established a barrier against other normal updates, but - * we also need to barrier against other journal_lock_updates() calls - * to make sure that we serialise special journal-locked operations - * too. - */ - mutex_lock(&journal->j_barrier); } /** @@ -475,14 +484,11 @@ void journal_lock_updates(journal_t *journal) * @journal: Journal to release the barrier on. * * Release a transaction barrier obtained with journal_lock_updates(). - * - * Should be called without the journal lock held. */ void journal_unlock_updates (journal_t *journal) { J_ASSERT(journal->j_barrier_count != 0); - mutex_unlock(&journal->j_barrier); spin_lock(&journal->j_state_lock); --journal->j_barrier_count; spin_unlock(&journal->j_state_lock); diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index 16a698bd906d..d49d202903fb 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -565,7 +565,7 @@ int jbd2_cleanup_journal_tail(journal_t *journal) * * Called with the journal locked. * Called with j_list_lock held. - * Returns number of bufers reaped (for debug) + * Returns number of buffers reaped (for debug) */ static int journal_clean_one_cp_list(struct journal_head *jh, int *released) diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 68d704db787f..5069b8475150 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -430,6 +430,12 @@ void jbd2_journal_commit_transaction(journal_t *journal) jbd_debug(3, "JBD2: commit phase 1\n"); /* + * Clear revoked flag to reflect there is no revoked buffers + * in the next transaction which is going to be started. + */ + jbd2_clear_buffer_revoked_flags(journal); + + /* * Switch to a new revoke table. */ jbd2_journal_switch_revoke_table(journal); diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 0fa0123151d3..c0a5f9f1b127 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -173,7 +173,7 @@ loop: */ jbd_debug(1, "Now suspending kjournald2\n"); write_unlock(&journal->j_state_lock); - refrigerator(); + try_to_freeze(); write_lock(&journal->j_state_lock); } else { /* diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index 69fd93588118..30b2867d6cc9 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -47,6 +47,10 @@ * overwriting the new data. We don't even need to clear the revoke * bit here. * + * We cache revoke status of a buffer in the current transaction in b_states + * bits. As the name says, revokevalid flag indicates that the cached revoke + * status of a buffer is valid and we can rely on the cached status. + * * Revoke information on buffers is a tri-state value: * * RevokeValid clear: no cached revoke status, need to look it up @@ -478,6 +482,36 @@ int jbd2_journal_cancel_revoke(handle_t *handle, struct journal_head *jh) return did_revoke; } +/* + * journal_clear_revoked_flag clears revoked flag of buffers in + * revoke table to reflect there is no revoked buffers in the next + * transaction which is going to be started. + */ +void jbd2_clear_buffer_revoked_flags(journal_t *journal) +{ + struct jbd2_revoke_table_s *revoke = journal->j_revoke; + int i = 0; + + for (i = 0; i < revoke->hash_size; i++) { + struct list_head *hash_list; + struct list_head *list_entry; + hash_list = &revoke->hash_table[i]; + + list_for_each(list_entry, hash_list) { + struct jbd2_revoke_record_s *record; + struct buffer_head *bh; + record = (struct jbd2_revoke_record_s *)list_entry; + bh = __find_get_block(journal->j_fs_dev, + record->blocknr, + journal->j_blocksize); + if (bh) { + clear_buffer_revoked(bh); + __brelse(bh); + } + } + } +} + /* journal_switch_revoke table select j_revoke for next transaction * we do not want to suspend any processing until all revokes are * written -bzzz diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index a0e41a4c080e..35ae096bed5d 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -517,12 +517,13 @@ void jbd2_journal_lock_updates(journal_t *journal) break; spin_lock(&transaction->t_handle_lock); + prepare_to_wait(&journal->j_wait_updates, &wait, + TASK_UNINTERRUPTIBLE); if (!atomic_read(&transaction->t_updates)) { spin_unlock(&transaction->t_handle_lock); + finish_wait(&journal->j_wait_updates, &wait); break; } - prepare_to_wait(&journal->j_wait_updates, &wait, - TASK_UNINTERRUPTIBLE); spin_unlock(&transaction->t_handle_lock); write_unlock(&journal->j_state_lock); schedule(); diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c index be6169bd8acd..973ac5822bd7 100644 --- a/fs/jffs2/dir.c +++ b/fs/jffs2/dir.c @@ -22,16 +22,16 @@ static int jffs2_readdir (struct file *, void *, filldir_t); -static int jffs2_create (struct inode *,struct dentry *,int, +static int jffs2_create (struct inode *,struct dentry *,umode_t, struct nameidata *); static struct dentry *jffs2_lookup (struct inode *,struct dentry *, struct nameidata *); static int jffs2_link (struct dentry *,struct inode *,struct dentry *); static int jffs2_unlink (struct inode *,struct dentry *); static int jffs2_symlink (struct inode *,struct dentry *,const char *); -static int jffs2_mkdir (struct inode *,struct dentry *,int); +static int jffs2_mkdir (struct inode *,struct dentry *,umode_t); static int jffs2_rmdir (struct inode *,struct dentry *); -static int jffs2_mknod (struct inode *,struct dentry *,int,dev_t); +static int jffs2_mknod (struct inode *,struct dentry *,umode_t,dev_t); static int jffs2_rename (struct inode *, struct dentry *, struct inode *, struct dentry *); @@ -169,8 +169,8 @@ static int jffs2_readdir(struct file *filp, void *dirent, filldir_t filldir) /***********************************************************************/ -static int jffs2_create(struct inode *dir_i, struct dentry *dentry, int mode, - struct nameidata *nd) +static int jffs2_create(struct inode *dir_i, struct dentry *dentry, + umode_t mode, struct nameidata *nd) { struct jffs2_raw_inode *ri; struct jffs2_inode_info *f, *dir_f; @@ -450,7 +450,7 @@ static int jffs2_symlink (struct inode *dir_i, struct dentry *dentry, const char } -static int jffs2_mkdir (struct inode *dir_i, struct dentry *dentry, int mode) +static int jffs2_mkdir (struct inode *dir_i, struct dentry *dentry, umode_t mode) { struct jffs2_inode_info *f, *dir_f; struct jffs2_sb_info *c; @@ -618,7 +618,7 @@ static int jffs2_rmdir (struct inode *dir_i, struct dentry *dentry) return ret; } -static int jffs2_mknod (struct inode *dir_i, struct dentry *dentry, int mode, dev_t rdev) +static int jffs2_mknod (struct inode *dir_i, struct dentry *dentry, umode_t mode, dev_t rdev) { struct jffs2_inode_info *f, *dir_f; struct jffs2_sb_info *c; diff --git a/fs/jffs2/erase.c b/fs/jffs2/erase.c index e513f1913c15..a01cdad6aad1 100644 --- a/fs/jffs2/erase.c +++ b/fs/jffs2/erase.c @@ -74,7 +74,7 @@ static void jffs2_erase_block(struct jffs2_sb_info *c, ((struct erase_priv_struct *)instr->priv)->jeb = jeb; ((struct erase_priv_struct *)instr->priv)->c = c; - ret = c->mtd->erase(c->mtd, instr); + ret = mtd_erase(c->mtd, instr); if (!ret) return; @@ -336,12 +336,11 @@ static int jffs2_block_check_erase(struct jffs2_sb_info *c, struct jffs2_erasebl uint32_t ofs; size_t retlen; int ret = -EIO; + unsigned long *wordebuf; - if (c->mtd->point) { - unsigned long *wordebuf; - - ret = c->mtd->point(c->mtd, jeb->offset, c->sector_size, - &retlen, &ebuf, NULL); + ret = mtd_point(c->mtd, jeb->offset, c->sector_size, &retlen, + &ebuf, NULL); + if (ret != -EOPNOTSUPP) { if (ret) { D1(printk(KERN_DEBUG "MTD point failed %d\n", ret)); goto do_flash_read; @@ -349,7 +348,7 @@ static int jffs2_block_check_erase(struct jffs2_sb_info *c, struct jffs2_erasebl if (retlen < c->sector_size) { /* Don't muck about if it won't let us point to the whole erase sector */ D1(printk(KERN_DEBUG "MTD point returned len too short: 0x%zx\n", retlen)); - c->mtd->unpoint(c->mtd, jeb->offset, retlen); + mtd_unpoint(c->mtd, jeb->offset, retlen); goto do_flash_read; } wordebuf = ebuf-sizeof(*wordebuf); @@ -358,7 +357,7 @@ static int jffs2_block_check_erase(struct jffs2_sb_info *c, struct jffs2_erasebl if (*++wordebuf != ~0) break; } while(--retlen); - c->mtd->unpoint(c->mtd, jeb->offset, c->sector_size); + mtd_unpoint(c->mtd, jeb->offset, c->sector_size); if (retlen) { printk(KERN_WARNING "Newly-erased block contained word 0x%lx at offset 0x%08tx\n", *wordebuf, jeb->offset + c->sector_size-retlen*sizeof(*wordebuf)); @@ -381,7 +380,7 @@ static int jffs2_block_check_erase(struct jffs2_sb_info *c, struct jffs2_erasebl *bad_offset = ofs; - ret = c->mtd->read(c->mtd, ofs, readlen, &retlen, ebuf); + ret = mtd_read(c->mtd, ofs, readlen, &retlen, ebuf); if (ret) { printk(KERN_WARNING "Read of newly-erased block at 0x%08x failed: %d. Putting on bad_list\n", ofs, ret); ret = -EIO; diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c index 4b8afe39a87f..2e0123867cb1 100644 --- a/fs/jffs2/fs.c +++ b/fs/jffs2/fs.c @@ -466,7 +466,6 @@ struct inode *jffs2_new_inode (struct inode *dir_i, umode_t mode, struct jffs2_r if (insert_inode_locked(inode) < 0) { make_bad_inode(inode); - unlock_new_inode(inode); iput(inode); return ERR_PTR(-EINVAL); } diff --git a/fs/jffs2/readinode.c b/fs/jffs2/readinode.c index ee57bac1ba6d..3093ac4fb24c 100644 --- a/fs/jffs2/readinode.c +++ b/fs/jffs2/readinode.c @@ -62,17 +62,15 @@ static int check_node_data(struct jffs2_sb_info *c, struct jffs2_tmp_dnode_info #ifndef __ECOS /* TODO: instead, incapsulate point() stuff to jffs2_flash_read(), * adding and jffs2_flash_read_end() interface. */ - if (c->mtd->point) { - err = c->mtd->point(c->mtd, ofs, len, &retlen, - (void **)&buffer, NULL); - if (!err && retlen < len) { - JFFS2_WARNING("MTD point returned len too short: %zu instead of %u.\n", retlen, tn->csize); - c->mtd->unpoint(c->mtd, ofs, retlen); - } else if (err) + err = mtd_point(c->mtd, ofs, len, &retlen, (void **)&buffer, NULL); + if (!err && retlen < len) { + JFFS2_WARNING("MTD point returned len too short: %zu instead of %u.\n", retlen, tn->csize); + mtd_unpoint(c->mtd, ofs, retlen); + } else if (err) { + if (err != -EOPNOTSUPP) JFFS2_WARNING("MTD point failed: error code %d.\n", err); - else - pointed = 1; /* succefully pointed to device */ - } + } else + pointed = 1; /* succefully pointed to device */ #endif if (!pointed) { @@ -101,7 +99,7 @@ static int check_node_data(struct jffs2_sb_info *c, struct jffs2_tmp_dnode_info kfree(buffer); #ifndef __ECOS else - c->mtd->unpoint(c->mtd, ofs, len); + mtd_unpoint(c->mtd, ofs, len); #endif if (crc != tn->data_crc) { @@ -137,7 +135,7 @@ free_out: kfree(buffer); #ifndef __ECOS else - c->mtd->unpoint(c->mtd, ofs, len); + mtd_unpoint(c->mtd, ofs, len); #endif return err; } diff --git a/fs/jffs2/scan.c b/fs/jffs2/scan.c index 28107ca136e4..f99464833bb2 100644 --- a/fs/jffs2/scan.c +++ b/fs/jffs2/scan.c @@ -97,15 +97,15 @@ int jffs2_scan_medium(struct jffs2_sb_info *c) size_t pointlen, try_size; if (c->mtd->point) { - ret = c->mtd->point(c->mtd, 0, c->mtd->size, &pointlen, - (void **)&flashbuf, NULL); + ret = mtd_point(c->mtd, 0, c->mtd->size, &pointlen, + (void **)&flashbuf, NULL); if (!ret && pointlen < c->mtd->size) { /* Don't muck about if it won't let us point to the whole flash */ D1(printk(KERN_DEBUG "MTD point returned len too short: 0x%zx\n", pointlen)); - c->mtd->unpoint(c->mtd, 0, pointlen); + mtd_unpoint(c->mtd, 0, pointlen); flashbuf = NULL; } - if (ret) + if (ret && ret != -EOPNOTSUPP) D1(printk(KERN_DEBUG "MTD point failed %d\n", ret)); } #endif @@ -273,7 +273,7 @@ int jffs2_scan_medium(struct jffs2_sb_info *c) kfree(flashbuf); #ifndef __ECOS else - c->mtd->unpoint(c->mtd, 0, c->mtd->size); + mtd_unpoint(c->mtd, 0, c->mtd->size); #endif kfree(s); return ret; @@ -455,7 +455,7 @@ static int jffs2_scan_eraseblock (struct jffs2_sb_info *c, struct jffs2_eraseblo if (jffs2_cleanmarker_oob(c)) { int ret; - if (c->mtd->block_isbad(c->mtd, jeb->offset)) + if (mtd_block_isbad(c->mtd, jeb->offset)) return BLK_STATE_BADBLOCK; ret = jffs2_check_nand_cleanmarker(c, jeb); diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index e7e974454115..f2d96b5e64f6 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -45,7 +45,6 @@ static struct inode *jffs2_alloc_inode(struct super_block *sb) static void jffs2_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(jffs2_inode_cachep, JFFS2_INODE_INFO(inode)); } @@ -97,9 +96,9 @@ static const char *jffs2_compr_name(unsigned int compr) } } -static int jffs2_show_options(struct seq_file *s, struct vfsmount *mnt) +static int jffs2_show_options(struct seq_file *s, struct dentry *root) { - struct jffs2_sb_info *c = JFFS2_SB_INFO(mnt->mnt_sb); + struct jffs2_sb_info *c = JFFS2_SB_INFO(root->d_sb); struct jffs2_mount_opts *opts = &c->mount_opts; if (opts->override_compr) @@ -336,9 +335,7 @@ static void jffs2_put_super (struct super_block *sb) jffs2_flash_cleanup(c); kfree(c->inocache_list); jffs2_clear_xattr_subsystem(c); - if (c->mtd->sync) - c->mtd->sync(c->mtd); - + mtd_sync(c->mtd); D1(printk(KERN_DEBUG "jffs2_put_super returning\n")); } diff --git a/fs/jffs2/wbuf.c b/fs/jffs2/wbuf.c index b09e51d2f81f..30e8f47e8a23 100644 --- a/fs/jffs2/wbuf.c +++ b/fs/jffs2/wbuf.c @@ -228,7 +228,7 @@ static int jffs2_verify_write(struct jffs2_sb_info *c, unsigned char *buf, size_t retlen; char *eccstr; - ret = c->mtd->read(c->mtd, ofs, c->wbuf_pagesize, &retlen, c->wbuf_verify); + ret = mtd_read(c->mtd, ofs, c->wbuf_pagesize, &retlen, c->wbuf_verify); if (ret && ret != -EUCLEAN && ret != -EBADMSG) { printk(KERN_WARNING "jffs2_verify_write(): Read back of page at %08x failed: %d\n", c->wbuf_ofs, ret); return ret; @@ -337,7 +337,8 @@ static void jffs2_wbuf_recover(struct jffs2_sb_info *c) } /* Do the read... */ - ret = c->mtd->read(c->mtd, start, c->wbuf_ofs - start, &retlen, buf); + ret = mtd_read(c->mtd, start, c->wbuf_ofs - start, &retlen, + buf); /* ECC recovered ? */ if ((ret == -EUCLEAN || ret == -EBADMSG) && @@ -413,13 +414,12 @@ static void jffs2_wbuf_recover(struct jffs2_sb_info *c) if (breakme++ == 20) { printk(KERN_NOTICE "Faking write error at 0x%08x\n", ofs); breakme = 0; - c->mtd->write(c->mtd, ofs, towrite, &retlen, - brokenbuf); + mtd_write(c->mtd, ofs, towrite, &retlen, brokenbuf); ret = -EIO; } else #endif - ret = c->mtd->write(c->mtd, ofs, towrite, &retlen, - rewrite_buf); + ret = mtd_write(c->mtd, ofs, towrite, &retlen, + rewrite_buf); if (ret || retlen != towrite || jffs2_verify_write(c, rewrite_buf, ofs)) { /* Argh. We tried. Really we did. */ @@ -619,13 +619,14 @@ static int __jffs2_flush_wbuf(struct jffs2_sb_info *c, int pad) if (breakme++ == 20) { printk(KERN_NOTICE "Faking write error at 0x%08x\n", c->wbuf_ofs); breakme = 0; - c->mtd->write(c->mtd, c->wbuf_ofs, c->wbuf_pagesize, &retlen, - brokenbuf); + mtd_write(c->mtd, c->wbuf_ofs, c->wbuf_pagesize, &retlen, + brokenbuf); ret = -EIO; } else #endif - ret = c->mtd->write(c->mtd, c->wbuf_ofs, c->wbuf_pagesize, &retlen, c->wbuf); + ret = mtd_write(c->mtd, c->wbuf_ofs, c->wbuf_pagesize, + &retlen, c->wbuf); if (ret) { printk(KERN_WARNING "jffs2_flush_wbuf(): Write failed with %d\n", ret); @@ -861,8 +862,8 @@ int jffs2_flash_writev(struct jffs2_sb_info *c, const struct kvec *invecs, v += wbuf_retlen; if (vlen >= c->wbuf_pagesize) { - ret = c->mtd->write(c->mtd, outvec_to, PAGE_DIV(vlen), - &wbuf_retlen, v); + ret = mtd_write(c->mtd, outvec_to, PAGE_DIV(vlen), + &wbuf_retlen, v); if (ret < 0 || wbuf_retlen != PAGE_DIV(vlen)) goto outfile; @@ -948,11 +949,11 @@ int jffs2_flash_read(struct jffs2_sb_info *c, loff_t ofs, size_t len, size_t *re int ret; if (!jffs2_is_writebuffered(c)) - return c->mtd->read(c->mtd, ofs, len, retlen, buf); + return mtd_read(c->mtd, ofs, len, retlen, buf); /* Read flash */ down_read(&c->wbuf_sem); - ret = c->mtd->read(c->mtd, ofs, len, retlen, buf); + ret = mtd_read(c->mtd, ofs, len, retlen, buf); if ( (ret == -EBADMSG || ret == -EUCLEAN) && (*retlen == len) ) { if (ret == -EBADMSG) @@ -1031,7 +1032,7 @@ int jffs2_check_oob_empty(struct jffs2_sb_info *c, ops.len = ops.ooboffs = ops.retlen = ops.oobretlen = 0; ops.datbuf = NULL; - ret = c->mtd->read_oob(c->mtd, jeb->offset, &ops); + ret = mtd_read_oob(c->mtd, jeb->offset, &ops); if (ret || ops.oobretlen != ops.ooblen) { printk(KERN_ERR "cannot read OOB for EB at %08x, requested %zd" " bytes, read %zd bytes, error %d\n", @@ -1074,7 +1075,7 @@ int jffs2_check_nand_cleanmarker(struct jffs2_sb_info *c, ops.len = ops.ooboffs = ops.retlen = ops.oobretlen = 0; ops.datbuf = NULL; - ret = c->mtd->read_oob(c->mtd, jeb->offset, &ops); + ret = mtd_read_oob(c->mtd, jeb->offset, &ops); if (ret || ops.oobretlen != ops.ooblen) { printk(KERN_ERR "cannot read OOB for EB at %08x, requested %zd" " bytes, read %zd bytes, error %d\n", @@ -1100,7 +1101,7 @@ int jffs2_write_nand_cleanmarker(struct jffs2_sb_info *c, ops.len = ops.ooboffs = ops.retlen = ops.oobretlen = 0; ops.datbuf = NULL; - ret = c->mtd->write_oob(c->mtd, jeb->offset, &ops); + ret = mtd_write_oob(c->mtd, jeb->offset, &ops); if (ret || ops.oobretlen != ops.ooblen) { printk(KERN_ERR "cannot write OOB for EB at %08x, requested %zd" " bytes, read %zd bytes, error %d\n", @@ -1129,11 +1130,8 @@ int jffs2_write_nand_badblock(struct jffs2_sb_info *c, struct jffs2_eraseblock * if( ++jeb->bad_count < MAX_ERASE_FAILURES) return 0; - if (!c->mtd->block_markbad) - return 1; // What else can we do? - printk(KERN_WARNING "JFFS2: marking eraseblock at %08x\n as bad", bad_offset); - ret = c->mtd->block_markbad(c->mtd, bad_offset); + ret = mtd_block_markbad(c->mtd, bad_offset); if (ret) { D1(printk(KERN_WARNING "jffs2_write_nand_badblock(): Write failed for block at %08x: error %d\n", jeb->offset, ret)); diff --git a/fs/jffs2/writev.c b/fs/jffs2/writev.c index b9276b11bac6..a1bda9dab3f8 100644 --- a/fs/jffs2/writev.c +++ b/fs/jffs2/writev.c @@ -13,30 +13,6 @@ #include <linux/mtd/mtd.h> #include "nodelist.h" -/* This ought to be in core MTD code. All registered MTD devices - without writev should have this put in place. Bug the MTD - maintainer */ -static inline int mtd_fake_writev(struct mtd_info *mtd, const struct kvec *vecs, - unsigned long count, loff_t to, size_t *retlen) -{ - unsigned long i; - size_t totlen = 0, thislen; - int ret = 0; - - for (i=0; i<count; i++) { - if (!vecs[i].iov_len) - continue; - ret = mtd->write(mtd, to, vecs[i].iov_len, &thislen, vecs[i].iov_base); - totlen += thislen; - if (ret || thislen != vecs[i].iov_len) - break; - to += vecs[i].iov_len; - } - if (retlen) - *retlen = totlen; - return ret; -} - int jffs2_flash_direct_writev(struct jffs2_sb_info *c, const struct kvec *vecs, unsigned long count, loff_t to, size_t *retlen) { @@ -50,18 +26,14 @@ int jffs2_flash_direct_writev(struct jffs2_sb_info *c, const struct kvec *vecs, } } - if (c->mtd->writev) - return c->mtd->writev(c->mtd, vecs, count, to, retlen); - else { - return mtd_fake_writev(c->mtd, vecs, count, to, retlen); - } + return mtd_writev(c->mtd, vecs, count, to, retlen); } int jffs2_flash_direct_write(struct jffs2_sb_info *c, loff_t ofs, size_t len, size_t *retlen, const u_char *buf) { int ret; - ret = c->mtd->write(c->mtd, ofs, len, retlen, buf); + ret = mtd_write(c->mtd, ofs, len, retlen, buf); if (jffs2_sum_active()) { struct kvec vecs[1]; diff --git a/fs/jfs/ioctl.c b/fs/jfs/ioctl.c index 6f98a1866776..f19d1e04a374 100644 --- a/fs/jfs/ioctl.c +++ b/fs/jfs/ioctl.c @@ -68,7 +68,7 @@ long jfs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) unsigned int oldflags; int err; - err = mnt_want_write(filp->f_path.mnt); + err = mnt_want_write_file(filp); if (err) return err; @@ -120,7 +120,7 @@ long jfs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) inode->i_ctime = CURRENT_TIME_SEC; mark_inode_dirty(inode); setflags_out: - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return err; } default: diff --git a/fs/jfs/jfs_logmgr.c b/fs/jfs/jfs_logmgr.c index cc5f811ed383..2eb952c41a69 100644 --- a/fs/jfs/jfs_logmgr.c +++ b/fs/jfs/jfs_logmgr.c @@ -2349,7 +2349,7 @@ int jfsIOWait(void *arg) if (freezing(current)) { spin_unlock_irq(&log_redrive_lock); - refrigerator(); + try_to_freeze(); } else { set_current_state(TASK_INTERRUPTIBLE); spin_unlock_irq(&log_redrive_lock); diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c index af9606057dde..bb8b661bcc50 100644 --- a/fs/jfs/jfs_txnmgr.c +++ b/fs/jfs/jfs_txnmgr.c @@ -2800,7 +2800,7 @@ int jfs_lazycommit(void *arg) if (freezing(current)) { LAZY_UNLOCK(flags); - refrigerator(); + try_to_freeze(); } else { DECLARE_WAITQUEUE(wq, current); @@ -2994,7 +2994,7 @@ int jfs_sync(void *arg) if (freezing(current)) { TXN_UNLOCK(); - refrigerator(); + try_to_freeze(); } else { set_current_state(TASK_INTERRUPTIBLE); TXN_UNLOCK(); diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c index a112ad96e474..5f7c160ea64f 100644 --- a/fs/jfs/namei.c +++ b/fs/jfs/namei.c @@ -72,7 +72,7 @@ static inline void free_ea_wmap(struct inode *inode) * RETURN: Errors from subroutines * */ -static int jfs_create(struct inode *dip, struct dentry *dentry, int mode, +static int jfs_create(struct inode *dip, struct dentry *dentry, umode_t mode, struct nameidata *nd) { int rc = 0; @@ -205,7 +205,7 @@ static int jfs_create(struct inode *dip, struct dentry *dentry, int mode, * note: * EACCESS: user needs search+write permission on the parent directory */ -static int jfs_mkdir(struct inode *dip, struct dentry *dentry, int mode) +static int jfs_mkdir(struct inode *dip, struct dentry *dentry, umode_t mode) { int rc = 0; tid_t tid; /* transaction id */ @@ -1353,7 +1353,7 @@ static int jfs_rename(struct inode *old_dir, struct dentry *old_dentry, * FUNCTION: Create a special file (device) */ static int jfs_mknod(struct inode *dir, struct dentry *dentry, - int mode, dev_t rdev) + umode_t mode, dev_t rdev) { struct jfs_inode_info *jfs_ip; struct btstack btstack; diff --git a/fs/jfs/super.c b/fs/jfs/super.c index a44eff076c17..682bca642f38 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -119,7 +119,6 @@ static void jfs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); struct jfs_inode_info *ji = JFS_IP(inode); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(jfs_inode_cachep, ji); } @@ -609,9 +608,9 @@ static int jfs_sync_fs(struct super_block *sb, int wait) return 0; } -static int jfs_show_options(struct seq_file *seq, struct vfsmount *vfs) +static int jfs_show_options(struct seq_file *seq, struct dentry *root) { - struct jfs_sb_info *sbi = JFS_SBI(vfs->mnt_sb); + struct jfs_sb_info *sbi = JFS_SBI(root->d_sb); if (sbi->uid != -1) seq_printf(seq, ",uid=%d", sbi->uid); diff --git a/fs/libfs.c b/fs/libfs.c index f6d411eef1e7..5b2dbb3ba4fc 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -12,7 +12,7 @@ #include <linux/mutex.h> #include <linux/exportfs.h> #include <linux/writeback.h> -#include <linux/buffer_head.h> +#include <linux/buffer_head.h> /* sync_mapping_buffers */ #include <asm/uaccess.h> diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c index 23d7451b2938..65ba36b80a9e 100644 --- a/fs/lockd/mon.c +++ b/fs/lockd/mon.c @@ -55,7 +55,7 @@ static DEFINE_SPINLOCK(nsm_lock); * Local NSM state */ u32 __read_mostly nsm_local_state; -int __read_mostly nsm_use_hostnames; +bool __read_mostly nsm_use_hostnames; static inline struct sockaddr *nsm_addr(const struct nsm_handle *nsm) { diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c index 1ca0679c80bf..2240d384d787 100644 --- a/fs/lockd/svcsubs.c +++ b/fs/lockd/svcsubs.c @@ -403,7 +403,7 @@ nlmsvc_match_sb(void *datap, struct nlm_file *file) { struct super_block *sb = datap; - return sb == file->f_file->f_path.mnt->mnt_sb; + return sb == file->f_file->f_path.dentry->d_sb; } /** diff --git a/fs/locks.c b/fs/locks.c index 3b0d05dcd7c1..637694bf3a03 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -1205,6 +1205,8 @@ int __break_lease(struct inode *inode, unsigned int mode) int want_write = (mode & O_ACCMODE) != O_RDONLY; new_fl = lease_alloc(NULL, want_write ? F_WRLCK : F_RDLCK); + if (IS_ERR(new_fl)) + return PTR_ERR(new_fl); lock_flocks(); @@ -1221,12 +1223,6 @@ int __break_lease(struct inode *inode, unsigned int mode) if (fl->fl_owner == current->files) i_have_this_lease = 1; - if (IS_ERR(new_fl) && !i_have_this_lease - && ((mode & O_NONBLOCK) == 0)) { - error = PTR_ERR(new_fl); - goto out; - } - break_time = 0; if (lease_break_time > 0) { break_time = jiffies + lease_break_time * HZ; @@ -1284,8 +1280,7 @@ restart: out: unlock_flocks(); - if (!IS_ERR(new_fl)) - locks_free_lock(new_fl); + locks_free_lock(new_fl); return error; } diff --git a/fs/logfs/dev_mtd.c b/fs/logfs/dev_mtd.c index 339e17e9133d..e97404d611e0 100644 --- a/fs/logfs/dev_mtd.c +++ b/fs/logfs/dev_mtd.c @@ -13,13 +13,14 @@ #define PAGE_OFS(ofs) ((ofs) & (PAGE_SIZE-1)) -static int mtd_read(struct super_block *sb, loff_t ofs, size_t len, void *buf) +static int logfs_mtd_read(struct super_block *sb, loff_t ofs, size_t len, + void *buf) { struct mtd_info *mtd = logfs_super(sb)->s_mtd; size_t retlen; int ret; - ret = mtd->read(mtd, ofs, len, &retlen, buf); + ret = mtd_read(mtd, ofs, len, &retlen, buf); BUG_ON(ret == -EINVAL); if (ret) return ret; @@ -31,7 +32,8 @@ static int mtd_read(struct super_block *sb, loff_t ofs, size_t len, void *buf) return 0; } -static int mtd_write(struct super_block *sb, loff_t ofs, size_t len, void *buf) +static int loffs_mtd_write(struct super_block *sb, loff_t ofs, size_t len, + void *buf) { struct logfs_super *super = logfs_super(sb); struct mtd_info *mtd = super->s_mtd; @@ -47,7 +49,7 @@ static int mtd_write(struct super_block *sb, loff_t ofs, size_t len, void *buf) BUG_ON(len > PAGE_CACHE_SIZE); page_start = ofs & PAGE_CACHE_MASK; page_end = PAGE_CACHE_ALIGN(ofs + len) - 1; - ret = mtd->write(mtd, ofs, len, &retlen, buf); + ret = mtd_write(mtd, ofs, len, &retlen, buf); if (ret || (retlen != len)) return -EIO; @@ -60,14 +62,15 @@ static int mtd_write(struct super_block *sb, loff_t ofs, size_t len, void *buf) * asynchronous properties. So just to prevent the first implementor of such * a thing from breaking logfs in 2350, we do the usual pointless dance to * declare a completion variable and wait for completion before returning - * from mtd_erase(). What an exercise in futility! + * from logfs_mtd_erase(). What an exercise in futility! */ static void logfs_erase_callback(struct erase_info *ei) { complete((struct completion *)ei->priv); } -static int mtd_erase_mapping(struct super_block *sb, loff_t ofs, size_t len) +static int logfs_mtd_erase_mapping(struct super_block *sb, loff_t ofs, + size_t len) { struct logfs_super *super = logfs_super(sb); struct address_space *mapping = super->s_mapping_inode->i_mapping; @@ -84,7 +87,7 @@ static int mtd_erase_mapping(struct super_block *sb, loff_t ofs, size_t len) return 0; } -static int mtd_erase(struct super_block *sb, loff_t ofs, size_t len, +static int logfs_mtd_erase(struct super_block *sb, loff_t ofs, size_t len, int ensure_write) { struct mtd_info *mtd = logfs_super(sb)->s_mtd; @@ -102,30 +105,29 @@ static int mtd_erase(struct super_block *sb, loff_t ofs, size_t len, ei.len = len; ei.callback = logfs_erase_callback; ei.priv = (long)&complete; - ret = mtd->erase(mtd, &ei); + ret = mtd_erase(mtd, &ei); if (ret) return -EIO; wait_for_completion(&complete); if (ei.state != MTD_ERASE_DONE) return -EIO; - return mtd_erase_mapping(sb, ofs, len); + return logfs_mtd_erase_mapping(sb, ofs, len); } -static void mtd_sync(struct super_block *sb) +static void logfs_mtd_sync(struct super_block *sb) { struct mtd_info *mtd = logfs_super(sb)->s_mtd; - if (mtd->sync) - mtd->sync(mtd); + mtd_sync(mtd); } -static int mtd_readpage(void *_sb, struct page *page) +static int logfs_mtd_readpage(void *_sb, struct page *page) { struct super_block *sb = _sb; int err; - err = mtd_read(sb, page->index << PAGE_SHIFT, PAGE_SIZE, + err = logfs_mtd_read(sb, page->index << PAGE_SHIFT, PAGE_SIZE, page_address(page)); if (err == -EUCLEAN || err == -EBADMSG) { /* -EBADMSG happens regularly on power failures */ @@ -143,18 +145,18 @@ static int mtd_readpage(void *_sb, struct page *page) return err; } -static struct page *mtd_find_first_sb(struct super_block *sb, u64 *ofs) +static struct page *logfs_mtd_find_first_sb(struct super_block *sb, u64 *ofs) { struct logfs_super *super = logfs_super(sb); struct address_space *mapping = super->s_mapping_inode->i_mapping; - filler_t *filler = mtd_readpage; + filler_t *filler = logfs_mtd_readpage; struct mtd_info *mtd = super->s_mtd; - if (!mtd->block_isbad) + if (!mtd_can_have_bb(mtd)) return NULL; *ofs = 0; - while (mtd->block_isbad(mtd, *ofs)) { + while (mtd_block_isbad(mtd, *ofs)) { *ofs += mtd->erasesize; if (*ofs >= mtd->size) return NULL; @@ -163,18 +165,18 @@ static struct page *mtd_find_first_sb(struct super_block *sb, u64 *ofs) return read_cache_page(mapping, *ofs >> PAGE_SHIFT, filler, sb); } -static struct page *mtd_find_last_sb(struct super_block *sb, u64 *ofs) +static struct page *logfs_mtd_find_last_sb(struct super_block *sb, u64 *ofs) { struct logfs_super *super = logfs_super(sb); struct address_space *mapping = super->s_mapping_inode->i_mapping; - filler_t *filler = mtd_readpage; + filler_t *filler = logfs_mtd_readpage; struct mtd_info *mtd = super->s_mtd; - if (!mtd->block_isbad) + if (!mtd_can_have_bb(mtd)) return NULL; *ofs = mtd->size - mtd->erasesize; - while (mtd->block_isbad(mtd, *ofs)) { + while (mtd_block_isbad(mtd, *ofs)) { *ofs -= mtd->erasesize; if (*ofs <= 0) return NULL; @@ -184,7 +186,7 @@ static struct page *mtd_find_last_sb(struct super_block *sb, u64 *ofs) return read_cache_page(mapping, *ofs >> PAGE_SHIFT, filler, sb); } -static int __mtd_writeseg(struct super_block *sb, u64 ofs, pgoff_t index, +static int __logfs_mtd_writeseg(struct super_block *sb, u64 ofs, pgoff_t index, size_t nr_pages) { struct logfs_super *super = logfs_super(sb); @@ -196,8 +198,8 @@ static int __mtd_writeseg(struct super_block *sb, u64 ofs, pgoff_t index, page = find_lock_page(mapping, index + i); BUG_ON(!page); - err = mtd_write(sb, page->index << PAGE_SHIFT, PAGE_SIZE, - page_address(page)); + err = loffs_mtd_write(sb, page->index << PAGE_SHIFT, PAGE_SIZE, + page_address(page)); unlock_page(page); page_cache_release(page); if (err) @@ -206,7 +208,7 @@ static int __mtd_writeseg(struct super_block *sb, u64 ofs, pgoff_t index, return 0; } -static void mtd_writeseg(struct super_block *sb, u64 ofs, size_t len) +static void logfs_mtd_writeseg(struct super_block *sb, u64 ofs, size_t len) { struct logfs_super *super = logfs_super(sb); int head; @@ -227,15 +229,15 @@ static void mtd_writeseg(struct super_block *sb, u64 ofs, size_t len) len += head; } len = PAGE_ALIGN(len); - __mtd_writeseg(sb, ofs, ofs >> PAGE_SHIFT, len >> PAGE_SHIFT); + __logfs_mtd_writeseg(sb, ofs, ofs >> PAGE_SHIFT, len >> PAGE_SHIFT); } -static void mtd_put_device(struct logfs_super *s) +static void logfs_mtd_put_device(struct logfs_super *s) { put_mtd_device(s->s_mtd); } -static int mtd_can_write_buf(struct super_block *sb, u64 ofs) +static int logfs_mtd_can_write_buf(struct super_block *sb, u64 ofs) { struct logfs_super *super = logfs_super(sb); void *buf; @@ -244,7 +246,7 @@ static int mtd_can_write_buf(struct super_block *sb, u64 ofs) buf = kmalloc(super->s_writesize, GFP_KERNEL); if (!buf) return -ENOMEM; - err = mtd_read(sb, ofs, super->s_writesize, buf); + err = logfs_mtd_read(sb, ofs, super->s_writesize, buf); if (err) goto out; if (memchr_inv(buf, 0xff, super->s_writesize)) @@ -255,14 +257,14 @@ out: } static const struct logfs_device_ops mtd_devops = { - .find_first_sb = mtd_find_first_sb, - .find_last_sb = mtd_find_last_sb, - .readpage = mtd_readpage, - .writeseg = mtd_writeseg, - .erase = mtd_erase, - .can_write_buf = mtd_can_write_buf, - .sync = mtd_sync, - .put_device = mtd_put_device, + .find_first_sb = logfs_mtd_find_first_sb, + .find_last_sb = logfs_mtd_find_last_sb, + .readpage = logfs_mtd_readpage, + .writeseg = logfs_mtd_writeseg, + .erase = logfs_mtd_erase, + .can_write_buf = logfs_mtd_can_write_buf, + .sync = logfs_mtd_sync, + .put_device = logfs_mtd_put_device, }; int logfs_get_sb_mtd(struct logfs_super *s, int mtdnr) diff --git a/fs/logfs/dir.c b/fs/logfs/dir.c index b7d7f67cee5a..501043e8966c 100644 --- a/fs/logfs/dir.c +++ b/fs/logfs/dir.c @@ -482,7 +482,7 @@ out: return ret; } -static int logfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int logfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct inode *inode; @@ -501,7 +501,7 @@ static int logfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) return __logfs_create(dir, dentry, inode, NULL, 0); } -static int logfs_create(struct inode *dir, struct dentry *dentry, int mode, +static int logfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { struct inode *inode; @@ -517,7 +517,7 @@ static int logfs_create(struct inode *dir, struct dentry *dentry, int mode, return __logfs_create(dir, dentry, inode, NULL, 0); } -static int logfs_mknod(struct inode *dir, struct dentry *dentry, int mode, +static int logfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) { struct inode *inode; diff --git a/fs/logfs/inode.c b/fs/logfs/inode.c index 7e441ad5f792..388df1aa35e5 100644 --- a/fs/logfs/inode.c +++ b/fs/logfs/inode.c @@ -144,7 +144,6 @@ struct inode *logfs_safe_iget(struct super_block *sb, ino_t ino, int *is_cached) static void logfs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(logfs_inode_cache, logfs_inode(inode)); } @@ -324,7 +323,7 @@ static void logfs_set_ino_generation(struct super_block *sb, mutex_unlock(&super->s_journal_mutex); } -struct inode *logfs_new_inode(struct inode *dir, int mode) +struct inode *logfs_new_inode(struct inode *dir, umode_t mode) { struct super_block *sb = dir->i_sb; struct inode *inode; diff --git a/fs/logfs/logfs.h b/fs/logfs/logfs.h index 398ecff6e548..926373866a55 100644 --- a/fs/logfs/logfs.h +++ b/fs/logfs/logfs.h @@ -520,7 +520,7 @@ extern const struct super_operations logfs_super_operations; struct inode *logfs_iget(struct super_block *sb, ino_t ino); struct inode *logfs_safe_iget(struct super_block *sb, ino_t ino, int *cookie); void logfs_safe_iput(struct inode *inode, int cookie); -struct inode *logfs_new_inode(struct inode *dir, int mode); +struct inode *logfs_new_inode(struct inode *dir, umode_t mode); struct inode *logfs_new_meta_inode(struct super_block *sb, u64 ino); struct inode *logfs_read_meta_inode(struct super_block *sb, u64 ino); int logfs_init_inode_cache(void); diff --git a/fs/minix/bitmap.c b/fs/minix/bitmap.c index 3f32bcb0d9bd..4bc50dac8e97 100644 --- a/fs/minix/bitmap.c +++ b/fs/minix/bitmap.c @@ -16,38 +16,26 @@ #include <linux/bitops.h> #include <linux/sched.h> -static const int nibblemap[] = { 4,3,3,2,3,2,2,1,3,2,2,1,2,1,1,0 }; - static DEFINE_SPINLOCK(bitmap_lock); -static unsigned long count_free(struct buffer_head *map[], unsigned numblocks, __u32 numbits) +/* + * bitmap consists of blocks filled with 16bit words + * bit set == busy, bit clear == free + * endianness is a mess, but for counting zero bits it really doesn't matter... + */ +static __u32 count_free(struct buffer_head *map[], unsigned blocksize, __u32 numbits) { - unsigned i, j, sum = 0; - struct buffer_head *bh; - - for (i=0; i<numblocks-1; i++) { - if (!(bh=map[i])) - return(0); - for (j=0; j<bh->b_size; j++) - sum += nibblemap[bh->b_data[j] & 0xf] - + nibblemap[(bh->b_data[j]>>4) & 0xf]; - } + __u32 sum = 0; + unsigned blocks = DIV_ROUND_UP(numbits, blocksize * 8); - if (numblocks==0 || !(bh=map[numblocks-1])) - return(0); - i = ((numbits - (numblocks-1) * bh->b_size * 8) / 16) * 2; - for (j=0; j<i; j++) { - sum += nibblemap[bh->b_data[j] & 0xf] - + nibblemap[(bh->b_data[j]>>4) & 0xf]; + while (blocks--) { + unsigned words = blocksize / 2; + __u16 *p = (__u16 *)(*map++)->b_data; + while (words--) + sum += 16 - hweight16(*p++); } - i = numbits%16; - if (i!=0) { - i = *(__u16 *)(&bh->b_data[j]) | ~((1<<i) - 1); - sum += nibblemap[i & 0xf] + nibblemap[(i>>4) & 0xf]; - sum += nibblemap[(i>>8) & 0xf] + nibblemap[(i>>12) & 0xf]; - } - return(sum); + return sum; } void minix_free_block(struct inode *inode, unsigned long block) @@ -105,10 +93,12 @@ int minix_new_block(struct inode * inode) return 0; } -unsigned long minix_count_free_blocks(struct minix_sb_info *sbi) +unsigned long minix_count_free_blocks(struct super_block *sb) { - return (count_free(sbi->s_zmap, sbi->s_zmap_blocks, - sbi->s_nzones - sbi->s_firstdatazone + 1) + struct minix_sb_info *sbi = minix_sb(sb); + u32 bits = sbi->s_nzones - (sbi->s_firstdatazone + 1); + + return (count_free(sbi->s_zmap, sb->s_blocksize, bits) << sbi->s_log_zone_size); } @@ -219,7 +209,7 @@ void minix_free_inode(struct inode * inode) mark_buffer_dirty(bh); } -struct inode *minix_new_inode(const struct inode *dir, int mode, int *error) +struct inode *minix_new_inode(const struct inode *dir, umode_t mode, int *error) { struct super_block *sb = dir->i_sb; struct minix_sb_info *sbi = minix_sb(sb); @@ -273,7 +263,10 @@ struct inode *minix_new_inode(const struct inode *dir, int mode, int *error) return inode; } -unsigned long minix_count_free_inodes(struct minix_sb_info *sbi) +unsigned long minix_count_free_inodes(struct super_block *sb) { - return count_free(sbi->s_imap, sbi->s_imap_blocks, sbi->s_ninodes + 1); + struct minix_sb_info *sbi = minix_sb(sb); + u32 bits = sbi->s_ninodes + 1; + + return count_free(sbi->s_imap, sb->s_blocksize, bits); } diff --git a/fs/minix/inode.c b/fs/minix/inode.c index 64cdcd662ffc..fa8b612b8ce2 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -71,7 +71,6 @@ static struct inode *minix_alloc_inode(struct super_block *sb) static void minix_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(minix_inode_cachep, minix_i(inode)); } @@ -263,6 +262,26 @@ static int minix_fill_super(struct super_block *s, void *data, int silent) goto out_no_root; } + /* Apparently minix can create filesystems that allocate more blocks for + * the bitmaps than needed. We simply ignore that, but verify it didn't + * create one with not enough blocks and bail out if so. + */ + block = minix_blocks_needed(sbi->s_ninodes, s->s_blocksize); + if (sbi->s_imap_blocks < block) { + printk("MINIX-fs: file system does not have enough " + "imap blocks allocated. Refusing to mount\n"); + goto out_iput; + } + + block = minix_blocks_needed( + (sbi->s_nzones - (sbi->s_firstdatazone + 1)), + s->s_blocksize); + if (sbi->s_zmap_blocks < block) { + printk("MINIX-fs: file system does not have enough " + "zmap blocks allocated. Refusing to mount.\n"); + goto out_iput; + } + ret = -ENOMEM; s->s_root = d_alloc_root(root_inode); if (!s->s_root) @@ -276,9 +295,10 @@ static int minix_fill_super(struct super_block *s, void *data, int silent) if (!(sbi->s_mount_state & MINIX_VALID_FS)) printk("MINIX-fs: mounting unchecked file system, " "running fsck is recommended\n"); - else if (sbi->s_mount_state & MINIX_ERROR_FS) + else if (sbi->s_mount_state & MINIX_ERROR_FS) printk("MINIX-fs: mounting file system with errors, " "running fsck is recommended\n"); + return 0; out_iput: @@ -339,10 +359,10 @@ static int minix_statfs(struct dentry *dentry, struct kstatfs *buf) buf->f_type = sb->s_magic; buf->f_bsize = sb->s_blocksize; buf->f_blocks = (sbi->s_nzones - sbi->s_firstdatazone) << sbi->s_log_zone_size; - buf->f_bfree = minix_count_free_blocks(sbi); + buf->f_bfree = minix_count_free_blocks(sb); buf->f_bavail = buf->f_bfree; buf->f_files = sbi->s_ninodes; - buf->f_ffree = minix_count_free_inodes(sbi); + buf->f_ffree = minix_count_free_inodes(sb); buf->f_namelen = sbi->s_namelen; buf->f_fsid.val[0] = (u32)id; buf->f_fsid.val[1] = (u32)(id >> 32); diff --git a/fs/minix/minix.h b/fs/minix/minix.h index 341e2122879a..c889ef0aa571 100644 --- a/fs/minix/minix.h +++ b/fs/minix/minix.h @@ -46,12 +46,12 @@ struct minix_sb_info { extern struct inode *minix_iget(struct super_block *, unsigned long); extern struct minix_inode * minix_V1_raw_inode(struct super_block *, ino_t, struct buffer_head **); extern struct minix2_inode * minix_V2_raw_inode(struct super_block *, ino_t, struct buffer_head **); -extern struct inode * minix_new_inode(const struct inode *, int, int *); +extern struct inode * minix_new_inode(const struct inode *, umode_t, int *); extern void minix_free_inode(struct inode * inode); -extern unsigned long minix_count_free_inodes(struct minix_sb_info *sbi); +extern unsigned long minix_count_free_inodes(struct super_block *sb); extern int minix_new_block(struct inode * inode); extern void minix_free_block(struct inode *inode, unsigned long block); -extern unsigned long minix_count_free_blocks(struct minix_sb_info *sbi); +extern unsigned long minix_count_free_blocks(struct super_block *sb); extern int minix_getattr(struct vfsmount *, struct dentry *, struct kstat *); extern int minix_prepare_chunk(struct page *page, loff_t pos, unsigned len); @@ -88,6 +88,11 @@ static inline struct minix_inode_info *minix_i(struct inode *inode) return list_entry(inode, struct minix_inode_info, vfs_inode); } +static inline unsigned minix_blocks_needed(unsigned bits, unsigned blocksize) +{ + return DIV_ROUND_UP(bits, blocksize * 8); +} + #if defined(CONFIG_MINIX_FS_NATIVE_ENDIAN) && \ defined(CONFIG_MINIX_FS_BIG_ENDIAN_16BIT_INDEXED) @@ -125,7 +130,7 @@ static inline int minix_find_first_zero_bit(const void *vaddr, unsigned size) if (!size) return 0; - size = (size >> 4) + ((size & 15) > 0); + size >>= 4; while (*p++ == 0xffff) { if (--size == 0) return (p - addr) << 4; diff --git a/fs/minix/namei.c b/fs/minix/namei.c index 6e6777f1b4b2..2f76e38c2065 100644 --- a/fs/minix/namei.c +++ b/fs/minix/namei.c @@ -36,7 +36,7 @@ static struct dentry *minix_lookup(struct inode * dir, struct dentry *dentry, st return NULL; } -static int minix_mknod(struct inode * dir, struct dentry *dentry, int mode, dev_t rdev) +static int minix_mknod(struct inode * dir, struct dentry *dentry, umode_t mode, dev_t rdev) { int error; struct inode *inode; @@ -54,7 +54,7 @@ static int minix_mknod(struct inode * dir, struct dentry *dentry, int mode, dev_ return error; } -static int minix_create(struct inode * dir, struct dentry *dentry, int mode, +static int minix_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { return minix_mknod(dir, dentry, mode, 0); @@ -103,7 +103,7 @@ static int minix_link(struct dentry * old_dentry, struct inode * dir, return add_nondir(dentry, inode); } -static int minix_mkdir(struct inode * dir, struct dentry *dentry, int mode) +static int minix_mkdir(struct inode * dir, struct dentry *dentry, umode_t mode) { struct inode * inode; int err = -EMLINK; diff --git a/fs/mount.h b/fs/mount.h new file mode 100644 index 000000000000..4ef36d93e5a2 --- /dev/null +++ b/fs/mount.h @@ -0,0 +1,76 @@ +#include <linux/mount.h> +#include <linux/seq_file.h> +#include <linux/poll.h> + +struct mnt_namespace { + atomic_t count; + struct mount * root; + struct list_head list; + wait_queue_head_t poll; + int event; +}; + +struct mnt_pcp { + int mnt_count; + int mnt_writers; +}; + +struct mount { + struct list_head mnt_hash; + struct mount *mnt_parent; + struct dentry *mnt_mountpoint; + struct vfsmount mnt; +#ifdef CONFIG_SMP + struct mnt_pcp __percpu *mnt_pcp; + atomic_t mnt_longterm; /* how many of the refs are longterm */ +#else + int mnt_count; + int mnt_writers; +#endif + struct list_head mnt_mounts; /* list of children, anchored here */ + struct list_head mnt_child; /* and going through their mnt_child */ + struct list_head mnt_instance; /* mount instance on sb->s_mounts */ + const char *mnt_devname; /* Name of device e.g. /dev/dsk/hda1 */ + struct list_head mnt_list; + struct list_head mnt_expire; /* link in fs-specific expiry list */ + struct list_head mnt_share; /* circular list of shared mounts */ + struct list_head mnt_slave_list;/* list of slave mounts */ + struct list_head mnt_slave; /* slave list entry */ + struct mount *mnt_master; /* slave is on master->mnt_slave_list */ + struct mnt_namespace *mnt_ns; /* containing namespace */ +#ifdef CONFIG_FSNOTIFY + struct hlist_head mnt_fsnotify_marks; + __u32 mnt_fsnotify_mask; +#endif + int mnt_id; /* mount identifier */ + int mnt_group_id; /* peer group identifier */ + int mnt_expiry_mark; /* true if marked for expiry */ + int mnt_pinned; + int mnt_ghosts; +}; + +static inline struct mount *real_mount(struct vfsmount *mnt) +{ + return container_of(mnt, struct mount, mnt); +} + +static inline int mnt_has_parent(struct mount *mnt) +{ + return mnt != mnt->mnt_parent; +} + +extern struct mount *__lookup_mnt(struct vfsmount *, struct dentry *, int); + +static inline void get_mnt_ns(struct mnt_namespace *ns) +{ + atomic_inc(&ns->count); +} + +struct proc_mounts { + struct seq_file m; /* must be the first element */ + struct mnt_namespace *ns; + struct path root; + int (*show)(struct seq_file *, struct vfsmount *); +}; + +extern const struct seq_operations mounts_op; diff --git a/fs/mpage.c b/fs/mpage.c index fdfae9fa98cd..643e9f55ef29 100644 --- a/fs/mpage.c +++ b/fs/mpage.c @@ -371,9 +371,6 @@ mpage_readpages(struct address_space *mapping, struct list_head *pages, sector_t last_block_in_bio = 0; struct buffer_head map_bh; unsigned long first_logical_block = 0; - struct blk_plug plug; - - blk_start_plug(&plug); map_bh.b_state = 0; map_bh.b_size = 0; @@ -395,7 +392,6 @@ mpage_readpages(struct address_space *mapping, struct list_head *pages, BUG_ON(!list_empty(pages)); if (bio) mpage_bio_submit(READ, bio); - blk_finish_plug(&plug); return 0; } EXPORT_SYMBOL(mpage_readpages); diff --git a/fs/namei.c b/fs/namei.c index 5008f01787f5..c283a1ec008e 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -36,6 +36,7 @@ #include <asm/uaccess.h> #include "internal.h" +#include "mount.h" /* [Feb-1997 T. Schoebel-Theuer] * Fundamental changes in the pathname lookup mechanisms (namei) @@ -676,36 +677,38 @@ follow_link(struct path *link, struct nameidata *nd, void **p) static int follow_up_rcu(struct path *path) { - struct vfsmount *parent; + struct mount *mnt = real_mount(path->mnt); + struct mount *parent; struct dentry *mountpoint; - parent = path->mnt->mnt_parent; - if (parent == path->mnt) + parent = mnt->mnt_parent; + if (&parent->mnt == path->mnt) return 0; - mountpoint = path->mnt->mnt_mountpoint; + mountpoint = mnt->mnt_mountpoint; path->dentry = mountpoint; - path->mnt = parent; + path->mnt = &parent->mnt; return 1; } int follow_up(struct path *path) { - struct vfsmount *parent; + struct mount *mnt = real_mount(path->mnt); + struct mount *parent; struct dentry *mountpoint; br_read_lock(vfsmount_lock); - parent = path->mnt->mnt_parent; - if (parent == path->mnt) { + parent = mnt->mnt_parent; + if (&parent->mnt == path->mnt) { br_read_unlock(vfsmount_lock); return 0; } - mntget(parent); - mountpoint = dget(path->mnt->mnt_mountpoint); + mntget(&parent->mnt); + mountpoint = dget(mnt->mnt_mountpoint); br_read_unlock(vfsmount_lock); dput(path->dentry); path->dentry = mountpoint; mntput(path->mnt); - path->mnt = parent; + path->mnt = &parent->mnt; return 1; } @@ -884,7 +887,7 @@ static bool __follow_mount_rcu(struct nameidata *nd, struct path *path, struct inode **inode) { for (;;) { - struct vfsmount *mounted; + struct mount *mounted; /* * Don't forget we might have a non-mountpoint managed dentry * that wants to block transit. @@ -898,8 +901,8 @@ static bool __follow_mount_rcu(struct nameidata *nd, struct path *path, mounted = __lookup_mnt(path->mnt, path->dentry, 1); if (!mounted) break; - path->mnt = mounted; - path->dentry = mounted->mnt_root; + path->mnt = &mounted->mnt; + path->dentry = mounted->mnt.mnt_root; nd->flags |= LOOKUP_JUMPED; nd->seq = read_seqcount_begin(&path->dentry->d_seq); /* @@ -915,12 +918,12 @@ static bool __follow_mount_rcu(struct nameidata *nd, struct path *path, static void follow_mount_rcu(struct nameidata *nd) { while (d_mountpoint(nd->path.dentry)) { - struct vfsmount *mounted; + struct mount *mounted; mounted = __lookup_mnt(nd->path.mnt, nd->path.dentry, 1); if (!mounted) break; - nd->path.mnt = mounted; - nd->path.dentry = mounted->mnt_root; + nd->path.mnt = &mounted->mnt; + nd->path.dentry = mounted->mnt.mnt_root; nd->seq = read_seqcount_begin(&nd->path.dentry->d_seq); } } @@ -1976,7 +1979,7 @@ void unlock_rename(struct dentry *p1, struct dentry *p2) } } -int vfs_create(struct inode *dir, struct dentry *dentry, int mode, +int vfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { int error = may_create(dir, dentry); @@ -2177,7 +2180,7 @@ static struct file *do_last(struct nameidata *nd, struct path *path, /* Negative dentry, just create the file */ if (!dentry->d_inode) { - int mode = op->mode; + umode_t mode = op->mode; if (!IS_POSIXACL(dir->d_inode)) mode &= ~current_umask(); /* @@ -2444,7 +2447,7 @@ struct dentry *user_path_create(int dfd, const char __user *pathname, struct pat } EXPORT_SYMBOL(user_path_create); -int vfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) +int vfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev) { int error = may_create(dir, dentry); @@ -2472,7 +2475,7 @@ int vfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) return error; } -static int may_mknod(mode_t mode) +static int may_mknod(umode_t mode) { switch (mode & S_IFMT) { case S_IFREG: @@ -2489,7 +2492,7 @@ static int may_mknod(mode_t mode) } } -SYSCALL_DEFINE4(mknodat, int, dfd, const char __user *, filename, int, mode, +SYSCALL_DEFINE4(mknodat, int, dfd, const char __user *, filename, umode_t, mode, unsigned, dev) { struct dentry *dentry; @@ -2536,12 +2539,12 @@ out_dput: return error; } -SYSCALL_DEFINE3(mknod, const char __user *, filename, int, mode, unsigned, dev) +SYSCALL_DEFINE3(mknod, const char __user *, filename, umode_t, mode, unsigned, dev) { return sys_mknodat(AT_FDCWD, filename, mode, dev); } -int vfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +int vfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { int error = may_create(dir, dentry); @@ -2562,7 +2565,7 @@ int vfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) return error; } -SYSCALL_DEFINE3(mkdirat, int, dfd, const char __user *, pathname, int, mode) +SYSCALL_DEFINE3(mkdirat, int, dfd, const char __user *, pathname, umode_t, mode) { struct dentry *dentry; struct path path; @@ -2590,7 +2593,7 @@ out_dput: return error; } -SYSCALL_DEFINE2(mkdir, const char __user *, pathname, int, mode) +SYSCALL_DEFINE2(mkdir, const char __user *, pathname, umode_t, mode) { return sys_mkdirat(AT_FDCWD, pathname, mode); } diff --git a/fs/namespace.c b/fs/namespace.c index e5e1c7d1839b..e6081996c9a2 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -9,30 +9,17 @@ */ #include <linux/syscalls.h> -#include <linux/slab.h> -#include <linux/sched.h> -#include <linux/spinlock.h> -#include <linux/percpu.h> -#include <linux/init.h> -#include <linux/kernel.h> -#include <linux/acct.h> +#include <linux/export.h> #include <linux/capability.h> -#include <linux/cpumask.h> -#include <linux/module.h> -#include <linux/sysfs.h> -#include <linux/seq_file.h> #include <linux/mnt_namespace.h> #include <linux/namei.h> -#include <linux/nsproxy.h> #include <linux/security.h> -#include <linux/mount.h> -#include <linux/ramfs.h> -#include <linux/log2.h> #include <linux/idr.h> -#include <linux/fs_struct.h> -#include <linux/fsnotify.h> -#include <asm/uaccess.h> -#include <asm/unistd.h> +#include <linux/acct.h> /* acct_auto_close_mnt */ +#include <linux/ramfs.h> /* init_rootfs */ +#include <linux/fs_struct.h> /* get_fs_root et.al. */ +#include <linux/fsnotify.h> /* fsnotify_vfsmount_delete */ +#include <linux/uaccess.h> #include "pnode.h" #include "internal.h" @@ -78,7 +65,7 @@ static inline unsigned long hash(struct vfsmount *mnt, struct dentry *dentry) * allocation is serialized by namespace_sem, but we need the spinlock to * serialize with freeing. */ -static int mnt_alloc_id(struct vfsmount *mnt) +static int mnt_alloc_id(struct mount *mnt) { int res; @@ -95,7 +82,7 @@ retry: return res; } -static void mnt_free_id(struct vfsmount *mnt) +static void mnt_free_id(struct mount *mnt) { int id = mnt->mnt_id; spin_lock(&mnt_id_lock); @@ -110,7 +97,7 @@ static void mnt_free_id(struct vfsmount *mnt) * * mnt_group_ida is protected by namespace_sem */ -static int mnt_alloc_group_id(struct vfsmount *mnt) +static int mnt_alloc_group_id(struct mount *mnt) { int res; @@ -129,7 +116,7 @@ static int mnt_alloc_group_id(struct vfsmount *mnt) /* * Release a peer group ID */ -void mnt_release_group_id(struct vfsmount *mnt) +void mnt_release_group_id(struct mount *mnt) { int id = mnt->mnt_group_id; ida_remove(&mnt_group_ida, id); @@ -141,7 +128,7 @@ void mnt_release_group_id(struct vfsmount *mnt) /* * vfsmount lock must be held for read */ -static inline void mnt_add_count(struct vfsmount *mnt, int n) +static inline void mnt_add_count(struct mount *mnt, int n) { #ifdef CONFIG_SMP this_cpu_add(mnt->mnt_pcp->mnt_count, n); @@ -152,35 +139,10 @@ static inline void mnt_add_count(struct vfsmount *mnt, int n) #endif } -static inline void mnt_set_count(struct vfsmount *mnt, int n) -{ -#ifdef CONFIG_SMP - this_cpu_write(mnt->mnt_pcp->mnt_count, n); -#else - mnt->mnt_count = n; -#endif -} - -/* - * vfsmount lock must be held for read - */ -static inline void mnt_inc_count(struct vfsmount *mnt) -{ - mnt_add_count(mnt, 1); -} - -/* - * vfsmount lock must be held for read - */ -static inline void mnt_dec_count(struct vfsmount *mnt) -{ - mnt_add_count(mnt, -1); -} - /* * vfsmount lock must be held for write */ -unsigned int mnt_get_count(struct vfsmount *mnt) +unsigned int mnt_get_count(struct mount *mnt) { #ifdef CONFIG_SMP unsigned int count = 0; @@ -196,9 +158,9 @@ unsigned int mnt_get_count(struct vfsmount *mnt) #endif } -static struct vfsmount *alloc_vfsmnt(const char *name) +static struct mount *alloc_vfsmnt(const char *name) { - struct vfsmount *mnt = kmem_cache_zalloc(mnt_cache, GFP_KERNEL); + struct mount *mnt = kmem_cache_zalloc(mnt_cache, GFP_KERNEL); if (mnt) { int err; @@ -277,7 +239,7 @@ int __mnt_is_readonly(struct vfsmount *mnt) } EXPORT_SYMBOL_GPL(__mnt_is_readonly); -static inline void mnt_inc_writers(struct vfsmount *mnt) +static inline void mnt_inc_writers(struct mount *mnt) { #ifdef CONFIG_SMP this_cpu_inc(mnt->mnt_pcp->mnt_writers); @@ -286,7 +248,7 @@ static inline void mnt_inc_writers(struct vfsmount *mnt) #endif } -static inline void mnt_dec_writers(struct vfsmount *mnt) +static inline void mnt_dec_writers(struct mount *mnt) { #ifdef CONFIG_SMP this_cpu_dec(mnt->mnt_pcp->mnt_writers); @@ -295,7 +257,7 @@ static inline void mnt_dec_writers(struct vfsmount *mnt) #endif } -static unsigned int mnt_get_writers(struct vfsmount *mnt) +static unsigned int mnt_get_writers(struct mount *mnt) { #ifdef CONFIG_SMP unsigned int count = 0; @@ -311,6 +273,15 @@ static unsigned int mnt_get_writers(struct vfsmount *mnt) #endif } +static int mnt_is_readonly(struct vfsmount *mnt) +{ + if (mnt->mnt_sb->s_readonly_remount) + return 1; + /* Order wrt setting s_flags/s_readonly_remount in do_remount() */ + smp_rmb(); + return __mnt_is_readonly(mnt); +} + /* * Most r/o checks on a fs are for operations that take * discrete amounts of time, like a write() or unlink(). @@ -321,7 +292,7 @@ static unsigned int mnt_get_writers(struct vfsmount *mnt) */ /** * mnt_want_write - get write access to a mount - * @mnt: the mount on which to take a write + * @m: the mount on which to take a write * * This tells the low-level filesystem that a write is * about to be performed to it, and makes sure that @@ -329,8 +300,9 @@ static unsigned int mnt_get_writers(struct vfsmount *mnt) * the write operation is finished, mnt_drop_write() * must be called. This is effectively a refcount. */ -int mnt_want_write(struct vfsmount *mnt) +int mnt_want_write(struct vfsmount *m) { + struct mount *mnt = real_mount(m); int ret = 0; preempt_disable(); @@ -341,7 +313,7 @@ int mnt_want_write(struct vfsmount *mnt) * incremented count after it has set MNT_WRITE_HOLD. */ smp_mb(); - while (mnt->mnt_flags & MNT_WRITE_HOLD) + while (mnt->mnt.mnt_flags & MNT_WRITE_HOLD) cpu_relax(); /* * After the slowpath clears MNT_WRITE_HOLD, mnt_is_readonly will @@ -349,12 +321,10 @@ int mnt_want_write(struct vfsmount *mnt) * MNT_WRITE_HOLD is cleared. */ smp_rmb(); - if (__mnt_is_readonly(mnt)) { + if (mnt_is_readonly(m)) { mnt_dec_writers(mnt); ret = -EROFS; - goto out; } -out: preempt_enable(); return ret; } @@ -378,7 +348,7 @@ int mnt_clone_write(struct vfsmount *mnt) if (__mnt_is_readonly(mnt)) return -EROFS; preempt_disable(); - mnt_inc_writers(mnt); + mnt_inc_writers(real_mount(mnt)); preempt_enable(); return 0; } @@ -412,17 +382,23 @@ EXPORT_SYMBOL_GPL(mnt_want_write_file); void mnt_drop_write(struct vfsmount *mnt) { preempt_disable(); - mnt_dec_writers(mnt); + mnt_dec_writers(real_mount(mnt)); preempt_enable(); } EXPORT_SYMBOL_GPL(mnt_drop_write); -static int mnt_make_readonly(struct vfsmount *mnt) +void mnt_drop_write_file(struct file *file) +{ + mnt_drop_write(file->f_path.mnt); +} +EXPORT_SYMBOL(mnt_drop_write_file); + +static int mnt_make_readonly(struct mount *mnt) { int ret = 0; br_write_lock(vfsmount_lock); - mnt->mnt_flags |= MNT_WRITE_HOLD; + mnt->mnt.mnt_flags |= MNT_WRITE_HOLD; /* * After storing MNT_WRITE_HOLD, we'll read the counters. This store * should be visible before we do. @@ -448,25 +424,61 @@ static int mnt_make_readonly(struct vfsmount *mnt) if (mnt_get_writers(mnt) > 0) ret = -EBUSY; else - mnt->mnt_flags |= MNT_READONLY; + mnt->mnt.mnt_flags |= MNT_READONLY; /* * MNT_READONLY must become visible before ~MNT_WRITE_HOLD, so writers * that become unheld will see MNT_READONLY. */ smp_wmb(); - mnt->mnt_flags &= ~MNT_WRITE_HOLD; + mnt->mnt.mnt_flags &= ~MNT_WRITE_HOLD; br_write_unlock(vfsmount_lock); return ret; } -static void __mnt_unmake_readonly(struct vfsmount *mnt) +static void __mnt_unmake_readonly(struct mount *mnt) { br_write_lock(vfsmount_lock); - mnt->mnt_flags &= ~MNT_READONLY; + mnt->mnt.mnt_flags &= ~MNT_READONLY; + br_write_unlock(vfsmount_lock); +} + +int sb_prepare_remount_readonly(struct super_block *sb) +{ + struct mount *mnt; + int err = 0; + + /* Racy optimization. Recheck the counter under MNT_WRITE_HOLD */ + if (atomic_long_read(&sb->s_remove_count)) + return -EBUSY; + + br_write_lock(vfsmount_lock); + list_for_each_entry(mnt, &sb->s_mounts, mnt_instance) { + if (!(mnt->mnt.mnt_flags & MNT_READONLY)) { + mnt->mnt.mnt_flags |= MNT_WRITE_HOLD; + smp_mb(); + if (mnt_get_writers(mnt) > 0) { + err = -EBUSY; + break; + } + } + } + if (!err && atomic_long_read(&sb->s_remove_count)) + err = -EBUSY; + + if (!err) { + sb->s_readonly_remount = 1; + smp_wmb(); + } + list_for_each_entry(mnt, &sb->s_mounts, mnt_instance) { + if (mnt->mnt.mnt_flags & MNT_WRITE_HOLD) + mnt->mnt.mnt_flags &= ~MNT_WRITE_HOLD; + } br_write_unlock(vfsmount_lock); + + return err; } -static void free_vfsmnt(struct vfsmount *mnt) +static void free_vfsmnt(struct mount *mnt) { kfree(mnt->mnt_devname); mnt_free_id(mnt); @@ -481,20 +493,20 @@ static void free_vfsmnt(struct vfsmount *mnt) * @dir. If @dir is set return the first mount else return the last mount. * vfsmount_lock must be held for read or write. */ -struct vfsmount *__lookup_mnt(struct vfsmount *mnt, struct dentry *dentry, +struct mount *__lookup_mnt(struct vfsmount *mnt, struct dentry *dentry, int dir) { struct list_head *head = mount_hashtable + hash(mnt, dentry); struct list_head *tmp = head; - struct vfsmount *p, *found = NULL; + struct mount *p, *found = NULL; for (;;) { tmp = dir ? tmp->next : tmp->prev; p = NULL; if (tmp == head) break; - p = list_entry(tmp, struct vfsmount, mnt_hash); - if (p->mnt_parent == mnt && p->mnt_mountpoint == dentry) { + p = list_entry(tmp, struct mount, mnt_hash); + if (&p->mnt_parent->mnt == mnt && p->mnt_mountpoint == dentry) { found = p; break; } @@ -508,16 +520,21 @@ struct vfsmount *__lookup_mnt(struct vfsmount *mnt, struct dentry *dentry, */ struct vfsmount *lookup_mnt(struct path *path) { - struct vfsmount *child_mnt; + struct mount *child_mnt; br_read_lock(vfsmount_lock); - if ((child_mnt = __lookup_mnt(path->mnt, path->dentry, 1))) - mntget(child_mnt); - br_read_unlock(vfsmount_lock); - return child_mnt; + child_mnt = __lookup_mnt(path->mnt, path->dentry, 1); + if (child_mnt) { + mnt_add_count(child_mnt, 1); + br_read_unlock(vfsmount_lock); + return &child_mnt->mnt; + } else { + br_read_unlock(vfsmount_lock); + return NULL; + } } -static inline int check_mnt(struct vfsmount *mnt) +static inline int check_mnt(struct mount *mnt) { return mnt->mnt_ns == current->nsproxy->mnt_ns; } @@ -548,12 +565,12 @@ static void __touch_mnt_namespace(struct mnt_namespace *ns) * Clear dentry's mounted state if it has no remaining mounts. * vfsmount_lock must be held for write. */ -static void dentry_reset_mounted(struct vfsmount *mnt, struct dentry *dentry) +static void dentry_reset_mounted(struct dentry *dentry) { unsigned u; for (u = 0; u < HASH_SIZE; u++) { - struct vfsmount *p; + struct mount *p; list_for_each_entry(p, &mount_hashtable[u], mnt_hash) { if (p->mnt_mountpoint == dentry) @@ -568,25 +585,26 @@ static void dentry_reset_mounted(struct vfsmount *mnt, struct dentry *dentry) /* * vfsmount lock must be held for write */ -static void detach_mnt(struct vfsmount *mnt, struct path *old_path) +static void detach_mnt(struct mount *mnt, struct path *old_path) { old_path->dentry = mnt->mnt_mountpoint; - old_path->mnt = mnt->mnt_parent; + old_path->mnt = &mnt->mnt_parent->mnt; mnt->mnt_parent = mnt; - mnt->mnt_mountpoint = mnt->mnt_root; + mnt->mnt_mountpoint = mnt->mnt.mnt_root; list_del_init(&mnt->mnt_child); list_del_init(&mnt->mnt_hash); - dentry_reset_mounted(old_path->mnt, old_path->dentry); + dentry_reset_mounted(old_path->dentry); } /* * vfsmount lock must be held for write */ -void mnt_set_mountpoint(struct vfsmount *mnt, struct dentry *dentry, - struct vfsmount *child_mnt) +void mnt_set_mountpoint(struct mount *mnt, struct dentry *dentry, + struct mount *child_mnt) { - child_mnt->mnt_parent = mntget(mnt); + mnt_add_count(mnt, 1); /* essentially, that's mntget */ child_mnt->mnt_mountpoint = dget(dentry); + child_mnt->mnt_parent = mnt; spin_lock(&dentry->d_lock); dentry->d_flags |= DCACHE_MOUNTED; spin_unlock(&dentry->d_lock); @@ -595,15 +613,15 @@ void mnt_set_mountpoint(struct vfsmount *mnt, struct dentry *dentry, /* * vfsmount lock must be held for write */ -static void attach_mnt(struct vfsmount *mnt, struct path *path) +static void attach_mnt(struct mount *mnt, struct path *path) { - mnt_set_mountpoint(path->mnt, path->dentry, mnt); + mnt_set_mountpoint(real_mount(path->mnt), path->dentry, mnt); list_add_tail(&mnt->mnt_hash, mount_hashtable + hash(path->mnt, path->dentry)); - list_add_tail(&mnt->mnt_child, &path->mnt->mnt_mounts); + list_add_tail(&mnt->mnt_child, &real_mount(path->mnt)->mnt_mounts); } -static inline void __mnt_make_longterm(struct vfsmount *mnt) +static inline void __mnt_make_longterm(struct mount *mnt) { #ifdef CONFIG_SMP atomic_inc(&mnt->mnt_longterm); @@ -611,7 +629,7 @@ static inline void __mnt_make_longterm(struct vfsmount *mnt) } /* needs vfsmount lock for write */ -static inline void __mnt_make_shortterm(struct vfsmount *mnt) +static inline void __mnt_make_shortterm(struct mount *mnt) { #ifdef CONFIG_SMP atomic_dec(&mnt->mnt_longterm); @@ -621,10 +639,10 @@ static inline void __mnt_make_shortterm(struct vfsmount *mnt) /* * vfsmount lock must be held for write */ -static void commit_tree(struct vfsmount *mnt) +static void commit_tree(struct mount *mnt) { - struct vfsmount *parent = mnt->mnt_parent; - struct vfsmount *m; + struct mount *parent = mnt->mnt_parent; + struct mount *m; LIST_HEAD(head); struct mnt_namespace *n = parent->mnt_ns; @@ -639,12 +657,12 @@ static void commit_tree(struct vfsmount *mnt) list_splice(&head, n->list.prev); list_add_tail(&mnt->mnt_hash, mount_hashtable + - hash(parent, mnt->mnt_mountpoint)); + hash(&parent->mnt, mnt->mnt_mountpoint)); list_add_tail(&mnt->mnt_child, &parent->mnt_mounts); touch_mnt_namespace(n); } -static struct vfsmount *next_mnt(struct vfsmount *p, struct vfsmount *root) +static struct mount *next_mnt(struct mount *p, struct mount *root) { struct list_head *next = p->mnt_mounts.next; if (next == &p->mnt_mounts) { @@ -657,14 +675,14 @@ static struct vfsmount *next_mnt(struct vfsmount *p, struct vfsmount *root) p = p->mnt_parent; } } - return list_entry(next, struct vfsmount, mnt_child); + return list_entry(next, struct mount, mnt_child); } -static struct vfsmount *skip_mnt_tree(struct vfsmount *p) +static struct mount *skip_mnt_tree(struct mount *p) { struct list_head *prev = p->mnt_mounts.prev; while (prev != &p->mnt_mounts) { - p = list_entry(prev, struct vfsmount, mnt_child); + p = list_entry(prev, struct mount, mnt_child); prev = p->mnt_mounts.prev; } return p; @@ -673,7 +691,7 @@ static struct vfsmount *skip_mnt_tree(struct vfsmount *p) struct vfsmount * vfs_kern_mount(struct file_system_type *type, int flags, const char *name, void *data) { - struct vfsmount *mnt; + struct mount *mnt; struct dentry *root; if (!type) @@ -684,7 +702,7 @@ vfs_kern_mount(struct file_system_type *type, int flags, const char *name, void return ERR_PTR(-ENOMEM); if (flags & MS_KERNMOUNT) - mnt->mnt_flags = MNT_INTERNAL; + mnt->mnt.mnt_flags = MNT_INTERNAL; root = mount_fs(type, flags, name, data); if (IS_ERR(root)) { @@ -692,19 +710,22 @@ vfs_kern_mount(struct file_system_type *type, int flags, const char *name, void return ERR_CAST(root); } - mnt->mnt_root = root; - mnt->mnt_sb = root->d_sb; - mnt->mnt_mountpoint = mnt->mnt_root; + mnt->mnt.mnt_root = root; + mnt->mnt.mnt_sb = root->d_sb; + mnt->mnt_mountpoint = mnt->mnt.mnt_root; mnt->mnt_parent = mnt; - return mnt; + br_write_lock(vfsmount_lock); + list_add_tail(&mnt->mnt_instance, &root->d_sb->s_mounts); + br_write_unlock(vfsmount_lock); + return &mnt->mnt; } EXPORT_SYMBOL_GPL(vfs_kern_mount); -static struct vfsmount *clone_mnt(struct vfsmount *old, struct dentry *root, +static struct mount *clone_mnt(struct mount *old, struct dentry *root, int flag) { - struct super_block *sb = old->mnt_sb; - struct vfsmount *mnt = alloc_vfsmnt(old->mnt_devname); + struct super_block *sb = old->mnt.mnt_sb; + struct mount *mnt = alloc_vfsmnt(old->mnt_devname); if (mnt) { if (flag & (CL_SLAVE | CL_PRIVATE)) @@ -718,12 +739,15 @@ static struct vfsmount *clone_mnt(struct vfsmount *old, struct dentry *root, goto out_free; } - mnt->mnt_flags = old->mnt_flags & ~MNT_WRITE_HOLD; + mnt->mnt.mnt_flags = old->mnt.mnt_flags & ~MNT_WRITE_HOLD; atomic_inc(&sb->s_active); - mnt->mnt_sb = sb; - mnt->mnt_root = dget(root); - mnt->mnt_mountpoint = mnt->mnt_root; + mnt->mnt.mnt_sb = sb; + mnt->mnt.mnt_root = dget(root); + mnt->mnt_mountpoint = mnt->mnt.mnt_root; mnt->mnt_parent = mnt; + br_write_lock(vfsmount_lock); + list_add_tail(&mnt->mnt_instance, &sb->s_mounts); + br_write_unlock(vfsmount_lock); if (flag & CL_SLAVE) { list_add(&mnt->mnt_slave, &old->mnt_slave_list); @@ -753,9 +777,10 @@ static struct vfsmount *clone_mnt(struct vfsmount *old, struct dentry *root, return NULL; } -static inline void mntfree(struct vfsmount *mnt) +static inline void mntfree(struct mount *mnt) { - struct super_block *sb = mnt->mnt_sb; + struct vfsmount *m = &mnt->mnt; + struct super_block *sb = m->mnt_sb; /* * This probably indicates that somebody messed @@ -768,32 +793,32 @@ static inline void mntfree(struct vfsmount *mnt) * so mnt_get_writers() below is safe. */ WARN_ON(mnt_get_writers(mnt)); - fsnotify_vfsmount_delete(mnt); - dput(mnt->mnt_root); + fsnotify_vfsmount_delete(m); + dput(m->mnt_root); free_vfsmnt(mnt); deactivate_super(sb); } -static void mntput_no_expire(struct vfsmount *mnt) +static void mntput_no_expire(struct mount *mnt) { put_again: #ifdef CONFIG_SMP br_read_lock(vfsmount_lock); if (likely(atomic_read(&mnt->mnt_longterm))) { - mnt_dec_count(mnt); + mnt_add_count(mnt, -1); br_read_unlock(vfsmount_lock); return; } br_read_unlock(vfsmount_lock); br_write_lock(vfsmount_lock); - mnt_dec_count(mnt); + mnt_add_count(mnt, -1); if (mnt_get_count(mnt)) { br_write_unlock(vfsmount_lock); return; } #else - mnt_dec_count(mnt); + mnt_add_count(mnt, -1); if (likely(mnt_get_count(mnt))) return; br_write_lock(vfsmount_lock); @@ -802,9 +827,10 @@ put_again: mnt_add_count(mnt, mnt->mnt_pinned + 1); mnt->mnt_pinned = 0; br_write_unlock(vfsmount_lock); - acct_auto_close_mnt(mnt); + acct_auto_close_mnt(&mnt->mnt); goto put_again; } + list_del(&mnt->mnt_instance); br_write_unlock(vfsmount_lock); mntfree(mnt); } @@ -812,10 +838,11 @@ put_again: void mntput(struct vfsmount *mnt) { if (mnt) { + struct mount *m = real_mount(mnt); /* avoid cacheline pingpong, hope gcc doesn't get "smart" */ - if (unlikely(mnt->mnt_expiry_mark)) - mnt->mnt_expiry_mark = 0; - mntput_no_expire(mnt); + if (unlikely(m->mnt_expiry_mark)) + m->mnt_expiry_mark = 0; + mntput_no_expire(m); } } EXPORT_SYMBOL(mntput); @@ -823,7 +850,7 @@ EXPORT_SYMBOL(mntput); struct vfsmount *mntget(struct vfsmount *mnt) { if (mnt) - mnt_inc_count(mnt); + mnt_add_count(real_mount(mnt), 1); return mnt; } EXPORT_SYMBOL(mntget); @@ -831,16 +858,17 @@ EXPORT_SYMBOL(mntget); void mnt_pin(struct vfsmount *mnt) { br_write_lock(vfsmount_lock); - mnt->mnt_pinned++; + real_mount(mnt)->mnt_pinned++; br_write_unlock(vfsmount_lock); } EXPORT_SYMBOL(mnt_pin); -void mnt_unpin(struct vfsmount *mnt) +void mnt_unpin(struct vfsmount *m) { + struct mount *mnt = real_mount(m); br_write_lock(vfsmount_lock); if (mnt->mnt_pinned) { - mnt_inc_count(mnt); + mnt_add_count(mnt, 1); mnt->mnt_pinned--; } br_write_unlock(vfsmount_lock); @@ -858,12 +886,12 @@ static inline void mangle(struct seq_file *m, const char *s) * * See also save_mount_options(). */ -int generic_show_options(struct seq_file *m, struct vfsmount *mnt) +int generic_show_options(struct seq_file *m, struct dentry *root) { const char *options; rcu_read_lock(); - options = rcu_dereference(mnt->mnt_sb->s_options); + options = rcu_dereference(root->d_sb->s_options); if (options != NULL && options[0]) { seq_putc(m, ','); @@ -907,10 +935,10 @@ void replace_mount_options(struct super_block *sb, char *options) EXPORT_SYMBOL(replace_mount_options); #ifdef CONFIG_PROC_FS -/* iterator */ +/* iterator; we want it to have access to namespace_sem, thus here... */ static void *m_start(struct seq_file *m, loff_t *pos) { - struct proc_mounts *p = m->private; + struct proc_mounts *p = container_of(m, struct proc_mounts, m); down_read(&namespace_sem); return seq_list_start(&p->ns->list, *pos); @@ -918,7 +946,7 @@ static void *m_start(struct seq_file *m, loff_t *pos) static void *m_next(struct seq_file *m, void *v, loff_t *pos) { - struct proc_mounts *p = m->private; + struct proc_mounts *p = container_of(m, struct proc_mounts, m); return seq_list_next(v, &p->ns->list, pos); } @@ -928,222 +956,18 @@ static void m_stop(struct seq_file *m, void *v) up_read(&namespace_sem); } -int mnt_had_events(struct proc_mounts *p) -{ - struct mnt_namespace *ns = p->ns; - int res = 0; - - br_read_lock(vfsmount_lock); - if (p->m.poll_event != ns->event) { - p->m.poll_event = ns->event; - res = 1; - } - br_read_unlock(vfsmount_lock); - - return res; -} - -struct proc_fs_info { - int flag; - const char *str; -}; - -static int show_sb_opts(struct seq_file *m, struct super_block *sb) +static int m_show(struct seq_file *m, void *v) { - static const struct proc_fs_info fs_info[] = { - { MS_SYNCHRONOUS, ",sync" }, - { MS_DIRSYNC, ",dirsync" }, - { MS_MANDLOCK, ",mand" }, - { 0, NULL } - }; - const struct proc_fs_info *fs_infop; - - for (fs_infop = fs_info; fs_infop->flag; fs_infop++) { - if (sb->s_flags & fs_infop->flag) - seq_puts(m, fs_infop->str); - } - - return security_sb_show_options(m, sb); -} - -static void show_mnt_opts(struct seq_file *m, struct vfsmount *mnt) -{ - static const struct proc_fs_info mnt_info[] = { - { MNT_NOSUID, ",nosuid" }, - { MNT_NODEV, ",nodev" }, - { MNT_NOEXEC, ",noexec" }, - { MNT_NOATIME, ",noatime" }, - { MNT_NODIRATIME, ",nodiratime" }, - { MNT_RELATIME, ",relatime" }, - { 0, NULL } - }; - const struct proc_fs_info *fs_infop; - - for (fs_infop = mnt_info; fs_infop->flag; fs_infop++) { - if (mnt->mnt_flags & fs_infop->flag) - seq_puts(m, fs_infop->str); - } -} - -static void show_type(struct seq_file *m, struct super_block *sb) -{ - mangle(m, sb->s_type->name); - if (sb->s_subtype && sb->s_subtype[0]) { - seq_putc(m, '.'); - mangle(m, sb->s_subtype); - } -} - -static int show_vfsmnt(struct seq_file *m, void *v) -{ - struct vfsmount *mnt = list_entry(v, struct vfsmount, mnt_list); - int err = 0; - struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt }; - - if (mnt->mnt_sb->s_op->show_devname) { - err = mnt->mnt_sb->s_op->show_devname(m, mnt); - if (err) - goto out; - } else { - mangle(m, mnt->mnt_devname ? mnt->mnt_devname : "none"); - } - seq_putc(m, ' '); - seq_path(m, &mnt_path, " \t\n\\"); - seq_putc(m, ' '); - show_type(m, mnt->mnt_sb); - seq_puts(m, __mnt_is_readonly(mnt) ? " ro" : " rw"); - err = show_sb_opts(m, mnt->mnt_sb); - if (err) - goto out; - show_mnt_opts(m, mnt); - if (mnt->mnt_sb->s_op->show_options) - err = mnt->mnt_sb->s_op->show_options(m, mnt); - seq_puts(m, " 0 0\n"); -out: - return err; + struct proc_mounts *p = container_of(m, struct proc_mounts, m); + struct mount *r = list_entry(v, struct mount, mnt_list); + return p->show(m, &r->mnt); } const struct seq_operations mounts_op = { .start = m_start, .next = m_next, .stop = m_stop, - .show = show_vfsmnt -}; - -static int show_mountinfo(struct seq_file *m, void *v) -{ - struct proc_mounts *p = m->private; - struct vfsmount *mnt = list_entry(v, struct vfsmount, mnt_list); - struct super_block *sb = mnt->mnt_sb; - struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt }; - struct path root = p->root; - int err = 0; - - seq_printf(m, "%i %i %u:%u ", mnt->mnt_id, mnt->mnt_parent->mnt_id, - MAJOR(sb->s_dev), MINOR(sb->s_dev)); - if (sb->s_op->show_path) - err = sb->s_op->show_path(m, mnt); - else - seq_dentry(m, mnt->mnt_root, " \t\n\\"); - if (err) - goto out; - seq_putc(m, ' '); - seq_path_root(m, &mnt_path, &root, " \t\n\\"); - if (root.mnt != p->root.mnt || root.dentry != p->root.dentry) { - /* - * Mountpoint is outside root, discard that one. Ugly, - * but less so than trying to do that in iterator in a - * race-free way (due to renames). - */ - return SEQ_SKIP; - } - seq_puts(m, mnt->mnt_flags & MNT_READONLY ? " ro" : " rw"); - show_mnt_opts(m, mnt); - - /* Tagged fields ("foo:X" or "bar") */ - if (IS_MNT_SHARED(mnt)) - seq_printf(m, " shared:%i", mnt->mnt_group_id); - if (IS_MNT_SLAVE(mnt)) { - int master = mnt->mnt_master->mnt_group_id; - int dom = get_dominating_id(mnt, &p->root); - seq_printf(m, " master:%i", master); - if (dom && dom != master) - seq_printf(m, " propagate_from:%i", dom); - } - if (IS_MNT_UNBINDABLE(mnt)) - seq_puts(m, " unbindable"); - - /* Filesystem specific data */ - seq_puts(m, " - "); - show_type(m, sb); - seq_putc(m, ' '); - if (sb->s_op->show_devname) - err = sb->s_op->show_devname(m, mnt); - else - mangle(m, mnt->mnt_devname ? mnt->mnt_devname : "none"); - if (err) - goto out; - seq_puts(m, sb->s_flags & MS_RDONLY ? " ro" : " rw"); - err = show_sb_opts(m, sb); - if (err) - goto out; - if (sb->s_op->show_options) - err = sb->s_op->show_options(m, mnt); - seq_putc(m, '\n'); -out: - return err; -} - -const struct seq_operations mountinfo_op = { - .start = m_start, - .next = m_next, - .stop = m_stop, - .show = show_mountinfo, -}; - -static int show_vfsstat(struct seq_file *m, void *v) -{ - struct vfsmount *mnt = list_entry(v, struct vfsmount, mnt_list); - struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt }; - int err = 0; - - /* device */ - if (mnt->mnt_sb->s_op->show_devname) { - seq_puts(m, "device "); - err = mnt->mnt_sb->s_op->show_devname(m, mnt); - } else { - if (mnt->mnt_devname) { - seq_puts(m, "device "); - mangle(m, mnt->mnt_devname); - } else - seq_puts(m, "no device"); - } - - /* mount point */ - seq_puts(m, " mounted on "); - seq_path(m, &mnt_path, " \t\n\\"); - seq_putc(m, ' '); - - /* file system type */ - seq_puts(m, "with fstype "); - show_type(m, mnt->mnt_sb); - - /* optional statistics */ - if (mnt->mnt_sb->s_op->show_stats) { - seq_putc(m, ' '); - if (!err) - err = mnt->mnt_sb->s_op->show_stats(m, mnt); - } - - seq_putc(m, '\n'); - return err; -} - -const struct seq_operations mountstats_op = { - .start = m_start, - .next = m_next, - .stop = m_stop, - .show = show_vfsstat, + .show = m_show, }; #endif /* CONFIG_PROC_FS */ @@ -1155,11 +979,13 @@ const struct seq_operations mountstats_op = { * open files, pwds, chroots or sub mounts that are * busy. */ -int may_umount_tree(struct vfsmount *mnt) +int may_umount_tree(struct vfsmount *m) { + struct mount *mnt = real_mount(m); int actual_refs = 0; int minimum_refs = 0; - struct vfsmount *p; + struct mount *p; + BUG_ON(!m); /* write lock needed for mnt_get_count */ br_write_lock(vfsmount_lock); @@ -1195,7 +1021,7 @@ int may_umount(struct vfsmount *mnt) int ret = 1; down_read(&namespace_sem); br_write_lock(vfsmount_lock); - if (propagate_mount_busy(mnt, 2)) + if (propagate_mount_busy(real_mount(mnt), 2)) ret = 0; br_write_unlock(vfsmount_lock); up_read(&namespace_sem); @@ -1206,25 +1032,25 @@ EXPORT_SYMBOL(may_umount); void release_mounts(struct list_head *head) { - struct vfsmount *mnt; + struct mount *mnt; while (!list_empty(head)) { - mnt = list_first_entry(head, struct vfsmount, mnt_hash); + mnt = list_first_entry(head, struct mount, mnt_hash); list_del_init(&mnt->mnt_hash); - if (mnt->mnt_parent != mnt) { + if (mnt_has_parent(mnt)) { struct dentry *dentry; - struct vfsmount *m; + struct mount *m; br_write_lock(vfsmount_lock); dentry = mnt->mnt_mountpoint; m = mnt->mnt_parent; - mnt->mnt_mountpoint = mnt->mnt_root; + mnt->mnt_mountpoint = mnt->mnt.mnt_root; mnt->mnt_parent = mnt; m->mnt_ghosts--; br_write_unlock(vfsmount_lock); dput(dentry); - mntput(m); + mntput(&m->mnt); } - mntput(mnt); + mntput(&mnt->mnt); } } @@ -1232,10 +1058,10 @@ void release_mounts(struct list_head *head) * vfsmount lock must be held for write * namespace_sem must be held for write */ -void umount_tree(struct vfsmount *mnt, int propagate, struct list_head *kill) +void umount_tree(struct mount *mnt, int propagate, struct list_head *kill) { LIST_HEAD(tmp_list); - struct vfsmount *p; + struct mount *p; for (p = mnt; p; p = next_mnt(p, mnt)) list_move(&p->mnt_hash, &tmp_list); @@ -1250,24 +1076,24 @@ void umount_tree(struct vfsmount *mnt, int propagate, struct list_head *kill) p->mnt_ns = NULL; __mnt_make_shortterm(p); list_del_init(&p->mnt_child); - if (p->mnt_parent != p) { + if (mnt_has_parent(p)) { p->mnt_parent->mnt_ghosts++; - dentry_reset_mounted(p->mnt_parent, p->mnt_mountpoint); + dentry_reset_mounted(p->mnt_mountpoint); } change_mnt_propagation(p, MS_PRIVATE); } list_splice(&tmp_list, kill); } -static void shrink_submounts(struct vfsmount *mnt, struct list_head *umounts); +static void shrink_submounts(struct mount *mnt, struct list_head *umounts); -static int do_umount(struct vfsmount *mnt, int flags) +static int do_umount(struct mount *mnt, int flags) { - struct super_block *sb = mnt->mnt_sb; + struct super_block *sb = mnt->mnt.mnt_sb; int retval; LIST_HEAD(umount_list); - retval = security_sb_umount(mnt, flags); + retval = security_sb_umount(&mnt->mnt, flags); if (retval) return retval; @@ -1278,7 +1104,7 @@ static int do_umount(struct vfsmount *mnt, int flags) * (2) the usage count == 1 [parent vfsmount] + 1 [sys_umount] */ if (flags & MNT_EXPIRE) { - if (mnt == current->fs->root.mnt || + if (&mnt->mnt == current->fs->root.mnt || flags & (MNT_FORCE | MNT_DETACH)) return -EINVAL; @@ -1320,7 +1146,7 @@ static int do_umount(struct vfsmount *mnt, int flags) * /reboot - static binary that would close all descriptors and * call reboot(9). Then init(8) could umount root and exec /reboot. */ - if (mnt == current->fs->root.mnt && !(flags & MNT_DETACH)) { + if (&mnt->mnt == current->fs->root.mnt && !(flags & MNT_DETACH)) { /* * Special case for "unmounting" root ... * we just try to remount it readonly. @@ -1362,6 +1188,7 @@ static int do_umount(struct vfsmount *mnt, int flags) SYSCALL_DEFINE2(umount, char __user *, name, int, flags) { struct path path; + struct mount *mnt; int retval; int lookup_flags = 0; @@ -1374,21 +1201,22 @@ SYSCALL_DEFINE2(umount, char __user *, name, int, flags) retval = user_path_at(AT_FDCWD, name, lookup_flags, &path); if (retval) goto out; + mnt = real_mount(path.mnt); retval = -EINVAL; if (path.dentry != path.mnt->mnt_root) goto dput_and_out; - if (!check_mnt(path.mnt)) + if (!check_mnt(mnt)) goto dput_and_out; retval = -EPERM; if (!capable(CAP_SYS_ADMIN)) goto dput_and_out; - retval = do_umount(path.mnt, flags); + retval = do_umount(mnt, flags); dput_and_out: /* we mustn't call path_put() as that would clear mnt_expiry_mark */ dput(path.dentry); - mntput_no_expire(path.mnt); + mntput_no_expire(mnt); out: return retval; } @@ -1423,10 +1251,10 @@ static int mount_is_safe(struct path *path) #endif } -struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry, +struct mount *copy_tree(struct mount *mnt, struct dentry *dentry, int flag) { - struct vfsmount *res, *p, *q, *r, *s; + struct mount *res, *p, *q, *r; struct path path; if (!(flag & CL_COPY_ALL) && IS_MNT_UNBINDABLE(mnt)) @@ -1439,6 +1267,7 @@ struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry, p = mnt; list_for_each_entry(r, &mnt->mnt_mounts, mnt_child) { + struct mount *s; if (!is_subdir(r->mnt_mountpoint, dentry)) continue; @@ -1452,9 +1281,9 @@ struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry, q = q->mnt_parent; } p = s; - path.mnt = q; + path.mnt = &q->mnt; path.dentry = p->mnt_mountpoint; - q = clone_mnt(p, p->mnt_root, flag); + q = clone_mnt(p, p->mnt.mnt_root, flag); if (!q) goto Enomem; br_write_lock(vfsmount_lock); @@ -1477,11 +1306,12 @@ Enomem: struct vfsmount *collect_mounts(struct path *path) { - struct vfsmount *tree; + struct mount *tree; down_write(&namespace_sem); - tree = copy_tree(path->mnt, path->dentry, CL_COPY_ALL | CL_PRIVATE); + tree = copy_tree(real_mount(path->mnt), path->dentry, + CL_COPY_ALL | CL_PRIVATE); up_write(&namespace_sem); - return tree; + return tree ? &tree->mnt : NULL; } void drop_collected_mounts(struct vfsmount *mnt) @@ -1489,7 +1319,7 @@ void drop_collected_mounts(struct vfsmount *mnt) LIST_HEAD(umount_list); down_write(&namespace_sem); br_write_lock(vfsmount_lock); - umount_tree(mnt, 0, &umount_list); + umount_tree(real_mount(mnt), 0, &umount_list); br_write_unlock(vfsmount_lock); up_write(&namespace_sem); release_mounts(&umount_list); @@ -1498,21 +1328,21 @@ void drop_collected_mounts(struct vfsmount *mnt) int iterate_mounts(int (*f)(struct vfsmount *, void *), void *arg, struct vfsmount *root) { - struct vfsmount *mnt; + struct mount *mnt; int res = f(root, arg); if (res) return res; - list_for_each_entry(mnt, &root->mnt_list, mnt_list) { - res = f(mnt, arg); + list_for_each_entry(mnt, &real_mount(root)->mnt_list, mnt_list) { + res = f(&mnt->mnt, arg); if (res) return res; } return 0; } -static void cleanup_group_ids(struct vfsmount *mnt, struct vfsmount *end) +static void cleanup_group_ids(struct mount *mnt, struct mount *end) { - struct vfsmount *p; + struct mount *p; for (p = mnt; p != end; p = next_mnt(p, mnt)) { if (p->mnt_group_id && !IS_MNT_SHARED(p)) @@ -1520,9 +1350,9 @@ static void cleanup_group_ids(struct vfsmount *mnt, struct vfsmount *end) } } -static int invent_group_ids(struct vfsmount *mnt, bool recurse) +static int invent_group_ids(struct mount *mnt, bool recurse) { - struct vfsmount *p; + struct mount *p; for (p = mnt; p; p = recurse ? next_mnt(p, mnt) : NULL) { if (!p->mnt_group_id && !IS_MNT_SHARED(p)) { @@ -1600,13 +1430,13 @@ static int invent_group_ids(struct vfsmount *mnt, bool recurse) * Must be called without spinlocks held, since this function can sleep * in allocations. */ -static int attach_recursive_mnt(struct vfsmount *source_mnt, +static int attach_recursive_mnt(struct mount *source_mnt, struct path *path, struct path *parent_path) { LIST_HEAD(tree_list); - struct vfsmount *dest_mnt = path->mnt; + struct mount *dest_mnt = real_mount(path->mnt); struct dentry *dest_dentry = path->dentry; - struct vfsmount *child, *p; + struct mount *child, *p; int err; if (IS_MNT_SHARED(dest_mnt)) { @@ -1627,7 +1457,7 @@ static int attach_recursive_mnt(struct vfsmount *source_mnt, if (parent_path) { detach_mnt(source_mnt, parent_path); attach_mnt(source_mnt, path); - touch_mnt_namespace(parent_path->mnt->mnt_ns); + touch_mnt_namespace(source_mnt->mnt_ns); } else { mnt_set_mountpoint(dest_mnt, dest_dentry, source_mnt); commit_tree(source_mnt); @@ -1675,13 +1505,13 @@ static void unlock_mount(struct path *path) mutex_unlock(&path->dentry->d_inode->i_mutex); } -static int graft_tree(struct vfsmount *mnt, struct path *path) +static int graft_tree(struct mount *mnt, struct path *path) { - if (mnt->mnt_sb->s_flags & MS_NOUSER) + if (mnt->mnt.mnt_sb->s_flags & MS_NOUSER) return -EINVAL; if (S_ISDIR(path->dentry->d_inode->i_mode) != - S_ISDIR(mnt->mnt_root->d_inode->i_mode)) + S_ISDIR(mnt->mnt.mnt_root->d_inode->i_mode)) return -ENOTDIR; if (d_unlinked(path->dentry)) @@ -1712,7 +1542,8 @@ static int flags_to_propagation_type(int flags) */ static int do_change_type(struct path *path, int flag) { - struct vfsmount *m, *mnt = path->mnt; + struct mount *m; + struct mount *mnt = real_mount(path->mnt); int recurse = flag & MS_REC; int type; int err = 0; @@ -1752,7 +1583,7 @@ static int do_loopback(struct path *path, char *old_name, { LIST_HEAD(umount_list); struct path old_path; - struct vfsmount *mnt = NULL; + struct mount *mnt = NULL, *old; int err = mount_is_safe(path); if (err) return err; @@ -1766,18 +1597,20 @@ static int do_loopback(struct path *path, char *old_name, if (err) goto out; + old = real_mount(old_path.mnt); + err = -EINVAL; - if (IS_MNT_UNBINDABLE(old_path.mnt)) + if (IS_MNT_UNBINDABLE(old)) goto out2; - if (!check_mnt(path->mnt) || !check_mnt(old_path.mnt)) + if (!check_mnt(real_mount(path->mnt)) || !check_mnt(old)) goto out2; err = -ENOMEM; if (recurse) - mnt = copy_tree(old_path.mnt, old_path.dentry, 0); + mnt = copy_tree(old, old_path.dentry, 0); else - mnt = clone_mnt(old_path.mnt, old_path.dentry, 0); + mnt = clone_mnt(old, old_path.dentry, 0); if (!mnt) goto out2; @@ -1807,9 +1640,9 @@ static int change_mount_flags(struct vfsmount *mnt, int ms_flags) return 0; if (readonly_request) - error = mnt_make_readonly(mnt); + error = mnt_make_readonly(real_mount(mnt)); else - __mnt_unmake_readonly(mnt); + __mnt_unmake_readonly(real_mount(mnt)); return error; } @@ -1823,11 +1656,12 @@ static int do_remount(struct path *path, int flags, int mnt_flags, { int err; struct super_block *sb = path->mnt->mnt_sb; + struct mount *mnt = real_mount(path->mnt); if (!capable(CAP_SYS_ADMIN)) return -EPERM; - if (!check_mnt(path->mnt)) + if (!check_mnt(mnt)) return -EINVAL; if (path->dentry != path->mnt->mnt_root) @@ -1844,22 +1678,22 @@ static int do_remount(struct path *path, int flags, int mnt_flags, err = do_remount_sb(sb, flags, data, 0); if (!err) { br_write_lock(vfsmount_lock); - mnt_flags |= path->mnt->mnt_flags & MNT_PROPAGATION_MASK; - path->mnt->mnt_flags = mnt_flags; + mnt_flags |= mnt->mnt.mnt_flags & MNT_PROPAGATION_MASK; + mnt->mnt.mnt_flags = mnt_flags; br_write_unlock(vfsmount_lock); } up_write(&sb->s_umount); if (!err) { br_write_lock(vfsmount_lock); - touch_mnt_namespace(path->mnt->mnt_ns); + touch_mnt_namespace(mnt->mnt_ns); br_write_unlock(vfsmount_lock); } return err; } -static inline int tree_contains_unbindable(struct vfsmount *mnt) +static inline int tree_contains_unbindable(struct mount *mnt) { - struct vfsmount *p; + struct mount *p; for (p = mnt; p; p = next_mnt(p, mnt)) { if (IS_MNT_UNBINDABLE(p)) return 1; @@ -1870,7 +1704,8 @@ static inline int tree_contains_unbindable(struct vfsmount *mnt) static int do_move_mount(struct path *path, char *old_name) { struct path old_path, parent_path; - struct vfsmount *p; + struct mount *p; + struct mount *old; int err = 0; if (!capable(CAP_SYS_ADMIN)) return -EPERM; @@ -1884,8 +1719,11 @@ static int do_move_mount(struct path *path, char *old_name) if (err < 0) goto out; + old = real_mount(old_path.mnt); + p = real_mount(path->mnt); + err = -EINVAL; - if (!check_mnt(path->mnt) || !check_mnt(old_path.mnt)) + if (!check_mnt(p) || !check_mnt(old)) goto out1; if (d_unlinked(path->dentry)) @@ -1895,7 +1733,7 @@ static int do_move_mount(struct path *path, char *old_name) if (old_path.dentry != old_path.mnt->mnt_root) goto out1; - if (old_path.mnt == old_path.mnt->mnt_parent) + if (!mnt_has_parent(old)) goto out1; if (S_ISDIR(path->dentry->d_inode->i_mode) != @@ -1904,28 +1742,26 @@ static int do_move_mount(struct path *path, char *old_name) /* * Don't move a mount residing in a shared parent. */ - if (old_path.mnt->mnt_parent && - IS_MNT_SHARED(old_path.mnt->mnt_parent)) + if (IS_MNT_SHARED(old->mnt_parent)) goto out1; /* * Don't move a mount tree containing unbindable mounts to a destination * mount which is shared. */ - if (IS_MNT_SHARED(path->mnt) && - tree_contains_unbindable(old_path.mnt)) + if (IS_MNT_SHARED(p) && tree_contains_unbindable(old)) goto out1; err = -ELOOP; - for (p = path->mnt; p->mnt_parent != p; p = p->mnt_parent) - if (p == old_path.mnt) + for (; mnt_has_parent(p); p = p->mnt_parent) + if (p == old) goto out1; - err = attach_recursive_mnt(old_path.mnt, path, &parent_path); + err = attach_recursive_mnt(old, path, &parent_path); if (err) goto out1; /* if the mount is moved, it should no longer be expire * automatically */ - list_del_init(&old_path.mnt->mnt_expire); + list_del_init(&old->mnt_expire); out1: unlock_mount(path); out: @@ -1958,7 +1794,7 @@ static struct vfsmount *fs_set_subtype(struct vfsmount *mnt, const char *fstype) return ERR_PTR(err); } -struct vfsmount * +static struct vfsmount * do_kern_mount(const char *fstype, int flags, const char *name, void *data) { struct file_system_type *type = get_fs_type(fstype); @@ -1972,12 +1808,11 @@ do_kern_mount(const char *fstype, int flags, const char *name, void *data) put_filesystem(type); return mnt; } -EXPORT_SYMBOL_GPL(do_kern_mount); /* * add a mount into a namespace's mount tree */ -static int do_add_mount(struct vfsmount *newmnt, struct path *path, int mnt_flags) +static int do_add_mount(struct mount *newmnt, struct path *path, int mnt_flags) { int err; @@ -1988,20 +1823,20 @@ static int do_add_mount(struct vfsmount *newmnt, struct path *path, int mnt_flag return err; err = -EINVAL; - if (!(mnt_flags & MNT_SHRINKABLE) && !check_mnt(path->mnt)) + if (!(mnt_flags & MNT_SHRINKABLE) && !check_mnt(real_mount(path->mnt))) goto unlock; /* Refuse the same filesystem on the same mount point */ err = -EBUSY; - if (path->mnt->mnt_sb == newmnt->mnt_sb && + if (path->mnt->mnt_sb == newmnt->mnt.mnt_sb && path->mnt->mnt_root == path->dentry) goto unlock; err = -EINVAL; - if (S_ISLNK(newmnt->mnt_root->d_inode->i_mode)) + if (S_ISLNK(newmnt->mnt.mnt_root->d_inode->i_mode)) goto unlock; - newmnt->mnt_flags = mnt_flags; + newmnt->mnt.mnt_flags = mnt_flags; err = graft_tree(newmnt, path); unlock: @@ -2030,7 +1865,7 @@ static int do_new_mount(struct path *path, char *type, int flags, if (IS_ERR(mnt)) return PTR_ERR(mnt); - err = do_add_mount(mnt, path, mnt_flags); + err = do_add_mount(real_mount(mnt), path, mnt_flags); if (err) mntput(mnt); return err; @@ -2038,11 +1873,12 @@ static int do_new_mount(struct path *path, char *type, int flags, int finish_automount(struct vfsmount *m, struct path *path) { + struct mount *mnt = real_mount(m); int err; /* The new mount record should have at least 2 refs to prevent it being * expired before we get a chance to add it */ - BUG_ON(mnt_get_count(m) < 2); + BUG_ON(mnt_get_count(mnt) < 2); if (m->mnt_sb == path->mnt->mnt_sb && m->mnt_root == path->dentry) { @@ -2050,15 +1886,15 @@ int finish_automount(struct vfsmount *m, struct path *path) goto fail; } - err = do_add_mount(m, path, path->mnt->mnt_flags | MNT_SHRINKABLE); + err = do_add_mount(mnt, path, path->mnt->mnt_flags | MNT_SHRINKABLE); if (!err) return 0; fail: /* remove m from any expiration list it may be on */ - if (!list_empty(&m->mnt_expire)) { + if (!list_empty(&mnt->mnt_expire)) { down_write(&namespace_sem); br_write_lock(vfsmount_lock); - list_del_init(&m->mnt_expire); + list_del_init(&mnt->mnt_expire); br_write_unlock(vfsmount_lock); up_write(&namespace_sem); } @@ -2077,7 +1913,7 @@ void mnt_set_expiry(struct vfsmount *mnt, struct list_head *expiry_list) down_write(&namespace_sem); br_write_lock(vfsmount_lock); - list_add_tail(&mnt->mnt_expire, expiry_list); + list_add_tail(&real_mount(mnt)->mnt_expire, expiry_list); br_write_unlock(vfsmount_lock); up_write(&namespace_sem); @@ -2091,7 +1927,7 @@ EXPORT_SYMBOL(mnt_set_expiry); */ void mark_mounts_for_expiry(struct list_head *mounts) { - struct vfsmount *mnt, *next; + struct mount *mnt, *next; LIST_HEAD(graveyard); LIST_HEAD(umounts); @@ -2114,7 +1950,7 @@ void mark_mounts_for_expiry(struct list_head *mounts) list_move(&mnt->mnt_expire, &graveyard); } while (!list_empty(&graveyard)) { - mnt = list_first_entry(&graveyard, struct vfsmount, mnt_expire); + mnt = list_first_entry(&graveyard, struct mount, mnt_expire); touch_mnt_namespace(mnt->mnt_ns); umount_tree(mnt, 1, &umounts); } @@ -2132,9 +1968,9 @@ EXPORT_SYMBOL_GPL(mark_mounts_for_expiry); * search the list of submounts for a given mountpoint, and move any * shrinkable submounts to the 'graveyard' list. */ -static int select_submounts(struct vfsmount *parent, struct list_head *graveyard) +static int select_submounts(struct mount *parent, struct list_head *graveyard) { - struct vfsmount *this_parent = parent; + struct mount *this_parent = parent; struct list_head *next; int found = 0; @@ -2143,10 +1979,10 @@ repeat: resume: while (next != &this_parent->mnt_mounts) { struct list_head *tmp = next; - struct vfsmount *mnt = list_entry(tmp, struct vfsmount, mnt_child); + struct mount *mnt = list_entry(tmp, struct mount, mnt_child); next = tmp->next; - if (!(mnt->mnt_flags & MNT_SHRINKABLE)) + if (!(mnt->mnt.mnt_flags & MNT_SHRINKABLE)) continue; /* * Descend a level if the d_mounts list is non-empty. @@ -2178,15 +2014,15 @@ resume: * * vfsmount_lock must be held for write */ -static void shrink_submounts(struct vfsmount *mnt, struct list_head *umounts) +static void shrink_submounts(struct mount *mnt, struct list_head *umounts) { LIST_HEAD(graveyard); - struct vfsmount *m; + struct mount *m; /* extract submounts of 'mountpoint' from the expiration list */ while (select_submounts(mnt, &graveyard)) { while (!list_empty(&graveyard)) { - m = list_first_entry(&graveyard, struct vfsmount, + m = list_first_entry(&graveyard, struct mount, mnt_expire); touch_mnt_namespace(m->mnt_ns); umount_tree(m, 1, umounts); @@ -2373,12 +2209,13 @@ static struct mnt_namespace *alloc_mnt_ns(void) void mnt_make_longterm(struct vfsmount *mnt) { - __mnt_make_longterm(mnt); + __mnt_make_longterm(real_mount(mnt)); } -void mnt_make_shortterm(struct vfsmount *mnt) +void mnt_make_shortterm(struct vfsmount *m) { #ifdef CONFIG_SMP + struct mount *mnt = real_mount(m); if (atomic_add_unless(&mnt->mnt_longterm, -1, 1)) return; br_write_lock(vfsmount_lock); @@ -2396,7 +2233,9 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, { struct mnt_namespace *new_ns; struct vfsmount *rootmnt = NULL, *pwdmnt = NULL; - struct vfsmount *p, *q; + struct mount *p, *q; + struct mount *old = mnt_ns->root; + struct mount *new; new_ns = alloc_mnt_ns(); if (IS_ERR(new_ns)) @@ -2404,15 +2243,15 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, down_write(&namespace_sem); /* First pass: copy the tree topology */ - new_ns->root = copy_tree(mnt_ns->root, mnt_ns->root->mnt_root, - CL_COPY_ALL | CL_EXPIRE); - if (!new_ns->root) { + new = copy_tree(old, old->mnt.mnt_root, CL_COPY_ALL | CL_EXPIRE); + if (!new) { up_write(&namespace_sem); kfree(new_ns); return ERR_PTR(-ENOMEM); } + new_ns->root = new; br_write_lock(vfsmount_lock); - list_add_tail(&new_ns->list, &new_ns->root->mnt_list); + list_add_tail(&new_ns->list, &new->mnt_list); br_write_unlock(vfsmount_lock); /* @@ -2420,27 +2259,27 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, * as belonging to new namespace. We have already acquired a private * fs_struct, so tsk->fs->lock is not needed. */ - p = mnt_ns->root; - q = new_ns->root; + p = old; + q = new; while (p) { q->mnt_ns = new_ns; __mnt_make_longterm(q); if (fs) { - if (p == fs->root.mnt) { - fs->root.mnt = mntget(q); + if (&p->mnt == fs->root.mnt) { + fs->root.mnt = mntget(&q->mnt); __mnt_make_longterm(q); - mnt_make_shortterm(p); - rootmnt = p; + mnt_make_shortterm(&p->mnt); + rootmnt = &p->mnt; } - if (p == fs->pwd.mnt) { - fs->pwd.mnt = mntget(q); + if (&p->mnt == fs->pwd.mnt) { + fs->pwd.mnt = mntget(&q->mnt); __mnt_make_longterm(q); - mnt_make_shortterm(p); - pwdmnt = p; + mnt_make_shortterm(&p->mnt); + pwdmnt = &p->mnt; } } - p = next_mnt(p, mnt_ns->root); - q = next_mnt(q, new_ns->root); + p = next_mnt(p, old); + q = next_mnt(q, new); } up_write(&namespace_sem); @@ -2473,20 +2312,50 @@ struct mnt_namespace *copy_mnt_ns(unsigned long flags, struct mnt_namespace *ns, * create_mnt_ns - creates a private namespace and adds a root filesystem * @mnt: pointer to the new root filesystem mountpoint */ -struct mnt_namespace *create_mnt_ns(struct vfsmount *mnt) +static struct mnt_namespace *create_mnt_ns(struct vfsmount *m) { - struct mnt_namespace *new_ns; - - new_ns = alloc_mnt_ns(); + struct mnt_namespace *new_ns = alloc_mnt_ns(); if (!IS_ERR(new_ns)) { + struct mount *mnt = real_mount(m); mnt->mnt_ns = new_ns; __mnt_make_longterm(mnt); new_ns->root = mnt; - list_add(&new_ns->list, &new_ns->root->mnt_list); + list_add(&new_ns->list, &mnt->mnt_list); + } else { + mntput(m); } return new_ns; } -EXPORT_SYMBOL(create_mnt_ns); + +struct dentry *mount_subtree(struct vfsmount *mnt, const char *name) +{ + struct mnt_namespace *ns; + struct super_block *s; + struct path path; + int err; + + ns = create_mnt_ns(mnt); + if (IS_ERR(ns)) + return ERR_CAST(ns); + + err = vfs_path_lookup(mnt->mnt_root, mnt, + name, LOOKUP_FOLLOW|LOOKUP_AUTOMOUNT, &path); + + put_mnt_ns(ns); + + if (err) + return ERR_PTR(err); + + /* trade a vfsmount reference for active sb one */ + s = path.mnt->mnt_sb; + atomic_inc(&s->s_active); + mntput(path.mnt); + /* lock the sucker */ + down_write(&s->s_umount); + /* ... and return the root of (sub)tree on it */ + return path.dentry; +} +EXPORT_SYMBOL(mount_subtree); SYSCALL_DEFINE5(mount, char __user *, dev_name, char __user *, dir_name, char __user *, type, unsigned long, flags, void __user *, data) @@ -2530,6 +2399,31 @@ out_type: } /* + * Return true if path is reachable from root + * + * namespace_sem or vfsmount_lock is held + */ +bool is_path_reachable(struct mount *mnt, struct dentry *dentry, + const struct path *root) +{ + while (&mnt->mnt != root->mnt && mnt_has_parent(mnt)) { + dentry = mnt->mnt_mountpoint; + mnt = mnt->mnt_parent; + } + return &mnt->mnt == root->mnt && is_subdir(dentry, root->dentry); +} + +int path_is_under(struct path *path1, struct path *path2) +{ + int res; + br_read_lock(vfsmount_lock); + res = is_path_reachable(real_mount(path1->mnt), path1->dentry, path2); + br_read_unlock(vfsmount_lock); + return res; +} +EXPORT_SYMBOL(path_is_under); + +/* * pivot_root Semantics: * Moves the root file system of the current process to the directory put_old, * makes new_root as the new root file system of the current process, and sets @@ -2557,8 +2451,8 @@ out_type: SYSCALL_DEFINE2(pivot_root, const char __user *, new_root, const char __user *, put_old) { - struct vfsmount *tmp; struct path new, old, parent_path, root_parent, root; + struct mount *new_mnt, *root_mnt; int error; if (!capable(CAP_SYS_ADMIN)) @@ -2582,11 +2476,13 @@ SYSCALL_DEFINE2(pivot_root, const char __user *, new_root, goto out3; error = -EINVAL; - if (IS_MNT_SHARED(old.mnt) || - IS_MNT_SHARED(new.mnt->mnt_parent) || - IS_MNT_SHARED(root.mnt->mnt_parent)) + new_mnt = real_mount(new.mnt); + root_mnt = real_mount(root.mnt); + if (IS_MNT_SHARED(real_mount(old.mnt)) || + IS_MNT_SHARED(new_mnt->mnt_parent) || + IS_MNT_SHARED(root_mnt->mnt_parent)) goto out4; - if (!check_mnt(root.mnt) || !check_mnt(new.mnt)) + if (!check_mnt(root_mnt) || !check_mnt(new_mnt)) goto out4; error = -ENOENT; if (d_unlinked(new.dentry)) @@ -2600,33 +2496,22 @@ SYSCALL_DEFINE2(pivot_root, const char __user *, new_root, error = -EINVAL; if (root.mnt->mnt_root != root.dentry) goto out4; /* not a mountpoint */ - if (root.mnt->mnt_parent == root.mnt) + if (!mnt_has_parent(root_mnt)) goto out4; /* not attached */ if (new.mnt->mnt_root != new.dentry) goto out4; /* not a mountpoint */ - if (new.mnt->mnt_parent == new.mnt) + if (!mnt_has_parent(new_mnt)) goto out4; /* not attached */ /* make sure we can reach put_old from new_root */ - tmp = old.mnt; - if (tmp != new.mnt) { - for (;;) { - if (tmp->mnt_parent == tmp) - goto out4; /* already mounted on put_old */ - if (tmp->mnt_parent == new.mnt) - break; - tmp = tmp->mnt_parent; - } - if (!is_subdir(tmp->mnt_mountpoint, new.dentry)) - goto out4; - } else if (!is_subdir(old.dentry, new.dentry)) + if (!is_path_reachable(real_mount(old.mnt), old.dentry, &new)) goto out4; br_write_lock(vfsmount_lock); - detach_mnt(new.mnt, &parent_path); - detach_mnt(root.mnt, &root_parent); + detach_mnt(new_mnt, &parent_path); + detach_mnt(root_mnt, &root_parent); /* mount old root on put_old */ - attach_mnt(root.mnt, &old); + attach_mnt(root_mnt, &old); /* mount new_root on / */ - attach_mnt(new.mnt, &root_parent); + attach_mnt(new_mnt, &root_parent); touch_mnt_namespace(current->nsproxy->mnt_ns); br_write_unlock(vfsmount_lock); chroot_fs_refs(&root, &new); @@ -2664,8 +2549,8 @@ static void __init init_mount_tree(void) init_task.nsproxy->mnt_ns = ns; get_mnt_ns(ns); - root.mnt = ns->root; - root.dentry = ns->root->mnt_root; + root.mnt = mnt; + root.dentry = mnt->mnt_root; set_fs_pwd(current->fs, &root); set_fs_root(current->fs, &root); @@ -2678,7 +2563,7 @@ void __init mnt_init(void) init_rwsem(&namespace_sem); - mnt_cache = kmem_cache_create("mnt_cache", sizeof(struct vfsmount), + mnt_cache = kmem_cache_create("mnt_cache", sizeof(struct mount), 0, SLAB_HWCACHE_ALIGN | SLAB_PANIC, NULL); mount_hashtable = (struct list_head *)__get_free_page(GFP_ATOMIC); @@ -2718,7 +2603,6 @@ void put_mnt_ns(struct mnt_namespace *ns) release_mounts(&umount_list); kfree(ns); } -EXPORT_SYMBOL(put_mnt_ns); struct vfsmount *kern_mount_data(struct file_system_type *type, void *data) { @@ -2744,3 +2628,8 @@ void kern_unmount(struct vfsmount *mnt) } } EXPORT_SYMBOL(kern_unmount); + +bool our_mnt(struct vfsmount *mnt) +{ + return check_mnt(real_mount(mnt)); +} diff --git a/fs/ncpfs/dir.c b/fs/ncpfs/dir.c index 9c51f621e901..aeed93a6bde0 100644 --- a/fs/ncpfs/dir.c +++ b/fs/ncpfs/dir.c @@ -30,15 +30,15 @@ static void ncp_do_readdir(struct file *, void *, filldir_t, static int ncp_readdir(struct file *, void *, filldir_t); -static int ncp_create(struct inode *, struct dentry *, int, struct nameidata *); +static int ncp_create(struct inode *, struct dentry *, umode_t, struct nameidata *); static struct dentry *ncp_lookup(struct inode *, struct dentry *, struct nameidata *); static int ncp_unlink(struct inode *, struct dentry *); -static int ncp_mkdir(struct inode *, struct dentry *, int); +static int ncp_mkdir(struct inode *, struct dentry *, umode_t); static int ncp_rmdir(struct inode *, struct dentry *); static int ncp_rename(struct inode *, struct dentry *, struct inode *, struct dentry *); static int ncp_mknod(struct inode * dir, struct dentry *dentry, - int mode, dev_t rdev); + umode_t mode, dev_t rdev); #if defined(CONFIG_NCPFS_EXTRAS) || defined(CONFIG_NCPFS_NFS_NS) extern int ncp_symlink(struct inode *, struct dentry *, const char *); #else @@ -919,7 +919,7 @@ out_close: goto out; } -int ncp_create_new(struct inode *dir, struct dentry *dentry, int mode, +int ncp_create_new(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev, __le32 attributes) { struct ncp_server *server = NCP_SERVER(dir); @@ -928,7 +928,7 @@ int ncp_create_new(struct inode *dir, struct dentry *dentry, int mode, int opmode; __u8 __name[NCP_MAXPATHLEN + 1]; - PPRINTK("ncp_create_new: creating %s/%s, mode=%x\n", + PPRINTK("ncp_create_new: creating %s/%s, mode=%hx\n", dentry->d_parent->d_name.name, dentry->d_name.name, mode); ncp_age_dentry(server, dentry); @@ -979,13 +979,13 @@ out: return error; } -static int ncp_create(struct inode *dir, struct dentry *dentry, int mode, +static int ncp_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { return ncp_create_new(dir, dentry, mode, 0, 0); } -static int ncp_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int ncp_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct ncp_entry_info finfo; struct ncp_server *server = NCP_SERVER(dir); @@ -1201,12 +1201,12 @@ out: } static int ncp_mknod(struct inode * dir, struct dentry *dentry, - int mode, dev_t rdev) + umode_t mode, dev_t rdev) { if (!new_valid_dev(rdev)) return -EINVAL; if (ncp_is_nfs_extras(NCP_SERVER(dir), NCP_FINFO(dir)->volNumber)) { - DPRINTK(KERN_DEBUG "ncp_mknod: mode = 0%o\n", mode); + DPRINTK(KERN_DEBUG "ncp_mknod: mode = 0%ho\n", mode); return ncp_create_new(dir, dentry, mode, rdev, 0); } return -EPERM; /* Strange, but true */ diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index 5b5fa33b6b9d..3d1e34f8a68e 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -44,7 +44,7 @@ static void ncp_evict_inode(struct inode *); static void ncp_put_super(struct super_block *); static int ncp_statfs(struct dentry *, struct kstatfs *); -static int ncp_show_options(struct seq_file *, struct vfsmount *); +static int ncp_show_options(struct seq_file *, struct dentry *); static struct kmem_cache * ncp_inode_cachep; @@ -60,7 +60,6 @@ static struct inode *ncp_alloc_inode(struct super_block *sb) static void ncp_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(ncp_inode_cachep, NCP_FINFO(inode)); } @@ -323,9 +322,9 @@ static void ncp_stop_tasks(struct ncp_server *server) { flush_work_sync(&server->timeout_tq); } -static int ncp_show_options(struct seq_file *seq, struct vfsmount *mnt) +static int ncp_show_options(struct seq_file *seq, struct dentry *root) { - struct ncp_server *server = NCP_SBP(mnt->mnt_sb); + struct ncp_server *server = NCP_SBP(root->d_sb); unsigned int tmp; if (server->m.uid != 0) @@ -548,7 +547,7 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent) error = bdi_setup_and_register(&server->bdi, "ncpfs", BDI_CAP_MAP_COPY); if (error) - goto out_bdi; + goto out_fput; server->ncp_filp = ncp_filp; server->ncp_sock = sock; @@ -559,7 +558,7 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent) error = -EBADF; server->info_filp = fget(data.info_fd); if (!server->info_filp) - goto out_fput; + goto out_bdi; error = -ENOTSOCK; sock_inode = server->info_filp->f_path.dentry->d_inode; if (!S_ISSOCK(sock_inode->i_mode)) @@ -746,9 +745,9 @@ out_nls: out_fput2: if (server->info_filp) fput(server->info_filp); -out_fput: - bdi_destroy(&server->bdi); out_bdi: + bdi_destroy(&server->bdi); +out_fput: /* 23/12/1998 Marcin Dalecki <dalecki@cs.net.pl>: * * The previously used put_filp(ncp_filp); was bogus, since diff --git a/fs/ncpfs/ioctl.c b/fs/ncpfs/ioctl.c index 790e92a9ec63..6958adfaff08 100644 --- a/fs/ncpfs/ioctl.c +++ b/fs/ncpfs/ioctl.c @@ -901,7 +901,7 @@ long ncp_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) ret = __ncp_ioctl(inode, cmd, arg); outDropWrite: if (need_drop_write) - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); out: return ret; } diff --git a/fs/ncpfs/ncplib_kernel.h b/fs/ncpfs/ncplib_kernel.h index 09881e6aa5ad..32c06587351a 100644 --- a/fs/ncpfs/ncplib_kernel.h +++ b/fs/ncpfs/ncplib_kernel.h @@ -114,7 +114,7 @@ int ncp_dirhandle_alloc(struct ncp_server *, __u8 vol, __le32 dirent, __u8 *dirh int ncp_dirhandle_free(struct ncp_server *, __u8 dirhandle); int ncp_create_new(struct inode *dir, struct dentry *dentry, - int mode, dev_t rdev, __le32 attributes); + umode_t mode, dev_t rdev, __le32 attributes); static inline int ncp_is_nfs_extras(struct ncp_server* server, unsigned int volnum) { #ifdef CONFIG_NCPFS_NFS_NS diff --git a/fs/ncpfs/symlink.c b/fs/ncpfs/symlink.c index 661f861d80c6..52439ddc8de0 100644 --- a/fs/ncpfs/symlink.c +++ b/fs/ncpfs/symlink.c @@ -108,7 +108,7 @@ int ncp_symlink(struct inode *dir, struct dentry *dentry, const char *symname) { char *rawlink; int length, err, i, outlen; int kludge; - int mode; + umode_t mode; __le32 attr; unsigned int hdr; diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c index 281ae95932c9..48cfac31f64c 100644 --- a/fs/nfs/blocklayout/blocklayout.c +++ b/fs/nfs/blocklayout/blocklayout.c @@ -90,9 +90,9 @@ static int is_writable(struct pnfs_block_extent *be, sector_t isect) */ struct parallel_io { struct kref refcnt; - struct rpc_call_ops call_ops; - void (*pnfs_callback) (void *data); + void (*pnfs_callback) (void *data, int num_se); void *data; + int bse_count; }; static inline struct parallel_io *alloc_parallel(void *data) @@ -103,6 +103,7 @@ static inline struct parallel_io *alloc_parallel(void *data) if (rv) { rv->data = data; kref_init(&rv->refcnt); + rv->bse_count = 0; } return rv; } @@ -117,7 +118,7 @@ static void destroy_parallel(struct kref *kref) struct parallel_io *p = container_of(kref, struct parallel_io, refcnt); dprintk("%s enter\n", __func__); - p->pnfs_callback(p->data); + p->pnfs_callback(p->data, p->bse_count); kfree(p); } @@ -146,14 +147,19 @@ static struct bio *bl_alloc_init_bio(int npg, sector_t isect, { struct bio *bio; + npg = min(npg, BIO_MAX_PAGES); bio = bio_alloc(GFP_NOIO, npg); - if (!bio) - return NULL; + if (!bio && (current->flags & PF_MEMALLOC)) { + while (!bio && (npg /= 2)) + bio = bio_alloc(GFP_NOIO, npg); + } - bio->bi_sector = isect - be->be_f_offset + be->be_v_offset; - bio->bi_bdev = be->be_mdev; - bio->bi_end_io = end_io; - bio->bi_private = par; + if (bio) { + bio->bi_sector = isect - be->be_f_offset + be->be_v_offset; + bio->bi_bdev = be->be_mdev; + bio->bi_end_io = end_io; + bio->bi_private = par; + } return bio; } @@ -212,22 +218,15 @@ static void bl_read_cleanup(struct work_struct *work) } static void -bl_end_par_io_read(void *data) +bl_end_par_io_read(void *data, int unused) { struct nfs_read_data *rdata = data; + rdata->task.tk_status = rdata->pnfs_error; INIT_WORK(&rdata->task.u.tk_work, bl_read_cleanup); schedule_work(&rdata->task.u.tk_work); } -/* We don't want normal .rpc_call_done callback used, so we replace it - * with this stub. - */ -static void bl_rpc_do_nothing(struct rpc_task *task, void *calldata) -{ - return; -} - static enum pnfs_try_status bl_read_pagelist(struct nfs_read_data *rdata) { @@ -247,8 +246,6 @@ bl_read_pagelist(struct nfs_read_data *rdata) par = alloc_parallel(rdata); if (!par) goto use_mds; - par->call_ops = *rdata->mds_ops; - par->call_ops.rpc_call_done = bl_rpc_do_nothing; par->pnfs_callback = bl_end_par_io_read; /* At this point, we can no longer jump to use_mds */ @@ -322,6 +319,7 @@ static void mark_extents_written(struct pnfs_block_layout *bl, { sector_t isect, end; struct pnfs_block_extent *be; + struct pnfs_block_short_extent *se; dprintk("%s(%llu, %u)\n", __func__, offset, count); if (count == 0) @@ -334,8 +332,11 @@ static void mark_extents_written(struct pnfs_block_layout *bl, be = bl_find_get_extent(bl, isect, NULL); BUG_ON(!be); /* FIXME */ len = min(end, be->be_f_offset + be->be_length) - isect; - if (be->be_state == PNFS_BLOCK_INVALID_DATA) - bl_mark_for_commit(be, isect, len); /* What if fails? */ + if (be->be_state == PNFS_BLOCK_INVALID_DATA) { + se = bl_pop_one_short_extent(be->be_inval); + BUG_ON(!se); + bl_mark_for_commit(be, isect, len, se); + } isect += len; bl_put_extent(be); } @@ -357,7 +358,8 @@ static void bl_end_io_write_zero(struct bio *bio, int err) end_page_writeback(page); page_cache_release(page); } while (bvec >= bio->bi_io_vec); - if (!uptodate) { + + if (unlikely(!uptodate)) { if (!wdata->pnfs_error) wdata->pnfs_error = -EIO; pnfs_set_lo_fail(wdata->lseg); @@ -366,7 +368,6 @@ static void bl_end_io_write_zero(struct bio *bio, int err) put_parallel(par); } -/* This is basically copied from mpage_end_io_read */ static void bl_end_io_write(struct bio *bio, int err) { struct parallel_io *par = bio->bi_private; @@ -392,7 +393,7 @@ static void bl_write_cleanup(struct work_struct *work) dprintk("%s enter\n", __func__); task = container_of(work, struct rpc_task, u.tk_work); wdata = container_of(task, struct nfs_write_data, task); - if (!wdata->pnfs_error) { + if (likely(!wdata->pnfs_error)) { /* Marks for LAYOUTCOMMIT */ mark_extents_written(BLK_LSEG2EXT(wdata->lseg), wdata->args.offset, wdata->args.count); @@ -401,11 +402,16 @@ static void bl_write_cleanup(struct work_struct *work) } /* Called when last of bios associated with a bl_write_pagelist call finishes */ -static void bl_end_par_io_write(void *data) +static void bl_end_par_io_write(void *data, int num_se) { struct nfs_write_data *wdata = data; - wdata->task.tk_status = 0; + if (unlikely(wdata->pnfs_error)) { + bl_free_short_extents(&BLK_LSEG2EXT(wdata->lseg)->bl_inval, + num_se); + } + + wdata->task.tk_status = wdata->pnfs_error; wdata->verf.committed = NFS_FILE_SYNC; INIT_WORK(&wdata->task.u.tk_work, bl_write_cleanup); schedule_work(&wdata->task.u.tk_work); @@ -484,6 +490,55 @@ cleanup: return ret; } +/* Find or create a zeroing page marked being writeback. + * Return ERR_PTR on error, NULL to indicate skip this page and page itself + * to indicate write out. + */ +static struct page * +bl_find_get_zeroing_page(struct inode *inode, pgoff_t index, + struct pnfs_block_extent *cow_read) +{ + struct page *page; + int locked = 0; + page = find_get_page(inode->i_mapping, index); + if (page) + goto check_page; + + page = find_or_create_page(inode->i_mapping, index, GFP_NOFS); + if (unlikely(!page)) { + dprintk("%s oom\n", __func__); + return ERR_PTR(-ENOMEM); + } + locked = 1; + +check_page: + /* PageDirty: Other will write this out + * PageWriteback: Other is writing this out + * PageUptodate: It was read before + */ + if (PageDirty(page) || PageWriteback(page)) { + print_page(page); + if (locked) + unlock_page(page); + page_cache_release(page); + return NULL; + } + + if (!locked) { + lock_page(page); + locked = 1; + goto check_page; + } + if (!PageUptodate(page)) { + /* New page, readin or zero it */ + init_page_for_write(page, cow_read); + } + set_page_writeback(page); + unlock_page(page); + + return page; +} + static enum pnfs_try_status bl_write_pagelist(struct nfs_write_data *wdata, int sync) { @@ -508,9 +563,7 @@ bl_write_pagelist(struct nfs_write_data *wdata, int sync) */ par = alloc_parallel(wdata); if (!par) - return PNFS_NOT_ATTEMPTED; - par->call_ops = *wdata->mds_ops; - par->call_ops.rpc_call_done = bl_rpc_do_nothing; + goto out_mds; par->pnfs_callback = bl_end_par_io_write; /* At this point, have to be more careful with error handling */ @@ -518,12 +571,15 @@ bl_write_pagelist(struct nfs_write_data *wdata, int sync) be = bl_find_get_extent(BLK_LSEG2EXT(wdata->lseg), isect, &cow_read); if (!be || !is_writable(be, isect)) { dprintk("%s no matching extents!\n", __func__); - wdata->pnfs_error = -EINVAL; - goto out; + goto out_mds; } /* First page inside INVALID extent */ if (be->be_state == PNFS_BLOCK_INVALID_DATA) { + if (likely(!bl_push_one_short_extent(be->be_inval))) + par->bse_count++; + else + goto out_mds; temp = offset >> PAGE_CACHE_SHIFT; npg_zero = do_div(temp, npg_per_block); isect = (sector_t) (((offset - npg_zero * PAGE_CACHE_SIZE) & @@ -543,36 +599,16 @@ fill_invalid_ext: dprintk("%s zero %dth page: index %lu isect %llu\n", __func__, npg_zero, index, (unsigned long long)isect); - page = - find_or_create_page(wdata->inode->i_mapping, index, - GFP_NOFS); - if (!page) { - dprintk("%s oom\n", __func__); - wdata->pnfs_error = -ENOMEM; + page = bl_find_get_zeroing_page(wdata->inode, index, + cow_read); + if (unlikely(IS_ERR(page))) { + wdata->pnfs_error = PTR_ERR(page); goto out; - } - - /* PageDirty: Other will write this out - * PageWriteback: Other is writing this out - * PageUptodate: It was read before - * sector_initialized: already written out - */ - if (PageDirty(page) || PageWriteback(page)) { - print_page(page); - unlock_page(page); - page_cache_release(page); + } else if (page == NULL) goto next_page; - } - if (!PageUptodate(page)) { - /* New page, readin or zero it */ - init_page_for_write(page, cow_read); - } - set_page_writeback(page); - unlock_page(page); ret = bl_mark_sectors_init(be->be_inval, isect, - PAGE_CACHE_SECTORS, - NULL); + PAGE_CACHE_SECTORS); if (unlikely(ret)) { dprintk("%s bl_mark_sectors_init fail %d\n", __func__, ret); @@ -581,6 +617,19 @@ fill_invalid_ext: wdata->pnfs_error = ret; goto out; } + if (likely(!bl_push_one_short_extent(be->be_inval))) + par->bse_count++; + else { + end_page_writeback(page); + page_cache_release(page); + wdata->pnfs_error = -ENOMEM; + goto out; + } + /* FIXME: This should be done in bi_end_io */ + mark_extents_written(BLK_LSEG2EXT(wdata->lseg), + page->index << PAGE_CACHE_SHIFT, + PAGE_CACHE_SIZE); + bio = bl_add_page_to_bio(bio, npg_zero, WRITE, isect, page, be, bl_end_io_write_zero, par); @@ -589,10 +638,6 @@ fill_invalid_ext: bio = NULL; goto out; } - /* FIXME: This should be done in bi_end_io */ - mark_extents_written(BLK_LSEG2EXT(wdata->lseg), - page->index << PAGE_CACHE_SHIFT, - PAGE_CACHE_SIZE); next_page: isect += PAGE_CACHE_SECTORS; extent_length -= PAGE_CACHE_SECTORS; @@ -616,13 +661,21 @@ next_page: wdata->pnfs_error = -EINVAL; goto out; } + if (be->be_state == PNFS_BLOCK_INVALID_DATA) { + if (likely(!bl_push_one_short_extent( + be->be_inval))) + par->bse_count++; + else { + wdata->pnfs_error = -ENOMEM; + goto out; + } + } extent_length = be->be_length - (isect - be->be_f_offset); } if (be->be_state == PNFS_BLOCK_INVALID_DATA) { ret = bl_mark_sectors_init(be->be_inval, isect, - PAGE_CACHE_SECTORS, - NULL); + PAGE_CACHE_SECTORS); if (unlikely(ret)) { dprintk("%s bl_mark_sectors_init fail %d\n", __func__, ret); @@ -664,6 +717,10 @@ out: bl_submit_bio(WRITE, bio); put_parallel(par); return PNFS_ATTEMPTED; +out_mds: + bl_put_extent(be); + kfree(par); + return PNFS_NOT_ATTEMPTED; } /* FIXME - range ignored */ @@ -690,11 +747,17 @@ static void release_inval_marks(struct pnfs_inval_markings *marks) { struct pnfs_inval_tracking *pos, *temp; + struct pnfs_block_short_extent *se, *stemp; list_for_each_entry_safe(pos, temp, &marks->im_tree.mtt_stub, it_link) { list_del(&pos->it_link); kfree(pos); } + + list_for_each_entry_safe(se, stemp, &marks->im_extents, bse_node) { + list_del(&se->bse_node); + kfree(se); + } return; } @@ -779,16 +842,13 @@ bl_cleanup_layoutcommit(struct nfs4_layoutcommit_data *lcdata) static void free_blk_mountid(struct block_mount_id *mid) { if (mid) { - struct pnfs_block_dev *dev; - spin_lock(&mid->bm_lock); - while (!list_empty(&mid->bm_devlist)) { - dev = list_first_entry(&mid->bm_devlist, - struct pnfs_block_dev, - bm_node); + struct pnfs_block_dev *dev, *tmp; + + /* No need to take bm_lock as we are last user freeing bm_devlist */ + list_for_each_entry_safe(dev, tmp, &mid->bm_devlist, bm_node) { list_del(&dev->bm_node); bl_free_block_dev(dev); } - spin_unlock(&mid->bm_lock); kfree(mid); } } diff --git a/fs/nfs/blocklayout/blocklayout.h b/fs/nfs/blocklayout/blocklayout.h index 42acf7ef5992..e31a2df28e70 100644 --- a/fs/nfs/blocklayout/blocklayout.h +++ b/fs/nfs/blocklayout/blocklayout.h @@ -70,6 +70,7 @@ struct pnfs_inval_markings { spinlock_t im_lock; struct my_tree im_tree; /* Sectors that need LAYOUTCOMMIT */ sector_t im_block_size; /* Server blocksize in sectors */ + struct list_head im_extents; /* Short extents for INVAL->RW conversion */ }; struct pnfs_inval_tracking { @@ -105,6 +106,7 @@ BL_INIT_INVAL_MARKS(struct pnfs_inval_markings *marks, sector_t blocksize) { spin_lock_init(&marks->im_lock); INIT_LIST_HEAD(&marks->im_tree.mtt_stub); + INIT_LIST_HEAD(&marks->im_extents); marks->im_block_size = blocksize; marks->im_tree.mtt_step_size = min((sector_t)PAGE_CACHE_SECTORS, blocksize); @@ -186,8 +188,7 @@ struct pnfs_block_extent * bl_find_get_extent(struct pnfs_block_layout *bl, sector_t isect, struct pnfs_block_extent **cow_read); int bl_mark_sectors_init(struct pnfs_inval_markings *marks, - sector_t offset, sector_t length, - sector_t **pages); + sector_t offset, sector_t length); void bl_put_extent(struct pnfs_block_extent *be); struct pnfs_block_extent *bl_alloc_extent(void); int bl_is_sector_init(struct pnfs_inval_markings *marks, sector_t isect); @@ -200,6 +201,11 @@ void clean_pnfs_block_layoutupdate(struct pnfs_block_layout *bl, int bl_add_merge_extent(struct pnfs_block_layout *bl, struct pnfs_block_extent *new); int bl_mark_for_commit(struct pnfs_block_extent *be, - sector_t offset, sector_t length); + sector_t offset, sector_t length, + struct pnfs_block_short_extent *new); +int bl_push_one_short_extent(struct pnfs_inval_markings *marks); +struct pnfs_block_short_extent * +bl_pop_one_short_extent(struct pnfs_inval_markings *marks); +void bl_free_short_extents(struct pnfs_inval_markings *marks, int num_to_free); #endif /* FS_NFS_NFS4BLOCKLAYOUT_H */ diff --git a/fs/nfs/blocklayout/extents.c b/fs/nfs/blocklayout/extents.c index 19fa7b0b8c00..1abac09f7cd5 100644 --- a/fs/nfs/blocklayout/extents.c +++ b/fs/nfs/blocklayout/extents.c @@ -110,13 +110,7 @@ static int _add_entry(struct my_tree *tree, u64 s, int32_t tag, return 0; } else { struct pnfs_inval_tracking *new; - if (storage) - new = storage; - else { - new = kmalloc(sizeof(*new), GFP_NOFS); - if (!new) - return -ENOMEM; - } + new = storage; new->it_sector = s; new->it_tags = (1 << tag); list_add(&new->it_link, &pos->it_link); @@ -139,11 +133,13 @@ static int _set_range(struct my_tree *tree, int32_t tag, u64 s, u64 length) } /* Ensure that future operations on given range of tree will not malloc */ -static int _preload_range(struct my_tree *tree, u64 offset, u64 length) +static int _preload_range(struct pnfs_inval_markings *marks, + u64 offset, u64 length) { u64 start, end, s; int count, i, used = 0, status = -ENOMEM; struct pnfs_inval_tracking **storage; + struct my_tree *tree = &marks->im_tree; dprintk("%s(%llu, %llu) enter\n", __func__, offset, length); start = normalize(offset, tree->mtt_step_size); @@ -161,12 +157,11 @@ static int _preload_range(struct my_tree *tree, u64 offset, u64 length) goto out_cleanup; } - /* Now need lock - HOW??? */ - + spin_lock_bh(&marks->im_lock); for (s = start; s < end; s += tree->mtt_step_size) used += _add_entry(tree, s, INTERNAL_EXISTS, storage[used]); + spin_unlock_bh(&marks->im_lock); - /* Unlock - HOW??? */ status = 0; out_cleanup: @@ -179,41 +174,14 @@ static int _preload_range(struct my_tree *tree, u64 offset, u64 length) return status; } -static void set_needs_init(sector_t *array, sector_t offset) -{ - sector_t *p = array; - - dprintk("%s enter\n", __func__); - if (!p) - return; - while (*p < offset) - p++; - if (*p == offset) - return; - else if (*p == ~0) { - *p++ = offset; - *p = ~0; - return; - } else { - sector_t *save = p; - dprintk("%s Adding %llu\n", __func__, (u64)offset); - while (*p != ~0) - p++; - p++; - memmove(save + 1, save, (char *)p - (char *)save); - *save = offset; - return; - } -} - /* We are relying on page lock to serialize this */ int bl_is_sector_init(struct pnfs_inval_markings *marks, sector_t isect) { int rv; - spin_lock(&marks->im_lock); + spin_lock_bh(&marks->im_lock); rv = _has_tag(&marks->im_tree, isect, EXTENT_INITIALIZED); - spin_unlock(&marks->im_lock); + spin_unlock_bh(&marks->im_lock); return rv; } @@ -253,78 +221,39 @@ static int is_range_written(struct pnfs_inval_markings *marks, { int rv; - spin_lock(&marks->im_lock); + spin_lock_bh(&marks->im_lock); rv = _range_has_tag(&marks->im_tree, start, end, EXTENT_WRITTEN); - spin_unlock(&marks->im_lock); + spin_unlock_bh(&marks->im_lock); return rv; } /* Marks sectors in [offest, offset_length) as having been initialized. * All lengths are step-aligned, where step is min(pagesize, blocksize). - * Notes where partial block is initialized, and helps prepare it for - * complete initialization later. + * Currently assumes offset is page-aligned */ -/* Currently assumes offset is page-aligned */ int bl_mark_sectors_init(struct pnfs_inval_markings *marks, - sector_t offset, sector_t length, - sector_t **pages) + sector_t offset, sector_t length) { - sector_t s, start, end; - sector_t *array = NULL; /* Pages to mark */ + sector_t start, end; dprintk("%s(offset=%llu,len=%llu) enter\n", __func__, (u64)offset, (u64)length); - s = max((sector_t) 3, - 2 * (marks->im_block_size / (PAGE_CACHE_SECTORS))); - dprintk("%s set max=%llu\n", __func__, (u64)s); - if (pages) { - array = kmalloc(s * sizeof(sector_t), GFP_NOFS); - if (!array) - goto outerr; - array[0] = ~0; - } start = normalize(offset, marks->im_block_size); end = normalize_up(offset + length, marks->im_block_size); - if (_preload_range(&marks->im_tree, start, end - start)) + if (_preload_range(marks, start, end - start)) goto outerr; - spin_lock(&marks->im_lock); - - for (s = normalize_up(start, PAGE_CACHE_SECTORS); - s < offset; s += PAGE_CACHE_SECTORS) { - dprintk("%s pre-area pages\n", __func__); - /* Portion of used block is not initialized */ - if (!_has_tag(&marks->im_tree, s, EXTENT_INITIALIZED)) - set_needs_init(array, s); - } + spin_lock_bh(&marks->im_lock); if (_set_range(&marks->im_tree, EXTENT_INITIALIZED, offset, length)) goto out_unlock; - for (s = normalize_up(offset + length, PAGE_CACHE_SECTORS); - s < end; s += PAGE_CACHE_SECTORS) { - dprintk("%s post-area pages\n", __func__); - if (!_has_tag(&marks->im_tree, s, EXTENT_INITIALIZED)) - set_needs_init(array, s); - } - - spin_unlock(&marks->im_lock); + spin_unlock_bh(&marks->im_lock); - if (pages) { - if (array[0] == ~0) { - kfree(array); - *pages = NULL; - } else - *pages = array; - } return 0; - out_unlock: - spin_unlock(&marks->im_lock); - outerr: - if (pages) { - kfree(array); - *pages = NULL; - } +out_unlock: + spin_unlock_bh(&marks->im_lock); +outerr: return -ENOMEM; } @@ -338,9 +267,9 @@ static int mark_written_sectors(struct pnfs_inval_markings *marks, dprintk("%s(offset=%llu,len=%llu) enter\n", __func__, (u64)offset, (u64)length); - spin_lock(&marks->im_lock); + spin_lock_bh(&marks->im_lock); status = _set_range(&marks->im_tree, EXTENT_WRITTEN, offset, length); - spin_unlock(&marks->im_lock); + spin_unlock_bh(&marks->im_lock); return status; } @@ -440,20 +369,18 @@ static void add_to_commitlist(struct pnfs_block_layout *bl, /* Note the range described by offset, length is guaranteed to be contained * within be. + * new will be freed, either by this function or add_to_commitlist if they + * decide not to use it, or after LAYOUTCOMMIT uses it in the commitlist. */ int bl_mark_for_commit(struct pnfs_block_extent *be, - sector_t offset, sector_t length) + sector_t offset, sector_t length, + struct pnfs_block_short_extent *new) { sector_t new_end, end = offset + length; - struct pnfs_block_short_extent *new; struct pnfs_block_layout *bl = container_of(be->be_inval, struct pnfs_block_layout, bl_inval); - new = kmalloc(sizeof(*new), GFP_NOFS); - if (!new) - return -ENOMEM; - mark_written_sectors(be->be_inval, offset, length); /* We want to add the range to commit list, but it must be * block-normalized, and verified that the normalized range has @@ -483,9 +410,6 @@ int bl_mark_for_commit(struct pnfs_block_extent *be, new->bse_mdev = be->be_mdev; spin_lock(&bl->bl_ext_lock); - /* new will be freed, either by add_to_commitlist if it decides not - * to use it, or after LAYOUTCOMMIT uses it in the commitlist. - */ add_to_commitlist(bl, new); spin_unlock(&bl->bl_ext_lock); return 0; @@ -933,3 +857,53 @@ clean_pnfs_block_layoutupdate(struct pnfs_block_layout *bl, } } } + +int bl_push_one_short_extent(struct pnfs_inval_markings *marks) +{ + struct pnfs_block_short_extent *new; + + new = kmalloc(sizeof(*new), GFP_NOFS); + if (unlikely(!new)) + return -ENOMEM; + + spin_lock_bh(&marks->im_lock); + list_add(&new->bse_node, &marks->im_extents); + spin_unlock_bh(&marks->im_lock); + + return 0; +} + +struct pnfs_block_short_extent * +bl_pop_one_short_extent(struct pnfs_inval_markings *marks) +{ + struct pnfs_block_short_extent *rv = NULL; + + spin_lock_bh(&marks->im_lock); + if (!list_empty(&marks->im_extents)) { + rv = list_entry((&marks->im_extents)->next, + struct pnfs_block_short_extent, bse_node); + list_del_init(&rv->bse_node); + } + spin_unlock_bh(&marks->im_lock); + + return rv; +} + +void bl_free_short_extents(struct pnfs_inval_markings *marks, int num_to_free) +{ + struct pnfs_block_short_extent *se = NULL, *tmp; + + if (num_to_free <= 0) + return; + + spin_lock(&marks->im_lock); + list_for_each_entry_safe(se, tmp, &marks->im_extents, bse_node) { + list_del(&se->bse_node); + kfree(se); + if (--num_to_free == 0) + break; + } + spin_unlock(&marks->im_lock); + + BUG_ON(num_to_free > 0); +} diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h index 07df5f1d85e5..c89d3b9e483c 100644 --- a/fs/nfs/callback.h +++ b/fs/nfs/callback.h @@ -162,7 +162,7 @@ struct cb_layoutrecallargs { }; }; -extern unsigned nfs4_callback_layoutrecall( +extern __be32 nfs4_callback_layoutrecall( struct cb_layoutrecallargs *args, void *dummy, struct cb_process_state *cps); diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index 43926add945b..54cea8ad5a76 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -339,7 +339,7 @@ validate_seqid(struct nfs4_slot_table *tbl, struct cb_sequenceargs * args) dprintk("%s enter. slotid %d seqid %d\n", __func__, args->csa_slotid, args->csa_sequenceid); - if (args->csa_slotid > NFS41_BC_MAX_CALLBACKS) + if (args->csa_slotid >= NFS41_BC_MAX_CALLBACKS) return htonl(NFS4ERR_BADSLOT); slot = tbl->slots + args->csa_slotid; diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c index 726e59a9e50f..d50b2742f23b 100644 --- a/fs/nfs/callback_xdr.c +++ b/fs/nfs/callback_xdr.c @@ -305,6 +305,10 @@ __be32 decode_devicenotify_args(struct svc_rqst *rqstp, n = ntohl(*p++); if (n <= 0) goto out; + if (n > ULONG_MAX / sizeof(*args->devs)) { + status = htonl(NFS4ERR_BADXDR); + goto out; + } args->devs = kmalloc(n * sizeof(*args->devs), GFP_KERNEL); if (!args->devs) { diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 873bf00d51a2..31778f74357d 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -84,7 +84,7 @@ retry: /* * Turn off NFSv4 uid/gid mapping when using AUTH_SYS */ -static int nfs4_disable_idmapping = 0; +static bool nfs4_disable_idmapping = true; /* * RPC cruft for NFS @@ -185,7 +185,7 @@ static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_ clp->cl_minorversion = cl_init->minorversion; clp->cl_mvops = nfs_v4_minor_ops[cl_init->minorversion]; #endif - cred = rpc_lookup_machine_cred(); + cred = rpc_lookup_machine_cred("*"); if (!IS_ERR(cred)) clp->cl_machine_cred = cred; nfs_fscache_get_client_cookie(clp); @@ -250,6 +250,11 @@ static void pnfs_init_server(struct nfs_server *server) rpc_init_wait_queue(&server->roc_rpcwaitq, "pNFS ROC"); } +static void nfs4_destroy_server(struct nfs_server *server) +{ + nfs4_purge_state_owners(server); +} + #else static void nfs4_shutdown_client(struct nfs_client *clp) { @@ -1065,6 +1070,7 @@ static struct nfs_server *nfs_alloc_server(void) INIT_LIST_HEAD(&server->master_link); INIT_LIST_HEAD(&server->delegations); INIT_LIST_HEAD(&server->layouts); + INIT_LIST_HEAD(&server->state_owners_lru); atomic_set(&server->active, 0); @@ -1538,6 +1544,7 @@ static int nfs4_server_common_setup(struct nfs_server *server, nfs_server_insert_lists(server); server->mount_time = jiffies; + server->destroy = nfs4_destroy_server; out: nfs_free_fattr(fattr); return error; @@ -1719,6 +1726,7 @@ struct nfs_server *nfs_clone_server(struct nfs_server *source, /* Copy data from the source */ server->nfs_client = source->nfs_client; + server->destroy = source->destroy; atomic_inc(&server->nfs_client->cl_count); nfs_server_copy_userdata(server, source); diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index b238d95ac48c..fd9a872fada0 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -47,13 +47,13 @@ static int nfs_opendir(struct inode *, struct file *); static int nfs_closedir(struct inode *, struct file *); static int nfs_readdir(struct file *, void *, filldir_t); static struct dentry *nfs_lookup(struct inode *, struct dentry *, struct nameidata *); -static int nfs_create(struct inode *, struct dentry *, int, struct nameidata *); -static int nfs_mkdir(struct inode *, struct dentry *, int); +static int nfs_create(struct inode *, struct dentry *, umode_t, struct nameidata *); +static int nfs_mkdir(struct inode *, struct dentry *, umode_t); static int nfs_rmdir(struct inode *, struct dentry *); static int nfs_unlink(struct inode *, struct dentry *); static int nfs_symlink(struct inode *, struct dentry *, const char *); static int nfs_link(struct dentry *, struct inode *, struct dentry *); -static int nfs_mknod(struct inode *, struct dentry *, int, dev_t); +static int nfs_mknod(struct inode *, struct dentry *, umode_t, dev_t); static int nfs_rename(struct inode *, struct dentry *, struct inode *, struct dentry *); static int nfs_fsync_dir(struct file *, loff_t, loff_t, int); @@ -112,7 +112,7 @@ const struct inode_operations nfs3_dir_inode_operations = { #ifdef CONFIG_NFS_V4 static struct dentry *nfs_atomic_lookup(struct inode *, struct dentry *, struct nameidata *); -static int nfs_open_create(struct inode *dir, struct dentry *dentry, int mode, struct nameidata *nd); +static int nfs_open_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd); const struct inode_operations nfs4_dir_inode_operations = { .create = nfs_open_create, .lookup = nfs_atomic_lookup, @@ -1368,18 +1368,7 @@ static fmode_t flags_to_mode(int flags) static struct nfs_open_context *create_nfs_open_context(struct dentry *dentry, int open_flags) { - struct nfs_open_context *ctx; - struct rpc_cred *cred; - fmode_t fmode = flags_to_mode(open_flags); - - cred = rpc_lookup_cred(); - if (IS_ERR(cred)) - return ERR_CAST(cred); - ctx = alloc_nfs_open_context(dentry, cred, fmode); - put_rpccred(cred); - if (ctx == NULL) - return ERR_PTR(-ENOMEM); - return ctx; + return alloc_nfs_open_context(dentry, flags_to_mode(open_flags)); } static int do_open(struct inode *inode, struct file *filp) @@ -1468,12 +1457,12 @@ static struct dentry *nfs_atomic_lookup(struct inode *dir, struct dentry *dentry res = NULL; goto out; /* This turned out not to be a regular file */ + case -EISDIR: case -ENOTDIR: goto no_open; case -ELOOP: if (!(nd->intent.open.flags & O_NOFOLLOW)) goto no_open; - /* case -EISDIR: */ /* case -EINVAL: */ default: res = ERR_CAST(inode); @@ -1584,8 +1573,8 @@ no_open: return nfs_lookup_revalidate(dentry, nd); } -static int nfs_open_create(struct inode *dir, struct dentry *dentry, int mode, - struct nameidata *nd) +static int nfs_open_create(struct inode *dir, struct dentry *dentry, + umode_t mode, struct nameidata *nd) { struct nfs_open_context *ctx = NULL; struct iattr attr; @@ -1675,8 +1664,8 @@ out_error: * that the operation succeeded on the server, but an error in the * reply path made it appear to have failed. */ -static int nfs_create(struct inode *dir, struct dentry *dentry, int mode, - struct nameidata *nd) +static int nfs_create(struct inode *dir, struct dentry *dentry, + umode_t mode, struct nameidata *nd) { struct iattr attr; int error; @@ -1704,7 +1693,7 @@ out_err: * See comments for nfs_proc_create regarding failed operations. */ static int -nfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev) +nfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) { struct iattr attr; int status; @@ -1730,7 +1719,7 @@ out_err: /* * See comments for nfs_proc_create regarding failed operations. */ -static int nfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int nfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct iattr attr; int error; diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 0a1f8312b4dc..c43a452f7da2 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -40,48 +40,8 @@ #define NFSDBG_FACILITY NFSDBG_FILE -static int nfs_file_open(struct inode *, struct file *); -static int nfs_file_release(struct inode *, struct file *); -static loff_t nfs_file_llseek(struct file *file, loff_t offset, int origin); -static int nfs_file_mmap(struct file *, struct vm_area_struct *); -static ssize_t nfs_file_splice_read(struct file *filp, loff_t *ppos, - struct pipe_inode_info *pipe, - size_t count, unsigned int flags); -static ssize_t nfs_file_read(struct kiocb *, const struct iovec *iov, - unsigned long nr_segs, loff_t pos); -static ssize_t nfs_file_splice_write(struct pipe_inode_info *pipe, - struct file *filp, loff_t *ppos, - size_t count, unsigned int flags); -static ssize_t nfs_file_write(struct kiocb *, const struct iovec *iov, - unsigned long nr_segs, loff_t pos); -static int nfs_file_flush(struct file *, fl_owner_t id); -static int nfs_file_fsync(struct file *, loff_t, loff_t, int datasync); -static int nfs_check_flags(int flags); -static int nfs_lock(struct file *filp, int cmd, struct file_lock *fl); -static int nfs_flock(struct file *filp, int cmd, struct file_lock *fl); -static int nfs_setlease(struct file *file, long arg, struct file_lock **fl); - static const struct vm_operations_struct nfs_file_vm_ops; -const struct file_operations nfs_file_operations = { - .llseek = nfs_file_llseek, - .read = do_sync_read, - .write = do_sync_write, - .aio_read = nfs_file_read, - .aio_write = nfs_file_write, - .mmap = nfs_file_mmap, - .open = nfs_file_open, - .flush = nfs_file_flush, - .release = nfs_file_release, - .fsync = nfs_file_fsync, - .lock = nfs_lock, - .flock = nfs_flock, - .splice_read = nfs_file_splice_read, - .splice_write = nfs_file_splice_write, - .check_flags = nfs_check_flags, - .setlease = nfs_setlease, -}; - const struct inode_operations nfs_file_inode_operations = { .permission = nfs_permission, .getattr = nfs_getattr, @@ -187,7 +147,7 @@ static loff_t nfs_file_llseek(struct file *filp, loff_t offset, int origin) * origin == SEEK_END || SEEK_DATA || SEEK_HOLE => we must revalidate * the cached file length */ - if (origin != SEEK_SET || origin != SEEK_CUR) { + if (origin != SEEK_SET && origin != SEEK_CUR) { struct inode *inode = filp->f_mapping->host; int retval = nfs_revalidate_file_size(inode, filp); @@ -312,13 +272,13 @@ nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync) datasync); ret = filemap_write_and_wait_range(inode->i_mapping, start, end); - if (ret) - return ret; mutex_lock(&inode->i_mutex); nfs_inc_stats(inode, NFSIOS_VFSFSYNC); have_error = test_and_clear_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags); status = nfs_commit_inode(inode, FLUSH_SYNC); + if (status >= 0 && ret < 0) + status = ret; have_error |= test_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags); if (have_error) ret = xchg(&ctx->error, 0); @@ -886,3 +846,54 @@ static int nfs_setlease(struct file *file, long arg, struct file_lock **fl) file->f_path.dentry->d_name.name, arg); return -EINVAL; } + +const struct file_operations nfs_file_operations = { + .llseek = nfs_file_llseek, + .read = do_sync_read, + .write = do_sync_write, + .aio_read = nfs_file_read, + .aio_write = nfs_file_write, + .mmap = nfs_file_mmap, + .open = nfs_file_open, + .flush = nfs_file_flush, + .release = nfs_file_release, + .fsync = nfs_file_fsync, + .lock = nfs_lock, + .flock = nfs_flock, + .splice_read = nfs_file_splice_read, + .splice_write = nfs_file_splice_write, + .check_flags = nfs_check_flags, + .setlease = nfs_setlease, +}; + +#ifdef CONFIG_NFS_V4 +static int +nfs4_file_open(struct inode *inode, struct file *filp) +{ + /* + * NFSv4 opens are handled in d_lookup and d_revalidate. If we get to + * this point, then something is very wrong + */ + dprintk("NFS: %s called! inode=%p filp=%p\n", __func__, inode, filp); + return -ENOTDIR; +} + +const struct file_operations nfs4_file_operations = { + .llseek = nfs_file_llseek, + .read = do_sync_read, + .write = do_sync_write, + .aio_read = nfs_file_read, + .aio_write = nfs_file_write, + .mmap = nfs_file_mmap, + .open = nfs4_file_open, + .flush = nfs_file_flush, + .release = nfs_file_release, + .fsync = nfs_file_fsync, + .lock = nfs_lock, + .flock = nfs_flock, + .splice_read = nfs_file_splice_read, + .splice_write = nfs_file_splice_write, + .check_flags = nfs_check_flags, + .setlease = nfs_setlease, +}; +#endif /* CONFIG_NFS_V4 */ diff --git a/fs/nfs/idmap.c b/fs/nfs/idmap.c index 47d1c6ff2d8e..2c05f1991e1e 100644 --- a/fs/nfs/idmap.c +++ b/fs/nfs/idmap.c @@ -38,6 +38,89 @@ #include <linux/kernel.h> #include <linux/slab.h> #include <linux/nfs_idmap.h> +#include <linux/nfs_fs.h> + +/** + * nfs_fattr_init_names - initialise the nfs_fattr owner_name/group_name fields + * @fattr: fully initialised struct nfs_fattr + * @owner_name: owner name string cache + * @group_name: group name string cache + */ +void nfs_fattr_init_names(struct nfs_fattr *fattr, + struct nfs4_string *owner_name, + struct nfs4_string *group_name) +{ + fattr->owner_name = owner_name; + fattr->group_name = group_name; +} + +static void nfs_fattr_free_owner_name(struct nfs_fattr *fattr) +{ + fattr->valid &= ~NFS_ATTR_FATTR_OWNER_NAME; + kfree(fattr->owner_name->data); +} + +static void nfs_fattr_free_group_name(struct nfs_fattr *fattr) +{ + fattr->valid &= ~NFS_ATTR_FATTR_GROUP_NAME; + kfree(fattr->group_name->data); +} + +static bool nfs_fattr_map_owner_name(struct nfs_server *server, struct nfs_fattr *fattr) +{ + struct nfs4_string *owner = fattr->owner_name; + __u32 uid; + + if (!(fattr->valid & NFS_ATTR_FATTR_OWNER_NAME)) + return false; + if (nfs_map_name_to_uid(server, owner->data, owner->len, &uid) == 0) { + fattr->uid = uid; + fattr->valid |= NFS_ATTR_FATTR_OWNER; + } + return true; +} + +static bool nfs_fattr_map_group_name(struct nfs_server *server, struct nfs_fattr *fattr) +{ + struct nfs4_string *group = fattr->group_name; + __u32 gid; + + if (!(fattr->valid & NFS_ATTR_FATTR_GROUP_NAME)) + return false; + if (nfs_map_group_to_gid(server, group->data, group->len, &gid) == 0) { + fattr->gid = gid; + fattr->valid |= NFS_ATTR_FATTR_GROUP; + } + return true; +} + +/** + * nfs_fattr_free_names - free up the NFSv4 owner and group strings + * @fattr: a fully initialised nfs_fattr structure + */ +void nfs_fattr_free_names(struct nfs_fattr *fattr) +{ + if (fattr->valid & NFS_ATTR_FATTR_OWNER_NAME) + nfs_fattr_free_owner_name(fattr); + if (fattr->valid & NFS_ATTR_FATTR_GROUP_NAME) + nfs_fattr_free_group_name(fattr); +} + +/** + * nfs_fattr_map_and_free_names - map owner/group strings into uid/gid and free + * @server: pointer to the filesystem nfs_server structure + * @fattr: a fully initialised nfs_fattr structure + * + * This helper maps the cached NFSv4 owner/group strings in fattr into + * their numeric uid/gid equivalents, and then frees the cached strings. + */ +void nfs_fattr_map_and_free_names(struct nfs_server *server, struct nfs_fattr *fattr) +{ + if (nfs_fattr_map_owner_name(server, fattr)) + nfs_fattr_free_owner_name(fattr); + if (nfs_fattr_map_group_name(server, fattr)) + nfs_fattr_free_group_name(fattr); +} static int nfs_map_string_to_numeric(const char *name, size_t namelen, __u32 *res) { diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index c07a55aec838..f649fba8c384 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -38,6 +38,7 @@ #include <linux/nfs_xdr.h> #include <linux/slab.h> #include <linux/compat.h> +#include <linux/freezer.h> #include <asm/system.h> #include <asm/uaccess.h> @@ -56,7 +57,7 @@ #define NFS_64_BIT_INODE_NUMBERS_ENABLED 1 /* Default is to see 64-bit inode numbers */ -static int enable_ino64 = NFS_64_BIT_INODE_NUMBERS_ENABLED; +static bool enable_ino64 = NFS_64_BIT_INODE_NUMBERS_ENABLED; static void nfs_invalidate_inode(struct inode *); static int nfs_update_inode(struct inode *, struct nfs_fattr *); @@ -77,7 +78,7 @@ int nfs_wait_bit_killable(void *word) { if (fatal_signal_pending(current)) return -ERESTARTSYS; - schedule(); + freezable_schedule(); return 0; } @@ -291,7 +292,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) */ inode->i_op = NFS_SB(sb)->nfs_client->rpc_ops->file_inode_ops; if (S_ISREG(inode->i_mode)) { - inode->i_fop = &nfs_file_operations; + inode->i_fop = NFS_SB(sb)->nfs_client->rpc_ops->file_ops; inode->i_data.a_ops = &nfs_file_aops; inode->i_data.backing_dev_info = &NFS_SB(sb)->backing_dev_info; } else if (S_ISDIR(inode->i_mode)) { @@ -629,23 +630,28 @@ void nfs_close_context(struct nfs_open_context *ctx, int is_sync) nfs_revalidate_inode(server, inode); } -struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry, struct rpc_cred *cred, fmode_t f_mode) +struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry, fmode_t f_mode) { struct nfs_open_context *ctx; + struct rpc_cred *cred = rpc_lookup_cred(); + if (IS_ERR(cred)) + return ERR_CAST(cred); ctx = kmalloc(sizeof(*ctx), GFP_KERNEL); - if (ctx != NULL) { - nfs_sb_active(dentry->d_sb); - ctx->dentry = dget(dentry); - ctx->cred = get_rpccred(cred); - ctx->state = NULL; - ctx->mode = f_mode; - ctx->flags = 0; - ctx->error = 0; - nfs_init_lock_context(&ctx->lock_context); - ctx->lock_context.open_context = ctx; - INIT_LIST_HEAD(&ctx->list); + if (!ctx) { + put_rpccred(cred); + return ERR_PTR(-ENOMEM); } + nfs_sb_active(dentry->d_sb); + ctx->dentry = dget(dentry); + ctx->cred = cred; + ctx->state = NULL; + ctx->mode = f_mode; + ctx->flags = 0; + ctx->error = 0; + nfs_init_lock_context(&ctx->lock_context); + ctx->lock_context.open_context = ctx; + INIT_LIST_HEAD(&ctx->list); return ctx; } @@ -738,15 +744,10 @@ static void nfs_file_clear_open_context(struct file *filp) int nfs_open(struct inode *inode, struct file *filp) { struct nfs_open_context *ctx; - struct rpc_cred *cred; - cred = rpc_lookup_cred(); - if (IS_ERR(cred)) - return PTR_ERR(cred); - ctx = alloc_nfs_open_context(filp->f_path.dentry, cred, filp->f_mode); - put_rpccred(cred); - if (ctx == NULL) - return -ENOMEM; + ctx = alloc_nfs_open_context(filp->f_path.dentry, filp->f_mode); + if (IS_ERR(ctx)) + return PTR_ERR(ctx); nfs_file_set_open_context(filp, ctx); put_nfs_open_context(ctx); nfs_fscache_set_inode_cookie(inode, filp); @@ -1019,6 +1020,8 @@ void nfs_fattr_init(struct nfs_fattr *fattr) fattr->valid = 0; fattr->time_start = jiffies; fattr->gencount = nfs_inc_attr_generation_counter(); + fattr->owner_name = NULL; + fattr->group_name = NULL; } struct nfs_fattr *nfs_alloc_fattr(void) @@ -1464,7 +1467,6 @@ struct inode *nfs_alloc_inode(struct super_block *sb) static void nfs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(nfs_inode_cachep, NFS_I(inode)); } diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index c1a1bd8ddf1c..8102db9b926c 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -299,12 +299,16 @@ extern void nfs_read_prepare(struct rpc_task *task, void *calldata); extern int nfs_generic_pagein(struct nfs_pageio_descriptor *desc, struct list_head *head); +extern void nfs_pageio_init_read_mds(struct nfs_pageio_descriptor *pgio, + struct inode *inode); extern void nfs_pageio_reset_read_mds(struct nfs_pageio_descriptor *pgio); extern void nfs_readdata_release(struct nfs_read_data *rdata); /* write.c */ extern int nfs_generic_flush(struct nfs_pageio_descriptor *desc, struct list_head *head); +extern void nfs_pageio_init_write_mds(struct nfs_pageio_descriptor *pgio, + struct inode *inode, int ioflags); extern void nfs_pageio_reset_write_mds(struct nfs_pageio_descriptor *pgio); extern void nfs_writedata_release(struct nfs_write_data *wdata); extern void nfs_commit_free(struct nfs_write_data *p); @@ -328,7 +332,7 @@ void nfs_commit_release_pages(struct nfs_write_data *data); #ifdef CONFIG_MIGRATION extern int nfs_migrate_page(struct address_space *, - struct page *, struct page *); + struct page *, struct page *, enum migrate_mode); #else #define nfs_migrate_page NULL #endif diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 85f1690ca08c..91943953a370 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -17,6 +17,7 @@ #include <linux/nfs_page.h> #include <linux/lockd/bind.h> #include <linux/nfs_mount.h> +#include <linux/freezer.h> #include "iostat.h" #include "internal.h" @@ -32,7 +33,7 @@ nfs3_rpc_wrapper(struct rpc_clnt *clnt, struct rpc_message *msg, int flags) res = rpc_call_sync(clnt, msg, flags); if (res != -EJUKEBOX && res != -EKEYEXPIRED) break; - schedule_timeout_killable(NFS_JUKEBOX_RETRY_TIME); + freezable_schedule_timeout_killable(NFS_JUKEBOX_RETRY_TIME); res = -ERESTARTSYS; } while (!fatal_signal_pending(current)); return res; @@ -853,6 +854,7 @@ const struct nfs_rpc_ops nfs_v3_clientops = { .dentry_ops = &nfs_dentry_operations, .dir_inode_ops = &nfs3_dir_inode_operations, .file_inode_ops = &nfs3_file_inode_operations, + .file_ops = &nfs_file_operations, .getroot = nfs3_proc_get_root, .getattr = nfs3_proc_getattr, .setattr = nfs3_proc_setattr, diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index 693ae22f8731..4d7d0aedc101 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -94,6 +94,8 @@ struct nfs_unique_id { struct nfs4_state_owner { struct nfs_unique_id so_owner_id; struct nfs_server *so_server; + struct list_head so_lru; + unsigned long so_expires; struct rb_node so_server_node; struct rpc_cred *so_cred; /* Associated cred */ @@ -319,6 +321,7 @@ static inline void nfs4_schedule_session_recovery(struct nfs4_session *session) extern struct nfs4_state_owner * nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *); extern void nfs4_put_state_owner(struct nfs4_state_owner *); +extern void nfs4_purge_state_owners(struct nfs_server *); extern struct nfs4_state * nfs4_get_open_state(struct inode *, struct nfs4_state_owner *); extern void nfs4_put_open_state(struct nfs4_state *); extern void nfs4_close_state(struct nfs4_state *, fmode_t); diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/nfs4filelayout.c index a62d36b9a99e..71ec08617e23 100644 --- a/fs/nfs/nfs4filelayout.c +++ b/fs/nfs/nfs4filelayout.c @@ -49,13 +49,14 @@ filelayout_get_dense_offset(struct nfs4_filelayout_segment *flseg, loff_t offset) { u32 stripe_width = flseg->stripe_unit * flseg->dsaddr->stripe_count; - u64 tmp; + u64 stripe_no; + u32 rem; offset -= flseg->pattern_offset; - tmp = offset; - do_div(tmp, stripe_width); + stripe_no = div_u64(offset, stripe_width); + div_u64_rem(offset, flseg->stripe_unit, &rem); - return tmp * flseg->stripe_unit + do_div(offset, flseg->stripe_unit); + return stripe_no * flseg->stripe_unit + rem; } /* This function is used by the layout driver to calculate the diff --git a/fs/nfs/nfs4filelayoutdev.c b/fs/nfs/nfs4filelayoutdev.c index ed388aae9689..8ae91908f5aa 100644 --- a/fs/nfs/nfs4filelayoutdev.c +++ b/fs/nfs/nfs4filelayoutdev.c @@ -382,7 +382,7 @@ decode_ds_addr(struct xdr_stream *streamp, gfp_t gfp_flags) { struct nfs4_pnfs_ds_addr *da = NULL; char *buf, *portstr; - u32 port; + __be16 port; int nlen, rlen; int tmp[2]; __be32 *p; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index b60fddf606f7..f0c849c98fe4 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -39,6 +39,8 @@ #include <linux/delay.h> #include <linux/errno.h> #include <linux/string.h> +#include <linux/ratelimit.h> +#include <linux/printk.h> #include <linux/slab.h> #include <linux/sunrpc/clnt.h> #include <linux/sunrpc/gss_api.h> @@ -50,9 +52,11 @@ #include <linux/namei.h> #include <linux/mount.h> #include <linux/module.h> +#include <linux/nfs_idmap.h> #include <linux/sunrpc/bc_xprt.h> #include <linux/xattr.h> #include <linux/utsname.h> +#include <linux/freezer.h> #include "nfs4_fs.h" #include "delegation.h" @@ -241,7 +245,7 @@ static int nfs4_delay(struct rpc_clnt *clnt, long *timeout) *timeout = NFS4_POLL_RETRY_MIN; if (*timeout > NFS4_POLL_RETRY_MAX) *timeout = NFS4_POLL_RETRY_MAX; - schedule_timeout_killable(*timeout); + freezable_schedule_timeout_killable(*timeout); if (fatal_signal_pending(current)) res = -ERESTARTSYS; *timeout <<= 1; @@ -361,9 +365,8 @@ static void renew_lease(const struct nfs_server *server, unsigned long timestamp * Must be called while holding tbl->slot_tbl_lock */ static void -nfs4_free_slot(struct nfs4_slot_table *tbl, struct nfs4_slot *free_slot) +nfs4_free_slot(struct nfs4_slot_table *tbl, u8 free_slotid) { - int free_slotid = free_slot - tbl->slots; int slotid = free_slotid; BUG_ON(slotid < 0 || slotid >= NFS4_MAX_SLOT_TABLE); @@ -428,7 +431,7 @@ static void nfs41_sequence_free_slot(struct nfs4_sequence_res *res) } spin_lock(&tbl->slot_tbl_lock); - nfs4_free_slot(tbl, res->sr_slot); + nfs4_free_slot(tbl, res->sr_slot - tbl->slots); nfs4_check_drain_fc_complete(res->sr_session); spin_unlock(&tbl->slot_tbl_lock); res->sr_slot = NULL; @@ -551,13 +554,10 @@ int nfs41_setup_sequence(struct nfs4_session *session, spin_lock(&tbl->slot_tbl_lock); if (test_bit(NFS4_SESSION_DRAINING, &session->session_state) && !rpc_task_has_priority(task, RPC_PRIORITY_PRIVILEGED)) { - /* - * The state manager will wait until the slot table is empty. - * Schedule the reset thread - */ + /* The state manager will wait until the slot table is empty */ rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL); spin_unlock(&tbl->slot_tbl_lock); - dprintk("%s Schedule Session Reset\n", __func__); + dprintk("%s session is draining\n", __func__); return -EAGAIN; } @@ -762,6 +762,8 @@ struct nfs4_opendata { struct nfs_openres o_res; struct nfs_open_confirmargs c_arg; struct nfs_open_confirmres c_res; + struct nfs4_string owner_name; + struct nfs4_string group_name; struct nfs_fattr f_attr; struct nfs_fattr dir_attr; struct dentry *dir; @@ -785,6 +787,7 @@ static void nfs4_init_opendata_res(struct nfs4_opendata *p) p->o_res.server = p->o_arg.server; nfs_fattr_init(&p->f_attr); nfs_fattr_init(&p->dir_attr); + nfs_fattr_init_names(&p->f_attr, &p->owner_name, &p->group_name); } static struct nfs4_opendata *nfs4_opendata_alloc(struct dentry *dentry, @@ -816,6 +819,7 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct dentry *dentry, p->o_arg.name = &dentry->d_name; p->o_arg.server = server; p->o_arg.bitmask = server->attr_bitmask; + p->o_arg.dir_bitmask = server->cache_consistency_bitmask; p->o_arg.claim = NFS4_OPEN_CLAIM_NULL; if (flags & O_CREAT) { u32 *s; @@ -852,6 +856,7 @@ static void nfs4_opendata_free(struct kref *kref) dput(p->dir); dput(p->dentry); nfs_sb_deactive(sb); + nfs_fattr_free_names(&p->f_attr); kfree(p); } @@ -894,6 +899,8 @@ out: static int can_open_delegated(struct nfs_delegation *delegation, fmode_t fmode) { + if (delegation == NULL) + return 0; if ((delegation->type & fmode) != fmode) return 0; if (test_bit(NFS_DELEGATION_NEED_RECLAIM, &delegation->flags)) @@ -1036,8 +1043,7 @@ static struct nfs4_state *nfs4_try_open_cached(struct nfs4_opendata *opendata) } rcu_read_lock(); delegation = rcu_dereference(nfsi->delegation); - if (delegation == NULL || - !can_open_delegated(delegation, fmode)) { + if (!can_open_delegated(delegation, fmode)) { rcu_read_unlock(); break; } @@ -1091,7 +1097,12 @@ static struct nfs4_state *nfs4_opendata_to_nfs4_state(struct nfs4_opendata *data if (delegation) delegation_flags = delegation->flags; rcu_read_unlock(); - if ((delegation_flags & 1UL<<NFS_DELEGATION_NEED_RECLAIM) == 0) + if (data->o_arg.claim == NFS4_OPEN_CLAIM_DELEGATE_CUR) { + pr_err_ratelimited("NFS: Broken NFSv4 server %s is " + "returning a delegation for " + "OPEN(CLAIM_DELEGATE_CUR)\n", + NFS_CLIENT(inode)->cl_server); + } else if ((delegation_flags & 1UL<<NFS_DELEGATION_NEED_RECLAIM) == 0) nfs_inode_set_delegation(state->inode, data->owner->so_cred, &data->o_res); @@ -1423,11 +1434,9 @@ static void nfs4_open_prepare(struct rpc_task *task, void *calldata) goto out_no_action; rcu_read_lock(); delegation = rcu_dereference(NFS_I(data->state->inode)->delegation); - if (delegation != NULL && - test_bit(NFS_DELEGATION_NEED_RECLAIM, &delegation->flags) == 0) { - rcu_read_unlock(); - goto out_no_action; - } + if (data->o_arg.claim != NFS4_OPEN_CLAIM_DELEGATE_CUR && + can_open_delegated(delegation, data->o_arg.fmode)) + goto unlock_no_action; rcu_read_unlock(); } /* Update sequence id. */ @@ -1444,6 +1453,8 @@ static void nfs4_open_prepare(struct rpc_task *task, void *calldata) return; rpc_call_start(task); return; +unlock_no_action: + rcu_read_unlock(); out_no_action: task->tk_action = NULL; @@ -1570,6 +1581,8 @@ static int _nfs4_recover_proc_open(struct nfs4_opendata *data) if (status != 0 || !data->rpc_done) return status; + nfs_fattr_map_and_free_names(NFS_SERVER(dir), &data->f_attr); + nfs_refresh_inode(dir, o_res->dir_attr); if (o_res->rflags & NFS4_OPEN_RESULT_CONFIRM) { @@ -1602,6 +1615,8 @@ static int _nfs4_proc_open(struct nfs4_opendata *data) return status; } + nfs_fattr_map_and_free_names(server, &data->f_attr); + if (o_arg->open_flags & O_CREAT) { update_changeattr(dir, &o_res->cinfo); nfs_post_op_update_inode(dir, o_res->dir_attr); @@ -2464,8 +2479,7 @@ static int nfs4_proc_lookup(struct rpc_clnt *clnt, struct inode *dir, struct qst case -NFS4ERR_BADNAME: return -ENOENT; case -NFS4ERR_MOVED: - err = nfs4_get_referral(dir, name, fattr, fhandle); - break; + return nfs4_get_referral(dir, name, fattr, fhandle); case -NFS4ERR_WRONGSEC: nfs_fixup_secinfo_attributes(fattr, fhandle); } @@ -3423,19 +3437,6 @@ static inline int nfs4_server_supports_acls(struct nfs_server *server) */ #define NFS4ACL_MAXPAGES (XATTR_SIZE_MAX >> PAGE_CACHE_SHIFT) -static void buf_to_pages(const void *buf, size_t buflen, - struct page **pages, unsigned int *pgbase) -{ - const void *p = buf; - - *pgbase = offset_in_page(buf); - p -= *pgbase; - while (p < buf + buflen) { - *(pages++) = virt_to_page(p); - p += PAGE_CACHE_SIZE; - } -} - static int buf_to_pages_noslab(const void *buf, size_t buflen, struct page **pages, unsigned int *pgbase) { @@ -3532,9 +3533,19 @@ out: nfs4_set_cached_acl(inode, acl); } +/* + * The getxattr API returns the required buffer length when called with a + * NULL buf. The NFSv4 acl tool then calls getxattr again after allocating + * the required buf. On a NULL buf, we send a page of data to the server + * guessing that the ACL request can be serviced by a page. If so, we cache + * up to the page of ACL data, and the 2nd call to getxattr is serviced by + * the cache. If not so, we throw away the page, and cache the required + * length. The next getxattr call will then produce another round trip to + * the server, this time with the input buf of the required size. + */ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t buflen) { - struct page *pages[NFS4ACL_MAXPAGES]; + struct page *pages[NFS4ACL_MAXPAGES] = {NULL, }; struct nfs_getaclargs args = { .fh = NFS_FH(inode), .acl_pages = pages, @@ -3549,41 +3560,60 @@ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bu .rpc_argp = &args, .rpc_resp = &res, }; - struct page *localpage = NULL; - int ret; + int ret = -ENOMEM, npages, i, acl_len = 0; - if (buflen < PAGE_SIZE) { - /* As long as we're doing a round trip to the server anyway, - * let's be prepared for a page of acl data. */ - localpage = alloc_page(GFP_KERNEL); - resp_buf = page_address(localpage); - if (localpage == NULL) - return -ENOMEM; - args.acl_pages[0] = localpage; - args.acl_pgbase = 0; - args.acl_len = PAGE_SIZE; - } else { - resp_buf = buf; - buf_to_pages(buf, buflen, args.acl_pages, &args.acl_pgbase); + npages = (buflen + PAGE_SIZE - 1) >> PAGE_SHIFT; + /* As long as we're doing a round trip to the server anyway, + * let's be prepared for a page of acl data. */ + if (npages == 0) + npages = 1; + + for (i = 0; i < npages; i++) { + pages[i] = alloc_page(GFP_KERNEL); + if (!pages[i]) + goto out_free; + } + if (npages > 1) { + /* for decoding across pages */ + args.acl_scratch = alloc_page(GFP_KERNEL); + if (!args.acl_scratch) + goto out_free; } - ret = nfs4_call_sync(NFS_SERVER(inode)->client, NFS_SERVER(inode), &msg, &args.seq_args, &res.seq_res, 0); + args.acl_len = npages * PAGE_SIZE; + args.acl_pgbase = 0; + /* Let decode_getfacl know not to fail if the ACL data is larger than + * the page we send as a guess */ + if (buf == NULL) + res.acl_flags |= NFS4_ACL_LEN_REQUEST; + resp_buf = page_address(pages[0]); + + dprintk("%s buf %p buflen %zu npages %d args.acl_len %zu\n", + __func__, buf, buflen, npages, args.acl_len); + ret = nfs4_call_sync(NFS_SERVER(inode)->client, NFS_SERVER(inode), + &msg, &args.seq_args, &res.seq_res, 0); if (ret) goto out_free; - if (res.acl_len > args.acl_len) - nfs4_write_cached_acl(inode, NULL, res.acl_len); + + acl_len = res.acl_len - res.acl_data_offset; + if (acl_len > args.acl_len) + nfs4_write_cached_acl(inode, NULL, acl_len); else - nfs4_write_cached_acl(inode, resp_buf, res.acl_len); + nfs4_write_cached_acl(inode, resp_buf + res.acl_data_offset, + acl_len); if (buf) { ret = -ERANGE; - if (res.acl_len > buflen) + if (acl_len > buflen) goto out_free; - if (localpage) - memcpy(buf, resp_buf, res.acl_len); + _copy_from_pages(buf, pages, res.acl_data_offset, + res.acl_len); } - ret = res.acl_len; + ret = acl_len; out_free: - if (localpage) - __free_page(localpage); + for (i = 0; i < npages; i++) + if (pages[i]) + __free_page(pages[i]); + if (args.acl_scratch) + __free_page(args.acl_scratch); return ret; } @@ -3614,6 +3644,8 @@ static ssize_t nfs4_proc_get_acl(struct inode *inode, void *buf, size_t buflen) nfs_zap_acl_cache(inode); ret = nfs4_read_cached_acl(inode, buf, buflen); if (ret != -ENOENT) + /* -ENOENT is returned if there is no ACL or if there is an ACL + * but no cached acl data, just the acl length */ return ret; return nfs4_get_acl_uncached(inode, buf, buflen); } @@ -3951,7 +3983,7 @@ int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4 static unsigned long nfs4_set_lock_task_retry(unsigned long timeout) { - schedule_timeout_killable(timeout); + freezable_schedule_timeout_killable(timeout); timeout <<= 1; if (timeout > NFS4_LOCK_MAXTIMEOUT) return NFS4_LOCK_MAXTIMEOUT; @@ -5014,23 +5046,6 @@ out: return ret; } -/* - * Reset the forechannel and backchannel slot tables - */ -static int nfs4_reset_slot_tables(struct nfs4_session *session) -{ - int status; - - status = nfs4_reset_slot_table(&session->fc_slot_table, - session->fc_attrs.max_reqs, 1); - if (status) - return status; - - status = nfs4_reset_slot_table(&session->bc_slot_table, - session->bc_attrs.max_reqs, 0); - return status; -} - /* Destroy the slot table */ static void nfs4_destroy_slot_tables(struct nfs4_session *session) { @@ -5076,29 +5091,35 @@ out: } /* - * Initialize the forechannel and backchannel tables + * Initialize or reset the forechannel and backchannel tables */ -static int nfs4_init_slot_tables(struct nfs4_session *session) +static int nfs4_setup_session_slot_tables(struct nfs4_session *ses) { struct nfs4_slot_table *tbl; - int status = 0; + int status; - tbl = &session->fc_slot_table; + dprintk("--> %s\n", __func__); + /* Fore channel */ + tbl = &ses->fc_slot_table; if (tbl->slots == NULL) { - status = nfs4_init_slot_table(tbl, - session->fc_attrs.max_reqs, 1); + status = nfs4_init_slot_table(tbl, ses->fc_attrs.max_reqs, 1); + if (status) /* -ENOMEM */ + return status; + } else { + status = nfs4_reset_slot_table(tbl, ses->fc_attrs.max_reqs, 1); if (status) return status; } - - tbl = &session->bc_slot_table; + /* Back channel */ + tbl = &ses->bc_slot_table; if (tbl->slots == NULL) { - status = nfs4_init_slot_table(tbl, - session->bc_attrs.max_reqs, 0); + status = nfs4_init_slot_table(tbl, ses->bc_attrs.max_reqs, 0); if (status) - nfs4_destroy_slot_tables(session); - } - + /* Fore and back channel share a connection so get + * both slot tables or neither */ + nfs4_destroy_slot_tables(ses); + } else + status = nfs4_reset_slot_table(tbl, ses->bc_attrs.max_reqs, 0); return status; } @@ -5286,13 +5307,9 @@ int nfs4_proc_create_session(struct nfs_client *clp) if (status) goto out; - /* Init and reset the fore channel */ - status = nfs4_init_slot_tables(session); - dprintk("slot table initialization returned %d\n", status); - if (status) - goto out; - status = nfs4_reset_slot_tables(session); - dprintk("slot table reset returned %d\n", status); + /* Init or reset the session slot tables */ + status = nfs4_setup_session_slot_tables(session); + dprintk("slot table setup returned %d\n", status); if (status) goto out; @@ -6253,6 +6270,7 @@ const struct nfs_rpc_ops nfs_v4_clientops = { .dentry_ops = &nfs4_dentry_operations, .dir_inode_ops = &nfs4_dir_inode_operations, .file_inode_ops = &nfs4_file_inode_operations, + .file_ops = &nfs4_file_operations, .getroot = nfs4_proc_get_root, .getattr = nfs4_proc_getattr, .setattr = nfs4_proc_setattr, diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 39914be40b03..a53f33b4ac3a 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -49,6 +49,7 @@ #include <linux/ratelimit.h> #include <linux/workqueue.h> #include <linux/bitops.h> +#include <linux/jiffies.h> #include "nfs4_fs.h" #include "callback.h" @@ -377,31 +378,24 @@ nfs4_find_state_owner_locked(struct nfs_server *server, struct rpc_cred *cred) { struct rb_node **p = &server->state_owners.rb_node, *parent = NULL; - struct nfs4_state_owner *sp, *res = NULL; + struct nfs4_state_owner *sp; while (*p != NULL) { parent = *p; sp = rb_entry(parent, struct nfs4_state_owner, so_server_node); - if (server < sp->so_server) { - p = &parent->rb_left; - continue; - } - if (server > sp->so_server) { - p = &parent->rb_right; - continue; - } if (cred < sp->so_cred) p = &parent->rb_left; else if (cred > sp->so_cred) p = &parent->rb_right; else { + if (!list_empty(&sp->so_lru)) + list_del_init(&sp->so_lru); atomic_inc(&sp->so_count); - res = sp; - break; + return sp; } } - return res; + return NULL; } static struct nfs4_state_owner * @@ -421,6 +415,8 @@ nfs4_insert_state_owner_locked(struct nfs4_state_owner *new) else if (new->so_cred > sp->so_cred) p = &parent->rb_right; else { + if (!list_empty(&sp->so_lru)) + list_del_init(&sp->so_lru); atomic_inc(&sp->so_count); return sp; } @@ -462,6 +458,7 @@ nfs4_alloc_state_owner(void) spin_lock_init(&sp->so_sequence.lock); INIT_LIST_HEAD(&sp->so_sequence.list); atomic_set(&sp->so_count, 1); + INIT_LIST_HEAD(&sp->so_lru); return sp; } @@ -479,6 +476,38 @@ nfs4_drop_state_owner(struct nfs4_state_owner *sp) } } +static void nfs4_free_state_owner(struct nfs4_state_owner *sp) +{ + rpc_destroy_wait_queue(&sp->so_sequence.wait); + put_rpccred(sp->so_cred); + kfree(sp); +} + +static void nfs4_gc_state_owners(struct nfs_server *server) +{ + struct nfs_client *clp = server->nfs_client; + struct nfs4_state_owner *sp, *tmp; + unsigned long time_min, time_max; + LIST_HEAD(doomed); + + spin_lock(&clp->cl_lock); + time_max = jiffies; + time_min = (long)time_max - (long)clp->cl_lease_time; + list_for_each_entry_safe(sp, tmp, &server->state_owners_lru, so_lru) { + /* NB: LRU is sorted so that oldest is at the head */ + if (time_in_range(sp->so_expires, time_min, time_max)) + break; + list_move(&sp->so_lru, &doomed); + nfs4_remove_state_owner_locked(sp); + } + spin_unlock(&clp->cl_lock); + + list_for_each_entry_safe(sp, tmp, &doomed, so_lru) { + list_del(&sp->so_lru); + nfs4_free_state_owner(sp); + } +} + /** * nfs4_get_state_owner - Look up a state owner given a credential * @server: nfs_server to search @@ -496,10 +525,10 @@ struct nfs4_state_owner *nfs4_get_state_owner(struct nfs_server *server, sp = nfs4_find_state_owner_locked(server, cred); spin_unlock(&clp->cl_lock); if (sp != NULL) - return sp; + goto out; new = nfs4_alloc_state_owner(); if (new == NULL) - return NULL; + goto out; new->so_server = server; new->so_cred = cred; spin_lock(&clp->cl_lock); @@ -511,26 +540,58 @@ struct nfs4_state_owner *nfs4_get_state_owner(struct nfs_server *server, rpc_destroy_wait_queue(&new->so_sequence.wait); kfree(new); } +out: + nfs4_gc_state_owners(server); return sp; } /** * nfs4_put_state_owner - Release a nfs4_state_owner * @sp: state owner data to release - * */ void nfs4_put_state_owner(struct nfs4_state_owner *sp) { - struct nfs_client *clp = sp->so_server->nfs_client; - struct rpc_cred *cred = sp->so_cred; + struct nfs_server *server = sp->so_server; + struct nfs_client *clp = server->nfs_client; if (!atomic_dec_and_lock(&sp->so_count, &clp->cl_lock)) return; - nfs4_remove_state_owner_locked(sp); + + if (!RB_EMPTY_NODE(&sp->so_server_node)) { + sp->so_expires = jiffies; + list_add_tail(&sp->so_lru, &server->state_owners_lru); + spin_unlock(&clp->cl_lock); + } else { + nfs4_remove_state_owner_locked(sp); + spin_unlock(&clp->cl_lock); + nfs4_free_state_owner(sp); + } +} + +/** + * nfs4_purge_state_owners - Release all cached state owners + * @server: nfs_server with cached state owners to release + * + * Called at umount time. Remaining state owners will be on + * the LRU with ref count of zero. + */ +void nfs4_purge_state_owners(struct nfs_server *server) +{ + struct nfs_client *clp = server->nfs_client; + struct nfs4_state_owner *sp, *tmp; + LIST_HEAD(doomed); + + spin_lock(&clp->cl_lock); + list_for_each_entry_safe(sp, tmp, &server->state_owners_lru, so_lru) { + list_move(&sp->so_lru, &doomed); + nfs4_remove_state_owner_locked(sp); + } spin_unlock(&clp->cl_lock); - rpc_destroy_wait_queue(&sp->so_sequence.wait); - put_rpccred(cred); - kfree(sp); + + list_for_each_entry_safe(sp, tmp, &doomed, so_lru) { + list_del(&sp->so_lru); + nfs4_free_state_owner(sp); + } } static struct nfs4_state * @@ -1156,11 +1217,13 @@ restart: if (status >= 0) { status = nfs4_reclaim_locks(state, ops); if (status >= 0) { + spin_lock(&state->state_lock); list_for_each_entry(lock, &state->lock_states, ls_locks) { if (!(lock->ls_flags & NFS_LOCK_INITIALIZED)) printk("%s: Lock reclaim failed!\n", __func__); } + spin_unlock(&state->state_lock); nfs4_put_open_state(state); goto restart; } @@ -1224,10 +1287,12 @@ static void nfs4_clear_open_state(struct nfs4_state *state) clear_bit(NFS_O_RDONLY_STATE, &state->flags); clear_bit(NFS_O_WRONLY_STATE, &state->flags); clear_bit(NFS_O_RDWR_STATE, &state->flags); + spin_lock(&state->state_lock); list_for_each_entry(lock, &state->lock_states, ls_locks) { lock->ls_seqid.flags = 0; lock->ls_flags &= ~NFS_LOCK_INITIALIZED; } + spin_unlock(&state->state_lock); } static void nfs4_reset_seqids(struct nfs_server *server, @@ -1350,12 +1415,14 @@ static void nfs4_warn_keyexpired(const char *s) static int nfs4_recovery_handle_error(struct nfs_client *clp, int error) { switch (error) { + case 0: + break; case -NFS4ERR_CB_PATH_DOWN: nfs_handle_cb_pathdown(clp); - return 0; + break; case -NFS4ERR_NO_GRACE: nfs4_state_end_reclaim_reboot(clp); - return 0; + break; case -NFS4ERR_STALE_CLIENTID: case -NFS4ERR_LEASE_MOVED: set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state); @@ -1375,13 +1442,15 @@ static int nfs4_recovery_handle_error(struct nfs_client *clp, int error) case -NFS4ERR_SEQ_MISORDERED: set_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state); /* Zero session reset errors */ - return 0; + break; case -EKEYEXPIRED: /* Nothing we can do */ nfs4_warn_keyexpired(clp->cl_hostname); - return 0; + break; + default: + return error; } - return error; + return 0; } static int nfs4_do_reclaim(struct nfs_client *clp, const struct nfs4_state_recovery_ops *ops) @@ -1394,6 +1463,7 @@ static int nfs4_do_reclaim(struct nfs_client *clp, const struct nfs4_state_recov restart: rcu_read_lock(); list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { + nfs4_purge_state_owners(server); spin_lock(&clp->cl_lock); for (pos = rb_first(&server->state_owners); pos != NULL; @@ -1428,7 +1498,7 @@ static int nfs4_check_lease(struct nfs_client *clp) struct rpc_cred *cred; const struct nfs4_state_maintenance_ops *ops = clp->cl_mvops->state_renewal_ops; - int status = -NFS4ERR_EXPIRED; + int status; /* Is the client already known to have an expired lease? */ if (test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state)) @@ -1438,6 +1508,7 @@ static int nfs4_check_lease(struct nfs_client *clp) spin_unlock(&clp->cl_lock); if (cred == NULL) { cred = nfs4_get_setclientid_cred(clp); + status = -ENOKEY; if (cred == NULL) goto out; } @@ -1525,16 +1596,16 @@ void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags) { if (!flags) return; - else if (flags & SEQ4_STATUS_RESTART_RECLAIM_NEEDED) + if (flags & SEQ4_STATUS_RESTART_RECLAIM_NEEDED) nfs41_handle_server_reboot(clp); - else if (flags & (SEQ4_STATUS_EXPIRED_ALL_STATE_REVOKED | + if (flags & (SEQ4_STATUS_EXPIRED_ALL_STATE_REVOKED | SEQ4_STATUS_EXPIRED_SOME_STATE_REVOKED | SEQ4_STATUS_ADMIN_STATE_REVOKED | SEQ4_STATUS_LEASE_MOVED)) nfs41_handle_state_revoked(clp); - else if (flags & SEQ4_STATUS_RECALLABLE_STATE_REVOKED) + if (flags & SEQ4_STATUS_RECALLABLE_STATE_REVOKED) nfs41_handle_recallable_state_revoked(clp); - else if (flags & (SEQ4_STATUS_CB_PATH_DOWN | + if (flags & (SEQ4_STATUS_CB_PATH_DOWN | SEQ4_STATUS_BACKCHANNEL_FAULT | SEQ4_STATUS_CB_PATH_DOWN_SESSION)) nfs41_handle_cb_path_down(clp); @@ -1662,10 +1733,10 @@ static void nfs4_state_manager(struct nfs_client *clp) if (test_and_clear_bit(NFS4CLNT_CHECK_LEASE, &clp->cl_state)) { status = nfs4_check_lease(clp); + if (status < 0) + goto out_error; if (test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state)) continue; - if (status < 0 && status != -NFS4ERR_CB_PATH_DOWN) - goto out_error; } /* Initialize or reset the session */ diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index e6161b213ed1..95e92e438407 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -2298,7 +2298,7 @@ static void nfs4_xdr_enc_open(struct rpc_rqst *req, struct xdr_stream *xdr, encode_getfh(xdr, &hdr); encode_getfattr(xdr, args->bitmask, &hdr); encode_restorefh(xdr, &hdr); - encode_getfattr(xdr, args->bitmask, &hdr); + encode_getfattr(xdr, args->dir_bitmask, &hdr); encode_nops(&hdr); } @@ -2517,11 +2517,13 @@ static void nfs4_xdr_enc_getacl(struct rpc_rqst *req, struct xdr_stream *xdr, encode_compound_hdr(xdr, req, &hdr); encode_sequence(xdr, &args->seq_args, &hdr); encode_putfh(xdr, args->fh, &hdr); - replen = hdr.replen + op_decode_hdr_maxsz + nfs4_fattr_bitmap_maxsz + 1; + replen = hdr.replen + op_decode_hdr_maxsz + 1; encode_getattr_two(xdr, FATTR4_WORD0_ACL, 0, &hdr); xdr_inline_pages(&req->rq_rcv_buf, replen << 2, args->acl_pages, args->acl_pgbase, args->acl_len); + xdr_set_scratch_buffer(xdr, page_address(args->acl_scratch), PAGE_SIZE); + encode_nops(&hdr); } @@ -3790,7 +3792,8 @@ out_overflow: } static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, - const struct nfs_server *server, uint32_t *uid, int may_sleep) + const struct nfs_server *server, uint32_t *uid, + struct nfs4_string *owner_name) { uint32_t len; __be32 *p; @@ -3807,8 +3810,12 @@ static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, p = xdr_inline_decode(xdr, len); if (unlikely(!p)) goto out_overflow; - if (!may_sleep) { - /* do nothing */ + if (owner_name != NULL) { + owner_name->data = kmemdup(p, len, GFP_NOWAIT); + if (owner_name->data != NULL) { + owner_name->len = len; + ret = NFS_ATTR_FATTR_OWNER_NAME; + } } else if (len < XDR_MAX_NETOBJ) { if (nfs_map_name_to_uid(server, (char *)p, len, uid) == 0) ret = NFS_ATTR_FATTR_OWNER; @@ -3828,7 +3835,8 @@ out_overflow: } static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, - const struct nfs_server *server, uint32_t *gid, int may_sleep) + const struct nfs_server *server, uint32_t *gid, + struct nfs4_string *group_name) { uint32_t len; __be32 *p; @@ -3845,8 +3853,12 @@ static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, p = xdr_inline_decode(xdr, len); if (unlikely(!p)) goto out_overflow; - if (!may_sleep) { - /* do nothing */ + if (group_name != NULL) { + group_name->data = kmemdup(p, len, GFP_NOWAIT); + if (group_name->data != NULL) { + group_name->len = len; + ret = NFS_ATTR_FATTR_GROUP_NAME; + } } else if (len < XDR_MAX_NETOBJ) { if (nfs_map_group_to_gid(server, (char *)p, len, gid) == 0) ret = NFS_ATTR_FATTR_GROUP; @@ -4283,7 +4295,7 @@ xdr_error: static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_fattr *fattr, struct nfs_fh *fh, - const struct nfs_server *server, int may_sleep) + const struct nfs_server *server) { int status; umode_t fmode = 0; @@ -4350,12 +4362,12 @@ static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap, goto xdr_error; fattr->valid |= status; - status = decode_attr_owner(xdr, bitmap, server, &fattr->uid, may_sleep); + status = decode_attr_owner(xdr, bitmap, server, &fattr->uid, fattr->owner_name); if (status < 0) goto xdr_error; fattr->valid |= status; - status = decode_attr_group(xdr, bitmap, server, &fattr->gid, may_sleep); + status = decode_attr_group(xdr, bitmap, server, &fattr->gid, fattr->group_name); if (status < 0) goto xdr_error; fattr->valid |= status; @@ -4396,7 +4408,7 @@ xdr_error: } static int decode_getfattr_generic(struct xdr_stream *xdr, struct nfs_fattr *fattr, - struct nfs_fh *fh, const struct nfs_server *server, int may_sleep) + struct nfs_fh *fh, const struct nfs_server *server) { __be32 *savep; uint32_t attrlen, @@ -4415,7 +4427,7 @@ static int decode_getfattr_generic(struct xdr_stream *xdr, struct nfs_fattr *fat if (status < 0) goto xdr_error; - status = decode_getfattr_attrs(xdr, bitmap, fattr, fh, server, may_sleep); + status = decode_getfattr_attrs(xdr, bitmap, fattr, fh, server); if (status < 0) goto xdr_error; @@ -4426,9 +4438,9 @@ xdr_error: } static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, - const struct nfs_server *server, int may_sleep) + const struct nfs_server *server) { - return decode_getfattr_generic(xdr, fattr, NULL, server, may_sleep); + return decode_getfattr_generic(xdr, fattr, NULL, server); } /* @@ -4957,17 +4969,18 @@ decode_restorefh(struct xdr_stream *xdr) } static int decode_getacl(struct xdr_stream *xdr, struct rpc_rqst *req, - size_t *acl_len) + struct nfs_getaclres *res) { - __be32 *savep; + __be32 *savep, *bm_p; uint32_t attrlen, bitmap[3] = {0}; struct kvec *iov = req->rq_rcv_buf.head; int status; - *acl_len = 0; + res->acl_len = 0; if ((status = decode_op_hdr(xdr, OP_GETATTR)) != 0) goto out; + bm_p = xdr->p; if ((status = decode_attr_bitmap(xdr, bitmap)) != 0) goto out; if ((status = decode_attr_length(xdr, &attrlen, &savep)) != 0) @@ -4979,18 +4992,30 @@ static int decode_getacl(struct xdr_stream *xdr, struct rpc_rqst *req, size_t hdrlen; u32 recvd; + /* The bitmap (xdr len + bitmaps) and the attr xdr len words + * are stored with the acl data to handle the problem of + * variable length bitmaps.*/ + xdr->p = bm_p; + res->acl_data_offset = be32_to_cpup(bm_p) + 2; + res->acl_data_offset <<= 2; + /* We ignore &savep and don't do consistency checks on * the attr length. Let userspace figure it out.... */ hdrlen = (u8 *)xdr->p - (u8 *)iov->iov_base; + attrlen += res->acl_data_offset; recvd = req->rq_rcv_buf.len - hdrlen; if (attrlen > recvd) { - dprintk("NFS: server cheating in getattr" - " acl reply: attrlen %u > recvd %u\n", + if (res->acl_flags & NFS4_ACL_LEN_REQUEST) { + /* getxattr interface called with a NULL buf */ + res->acl_len = attrlen; + goto out; + } + dprintk("NFS: acl reply: attrlen %u > recvd %u\n", attrlen, recvd); return -EINVAL; } xdr_read_pages(xdr, attrlen); - *acl_len = attrlen; + res->acl_len = attrlen; } else status = -EOPNOTSUPP; @@ -5696,8 +5721,7 @@ static int nfs4_xdr_dec_open_downgrade(struct rpc_rqst *rqstp, status = decode_open_downgrade(xdr, res); if (status != 0) goto out; - decode_getfattr(xdr, res->fattr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)); + decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -5723,8 +5747,7 @@ static int nfs4_xdr_dec_access(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_access(xdr, res); if (status != 0) goto out; - decode_getfattr(xdr, res->fattr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)); + decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -5753,8 +5776,7 @@ static int nfs4_xdr_dec_lookup(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_getfh(xdr, res->fh); if (status) goto out; - status = decode_getfattr(xdr, res->fattr, res->server - ,!RPC_IS_ASYNC(rqstp->rq_task)); + status = decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -5780,8 +5802,7 @@ static int nfs4_xdr_dec_lookup_root(struct rpc_rqst *rqstp, goto out; status = decode_getfh(xdr, res->fh); if (status == 0) - status = decode_getfattr(xdr, res->fattr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)); + status = decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -5807,8 +5828,7 @@ static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_remove(xdr, &res->cinfo); if (status) goto out; - decode_getfattr(xdr, res->dir_attr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)); + decode_getfattr(xdr, res->dir_attr, res->server); out: return status; } @@ -5841,14 +5861,12 @@ static int nfs4_xdr_dec_rename(struct rpc_rqst *rqstp, struct xdr_stream *xdr, if (status) goto out; /* Current FH is target directory */ - if (decode_getfattr(xdr, res->new_fattr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)) != 0) + if (decode_getfattr(xdr, res->new_fattr, res->server)) goto out; status = decode_restorefh(xdr); if (status) goto out; - decode_getfattr(xdr, res->old_fattr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)); + decode_getfattr(xdr, res->old_fattr, res->server); out: return status; } @@ -5884,14 +5902,12 @@ static int nfs4_xdr_dec_link(struct rpc_rqst *rqstp, struct xdr_stream *xdr, * Note order: OP_LINK leaves the directory as the current * filehandle. */ - if (decode_getfattr(xdr, res->dir_attr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)) != 0) + if (decode_getfattr(xdr, res->dir_attr, res->server)) goto out; status = decode_restorefh(xdr); if (status) goto out; - decode_getfattr(xdr, res->fattr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)); + decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -5923,14 +5939,12 @@ static int nfs4_xdr_dec_create(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_getfh(xdr, res->fh); if (status) goto out; - if (decode_getfattr(xdr, res->fattr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)) != 0) + if (decode_getfattr(xdr, res->fattr, res->server)) goto out; status = decode_restorefh(xdr); if (status) goto out; - decode_getfattr(xdr, res->dir_fattr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)); + decode_getfattr(xdr, res->dir_fattr, res->server); out: return status; } @@ -5962,8 +5976,7 @@ static int nfs4_xdr_dec_getattr(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_putfh(xdr); if (status) goto out; - status = decode_getfattr(xdr, res->fattr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)); + status = decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -6028,7 +6041,7 @@ nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_putfh(xdr); if (status) goto out; - status = decode_getacl(xdr, rqstp, &res->acl_len); + status = decode_getacl(xdr, rqstp, res); out: return status; @@ -6061,8 +6074,7 @@ static int nfs4_xdr_dec_close(struct rpc_rqst *rqstp, struct xdr_stream *xdr, * an ESTALE error. Shouldn't be a problem, * though, since fattr->valid will remain unset. */ - decode_getfattr(xdr, res->fattr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)); + decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -6093,13 +6105,11 @@ static int nfs4_xdr_dec_open(struct rpc_rqst *rqstp, struct xdr_stream *xdr, goto out; if (decode_getfh(xdr, &res->fh) != 0) goto out; - if (decode_getfattr(xdr, res->f_attr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)) != 0) + if (decode_getfattr(xdr, res->f_attr, res->server) != 0) goto out; if (decode_restorefh(xdr) != 0) goto out; - decode_getfattr(xdr, res->dir_attr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)); + decode_getfattr(xdr, res->dir_attr, res->server); out: return status; } @@ -6147,8 +6157,7 @@ static int nfs4_xdr_dec_open_noattr(struct rpc_rqst *rqstp, status = decode_open(xdr, res); if (status) goto out; - decode_getfattr(xdr, res->f_attr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)); + decode_getfattr(xdr, res->f_attr, res->server); out: return status; } @@ -6175,8 +6184,7 @@ static int nfs4_xdr_dec_setattr(struct rpc_rqst *rqstp, status = decode_setattr(xdr); if (status) goto out; - decode_getfattr(xdr, res->fattr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)); + decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -6356,8 +6364,7 @@ static int nfs4_xdr_dec_write(struct rpc_rqst *rqstp, struct xdr_stream *xdr, if (status) goto out; if (res->fattr) - decode_getfattr(xdr, res->fattr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)); + decode_getfattr(xdr, res->fattr, res->server); if (!status) status = res->count; out: @@ -6386,8 +6393,7 @@ static int nfs4_xdr_dec_commit(struct rpc_rqst *rqstp, struct xdr_stream *xdr, if (status) goto out; if (res->fattr) - decode_getfattr(xdr, res->fattr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)); + decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -6546,8 +6552,7 @@ static int nfs4_xdr_dec_delegreturn(struct rpc_rqst *rqstp, status = decode_delegreturn(xdr); if (status != 0) goto out; - decode_getfattr(xdr, res->fattr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)); + decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -6576,8 +6581,7 @@ static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, goto out; xdr_enter_page(xdr, PAGE_SIZE); status = decode_getfattr(xdr, &res->fs_locations->fattr, - res->fs_locations->server, - !RPC_IS_ASYNC(req->rq_task)); + res->fs_locations->server); out: return status; } @@ -6826,8 +6830,7 @@ static int nfs4_xdr_dec_layoutcommit(struct rpc_rqst *rqstp, status = decode_layoutcommit(xdr, rqstp, res); if (status) goto out; - decode_getfattr(xdr, res->fattr, res->server, - !RPC_IS_ASYNC(rqstp->rq_task)); + decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -6958,7 +6961,7 @@ int nfs4_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, goto out_overflow; if (decode_getfattr_attrs(xdr, bitmap, entry->fattr, entry->fh, - entry->server, 1) < 0) + entry->server) < 0) goto out_overflow; if (entry->fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID) entry->ino = entry->fattr->mounted_on_fileid; diff --git a/fs/nfs/objlayout/objio_osd.c b/fs/nfs/objlayout/objio_osd.c index c807ab93140e..55d01280a609 100644 --- a/fs/nfs/objlayout/objio_osd.c +++ b/fs/nfs/objlayout/objio_osd.c @@ -551,7 +551,8 @@ static const struct nfs_pageio_ops objio_pg_write_ops = { static struct pnfs_layoutdriver_type objlayout_type = { .id = LAYOUT_OSD2_OBJECTS, .name = "LAYOUT_OSD2_OBJECTS", - .flags = PNFS_LAYOUTRET_ON_SETATTR, + .flags = PNFS_LAYOUTRET_ON_SETATTR | + PNFS_LAYOUTRET_ON_ERROR, .alloc_layout_hdr = objlayout_alloc_layout_hdr, .free_layout_hdr = objlayout_free_layout_hdr, diff --git a/fs/nfs/objlayout/objlayout.c b/fs/nfs/objlayout/objlayout.c index 72074e3a04f9..b3c29039f5b8 100644 --- a/fs/nfs/objlayout/objlayout.c +++ b/fs/nfs/objlayout/objlayout.c @@ -254,6 +254,8 @@ objlayout_read_done(struct objlayout_io_res *oir, ssize_t status, bool sync) oir->status = rdata->task.tk_status = status; if (status >= 0) rdata->res.count = status; + else + rdata->pnfs_error = status; objlayout_iodone(oir); /* must not use oir after this point */ @@ -334,6 +336,8 @@ objlayout_write_done(struct objlayout_io_res *oir, ssize_t status, bool sync) if (status >= 0) { wdata->res.count = status; wdata->verf.committed = oir->committed; + } else { + wdata->pnfs_error = status; } objlayout_iodone(oir); /* must not use oir after this point */ diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index baf73536bc04..17149a490065 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -1166,6 +1166,33 @@ pnfs_generic_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, } EXPORT_SYMBOL_GPL(pnfs_generic_pg_test); +static int pnfs_write_done_resend_to_mds(struct inode *inode, struct list_head *head) +{ + struct nfs_pageio_descriptor pgio; + LIST_HEAD(failed); + + /* Resend all requests through the MDS */ + nfs_pageio_init_write_mds(&pgio, inode, FLUSH_STABLE); + while (!list_empty(head)) { + struct nfs_page *req = nfs_list_entry(head->next); + + nfs_list_remove_request(req); + if (!nfs_pageio_add_request(&pgio, req)) + nfs_list_add_request(req, &failed); + } + nfs_pageio_complete(&pgio); + + if (!list_empty(&failed)) { + /* For some reason our attempt to resend pages. Mark the + * overall send request as having failed, and let + * nfs_writeback_release_full deal with the error. + */ + list_move(&failed, head); + return -EIO; + } + return 0; +} + /* * Called by non rpc-based layout drivers */ @@ -1175,9 +1202,17 @@ void pnfs_ld_write_done(struct nfs_write_data *data) pnfs_set_layoutcommit(data); data->mds_ops->rpc_call_done(&data->task, data); } else { - put_lseg(data->lseg); - data->lseg = NULL; dprintk("pnfs write error = %d\n", data->pnfs_error); + if (NFS_SERVER(data->inode)->pnfs_curr_ld->flags & + PNFS_LAYOUTRET_ON_ERROR) { + /* Don't lo_commit on error, Server will needs to + * preform a file recovery. + */ + clear_bit(NFS_INO_LAYOUTCOMMIT, + &NFS_I(data->inode)->flags); + pnfs_return_layout(data->inode); + } + data->task.tk_status = pnfs_write_done_resend_to_mds(data->inode, &data->pages); } data->mds_ops->rpc_release(data); } @@ -1260,6 +1295,28 @@ pnfs_generic_pg_writepages(struct nfs_pageio_descriptor *desc) } EXPORT_SYMBOL_GPL(pnfs_generic_pg_writepages); +static void pnfs_ld_handle_read_error(struct nfs_read_data *data) +{ + struct nfs_pageio_descriptor pgio; + + put_lseg(data->lseg); + data->lseg = NULL; + dprintk("pnfs write error = %d\n", data->pnfs_error); + if (NFS_SERVER(data->inode)->pnfs_curr_ld->flags & + PNFS_LAYOUTRET_ON_ERROR) + pnfs_return_layout(data->inode); + + nfs_pageio_init_read_mds(&pgio, data->inode); + + while (!list_empty(&data->pages)) { + struct nfs_page *req = nfs_list_entry(data->pages.next); + + nfs_list_remove_request(req); + nfs_pageio_add_request(&pgio, req); + } + nfs_pageio_complete(&pgio); +} + /* * Called by non rpc-based layout drivers */ @@ -1268,11 +1325,8 @@ void pnfs_ld_read_done(struct nfs_read_data *data) if (likely(!data->pnfs_error)) { __nfs4_read_done_cb(data); data->mds_ops->rpc_call_done(&data->task, data); - } else { - put_lseg(data->lseg); - data->lseg = NULL; - dprintk("pnfs write error = %d\n", data->pnfs_error); - } + } else + pnfs_ld_handle_read_error(data); data->mds_ops->rpc_release(data); } EXPORT_SYMBOL_GPL(pnfs_ld_read_done); diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index 1509530cb111..53d593a0a4f2 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -68,6 +68,7 @@ enum { enum layoutdriver_policy_flags { /* Should the pNFS client commit and return the layout upon a setattr */ PNFS_LAYOUTRET_ON_SETATTR = 1 << 0, + PNFS_LAYOUTRET_ON_ERROR = 1 << 1, }; struct nfs4_deviceid_node; diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index ac40b8535d7e..0c672588fe5a 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -41,6 +41,7 @@ #include <linux/nfs_fs.h> #include <linux/nfs_page.h> #include <linux/lockd/bind.h> +#include <linux/freezer.h> #include "internal.h" #define NFSDBG_FACILITY NFSDBG_PROC @@ -59,7 +60,7 @@ nfs_rpc_wrapper(struct rpc_clnt *clnt, struct rpc_message *msg, int flags) res = rpc_call_sync(clnt, msg, flags); if (res != -EKEYEXPIRED) break; - schedule_timeout_killable(NFS_JUKEBOX_RETRY_TIME); + freezable_schedule_timeout_killable(NFS_JUKEBOX_RETRY_TIME); res = -ERESTARTSYS; } while (!fatal_signal_pending(current)); return res; @@ -710,6 +711,7 @@ const struct nfs_rpc_ops nfs_v2_clientops = { .dentry_ops = &nfs_dentry_operations, .dir_inode_ops = &nfs_dir_inode_operations, .file_inode_ops = &nfs_file_inode_operations, + .file_ops = &nfs_file_operations, .getroot = nfs_proc_get_root, .getattr = nfs_proc_getattr, .setattr = nfs_proc_setattr, diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 8b48ec63f722..cfa175c223dc 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -109,7 +109,7 @@ static void nfs_readpage_truncate_uninitialised_page(struct nfs_read_data *data) } } -static void nfs_pageio_init_read_mds(struct nfs_pageio_descriptor *pgio, +void nfs_pageio_init_read_mds(struct nfs_pageio_descriptor *pgio, struct inode *inode) { nfs_pageio_init(pgio, inode, &nfs_pageio_read_ops, @@ -534,23 +534,13 @@ static void nfs_readpage_result_full(struct rpc_task *task, void *calldata) static void nfs_readpage_release_full(void *calldata) { struct nfs_read_data *data = calldata; - struct nfs_pageio_descriptor pgio; - if (data->pnfs_error) { - nfs_pageio_init_read_mds(&pgio, data->inode); - pgio.pg_recoalesce = 1; - } while (!list_empty(&data->pages)) { struct nfs_page *req = nfs_list_entry(data->pages.next); nfs_list_remove_request(req); - if (!data->pnfs_error) - nfs_readpage_release(req); - else - nfs_pageio_add_request(&pgio, req); + nfs_readpage_release(req); } - if (data->pnfs_error) - nfs_pageio_complete(&pgio); nfs_readdata_release(calldata); } diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 480b3b6bf71e..3dfa4f112c0a 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -41,7 +41,6 @@ #include <linux/lockd/bind.h> #include <linux/seq_file.h> #include <linux/mount.h> -#include <linux/mnt_namespace.h> #include <linux/namei.h> #include <linux/nfs_idmap.h> #include <linux/vfs.h> @@ -263,10 +262,10 @@ static match_table_t nfs_local_lock_tokens = { static void nfs_umount_begin(struct super_block *); static int nfs_statfs(struct dentry *, struct kstatfs *); -static int nfs_show_options(struct seq_file *, struct vfsmount *); -static int nfs_show_devname(struct seq_file *, struct vfsmount *); -static int nfs_show_path(struct seq_file *, struct vfsmount *); -static int nfs_show_stats(struct seq_file *, struct vfsmount *); +static int nfs_show_options(struct seq_file *, struct dentry *); +static int nfs_show_devname(struct seq_file *, struct dentry *); +static int nfs_show_path(struct seq_file *, struct dentry *); +static int nfs_show_stats(struct seq_file *, struct dentry *); static struct dentry *nfs_fs_mount(struct file_system_type *, int, const char *, void *); static struct dentry *nfs_xdev_mount(struct file_system_type *fs_type, @@ -721,9 +720,9 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss, /* * Describe the mount options on this VFS mountpoint */ -static int nfs_show_options(struct seq_file *m, struct vfsmount *mnt) +static int nfs_show_options(struct seq_file *m, struct dentry *root) { - struct nfs_server *nfss = NFS_SB(mnt->mnt_sb); + struct nfs_server *nfss = NFS_SB(root->d_sb); nfs_show_mount_options(m, nfss, 0); @@ -761,14 +760,14 @@ static void show_pnfs(struct seq_file *m, struct nfs_server *server) {} #endif #endif -static int nfs_show_devname(struct seq_file *m, struct vfsmount *mnt) +static int nfs_show_devname(struct seq_file *m, struct dentry *root) { char *page = (char *) __get_free_page(GFP_KERNEL); char *devname, *dummy; int err = 0; if (!page) return -ENOMEM; - devname = nfs_path(&dummy, mnt->mnt_root, page, PAGE_SIZE); + devname = nfs_path(&dummy, root, page, PAGE_SIZE); if (IS_ERR(devname)) err = PTR_ERR(devname); else @@ -777,7 +776,7 @@ static int nfs_show_devname(struct seq_file *m, struct vfsmount *mnt) return err; } -static int nfs_show_path(struct seq_file *m, struct vfsmount *mnt) +static int nfs_show_path(struct seq_file *m, struct dentry *dentry) { seq_puts(m, "/"); return 0; @@ -786,10 +785,10 @@ static int nfs_show_path(struct seq_file *m, struct vfsmount *mnt) /* * Present statistical information for this VFS mountpoint */ -static int nfs_show_stats(struct seq_file *m, struct vfsmount *mnt) +static int nfs_show_stats(struct seq_file *m, struct dentry *root) { int i, cpu; - struct nfs_server *nfss = NFS_SB(mnt->mnt_sb); + struct nfs_server *nfss = NFS_SB(root->d_sb); struct rpc_auth *auth = nfss->client->cl_auth; struct nfs_iostats totals = { }; @@ -799,10 +798,10 @@ static int nfs_show_stats(struct seq_file *m, struct vfsmount *mnt) * Display all mount option settings */ seq_printf(m, "\n\topts:\t"); - seq_puts(m, mnt->mnt_sb->s_flags & MS_RDONLY ? "ro" : "rw"); - seq_puts(m, mnt->mnt_sb->s_flags & MS_SYNCHRONOUS ? ",sync" : ""); - seq_puts(m, mnt->mnt_sb->s_flags & MS_NOATIME ? ",noatime" : ""); - seq_puts(m, mnt->mnt_sb->s_flags & MS_NODIRATIME ? ",nodiratime" : ""); + seq_puts(m, root->d_sb->s_flags & MS_RDONLY ? "ro" : "rw"); + seq_puts(m, root->d_sb->s_flags & MS_SYNCHRONOUS ? ",sync" : ""); + seq_puts(m, root->d_sb->s_flags & MS_NOATIME ? ",noatime" : ""); + seq_puts(m, root->d_sb->s_flags & MS_NODIRATIME ? ",nodiratime" : ""); nfs_show_mount_options(m, nfss, 1); seq_printf(m, "\n\tage:\t%lu", (jiffies - nfss->mount_time) / HZ); @@ -909,10 +908,24 @@ static struct nfs_parsed_mount_data *nfs_alloc_parsed_mount_data(unsigned int ve data->auth_flavor_len = 1; data->version = version; data->minorversion = 0; + security_init_mnt_opts(&data->lsm_opts); } return data; } +static void nfs_free_parsed_mount_data(struct nfs_parsed_mount_data *data) +{ + if (data) { + kfree(data->client_address); + kfree(data->mount_server.hostname); + kfree(data->nfs_server.export_path); + kfree(data->nfs_server.hostname); + kfree(data->fscache_uniq); + security_free_mnt_opts(&data->lsm_opts); + kfree(data); + } +} + /* * Sanity-check a server address provided by the mount command. * @@ -2220,9 +2233,7 @@ static struct dentry *nfs_fs_mount(struct file_system_type *fs_type, data = nfs_alloc_parsed_mount_data(NFS_DEFAULT_VERSION); mntfh = nfs_alloc_fhandle(); if (data == NULL || mntfh == NULL) - goto out_free_fh; - - security_init_mnt_opts(&data->lsm_opts); + goto out; /* Validate the mount data */ error = nfs_validate_mount_data(raw_data, data, mntfh, dev_name); @@ -2234,8 +2245,6 @@ static struct dentry *nfs_fs_mount(struct file_system_type *fs_type, #ifdef CONFIG_NFS_V4 if (data->version == 4) { mntroot = nfs4_try_mount(flags, dev_name, data); - kfree(data->client_address); - kfree(data->nfs_server.export_path); goto out; } #endif /* CONFIG_NFS_V4 */ @@ -2290,13 +2299,8 @@ static struct dentry *nfs_fs_mount(struct file_system_type *fs_type, s->s_flags |= MS_ACTIVE; out: - kfree(data->nfs_server.hostname); - kfree(data->mount_server.hostname); - kfree(data->fscache_uniq); - security_free_mnt_opts(&data->lsm_opts); -out_free_fh: + nfs_free_parsed_mount_data(data); nfs_free_fhandle(mntfh); - kfree(data); return mntroot; out_err_nosb: @@ -2623,9 +2627,7 @@ nfs4_remote_mount(struct file_system_type *fs_type, int flags, mntfh = nfs_alloc_fhandle(); if (data == NULL || mntfh == NULL) - goto out_free_fh; - - security_init_mnt_opts(&data->lsm_opts); + goto out; /* Get a volume representation */ server = nfs4_create_server(data, mntfh); @@ -2677,13 +2679,10 @@ nfs4_remote_mount(struct file_system_type *fs_type, int flags, s->s_flags |= MS_ACTIVE; - security_free_mnt_opts(&data->lsm_opts); nfs_free_fhandle(mntfh); return mntroot; out: - security_free_mnt_opts(&data->lsm_opts); -out_free_fh: nfs_free_fhandle(mntfh); return ERR_PTR(error); @@ -2787,43 +2786,22 @@ static void nfs_referral_loop_unprotect(void) static struct dentry *nfs_follow_remote_path(struct vfsmount *root_mnt, const char *export_path) { - struct mnt_namespace *ns_private; - struct super_block *s; struct dentry *dentry; - struct path path; - int ret; - - ns_private = create_mnt_ns(root_mnt); - ret = PTR_ERR(ns_private); - if (IS_ERR(ns_private)) - goto out_mntput; + int err; - ret = nfs_referral_loop_protect(); - if (ret != 0) - goto out_put_mnt_ns; + if (IS_ERR(root_mnt)) + return ERR_CAST(root_mnt); - ret = vfs_path_lookup(root_mnt->mnt_root, root_mnt, - export_path, LOOKUP_FOLLOW|LOOKUP_AUTOMOUNT, &path); + err = nfs_referral_loop_protect(); + if (err) { + mntput(root_mnt); + return ERR_PTR(err); + } + dentry = mount_subtree(root_mnt, export_path); nfs_referral_loop_unprotect(); - put_mnt_ns(ns_private); - if (ret != 0) - goto out_err; - - s = path.mnt->mnt_sb; - atomic_inc(&s->s_active); - dentry = dget(path.dentry); - - path_put(&path); - down_write(&s->s_umount); return dentry; -out_put_mnt_ns: - put_mnt_ns(ns_private); -out_mntput: - mntput(root_mnt); -out_err: - return ERR_PTR(ret); } static struct dentry *nfs4_try_mount(int flags, const char *dev_name, @@ -2841,9 +2819,7 @@ static struct dentry *nfs4_try_mount(int flags, const char *dev_name, data->nfs_server.hostname); data->nfs_server.export_path = export_path; - res = ERR_CAST(root_mnt); - if (!IS_ERR(root_mnt)) - res = nfs_follow_remote_path(root_mnt, export_path); + res = nfs_follow_remote_path(root_mnt, export_path); dfprintk(MOUNT, "<-- nfs4_try_mount() = %ld%s\n", IS_ERR(res) ? PTR_ERR(res) : 0, @@ -2863,7 +2839,7 @@ static struct dentry *nfs4_mount(struct file_system_type *fs_type, data = nfs_alloc_parsed_mount_data(4); if (data == NULL) - goto out_free_data; + goto out; /* Validate the mount data */ error = nfs4_validate_mount_data(raw_data, data, dev_name); @@ -2877,12 +2853,7 @@ static struct dentry *nfs4_mount(struct file_system_type *fs_type, error = PTR_ERR(res); out: - kfree(data->client_address); - kfree(data->nfs_server.export_path); - kfree(data->nfs_server.hostname); - kfree(data->fscache_uniq); -out_free_data: - kfree(data); + nfs_free_parsed_mount_data(data); dprintk("<-- nfs4_mount() = %d%s\n", error, error != 0 ? " [error]" : ""); return res; @@ -3104,9 +3075,7 @@ static struct dentry *nfs4_referral_mount(struct file_system_type *fs_type, flags, data, data->hostname); data->mnt_path = export_path; - res = ERR_CAST(root_mnt); - if (!IS_ERR(root_mnt)) - res = nfs_follow_remote_path(root_mnt, export_path); + res = nfs_follow_remote_path(root_mnt, export_path); dprintk("<-- nfs4_referral_mount() = %ld%s\n", IS_ERR(res) ? PTR_ERR(res) : 0, IS_ERR(res) ? " [error]" : ""); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 1dda78db6a73..834f0fe96f89 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1052,7 +1052,7 @@ static const struct nfs_pageio_ops nfs_pageio_write_ops = { .pg_doio = nfs_generic_pg_writepages, }; -static void nfs_pageio_init_write_mds(struct nfs_pageio_descriptor *pgio, +void nfs_pageio_init_write_mds(struct nfs_pageio_descriptor *pgio, struct inode *inode, int ioflags) { nfs_pageio_init(pgio, inode, &nfs_pageio_write_ops, @@ -1166,13 +1166,7 @@ static void nfs_writeback_done_full(struct rpc_task *task, void *calldata) static void nfs_writeback_release_full(void *calldata) { struct nfs_write_data *data = calldata; - int ret, status = data->task.tk_status; - struct nfs_pageio_descriptor pgio; - - if (data->pnfs_error) { - nfs_pageio_init_write_mds(&pgio, data->inode, FLUSH_STABLE); - pgio.pg_recoalesce = 1; - } + int status = data->task.tk_status; /* Update attributes as result of writeback. */ while (!list_empty(&data->pages)) { @@ -1188,11 +1182,6 @@ static void nfs_writeback_release_full(void *calldata) req->wb_bytes, (long long)req_offset(req)); - if (data->pnfs_error) { - dprintk(", pnfs error = %d\n", data->pnfs_error); - goto next; - } - if (status < 0) { nfs_set_pageerror(page); nfs_context_set_write_error(req->wb_context, status); @@ -1212,19 +1201,7 @@ remove_request: next: nfs_clear_page_tag_locked(req); nfs_end_page_writeback(page); - if (data->pnfs_error) { - lock_page(page); - nfs_pageio_cond_complete(&pgio, page->index); - ret = nfs_page_async_flush(&pgio, page, 0); - if (ret) { - nfs_set_pageerror(page); - dprintk("rewrite to MDS error = %d\n", ret); - } - unlock_page(page); - } } - if (data->pnfs_error) - nfs_pageio_complete(&pgio); nfs_writedata_release(calldata); } @@ -1711,7 +1688,7 @@ out_error: #ifdef CONFIG_MIGRATION int nfs_migrate_page(struct address_space *mapping, struct page *newpage, - struct page *page) + struct page *page, enum migrate_mode mode) { /* * If PagePrivate is set, then the page is currently associated with @@ -1726,7 +1703,7 @@ int nfs_migrate_page(struct address_space *mapping, struct page *newpage, nfs_fscache_release_page(page, GFP_KERNEL); - return migrate_page(mapping, newpage, page); + return migrate_page(mapping, newpage, page, mode); } #endif diff --git a/fs/nfsd/Kconfig b/fs/nfsd/Kconfig index 10e6366608f2..8df1ea4a6ff9 100644 --- a/fs/nfsd/Kconfig +++ b/fs/nfsd/Kconfig @@ -80,3 +80,13 @@ config NFSD_V4 available from http://linux-nfs.org/. If unsure, say N. + +config NFSD_FAULT_INJECTION + bool "NFS server manual fault injection" + depends on NFSD_V4 && DEBUG_KERNEL + help + This option enables support for manually injecting faults + into the NFS server. This is intended to be used for + testing error recovery on the NFS client. + + If unsure, say N. diff --git a/fs/nfsd/Makefile b/fs/nfsd/Makefile index 9b118ee20193..af32ef06b4fe 100644 --- a/fs/nfsd/Makefile +++ b/fs/nfsd/Makefile @@ -6,6 +6,7 @@ obj-$(CONFIG_NFSD) += nfsd.o nfsd-y := nfssvc.o nfsctl.o nfsproc.o nfsfh.o vfs.o \ export.o auth.o lockd.o nfscache.o nfsxdr.o stats.o +nfsd-$(CONFIG_NFSD_FAULT_INJECTION) += fault_inject.o nfsd-$(CONFIG_NFSD_V2_ACL) += nfs2acl.o nfsd-$(CONFIG_NFSD_V3) += nfs3proc.o nfs3xdr.o nfsd-$(CONFIG_NFSD_V3_ACL) += nfs3acl.o diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index 62f3b9074e84..cf8a6bd062fa 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -87,7 +87,7 @@ static int expkey_parse(struct cache_detail *cd, char *mesg, int mlen) struct svc_expkey key; struct svc_expkey *ek = NULL; - if (mesg[mlen-1] != '\n') + if (mlen < 1 || mesg[mlen-1] != '\n') return -EINVAL; mesg[mlen-1] = 0; @@ -1226,12 +1226,12 @@ nfsd_export_init(void) int rv; dprintk("nfsd: initializing export module.\n"); - rv = cache_register(&svc_export_cache); + rv = cache_register_net(&svc_export_cache, &init_net); if (rv) return rv; - rv = cache_register(&svc_expkey_cache); + rv = cache_register_net(&svc_expkey_cache, &init_net); if (rv) - cache_unregister(&svc_export_cache); + cache_unregister_net(&svc_export_cache, &init_net); return rv; } @@ -1255,8 +1255,8 @@ nfsd_export_shutdown(void) dprintk("nfsd: shutting down export module.\n"); - cache_unregister(&svc_expkey_cache); - cache_unregister(&svc_export_cache); + cache_unregister_net(&svc_expkey_cache, &init_net); + cache_unregister_net(&svc_export_cache, &init_net); svcauth_unix_purge(); dprintk("nfsd: export shutdown complete.\n"); diff --git a/fs/nfsd/fault_inject.c b/fs/nfsd/fault_inject.c new file mode 100644 index 000000000000..ce7f0758d84c --- /dev/null +++ b/fs/nfsd/fault_inject.c @@ -0,0 +1,91 @@ +/* + * Copyright (c) 2011 Bryan Schumaker <bjschuma@netapp.com> + * + * Uses debugfs to create fault injection points for client testing + */ + +#include <linux/types.h> +#include <linux/fs.h> +#include <linux/debugfs.h> +#include <linux/module.h> + +#include "state.h" +#include "fault_inject.h" + +struct nfsd_fault_inject_op { + char *file; + void (*func)(u64); +}; + +static struct nfsd_fault_inject_op inject_ops[] = { + { + .file = "forget_clients", + .func = nfsd_forget_clients, + }, + { + .file = "forget_locks", + .func = nfsd_forget_locks, + }, + { + .file = "forget_openowners", + .func = nfsd_forget_openowners, + }, + { + .file = "forget_delegations", + .func = nfsd_forget_delegations, + }, + { + .file = "recall_delegations", + .func = nfsd_recall_delegations, + }, +}; + +static long int NUM_INJECT_OPS = sizeof(inject_ops) / sizeof(struct nfsd_fault_inject_op); +static struct dentry *debug_dir; + +static int nfsd_inject_set(void *op_ptr, u64 val) +{ + struct nfsd_fault_inject_op *op = op_ptr; + + if (val == 0) + printk(KERN_INFO "NFSD Fault Injection: %s (all)", op->file); + else + printk(KERN_INFO "NFSD Fault Injection: %s (n = %llu)", op->file, val); + + op->func(val); + return 0; +} + +static int nfsd_inject_get(void *data, u64 *val) +{ + return 0; +} + +DEFINE_SIMPLE_ATTRIBUTE(fops_nfsd, nfsd_inject_get, nfsd_inject_set, "%llu\n"); + +void nfsd_fault_inject_cleanup(void) +{ + debugfs_remove_recursive(debug_dir); +} + +int nfsd_fault_inject_init(void) +{ + unsigned int i; + struct nfsd_fault_inject_op *op; + mode_t mode = S_IFREG | S_IRUSR | S_IWUSR; + + debug_dir = debugfs_create_dir("nfsd", NULL); + if (!debug_dir) + goto fail; + + for (i = 0; i < NUM_INJECT_OPS; i++) { + op = &inject_ops[i]; + if (!debugfs_create_file(op->file, mode, debug_dir, op, &fops_nfsd)) + goto fail; + } + return 0; + +fail: + nfsd_fault_inject_cleanup(); + return -ENOMEM; +} diff --git a/fs/nfsd/fault_inject.h b/fs/nfsd/fault_inject.h new file mode 100644 index 000000000000..90bd0570956c --- /dev/null +++ b/fs/nfsd/fault_inject.h @@ -0,0 +1,28 @@ +/* + * Copyright (c) 2011 Bryan Schumaker <bjschuma@netapp.com> + * + * Function definitions for fault injection + */ + +#ifndef LINUX_NFSD_FAULT_INJECT_H +#define LINUX_NFSD_FAULT_INJECT_H + +#ifdef CONFIG_NFSD_FAULT_INJECTION +int nfsd_fault_inject_init(void); +void nfsd_fault_inject_cleanup(void); +void nfsd_forget_clients(u64); +void nfsd_forget_locks(u64); +void nfsd_forget_openowners(u64); +void nfsd_forget_delegations(u64); +void nfsd_recall_delegations(u64); +#else /* CONFIG_NFSD_FAULT_INJECTION */ +static inline int nfsd_fault_inject_init(void) { return 0; } +static inline void nfsd_fault_inject_cleanup(void) {} +static inline void nfsd_forget_clients(u64 num) {} +static inline void nfsd_forget_locks(u64 num) {} +static inline void nfsd_forget_openowners(u64 num) {} +static inline void nfsd_forget_delegations(u64 num) {} +static inline void nfsd_recall_delegations(u64 num) {} +#endif /* CONFIG_NFSD_FAULT_INJECTION */ + +#endif /* LINUX_NFSD_FAULT_INJECT_H */ diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index 7748d6a18d97..6f3ebb48b12f 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -718,7 +718,7 @@ int set_callback_cred(void) { if (callback_cred) return 0; - callback_cred = rpc_lookup_machine_cred(); + callback_cred = rpc_lookup_machine_cred("nfs"); if (!callback_cred) return -ENOMEM; return 0; diff --git a/fs/nfsd/nfs4idmap.c b/fs/nfsd/nfs4idmap.c index 55780a22fdbd..94096273cd6c 100644 --- a/fs/nfsd/nfs4idmap.c +++ b/fs/nfsd/nfs4idmap.c @@ -36,6 +36,7 @@ #include <linux/seq_file.h> #include <linux/sched.h> #include <linux/slab.h> +#include <net/net_namespace.h> #include "idmap.h" #include "nfsd.h" @@ -466,20 +467,20 @@ nfsd_idmap_init(void) { int rv; - rv = cache_register(&idtoname_cache); + rv = cache_register_net(&idtoname_cache, &init_net); if (rv) return rv; - rv = cache_register(&nametoid_cache); + rv = cache_register_net(&nametoid_cache, &init_net); if (rv) - cache_unregister(&idtoname_cache); + cache_unregister_net(&idtoname_cache, &init_net); return rv; } void nfsd_idmap_shutdown(void) { - cache_unregister(&idtoname_cache); - cache_unregister(&nametoid_cache); + cache_unregister_net(&idtoname_cache, &init_net); + cache_unregister_net(&nametoid_cache, &init_net); } static int diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index fa383361bc61..896da74ec563 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -266,10 +266,6 @@ do_open_fhandle(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_ { __be32 status; - /* Only reclaims from previously confirmed clients are valid */ - if ((status = nfs4_check_open_reclaim(&open->op_clientid))) - return status; - /* We don't know the target directory, and therefore can not * set the change info */ @@ -373,6 +369,9 @@ nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, break; case NFS4_OPEN_CLAIM_PREVIOUS: open->op_openowner->oo_flags |= NFS4_OO_CONFIRMED; + status = nfs4_check_open_reclaim(&open->op_clientid); + if (status) + goto out; case NFS4_OPEN_CLAIM_FH: case NFS4_OPEN_CLAIM_DELEG_CUR_FH: status = do_open_fhandle(rqstp, &cstate->current_fh, @@ -838,7 +837,7 @@ nfsd4_setattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, return status; } } - status = mnt_want_write(cstate->current_fh.fh_export->ex_path.mnt); + status = fh_want_write(&cstate->current_fh); if (status) return status; status = nfs_ok; @@ -856,7 +855,7 @@ nfsd4_setattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, status = nfsd_setattr(rqstp, &cstate->current_fh, &setattr->sa_iattr, 0, (time_t)0); out: - mnt_drop_write(cstate->current_fh.fh_export->ex_path.mnt); + fh_drop_write(&cstate->current_fh); return status; } diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index ed083b9a731b..0b3e875d1abd 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -117,8 +117,7 @@ out_no_tfm: return status; } -int -nfsd4_create_clid_dir(struct nfs4_client *clp) +void nfsd4_create_clid_dir(struct nfs4_client *clp) { const struct cred *original_cred; char *dname = clp->cl_recdir; @@ -127,13 +126,14 @@ nfsd4_create_clid_dir(struct nfs4_client *clp) dprintk("NFSD: nfsd4_create_clid_dir for \"%s\"\n", dname); - if (!rec_file || clp->cl_firststate) - return 0; - + if (clp->cl_firststate) + return; clp->cl_firststate = 1; + if (!rec_file) + return; status = nfs4_save_creds(&original_cred); if (status < 0) - return status; + return; dir = rec_file->f_path.dentry; /* lock the parent */ @@ -144,14 +144,21 @@ nfsd4_create_clid_dir(struct nfs4_client *clp) status = PTR_ERR(dentry); goto out_unlock; } - status = -EEXIST; if (dentry->d_inode) + /* + * In the 4.1 case, where we're called from + * reclaim_complete(), records from the previous reboot + * may still be left, so this is OK. + * + * In the 4.0 case, we should never get here; but we may + * as well be forgiving and just succeed silently. + */ goto out_put; - status = mnt_want_write(rec_file->f_path.mnt); + status = mnt_want_write_file(rec_file); if (status) goto out_put; status = vfs_mkdir(dir->d_inode, dentry, S_IRWXU); - mnt_drop_write(rec_file->f_path.mnt); + mnt_drop_write_file(rec_file); out_put: dput(dentry); out_unlock: @@ -164,7 +171,6 @@ out_unlock: " and is writeable", status, user_recovery_dirname); nfs4_reset_creds(original_cred); - return status; } typedef int (recdir_func)(struct dentry *, struct dentry *); @@ -268,7 +274,7 @@ nfsd4_remove_clid_dir(struct nfs4_client *clp) if (!rec_file || !clp->cl_firststate) return; - status = mnt_want_write(rec_file->f_path.mnt); + status = mnt_want_write_file(rec_file); if (status) goto out; clp->cl_firststate = 0; @@ -281,7 +287,7 @@ nfsd4_remove_clid_dir(struct nfs4_client *clp) nfs4_reset_creds(original_cred); if (status == 0) vfs_fsync(rec_file, 0); - mnt_drop_write(rec_file->f_path.mnt); + mnt_drop_write_file(rec_file); out: if (status) printk("NFSD: Failed to remove expired client state directory" @@ -311,13 +317,13 @@ nfsd4_recdir_purge_old(void) { if (!rec_file) return; - status = mnt_want_write(rec_file->f_path.mnt); + status = mnt_want_write_file(rec_file); if (status) goto out; status = nfsd4_list_rec_dir(purge_old); if (status == 0) vfs_fsync(rec_file, 0); - mnt_drop_write(rec_file->f_path.mnt); + mnt_drop_write_file(rec_file); out: if (status) printk("nfsd4: failed to purge old clients from recovery" diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 47e94e33a975..e8c98f009670 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -49,12 +49,20 @@ time_t nfsd4_lease = 90; /* default lease time */ time_t nfsd4_grace = 90; static time_t boot_time; -static stateid_t zerostateid; /* bits all 0 */ -static stateid_t onestateid; /* bits all 1 */ + +#define all_ones {{~0,~0},~0} +static const stateid_t one_stateid = { + .si_generation = ~0, + .si_opaque = all_ones, +}; +static const stateid_t zero_stateid = { + /* all fields zero */ +}; + static u64 current_sessionid = 1; -#define ZERO_STATEID(stateid) (!memcmp((stateid), &zerostateid, sizeof(stateid_t))) -#define ONE_STATEID(stateid) (!memcmp((stateid), &onestateid, sizeof(stateid_t))) +#define ZERO_STATEID(stateid) (!memcmp((stateid), &zero_stateid, sizeof(stateid_t))) +#define ONE_STATEID(stateid) (!memcmp((stateid), &one_stateid, sizeof(stateid_t))) /* forward declarations */ static int check_for_locks(struct nfs4_file *filp, struct nfs4_lockowner *lowner); @@ -133,21 +141,21 @@ unsigned int max_delegations; * Open owner state (share locks) */ -/* hash tables for open owners */ -#define OPEN_OWNER_HASH_BITS 8 -#define OPEN_OWNER_HASH_SIZE (1 << OPEN_OWNER_HASH_BITS) -#define OPEN_OWNER_HASH_MASK (OPEN_OWNER_HASH_SIZE - 1) +/* hash tables for lock and open owners */ +#define OWNER_HASH_BITS 8 +#define OWNER_HASH_SIZE (1 << OWNER_HASH_BITS) +#define OWNER_HASH_MASK (OWNER_HASH_SIZE - 1) -static unsigned int open_ownerstr_hashval(u32 clientid, struct xdr_netobj *ownername) +static unsigned int ownerstr_hashval(u32 clientid, struct xdr_netobj *ownername) { unsigned int ret; ret = opaque_hashval(ownername->data, ownername->len); ret += clientid; - return ret & OPEN_OWNER_HASH_MASK; + return ret & OWNER_HASH_MASK; } -static struct list_head open_ownerstr_hashtbl[OPEN_OWNER_HASH_SIZE]; +static struct list_head ownerstr_hashtbl[OWNER_HASH_SIZE]; /* hash table for nfs4_file */ #define FILE_HASH_BITS 8 @@ -514,6 +522,7 @@ static void unhash_lockowner(struct nfs4_lockowner *lo) list_del(&lo->lo_owner.so_strhash); list_del(&lo->lo_perstateid); + list_del(&lo->lo_owner_ino_hash); while (!list_empty(&lo->lo_owner.so_stateids)) { stp = list_first_entry(&lo->lo_owner.so_stateids, struct nfs4_ol_stateid, st_perstateowner); @@ -658,7 +667,7 @@ static int nfsd4_sanitize_slot_size(u32 size) /* * XXX: If we run out of reserved DRC memory we could (up to a point) * re-negotiate active sessions and reduce their slot usage to make - * rooom for new connections. For now we just fail the create session. + * room for new connections. For now we just fail the create session. */ static int nfsd4_get_drc_mem(int slotsize, u32 num) { @@ -985,12 +994,11 @@ static struct nfs4_client *alloc_client(struct xdr_netobj name) clp = kzalloc(sizeof(struct nfs4_client), GFP_KERNEL); if (clp == NULL) return NULL; - clp->cl_name.data = kmalloc(name.len, GFP_KERNEL); + clp->cl_name.data = kmemdup(name.data, name.len, GFP_KERNEL); if (clp->cl_name.data == NULL) { kfree(clp); return NULL; } - memcpy(clp->cl_name.data, name.data, name.len); clp->cl_name.len = name.len; return clp; } @@ -1058,7 +1066,6 @@ expire_client(struct nfs4_client *clp) spin_unlock(&recall_lock); while (!list_empty(&reaplist)) { dp = list_entry(reaplist.next, struct nfs4_delegation, dl_recall_lru); - list_del_init(&dp->dl_recall_lru); unhash_delegation(dp); } while (!list_empty(&clp->cl_openowners)) { @@ -2301,7 +2308,7 @@ nfsd4_free_slabs(void) nfsd4_free_slab(&deleg_slab); } -static int +int nfsd4_init_slabs(void) { openowner_slab = kmem_cache_create("nfsd4_openowners", @@ -2373,7 +2380,7 @@ static inline void *alloc_stateowner(struct kmem_cache *slab, struct xdr_netobj static void hash_openowner(struct nfs4_openowner *oo, struct nfs4_client *clp, unsigned int strhashval) { - list_add(&oo->oo_owner.so_strhash, &open_ownerstr_hashtbl[strhashval]); + list_add(&oo->oo_owner.so_strhash, &ownerstr_hashtbl[strhashval]); list_add(&oo->oo_perclient, &clp->cl_openowners); } @@ -2436,7 +2443,9 @@ find_openstateowner_str(unsigned int hashval, struct nfsd4_open *open) struct nfs4_stateowner *so; struct nfs4_openowner *oo; - list_for_each_entry(so, &open_ownerstr_hashtbl[hashval], so_strhash) { + list_for_each_entry(so, &ownerstr_hashtbl[hashval], so_strhash) { + if (!so->so_is_open_owner) + continue; if (same_owner_str(so, &open->op_owner, &open->op_clientid)) { oo = openowner(so); renew_client(oo->oo_owner.so_client); @@ -2580,7 +2589,7 @@ nfsd4_process_open1(struct nfsd4_compound_state *cstate, if (open->op_file == NULL) return nfserr_jukebox; - strhashval = open_ownerstr_hashval(clientid->cl_id, &open->op_owner); + strhashval = ownerstr_hashval(clientid->cl_id, &open->op_owner); oo = find_openstateowner_str(strhashval, open); open->op_openowner = oo; if (!oo) { @@ -3123,7 +3132,6 @@ nfs4_laundromat(void) spin_unlock(&recall_lock); list_for_each_safe(pos, next, &reaplist) { dp = list_entry (pos, struct nfs4_delegation, dl_recall_lru); - list_del_init(&dp->dl_recall_lru); unhash_delegation(dp); } test_val = nfsd4_lease; @@ -3718,13 +3726,11 @@ out: } -/* - * Lock owner state (byte-range locks) - */ #define LOFF_OVERFLOW(start, len) ((u64)(len) > ~(u64)(start)) -#define LOCK_HASH_BITS 8 -#define LOCK_HASH_SIZE (1 << LOCK_HASH_BITS) -#define LOCK_HASH_MASK (LOCK_HASH_SIZE - 1) + +#define LOCKOWNER_INO_HASH_BITS 8 +#define LOCKOWNER_INO_HASH_SIZE (1 << LOCKOWNER_INO_HASH_BITS) +#define LOCKOWNER_INO_HASH_MASK (LOCKOWNER_INO_HASH_SIZE - 1) static inline u64 end_offset(u64 start, u64 len) @@ -3746,16 +3752,14 @@ last_byte_offset(u64 start, u64 len) return end > start ? end - 1: NFS4_MAX_UINT64; } -static inline unsigned int -lock_ownerstr_hashval(struct inode *inode, u32 cl_id, - struct xdr_netobj *ownername) +static unsigned int lockowner_ino_hashval(struct inode *inode, u32 cl_id, struct xdr_netobj *ownername) { return (file_hashval(inode) + cl_id + opaque_hashval(ownername->data, ownername->len)) - & LOCK_HASH_MASK; + & LOCKOWNER_INO_HASH_MASK; } -static struct list_head lock_ownerstr_hashtbl[LOCK_HASH_SIZE]; +static struct list_head lockowner_ino_hashtbl[LOCKOWNER_INO_HASH_SIZE]; /* * TODO: Linux file offsets are _signed_ 64-bit quantities, which means that @@ -3809,23 +3813,39 @@ nevermind: deny->ld_type = NFS4_WRITE_LT; } +static bool same_lockowner_ino(struct nfs4_lockowner *lo, struct inode *inode, clientid_t *clid, struct xdr_netobj *owner) +{ + struct nfs4_ol_stateid *lst; + + if (!same_owner_str(&lo->lo_owner, owner, clid)) + return false; + lst = list_first_entry(&lo->lo_owner.so_stateids, + struct nfs4_ol_stateid, st_perstateowner); + return lst->st_file->fi_inode == inode; +} + static struct nfs4_lockowner * find_lockowner_str(struct inode *inode, clientid_t *clid, struct xdr_netobj *owner) { - unsigned int hashval = lock_ownerstr_hashval(inode, clid->cl_id, owner); - struct nfs4_stateowner *op; + unsigned int hashval = lockowner_ino_hashval(inode, clid->cl_id, owner); + struct nfs4_lockowner *lo; - list_for_each_entry(op, &lock_ownerstr_hashtbl[hashval], so_strhash) { - if (same_owner_str(op, owner, clid)) - return lockowner(op); + list_for_each_entry(lo, &lockowner_ino_hashtbl[hashval], lo_owner_ino_hash) { + if (same_lockowner_ino(lo, inode, clid, owner)) + return lo; } return NULL; } static void hash_lockowner(struct nfs4_lockowner *lo, unsigned int strhashval, struct nfs4_client *clp, struct nfs4_ol_stateid *open_stp) { - list_add(&lo->lo_owner.so_strhash, &lock_ownerstr_hashtbl[strhashval]); + struct inode *inode = open_stp->st_file->fi_inode; + unsigned int inohash = lockowner_ino_hashval(inode, + clp->cl_clientid.cl_id, &lo->lo_owner.so_owner); + + list_add(&lo->lo_owner.so_strhash, &ownerstr_hashtbl[strhashval]); + list_add(&lo->lo_owner_ino_hash, &lockowner_ino_hashtbl[inohash]); list_add(&lo->lo_perstateid, &open_stp->st_lockowners); } @@ -3834,7 +3854,7 @@ static void hash_lockowner(struct nfs4_lockowner *lo, unsigned int strhashval, s * Called in nfsd4_lock - therefore, OPEN and OPEN_CONFIRM (if needed) has * occurred. * - * strhashval = lock_ownerstr_hashval + * strhashval = ownerstr_hashval */ static struct nfs4_lockowner * @@ -3892,6 +3912,37 @@ static void get_lock_access(struct nfs4_ol_stateid *lock_stp, u32 access) __set_bit(access, &lock_stp->st_access_bmap); } +__be32 lookup_or_create_lock_state(struct nfsd4_compound_state *cstate, struct nfs4_ol_stateid *ost, struct nfsd4_lock *lock, struct nfs4_ol_stateid **lst, bool *new) +{ + struct nfs4_file *fi = ost->st_file; + struct nfs4_openowner *oo = openowner(ost->st_stateowner); + struct nfs4_client *cl = oo->oo_owner.so_client; + struct nfs4_lockowner *lo; + unsigned int strhashval; + + lo = find_lockowner_str(fi->fi_inode, &cl->cl_clientid, &lock->v.new.owner); + if (lo) { + if (!cstate->minorversion) + return nfserr_bad_seqid; + /* XXX: a lockowner always has exactly one stateid: */ + *lst = list_first_entry(&lo->lo_owner.so_stateids, + struct nfs4_ol_stateid, st_perstateowner); + return nfs_ok; + } + strhashval = ownerstr_hashval(cl->cl_clientid.cl_id, + &lock->v.new.owner); + lo = alloc_init_lock_stateowner(strhashval, cl, ost, lock); + if (lo == NULL) + return nfserr_jukebox; + *lst = alloc_init_lock_stateid(lo, fi, ost); + if (*lst == NULL) { + release_lockowner(lo); + return nfserr_jukebox; + } + *new = true; + return nfs_ok; +} + /* * LOCK operation */ @@ -3907,7 +3958,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct file_lock file_lock; struct file_lock conflock; __be32 status = 0; - unsigned int strhashval; + bool new_state = false; int lkflg; int err; @@ -3933,10 +3984,15 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, * lock stateid. */ struct nfs4_ol_stateid *open_stp = NULL; - + + if (nfsd4_has_session(cstate)) + /* See rfc 5661 18.10.3: given clientid is ignored: */ + memcpy(&lock->v.new.clientid, + &cstate->session->se_client->cl_clientid, + sizeof(clientid_t)); + status = nfserr_stale_clientid; - if (!nfsd4_has_session(cstate) && - STALE_CLIENTID(&lock->lk_new_clientid)) + if (STALE_CLIENTID(&lock->lk_new_clientid)) goto out; /* validate and update open stateid and open seqid */ @@ -3948,25 +4004,12 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, goto out; open_sop = openowner(open_stp->st_stateowner); status = nfserr_bad_stateid; - if (!nfsd4_has_session(cstate) && - !same_clid(&open_sop->oo_owner.so_client->cl_clientid, + if (!same_clid(&open_sop->oo_owner.so_client->cl_clientid, &lock->v.new.clientid)) goto out; - /* create lockowner and lock stateid */ - fp = open_stp->st_file; - strhashval = lock_ownerstr_hashval(fp->fi_inode, - open_sop->oo_owner.so_client->cl_clientid.cl_id, - &lock->v.new.owner); - /* XXX: Do we need to check for duplicate stateowners on - * the same file, or should they just be allowed (and - * create new stateids)? */ - status = nfserr_jukebox; - lock_sop = alloc_init_lock_stateowner(strhashval, - open_sop->oo_owner.so_client, open_stp, lock); - if (lock_sop == NULL) - goto out; - lock_stp = alloc_init_lock_stateid(lock_sop, fp, open_stp); - if (lock_stp == NULL) + status = lookup_or_create_lock_state(cstate, open_stp, lock, + &lock_stp, &new_state); + if (status) goto out; } else { /* lock (lock owner + lock stateid) already exists */ @@ -3976,10 +4019,9 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, NFS4_LOCK_STID, &lock_stp); if (status) goto out; - lock_sop = lockowner(lock_stp->st_stateowner); - fp = lock_stp->st_file; } - /* lock_sop and lock_stp have been created or found */ + lock_sop = lockowner(lock_stp->st_stateowner); + fp = lock_stp->st_file; lkflg = setlkflg(lock->lk_type); status = nfs4_check_openmode(lock_stp, lkflg); @@ -4054,7 +4096,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, break; } out: - if (status && lock->lk_is_new && lock_sop) + if (status && new_state) release_lockowner(lock_sop); if (!cstate->replay_owner) nfs4_unlock_state(); @@ -4251,7 +4293,7 @@ nfsd4_release_lockowner(struct svc_rqst *rqstp, struct nfs4_ol_stateid *stp; struct xdr_netobj *owner = &rlockowner->rl_owner; struct list_head matches; - int i; + unsigned int hashval = ownerstr_hashval(clid->cl_id, owner); __be32 status; dprintk("nfsd4_release_lockowner clientid: (%08x/%08x):\n", @@ -4266,22 +4308,19 @@ nfsd4_release_lockowner(struct svc_rqst *rqstp, nfs4_lock_state(); status = nfserr_locks_held; - /* XXX: we're doing a linear search through all the lockowners. - * Yipes! For now we'll just hope clients aren't really using - * release_lockowner much, but eventually we have to fix these - * data structures. */ INIT_LIST_HEAD(&matches); - for (i = 0; i < LOCK_HASH_SIZE; i++) { - list_for_each_entry(sop, &lock_ownerstr_hashtbl[i], so_strhash) { - if (!same_owner_str(sop, owner, clid)) - continue; - list_for_each_entry(stp, &sop->so_stateids, - st_perstateowner) { - lo = lockowner(sop); - if (check_for_locks(stp->st_file, lo)) - goto out; - list_add(&lo->lo_list, &matches); - } + + list_for_each_entry(sop, &ownerstr_hashtbl[hashval], so_strhash) { + if (sop->so_is_open_owner) + continue; + if (!same_owner_str(sop, owner, clid)) + continue; + list_for_each_entry(stp, &sop->so_stateids, + st_perstateowner) { + lo = lockowner(sop); + if (check_for_locks(stp->st_file, lo)) + goto out; + list_add(&lo->lo_list, &matches); } } /* Clients probably won't expect us to return with some (but not all) @@ -4394,16 +4433,127 @@ nfs4_check_open_reclaim(clientid_t *clid) return nfs4_find_reclaim_client(clid) ? nfs_ok : nfserr_reclaim_bad; } +#ifdef CONFIG_NFSD_FAULT_INJECTION + +void nfsd_forget_clients(u64 num) +{ + struct nfs4_client *clp, *next; + int count = 0; + + nfs4_lock_state(); + list_for_each_entry_safe(clp, next, &client_lru, cl_lru) { + nfsd4_remove_clid_dir(clp); + expire_client(clp); + if (++count == num) + break; + } + nfs4_unlock_state(); + + printk(KERN_INFO "NFSD: Forgot %d clients", count); +} + +static void release_lockowner_sop(struct nfs4_stateowner *sop) +{ + release_lockowner(lockowner(sop)); +} + +static void release_openowner_sop(struct nfs4_stateowner *sop) +{ + release_openowner(openowner(sop)); +} + +static int nfsd_release_n_owners(u64 num, bool is_open_owner, + void (*release_sop)(struct nfs4_stateowner *)) +{ + int i, count = 0; + struct nfs4_stateowner *sop, *next; + + for (i = 0; i < OWNER_HASH_SIZE; i++) { + list_for_each_entry_safe(sop, next, &ownerstr_hashtbl[i], so_strhash) { + if (sop->so_is_open_owner != is_open_owner) + continue; + release_sop(sop); + if (++count == num) + return count; + } + } + return count; +} + +void nfsd_forget_locks(u64 num) +{ + int count; + + nfs4_lock_state(); + count = nfsd_release_n_owners(num, false, release_lockowner_sop); + nfs4_unlock_state(); + + printk(KERN_INFO "NFSD: Forgot %d locks", count); +} + +void nfsd_forget_openowners(u64 num) +{ + int count; + + nfs4_lock_state(); + count = nfsd_release_n_owners(num, true, release_openowner_sop); + nfs4_unlock_state(); + + printk(KERN_INFO "NFSD: Forgot %d open owners", count); +} + +int nfsd_process_n_delegations(u64 num, void (*deleg_func)(struct nfs4_delegation *)) +{ + int i, count = 0; + struct nfs4_file *fp, *fnext; + struct nfs4_delegation *dp, *dnext; + + for (i = 0; i < FILE_HASH_SIZE; i++) { + list_for_each_entry_safe(fp, fnext, &file_hashtbl[i], fi_hash) { + list_for_each_entry_safe(dp, dnext, &fp->fi_delegations, dl_perfile) { + deleg_func(dp); + if (++count == num) + return count; + } + } + } + + return count; +} + +void nfsd_forget_delegations(u64 num) +{ + unsigned int count; + + nfs4_lock_state(); + count = nfsd_process_n_delegations(num, unhash_delegation); + nfs4_unlock_state(); + + printk(KERN_INFO "NFSD: Forgot %d delegations", count); +} + +void nfsd_recall_delegations(u64 num) +{ + unsigned int count; + + nfs4_lock_state(); + spin_lock(&recall_lock); + count = nfsd_process_n_delegations(num, nfsd_break_one_deleg); + spin_unlock(&recall_lock); + nfs4_unlock_state(); + + printk(KERN_INFO "NFSD: Recalled %d delegations", count); +} + +#endif /* CONFIG_NFSD_FAULT_INJECTION */ + /* initialization to perform at module load time: */ -int +void nfs4_state_init(void) { - int i, status; + int i; - status = nfsd4_init_slabs(); - if (status) - return status; for (i = 0; i < CLIENT_HASH_SIZE; i++) { INIT_LIST_HEAD(&conf_id_hashtbl[i]); INIT_LIST_HEAD(&conf_str_hashtbl[i]); @@ -4416,18 +4566,15 @@ nfs4_state_init(void) for (i = 0; i < FILE_HASH_SIZE; i++) { INIT_LIST_HEAD(&file_hashtbl[i]); } - for (i = 0; i < OPEN_OWNER_HASH_SIZE; i++) { - INIT_LIST_HEAD(&open_ownerstr_hashtbl[i]); - } - for (i = 0; i < LOCK_HASH_SIZE; i++) { - INIT_LIST_HEAD(&lock_ownerstr_hashtbl[i]); + for (i = 0; i < OWNER_HASH_SIZE; i++) { + INIT_LIST_HEAD(&ownerstr_hashtbl[i]); } - memset(&onestateid, ~0, sizeof(stateid_t)); + for (i = 0; i < LOCKOWNER_INO_HASH_SIZE; i++) + INIT_LIST_HEAD(&lockowner_ino_hashtbl[i]); INIT_LIST_HEAD(&close_lru); INIT_LIST_HEAD(&client_lru); INIT_LIST_HEAD(&del_recall_lru); reclaim_str_hashtbl_size = 0; - return 0; } static void @@ -4526,7 +4673,6 @@ __nfs4_state_shutdown(void) spin_unlock(&recall_lock); list_for_each_safe(pos, next, &reaplist) { dp = list_entry (pos, struct nfs4_delegation, dl_recall_lru); - list_del_init(&dp->dl_recall_lru); unhash_delegation(dp); } diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index b6fa792d6b85..0ec5a1b9700e 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -215,10 +215,9 @@ defer_free(struct nfsd4_compoundargs *argp, static char *savemem(struct nfsd4_compoundargs *argp, __be32 *p, int nbytes) { if (p == argp->tmp) { - p = kmalloc(nbytes, GFP_KERNEL); + p = kmemdup(argp->tmp, nbytes, GFP_KERNEL); if (!p) return NULL; - memcpy(p, argp->tmp, nbytes); } else { BUG_ON(p != argp->tmpp); argp->tmpp = NULL; diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index c45a2ea4a090..748eda93ce59 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -18,6 +18,7 @@ #include "idmap.h" #include "nfsd.h" #include "cache.h" +#include "fault_inject.h" /* * We have a single directory with several nodes in it. @@ -272,7 +273,7 @@ static ssize_t write_unlock_fs(struct file *file, char *buf, size_t size) * 2. Is that directory a mount point, or * 3. Is that directory the root of an exported file system? */ - error = nlmsvc_unlock_all_by_sb(path.mnt->mnt_sb); + error = nlmsvc_unlock_all_by_sb(path.dentry->d_sb); path_put(&path); return error; @@ -1128,9 +1129,13 @@ static int __init init_nfsd(void) int retval; printk(KERN_INFO "Installing knfsd (copyright (C) 1996 okir@monad.swb.de).\n"); - retval = nfs4_state_init(); /* nfs4 locking state */ + retval = nfsd4_init_slabs(); if (retval) return retval; + nfs4_state_init(); + retval = nfsd_fault_inject_init(); /* nfsd fault injection controls */ + if (retval) + goto out_free_slabs; nfsd_stat_init(); /* Statistics */ retval = nfsd_reply_cache_init(); if (retval) @@ -1161,6 +1166,8 @@ out_free_cache: nfsd_reply_cache_shutdown(); out_free_stat: nfsd_stat_shutdown(); + nfsd_fault_inject_cleanup(); +out_free_slabs: nfsd4_free_slabs(); return retval; } @@ -1175,6 +1182,7 @@ static void __exit exit_nfsd(void) nfsd_lockd_shutdown(); nfsd_idmap_shutdown(); nfsd4_free_slabs(); + nfsd_fault_inject_cleanup(); unregister_filesystem(&nfsd_fs_type); } diff --git a/fs/nfsd/nfsd.h b/fs/nfsd/nfsd.h index 58134a23fdfb..1d1e8589b4ce 100644 --- a/fs/nfsd/nfsd.h +++ b/fs/nfsd/nfsd.h @@ -104,14 +104,16 @@ static inline int nfsd_v4client(struct svc_rqst *rq) */ #ifdef CONFIG_NFSD_V4 extern unsigned int max_delegations; -int nfs4_state_init(void); +void nfs4_state_init(void); +int nfsd4_init_slabs(void); void nfsd4_free_slabs(void); int nfs4_state_start(void); void nfs4_state_shutdown(void); void nfs4_reset_lease(time_t leasetime); int nfs4_reset_recoverydir(char *recdir); #else -static inline int nfs4_state_init(void) { return 0; } +static inline void nfs4_state_init(void) { } +static inline int nfsd4_init_slabs(void) { return 0; } static inline void nfsd4_free_slabs(void) { } static inline int nfs4_state_start(void) { return 0; } static inline void nfs4_state_shutdown(void) { } @@ -338,15 +340,15 @@ static inline u32 nfsd_suppattrs2(u32 minorversion) } /* These will return ERR_INVAL if specified in GETATTR or READDIR. */ -#define NFSD_WRITEONLY_ATTRS_WORD1 \ -(FATTR4_WORD1_TIME_ACCESS_SET | FATTR4_WORD1_TIME_MODIFY_SET) +#define NFSD_WRITEONLY_ATTRS_WORD1 \ + (FATTR4_WORD1_TIME_ACCESS_SET | FATTR4_WORD1_TIME_MODIFY_SET) /* These are the only attrs allowed in CREATE/OPEN/SETATTR. */ -#define NFSD_WRITEABLE_ATTRS_WORD0 \ -(FATTR4_WORD0_SIZE | FATTR4_WORD0_ACL ) -#define NFSD_WRITEABLE_ATTRS_WORD1 \ -(FATTR4_WORD1_MODE | FATTR4_WORD1_OWNER | FATTR4_WORD1_OWNER_GROUP \ - | FATTR4_WORD1_TIME_ACCESS_SET | FATTR4_WORD1_TIME_MODIFY_SET) +#define NFSD_WRITEABLE_ATTRS_WORD0 \ + (FATTR4_WORD0_SIZE | FATTR4_WORD0_ACL) +#define NFSD_WRITEABLE_ATTRS_WORD1 \ + (FATTR4_WORD1_MODE | FATTR4_WORD1_OWNER | FATTR4_WORD1_OWNER_GROUP \ + | FATTR4_WORD1_TIME_ACCESS_SET | FATTR4_WORD1_TIME_MODIFY_SET) #define NFSD_WRITEABLE_ATTRS_WORD2 0 #define NFSD_SUPPATTR_EXCLCREAT_WORD0 \ diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index c763de5c1157..68454e75fce9 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -59,7 +59,7 @@ static int nfsd_acceptable(void *expv, struct dentry *dentry) * the write call). */ static inline __be32 -nfsd_mode_check(struct svc_rqst *rqstp, umode_t mode, int requested) +nfsd_mode_check(struct svc_rqst *rqstp, umode_t mode, umode_t requested) { mode &= S_IFMT; @@ -293,7 +293,7 @@ out: * include/linux/nfsd/nfsd.h. */ __be32 -fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access) +fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, umode_t type, int access) { struct svc_export *exp; struct dentry *dentry; diff --git a/fs/nfsd/nfsfh.h b/fs/nfsd/nfsfh.h index c16f8d8331b5..e5e6707ba687 100644 --- a/fs/nfsd/nfsfh.h +++ b/fs/nfsd/nfsfh.h @@ -102,7 +102,7 @@ extern char * SVCFH_fmt(struct svc_fh *fhp); /* * Function prototypes */ -__be32 fh_verify(struct svc_rqst *, struct svc_fh *, int, int); +__be32 fh_verify(struct svc_rqst *, struct svc_fh *, umode_t, int); __be32 fh_compose(struct svc_fh *, struct svc_export *, struct dentry *, struct svc_fh *); __be32 fh_update(struct svc_fh *); void fh_put(struct svc_fh *); diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h index a3cf38476a1b..ffb5df1db94f 100644 --- a/fs/nfsd/state.h +++ b/fs/nfsd/state.h @@ -366,6 +366,7 @@ struct nfs4_openowner { struct nfs4_lockowner { struct nfs4_stateowner lo_owner; /* must be first element */ + struct list_head lo_owner_ino_hash; /* hash by owner,file */ struct list_head lo_perstateid; /* for lockowners only */ struct list_head lo_list; /* for temporary uses */ }; @@ -482,7 +483,7 @@ extern void nfsd4_shutdown_recdir(void); extern int nfs4_client_to_reclaim(const char *name); extern int nfs4_has_reclaimed_state(const char *name, bool use_exchange_id); extern void nfsd4_recdir_purge_old(void); -extern int nfsd4_create_clid_dir(struct nfs4_client *clp); +extern void nfsd4_create_clid_dir(struct nfs4_client *clp); extern void nfsd4_remove_clid_dir(struct nfs4_client *clp); extern void release_session_client(struct nfsd4_session *); extern __be32 nfs4_validate_stateid(struct nfs4_client *, stateid_t *); diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 7a2e442623c8..edf6d3ed8777 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -307,7 +307,7 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, struct dentry *dentry; struct inode *inode; int accmode = NFSD_MAY_SATTR; - int ftype = 0; + umode_t ftype = 0; __be32 err; int host_err; int size_change = 0; @@ -594,8 +594,19 @@ nfsd4_get_nfs4_acl(struct svc_rqst *rqstp, struct dentry *dentry, struct nfs4_ac return error; } -#define NFSD_XATTR_JUNCTION_PREFIX XATTR_TRUSTED_PREFIX "junction." -#define NFSD_XATTR_JUNCTION_TYPE NFSD_XATTR_JUNCTION_PREFIX "type" +/* + * NFS junction information is stored in an extended attribute. + */ +#define NFSD_JUNCTION_XATTR_NAME XATTR_TRUSTED_PREFIX "junction.nfs" + +/** + * nfsd4_is_junction - Test if an object could be an NFS junction + * + * @dentry: object to test + * + * Returns 1 if "dentry" appears to contain NFS junction information. + * Otherwise 0 is returned. + */ int nfsd4_is_junction(struct dentry *dentry) { struct inode *inode = dentry->d_inode; @@ -606,7 +617,7 @@ int nfsd4_is_junction(struct dentry *dentry) return 0; if (!(inode->i_mode & S_ISVTX)) return 0; - if (vfs_getxattr(dentry, NFSD_XATTR_JUNCTION_TYPE, NULL, 0) <= 0) + if (vfs_getxattr(dentry, NFSD_JUNCTION_XATTR_NAME, NULL, 0) <= 0) return 0; return 1; } @@ -730,7 +741,7 @@ static int nfsd_open_break_lease(struct inode *inode, int access) * N.B. After this call fhp needs an fh_put */ __be32 -nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, +nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, umode_t type, int access, struct file **filp) { struct dentry *dentry; @@ -1300,7 +1311,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, goto out; } - host_err = mnt_want_write(fhp->fh_export->ex_path.mnt); + host_err = fh_want_write(fhp); if (host_err) goto out_nfserr; @@ -1325,7 +1336,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, break; } if (host_err < 0) { - mnt_drop_write(fhp->fh_export->ex_path.mnt); + fh_drop_write(fhp); goto out_nfserr; } @@ -1339,7 +1350,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, err2 = nfserrno(commit_metadata(fhp)); if (err2) err = err2; - mnt_drop_write(fhp->fh_export->ex_path.mnt); + fh_drop_write(fhp); /* * Update the file handle to get the new inode info. */ @@ -1430,7 +1441,7 @@ do_nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, v_atime = verifier[1]&0x7fffffff; } - host_err = mnt_want_write(fhp->fh_export->ex_path.mnt); + host_err = fh_want_write(fhp); if (host_err) goto out_nfserr; if (dchild->d_inode) { @@ -1469,13 +1480,13 @@ do_nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, case NFS3_CREATE_GUARDED: err = nfserr_exist; } - mnt_drop_write(fhp->fh_export->ex_path.mnt); + fh_drop_write(fhp); goto out; } host_err = vfs_create(dirp, dchild, iap->ia_mode, NULL); if (host_err < 0) { - mnt_drop_write(fhp->fh_export->ex_path.mnt); + fh_drop_write(fhp); goto out_nfserr; } if (created) @@ -1503,7 +1514,7 @@ do_nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, if (!err) err = nfserrno(commit_metadata(fhp)); - mnt_drop_write(fhp->fh_export->ex_path.mnt); + fh_drop_write(fhp); /* * Update the filehandle to get the new inode info. */ @@ -1600,7 +1611,7 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, if (IS_ERR(dnew)) goto out_nfserr; - host_err = mnt_want_write(fhp->fh_export->ex_path.mnt); + host_err = fh_want_write(fhp); if (host_err) goto out_nfserr; @@ -1621,7 +1632,7 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, err = nfserrno(commit_metadata(fhp)); fh_unlock(fhp); - mnt_drop_write(fhp->fh_export->ex_path.mnt); + fh_drop_write(fhp); cerr = fh_compose(resfhp, fhp->fh_export, dnew, fhp); dput(dnew); @@ -1674,7 +1685,7 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp, dold = tfhp->fh_dentry; - host_err = mnt_want_write(tfhp->fh_export->ex_path.mnt); + host_err = fh_want_write(tfhp); if (host_err) { err = nfserrno(host_err); goto out_dput; @@ -1699,7 +1710,7 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp, err = nfserrno(host_err); } out_drop_write: - mnt_drop_write(tfhp->fh_export->ex_path.mnt); + fh_drop_write(tfhp); out_dput: dput(dnew); out_unlock: @@ -1776,7 +1787,7 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, host_err = -EXDEV; if (ffhp->fh_export->ex_path.mnt != tfhp->fh_export->ex_path.mnt) goto out_dput_new; - host_err = mnt_want_write(ffhp->fh_export->ex_path.mnt); + host_err = fh_want_write(ffhp); if (host_err) goto out_dput_new; @@ -1795,7 +1806,7 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, host_err = commit_metadata(ffhp); } out_drop_write: - mnt_drop_write(ffhp->fh_export->ex_path.mnt); + fh_drop_write(ffhp); out_dput_new: dput(ndentry); out_dput_old: @@ -1854,7 +1865,7 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, if (!type) type = rdentry->d_inode->i_mode & S_IFMT; - host_err = mnt_want_write(fhp->fh_export->ex_path.mnt); + host_err = fh_want_write(fhp); if (host_err) goto out_put; @@ -1868,7 +1879,7 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, if (!host_err) host_err = commit_metadata(fhp); out_drop_write: - mnt_drop_write(fhp->fh_export->ex_path.mnt); + fh_drop_write(fhp); out_put: dput(rdentry); @@ -2270,7 +2281,7 @@ nfsd_set_posix_acl(struct svc_fh *fhp, int type, struct posix_acl *acl) } else size = 0; - error = mnt_want_write(fhp->fh_export->ex_path.mnt); + error = fh_want_write(fhp); if (error) goto getout; if (size) @@ -2284,7 +2295,7 @@ nfsd_set_posix_acl(struct svc_fh *fhp, int type, struct posix_acl *acl) error = 0; } } - mnt_drop_write(fhp->fh_export->ex_path.mnt); + fh_drop_write(fhp); getout: kfree(value); diff --git a/fs/nfsd/vfs.h b/fs/nfsd/vfs.h index 3f54ad03bb2b..1dcd238e11a0 100644 --- a/fs/nfsd/vfs.h +++ b/fs/nfsd/vfs.h @@ -66,7 +66,7 @@ __be32 do_nfsd_create(struct svc_rqst *, struct svc_fh *, __be32 nfsd_commit(struct svc_rqst *, struct svc_fh *, loff_t, unsigned long); #endif /* CONFIG_NFSD_V3 */ -__be32 nfsd_open(struct svc_rqst *, struct svc_fh *, int, +__be32 nfsd_open(struct svc_rqst *, struct svc_fh *, umode_t, int, struct file **); void nfsd_close(struct file *); __be32 nfsd_read(struct svc_rqst *, struct svc_fh *, @@ -106,4 +106,14 @@ struct posix_acl *nfsd_get_posix_acl(struct svc_fh *, int); int nfsd_set_posix_acl(struct svc_fh *, int, struct posix_acl *); #endif +static inline int fh_want_write(struct svc_fh *fh) +{ + return mnt_want_write(fh->fh_export->ex_path.mnt); +} + +static inline void fh_drop_write(struct svc_fh *fh) +{ + mnt_drop_write(fh->fh_export->ex_path.mnt); +} + #endif /* LINUX_NFSD_VFS_H */ diff --git a/fs/nilfs2/dir.c b/fs/nilfs2/dir.c index 3a1923943b14..ca35b3a46d17 100644 --- a/fs/nilfs2/dir.c +++ b/fs/nilfs2/dir.c @@ -251,7 +251,7 @@ nilfs_type_by_mode[S_IFMT >> S_SHIFT] = { static void nilfs_set_de_type(struct nilfs_dir_entry *de, struct inode *inode) { - mode_t mode = inode->i_mode; + umode_t mode = inode->i_mode; de->file_type = nilfs_type_by_mode[(mode & S_IFMT)>>S_SHIFT]; } diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c index b50ffb72e5b3..8f7b95ac1f7e 100644 --- a/fs/nilfs2/inode.c +++ b/fs/nilfs2/inode.c @@ -291,7 +291,7 @@ const struct address_space_operations nilfs_aops = { .is_partially_uptodate = block_is_partially_uptodate, }; -struct inode *nilfs_new_inode(struct inode *dir, int mode) +struct inode *nilfs_new_inode(struct inode *dir, umode_t mode) { struct super_block *sb = dir->i_sb; struct the_nilfs *nilfs = sb->s_fs_info; diff --git a/fs/nilfs2/ioctl.c b/fs/nilfs2/ioctl.c index 41d6743d303c..886649627c3d 100644 --- a/fs/nilfs2/ioctl.c +++ b/fs/nilfs2/ioctl.c @@ -27,7 +27,7 @@ #include <linux/uaccess.h> /* copy_from_user(), copy_to_user() */ #include <linux/vmalloc.h> #include <linux/compat.h> /* compat_ptr() */ -#include <linux/mount.h> /* mnt_want_write(), mnt_drop_write() */ +#include <linux/mount.h> /* mnt_want_write_file(), mnt_drop_write_file() */ #include <linux/buffer_head.h> #include <linux/nilfs2_fs.h> #include "nilfs.h" @@ -119,7 +119,7 @@ static int nilfs_ioctl_setflags(struct inode *inode, struct file *filp, if (get_user(flags, (int __user *)argp)) return -EFAULT; - ret = mnt_want_write(filp->f_path.mnt); + ret = mnt_want_write_file(filp); if (ret) return ret; @@ -154,7 +154,7 @@ static int nilfs_ioctl_setflags(struct inode *inode, struct file *filp, ret = nilfs_transaction_commit(inode->i_sb); out: mutex_unlock(&inode->i_mutex); - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return ret; } @@ -174,7 +174,7 @@ static int nilfs_ioctl_change_cpmode(struct inode *inode, struct file *filp, if (!capable(CAP_SYS_ADMIN)) return -EPERM; - ret = mnt_want_write(filp->f_path.mnt); + ret = mnt_want_write_file(filp); if (ret) return ret; @@ -194,7 +194,7 @@ static int nilfs_ioctl_change_cpmode(struct inode *inode, struct file *filp, up_read(&inode->i_sb->s_umount); out: - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return ret; } @@ -210,7 +210,7 @@ nilfs_ioctl_delete_checkpoint(struct inode *inode, struct file *filp, if (!capable(CAP_SYS_ADMIN)) return -EPERM; - ret = mnt_want_write(filp->f_path.mnt); + ret = mnt_want_write_file(filp); if (ret) return ret; @@ -225,7 +225,7 @@ nilfs_ioctl_delete_checkpoint(struct inode *inode, struct file *filp, else nilfs_transaction_commit(inode->i_sb); /* never fails */ out: - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return ret; } @@ -591,7 +591,7 @@ static int nilfs_ioctl_clean_segments(struct inode *inode, struct file *filp, if (!capable(CAP_SYS_ADMIN)) return -EPERM; - ret = mnt_want_write(filp->f_path.mnt); + ret = mnt_want_write_file(filp); if (ret) return ret; @@ -625,6 +625,9 @@ static int nilfs_ioctl_clean_segments(struct inode *inode, struct file *filp, if (argv[n].v_nmembs > nsegs * nilfs->ns_blocks_per_segment) goto out_free; + if (argv[n].v_nmembs >= UINT_MAX / argv[n].v_size) + goto out_free; + len = argv[n].v_size * argv[n].v_nmembs; base = (void __user *)(unsigned long)argv[n].v_base; if (len == 0) { @@ -672,7 +675,7 @@ out_free: vfree(kbufs[n]); kfree(kbufs[4]); out: - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return ret; } @@ -707,7 +710,7 @@ static int nilfs_ioctl_resize(struct inode *inode, struct file *filp, if (!capable(CAP_SYS_ADMIN)) goto out; - ret = mnt_want_write(filp->f_path.mnt); + ret = mnt_want_write_file(filp); if (ret) goto out; @@ -718,7 +721,7 @@ static int nilfs_ioctl_resize(struct inode *inode, struct file *filp, ret = nilfs_resize_fs(inode->i_sb, newsize); out_drop_write: - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); out: return ret; } @@ -842,6 +845,19 @@ long nilfs_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) case FS_IOC32_GETVERSION: cmd = FS_IOC_GETVERSION; break; + case NILFS_IOCTL_CHANGE_CPMODE: + case NILFS_IOCTL_DELETE_CHECKPOINT: + case NILFS_IOCTL_GET_CPINFO: + case NILFS_IOCTL_GET_CPSTAT: + case NILFS_IOCTL_GET_SUINFO: + case NILFS_IOCTL_GET_SUSTAT: + case NILFS_IOCTL_GET_VINFO: + case NILFS_IOCTL_GET_BDESCS: + case NILFS_IOCTL_CLEAN_SEGMENTS: + case NILFS_IOCTL_SYNC: + case NILFS_IOCTL_RESIZE: + case NILFS_IOCTL_SET_ALLOC_RANGE: + break; default: return -ENOIOCTLCMD; } diff --git a/fs/nilfs2/namei.c b/fs/nilfs2/namei.c index 768982de10e4..1cd3f624dffc 100644 --- a/fs/nilfs2/namei.c +++ b/fs/nilfs2/namei.c @@ -84,7 +84,7 @@ nilfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) * If the create succeeds, we fill in the inode information * with d_instantiate(). */ -static int nilfs_create(struct inode *dir, struct dentry *dentry, int mode, +static int nilfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { struct inode *inode; @@ -112,7 +112,7 @@ static int nilfs_create(struct inode *dir, struct dentry *dentry, int mode, } static int -nilfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev) +nilfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) { struct inode *inode; struct nilfs_transaction_info ti; @@ -213,7 +213,7 @@ static int nilfs_link(struct dentry *old_dentry, struct inode *dir, return err; } -static int nilfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int nilfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct inode *inode; struct nilfs_transaction_info ti; diff --git a/fs/nilfs2/nilfs.h b/fs/nilfs2/nilfs.h index 3777d138f895..250add84da76 100644 --- a/fs/nilfs2/nilfs.h +++ b/fs/nilfs2/nilfs.h @@ -246,7 +246,7 @@ int nilfs_ioctl_prepare_clean_segments(struct the_nilfs *, struct nilfs_argv *, /* inode.c */ void nilfs_inode_add_blocks(struct inode *inode, int n); void nilfs_inode_sub_blocks(struct inode *inode, int n); -extern struct inode *nilfs_new_inode(struct inode *, int); +extern struct inode *nilfs_new_inode(struct inode *, umode_t); extern void nilfs_free_inode(struct inode *); extern int nilfs_get_block(struct inode *, sector_t, struct buffer_head *, int); extern void nilfs_set_inode_flags(struct inode *); diff --git a/fs/nilfs2/segment.c b/fs/nilfs2/segment.c index bb24ab6c282f..0e72ad6f22aa 100644 --- a/fs/nilfs2/segment.c +++ b/fs/nilfs2/segment.c @@ -2470,7 +2470,7 @@ static int nilfs_segctor_thread(void *arg) if (freezing(current)) { spin_unlock(&sci->sc_state_lock); - refrigerator(); + try_to_freeze(); spin_lock(&sci->sc_state_lock); } else { DEFINE_WAIT(wait); diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c index 8351c44a7320..08e3d4f9df18 100644 --- a/fs/nilfs2/super.c +++ b/fs/nilfs2/super.c @@ -175,8 +175,6 @@ static void nilfs_i_callback(struct rcu_head *head) struct inode *inode = container_of(head, struct inode, i_rcu); struct nilfs_mdt_info *mdi = NILFS_MDT(inode); - INIT_LIST_HEAD(&inode->i_dentry); - if (mdi) { kfree(mdi->mi_bgl); /* kfree(NULL) is safe */ kfree(mdi); @@ -650,11 +648,11 @@ static int nilfs_statfs(struct dentry *dentry, struct kstatfs *buf) return 0; } -static int nilfs_show_options(struct seq_file *seq, struct vfsmount *vfs) +static int nilfs_show_options(struct seq_file *seq, struct dentry *dentry) { - struct super_block *sb = vfs->mnt_sb; + struct super_block *sb = dentry->d_sb; struct the_nilfs *nilfs = sb->s_fs_info; - struct nilfs_root *root = NILFS_I(vfs->mnt_root->d_inode)->i_root; + struct nilfs_root *root = NILFS_I(dentry->d_inode)->i_root; if (!nilfs_test_opt(nilfs, BARRIER)) seq_puts(seq, ",nobarrier"); diff --git a/fs/nls/nls_base.c b/fs/nls/nls_base.c index 44a88a9fa2c8..fea6bd5831dc 100644 --- a/fs/nls/nls_base.c +++ b/fs/nls/nls_base.c @@ -52,7 +52,7 @@ static const struct utf8_table utf8_table[] = #define SURROGATE_LOW 0x00000400 #define SURROGATE_BITS 0x000003ff -int utf8_to_utf32(const u8 *s, int len, unicode_t *pu) +int utf8_to_utf32(const u8 *s, int inlen, unicode_t *pu) { unsigned long l; int c0, c, nc; @@ -71,7 +71,7 @@ int utf8_to_utf32(const u8 *s, int len, unicode_t *pu) *pu = (unicode_t) l; return nc; } - if (len <= nc) + if (inlen <= nc) return -1; s++; c = (*s ^ 0x80) & 0xFF; @@ -83,7 +83,7 @@ int utf8_to_utf32(const u8 *s, int len, unicode_t *pu) } EXPORT_SYMBOL(utf8_to_utf32); -int utf32_to_utf8(unicode_t u, u8 *s, int maxlen) +int utf32_to_utf8(unicode_t u, u8 *s, int maxout) { unsigned long l; int c, nc; @@ -97,7 +97,7 @@ int utf32_to_utf8(unicode_t u, u8 *s, int maxlen) return -1; nc = 0; - for (t = utf8_table; t->cmask && maxlen; t++, maxlen--) { + for (t = utf8_table; t->cmask && maxout; t++, maxout--) { nc++; if (l <= t->lmask) { c = t->shift; @@ -114,34 +114,57 @@ int utf32_to_utf8(unicode_t u, u8 *s, int maxlen) } EXPORT_SYMBOL(utf32_to_utf8); -int utf8s_to_utf16s(const u8 *s, int len, wchar_t *pwcs) +static inline void put_utf16(wchar_t *s, unsigned c, enum utf16_endian endian) +{ + switch (endian) { + default: + *s = (wchar_t) c; + break; + case UTF16_LITTLE_ENDIAN: + *s = __cpu_to_le16(c); + break; + case UTF16_BIG_ENDIAN: + *s = __cpu_to_be16(c); + break; + } +} + +int utf8s_to_utf16s(const u8 *s, int inlen, enum utf16_endian endian, + wchar_t *pwcs, int maxout) { u16 *op; int size; unicode_t u; op = pwcs; - while (*s && len > 0) { + while (inlen > 0 && maxout > 0 && *s) { if (*s & 0x80) { - size = utf8_to_utf32(s, len, &u); + size = utf8_to_utf32(s, inlen, &u); if (size < 0) return -EINVAL; + s += size; + inlen -= size; if (u >= PLANE_SIZE) { + if (maxout < 2) + break; u -= PLANE_SIZE; - *op++ = (wchar_t) (SURROGATE_PAIR | - ((u >> 10) & SURROGATE_BITS)); - *op++ = (wchar_t) (SURROGATE_PAIR | + put_utf16(op++, SURROGATE_PAIR | + ((u >> 10) & SURROGATE_BITS), + endian); + put_utf16(op++, SURROGATE_PAIR | SURROGATE_LOW | - (u & SURROGATE_BITS)); + (u & SURROGATE_BITS), + endian); + maxout -= 2; } else { - *op++ = (wchar_t) u; + put_utf16(op++, u, endian); + maxout--; } - s += size; - len -= size; } else { - *op++ = *s++; - len--; + put_utf16(op++, *s++, endian); + inlen--; + maxout--; } } return op - pwcs; @@ -160,27 +183,27 @@ static inline unsigned long get_utf16(unsigned c, enum utf16_endian endian) } } -int utf16s_to_utf8s(const wchar_t *pwcs, int len, enum utf16_endian endian, - u8 *s, int maxlen) +int utf16s_to_utf8s(const wchar_t *pwcs, int inlen, enum utf16_endian endian, + u8 *s, int maxout) { u8 *op; int size; unsigned long u, v; op = s; - while (len > 0 && maxlen > 0) { + while (inlen > 0 && maxout > 0) { u = get_utf16(*pwcs, endian); if (!u) break; pwcs++; - len--; + inlen--; if (u > 0x7f) { if ((u & SURROGATE_MASK) == SURROGATE_PAIR) { if (u & SURROGATE_LOW) { /* Ignore character and move on */ continue; } - if (len <= 0) + if (inlen <= 0) break; v = get_utf16(*pwcs, endian); if ((v & SURROGATE_MASK) != SURROGATE_PAIR || @@ -191,18 +214,18 @@ int utf16s_to_utf8s(const wchar_t *pwcs, int len, enum utf16_endian endian, u = PLANE_SIZE + ((u & SURROGATE_BITS) << 10) + (v & SURROGATE_BITS); pwcs++; - len--; + inlen--; } - size = utf32_to_utf8(u, op, maxlen); + size = utf32_to_utf8(u, op, maxout); if (size == -1) { /* Ignore character and move on */ } else { op += size; - maxlen -= size; + maxout -= size; } } else { *op++ = (u8) u; - maxlen--; + maxout--; } } return op - s; diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c index 9fde1c00a296..3568c8a8b138 100644 --- a/fs/notify/fanotify/fanotify_user.c +++ b/fs/notify/fanotify/fanotify_user.c @@ -16,6 +16,8 @@ #include <asm/ioctls.h> +#include "../../mount.h" + #define FANOTIFY_DEFAULT_MAX_EVENTS 16384 #define FANOTIFY_DEFAULT_MAX_MARKS 8192 #define FANOTIFY_DEFAULT_MAX_LISTENERS 128 @@ -546,7 +548,7 @@ static int fanotify_remove_vfsmount_mark(struct fsnotify_group *group, removed = fanotify_mark_remove_from_mask(fsn_mark, mask, flags); fsnotify_put_mark(fsn_mark); - if (removed & mnt->mnt_fsnotify_mask) + if (removed & real_mount(mnt)->mnt_fsnotify_mask) fsnotify_recalc_vfsmount_mask(mnt); return 0; @@ -623,7 +625,7 @@ static int fanotify_add_vfsmount_mark(struct fsnotify_group *group, } added = fanotify_mark_add_to_mask(fsn_mark, mask, flags); - if (added & ~mnt->mnt_fsnotify_mask) + if (added & ~real_mount(mnt)->mnt_fsnotify_mask) fsnotify_recalc_vfsmount_mask(mnt); err: fsnotify_put_mark(fsn_mark); diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c index 79b47cbb5cd8..ccb14d3fc0de 100644 --- a/fs/notify/fsnotify.c +++ b/fs/notify/fsnotify.c @@ -26,6 +26,7 @@ #include <linux/fsnotify_backend.h> #include "fsnotify.h" +#include "../mount.h" /* * Clear all of the marks on an inode when it is being evicted from core @@ -205,13 +206,13 @@ int fsnotify(struct inode *to_tell, __u32 mask, void *data, int data_is, struct fsnotify_mark *inode_mark = NULL, *vfsmount_mark = NULL; struct fsnotify_group *inode_group, *vfsmount_group; struct fsnotify_event *event = NULL; - struct vfsmount *mnt; + struct mount *mnt; int idx, ret = 0; /* global tests shouldn't care about events on child only the specific event */ __u32 test_mask = (mask & ~FS_EVENT_ON_CHILD); if (data_is == FSNOTIFY_EVENT_PATH) - mnt = ((struct path *)data)->mnt; + mnt = real_mount(((struct path *)data)->mnt); else mnt = NULL; @@ -262,11 +263,11 @@ int fsnotify(struct inode *to_tell, __u32 mask, void *data, int data_is, /* we didn't use the vfsmount_mark */ vfsmount_group = NULL; } else if (vfsmount_group > inode_group) { - ret = send_to_group(to_tell, mnt, NULL, vfsmount_mark, mask, data, + ret = send_to_group(to_tell, &mnt->mnt, NULL, vfsmount_mark, mask, data, data_is, cookie, file_name, &event); inode_group = NULL; } else { - ret = send_to_group(to_tell, mnt, inode_mark, vfsmount_mark, + ret = send_to_group(to_tell, &mnt->mnt, inode_mark, vfsmount_mark, mask, data, data_is, cookie, file_name, &event); } diff --git a/fs/notify/mark.c b/fs/notify/mark.c index e14587d55689..f104d565b682 100644 --- a/fs/notify/mark.c +++ b/fs/notify/mark.c @@ -135,9 +135,6 @@ void fsnotify_destroy_mark(struct fsnotify_mark *mark) mark->flags &= ~FSNOTIFY_MARK_FLAG_ALIVE; - /* 1 from caller and 1 for being on i_list/g_list */ - BUG_ON(atomic_read(&mark->refcnt) < 2); - spin_lock(&group->mark_lock); if (mark->flags & FSNOTIFY_MARK_FLAG_INODE) { @@ -182,6 +179,11 @@ void fsnotify_destroy_mark(struct fsnotify_mark *mark) iput(inode); /* + * We don't necessarily have a ref on mark from caller so the above iput + * may have already destroyed it. Don't touch from now on. + */ + + /* * it's possible that this group tried to destroy itself, but this * this mark was simultaneously being freed by inode. If that's the * case, we finish freeing the group here. diff --git a/fs/notify/vfsmount_mark.c b/fs/notify/vfsmount_mark.c index 778fe6cae3b0..b7b4b0e8554f 100644 --- a/fs/notify/vfsmount_mark.c +++ b/fs/notify/vfsmount_mark.c @@ -28,15 +28,17 @@ #include <linux/fsnotify_backend.h> #include "fsnotify.h" +#include "../mount.h" void fsnotify_clear_marks_by_mount(struct vfsmount *mnt) { struct fsnotify_mark *mark, *lmark; struct hlist_node *pos, *n; + struct mount *m = real_mount(mnt); LIST_HEAD(free_list); spin_lock(&mnt->mnt_root->d_lock); - hlist_for_each_entry_safe(mark, pos, n, &mnt->mnt_fsnotify_marks, m.m_list) { + hlist_for_each_entry_safe(mark, pos, n, &m->mnt_fsnotify_marks, m.m_list) { list_add(&mark->m.free_m_list, &free_list); hlist_del_init_rcu(&mark->m.m_list); fsnotify_get_mark(mark); @@ -59,15 +61,16 @@ void fsnotify_clear_vfsmount_marks_by_group(struct fsnotify_group *group) */ static void fsnotify_recalc_vfsmount_mask_locked(struct vfsmount *mnt) { + struct mount *m = real_mount(mnt); struct fsnotify_mark *mark; struct hlist_node *pos; __u32 new_mask = 0; assert_spin_locked(&mnt->mnt_root->d_lock); - hlist_for_each_entry(mark, pos, &mnt->mnt_fsnotify_marks, m.m_list) + hlist_for_each_entry(mark, pos, &m->mnt_fsnotify_marks, m.m_list) new_mask |= mark->mask; - mnt->mnt_fsnotify_mask = new_mask; + m->mnt_fsnotify_mask = new_mask; } /* @@ -101,12 +104,13 @@ void fsnotify_destroy_vfsmount_mark(struct fsnotify_mark *mark) static struct fsnotify_mark *fsnotify_find_vfsmount_mark_locked(struct fsnotify_group *group, struct vfsmount *mnt) { + struct mount *m = real_mount(mnt); struct fsnotify_mark *mark; struct hlist_node *pos; assert_spin_locked(&mnt->mnt_root->d_lock); - hlist_for_each_entry(mark, pos, &mnt->mnt_fsnotify_marks, m.m_list) { + hlist_for_each_entry(mark, pos, &m->mnt_fsnotify_marks, m.m_list) { if (mark->group == group) { fsnotify_get_mark(mark); return mark; @@ -140,6 +144,7 @@ int fsnotify_add_vfsmount_mark(struct fsnotify_mark *mark, struct fsnotify_group *group, struct vfsmount *mnt, int allow_dups) { + struct mount *m = real_mount(mnt); struct fsnotify_mark *lmark; struct hlist_node *node, *last = NULL; int ret = 0; @@ -154,13 +159,13 @@ int fsnotify_add_vfsmount_mark(struct fsnotify_mark *mark, mark->m.mnt = mnt; /* is mark the first mark? */ - if (hlist_empty(&mnt->mnt_fsnotify_marks)) { - hlist_add_head_rcu(&mark->m.m_list, &mnt->mnt_fsnotify_marks); + if (hlist_empty(&m->mnt_fsnotify_marks)) { + hlist_add_head_rcu(&mark->m.m_list, &m->mnt_fsnotify_marks); goto out; } /* should mark be in the middle of the current list? */ - hlist_for_each_entry(lmark, node, &mnt->mnt_fsnotify_marks, m.m_list) { + hlist_for_each_entry(lmark, node, &m->mnt_fsnotify_marks, m.m_list) { last = node; if ((lmark->group == group) && !allow_dups) { diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c index 97e2dacbc867..2eaa66652944 100644 --- a/fs/ntfs/inode.c +++ b/fs/ntfs/inode.c @@ -335,7 +335,6 @@ struct inode *ntfs_alloc_big_inode(struct super_block *sb) static void ntfs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(ntfs_big_inode_cache, NTFS_I(inode)); } @@ -2301,16 +2300,16 @@ void ntfs_evict_big_inode(struct inode *vi) /** * ntfs_show_options - show mount options in /proc/mounts * @sf: seq_file in which to write our mount options - * @mnt: vfs mount whose mount options to display + * @root: root of the mounted tree whose mount options to display * * Called by the VFS once for each mounted ntfs volume when someone reads * /proc/mounts in order to display the NTFS specific mount options of each - * mount. The mount options of the vfs mount @mnt are written to the seq file + * mount. The mount options of fs specified by @root are written to the seq file * @sf and success is returned. */ -int ntfs_show_options(struct seq_file *sf, struct vfsmount *mnt) +int ntfs_show_options(struct seq_file *sf, struct dentry *root) { - ntfs_volume *vol = NTFS_SB(mnt->mnt_sb); + ntfs_volume *vol = NTFS_SB(root->d_sb); int i; seq_printf(sf, ",uid=%i", vol->uid); diff --git a/fs/ntfs/inode.h b/fs/ntfs/inode.h index fe8e7e928889..db29695f845c 100644 --- a/fs/ntfs/inode.h +++ b/fs/ntfs/inode.h @@ -298,7 +298,7 @@ extern void ntfs_clear_extent_inode(ntfs_inode *ni); extern int ntfs_read_inode_mount(struct inode *vi); -extern int ntfs_show_options(struct seq_file *sf, struct vfsmount *mnt); +extern int ntfs_show_options(struct seq_file *sf, struct dentry *root); #ifdef NTFS_RW diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c index b52706da4645..5a4a8af5c406 100644 --- a/fs/ntfs/super.c +++ b/fs/ntfs/super.c @@ -104,7 +104,7 @@ static bool parse_options(ntfs_volume *vol, char *opt) int errors = 0, sloppy = 0; uid_t uid = (uid_t)-1; gid_t gid = (gid_t)-1; - mode_t fmask = (mode_t)-1, dmask = (mode_t)-1; + umode_t fmask = (umode_t)-1, dmask = (umode_t)-1; int mft_zone_multiplier = -1, on_errors = -1; int show_sys_files = -1, case_sensitive = -1, disable_sparse = -1; struct nls_table *nls_map = NULL, *old_nls; @@ -287,9 +287,9 @@ no_mount_options: vol->uid = uid; if (gid != (gid_t)-1) vol->gid = gid; - if (fmask != (mode_t)-1) + if (fmask != (umode_t)-1) vol->fmask = fmask; - if (dmask != (mode_t)-1) + if (dmask != (umode_t)-1) vol->dmask = dmask; if (show_sys_files != -1) { if (show_sys_files) @@ -3198,7 +3198,7 @@ MODULE_DESCRIPTION("NTFS 1.2/3.x driver - Copyright (c) 2001-2011 Anton Altaparm MODULE_VERSION(NTFS_VERSION); MODULE_LICENSE("GPL"); #ifdef DEBUG -module_param(debug_msgs, bool, 0); +module_param(debug_msgs, bint, 0); MODULE_PARM_DESC(debug_msgs, "Enable debug messages."); #endif diff --git a/fs/ntfs/volume.h b/fs/ntfs/volume.h index 406ab55dfb32..15e3ba8d521a 100644 --- a/fs/ntfs/volume.h +++ b/fs/ntfs/volume.h @@ -48,8 +48,8 @@ typedef struct { unsigned long flags; /* Miscellaneous flags, see below. */ uid_t uid; /* uid that files will be mounted as. */ gid_t gid; /* gid that files will be mounted as. */ - mode_t fmask; /* The mask for file permissions. */ - mode_t dmask; /* The mask for directory + umode_t fmask; /* The mask for file permissions. */ + umode_t dmask; /* The mask for directory permissions. */ u8 mft_zone_multiplier; /* Initial mft zone multiplier. */ u8 on_errors; /* What to do on filesystem errors. */ diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index ed553c60de82..3165aebb43c8 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -5699,7 +5699,7 @@ int ocfs2_remove_btree_range(struct inode *inode, OCFS2_JOURNAL_ACCESS_WRITE); if (ret) { mlog_errno(ret); - goto out; + goto out_commit; } dquot_free_space_nodirty(inode, diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index c1efe939c774..78b68af3b0e3 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -290,7 +290,15 @@ static int ocfs2_readpage(struct file *file, struct page *page) } if (down_read_trylock(&oi->ip_alloc_sem) == 0) { + /* + * Unlock the page and cycle ip_alloc_sem so that we don't + * busyloop waiting for ip_alloc_sem to unlock + */ ret = AOP_TRUNCATED_PAGE; + unlock_page(page); + unlock = 0; + down_read(&oi->ip_alloc_sem); + up_read(&oi->ip_alloc_sem); goto out_inode_unlock; } @@ -563,6 +571,7 @@ static void ocfs2_dio_end_io(struct kiocb *iocb, { struct inode *inode = iocb->ki_filp->f_path.dentry->d_inode; int level; + wait_queue_head_t *wq = ocfs2_ioend_wq(inode); /* this io's submitter should not have unlocked this before we could */ BUG_ON(!ocfs2_iocb_is_rw_locked(iocb)); @@ -570,6 +579,15 @@ static void ocfs2_dio_end_io(struct kiocb *iocb, if (ocfs2_iocb_is_sem_locked(iocb)) ocfs2_iocb_clear_sem_locked(iocb); + if (ocfs2_iocb_is_unaligned_aio(iocb)) { + ocfs2_iocb_clear_unaligned_aio(iocb); + + if (atomic_dec_and_test(&OCFS2_I(inode)->ip_unaligned_aio) && + waitqueue_active(wq)) { + wake_up_all(wq); + } + } + ocfs2_iocb_clear_rw_locked(iocb); level = ocfs2_iocb_rw_locked_level(iocb); @@ -863,6 +881,12 @@ struct ocfs2_write_ctxt { struct page *w_target_page; /* + * w_target_locked is used for page_mkwrite path indicating no unlocking + * against w_target_page in ocfs2_write_end_nolock. + */ + unsigned int w_target_locked:1; + + /* * ocfs2_write_end() uses this to know what the real range to * write in the target should be. */ @@ -895,6 +919,24 @@ void ocfs2_unlock_and_free_pages(struct page **pages, int num_pages) static void ocfs2_free_write_ctxt(struct ocfs2_write_ctxt *wc) { + int i; + + /* + * w_target_locked is only set to true in the page_mkwrite() case. + * The intent is to allow us to lock the target page from write_begin() + * to write_end(). The caller must hold a ref on w_target_page. + */ + if (wc->w_target_locked) { + BUG_ON(!wc->w_target_page); + for (i = 0; i < wc->w_num_pages; i++) { + if (wc->w_target_page == wc->w_pages[i]) { + wc->w_pages[i] = NULL; + break; + } + } + mark_page_accessed(wc->w_target_page); + page_cache_release(wc->w_target_page); + } ocfs2_unlock_and_free_pages(wc->w_pages, wc->w_num_pages); brelse(wc->w_di_bh); @@ -1132,20 +1174,17 @@ static int ocfs2_grab_pages_for_write(struct address_space *mapping, */ lock_page(mmap_page); + /* Exit and let the caller retry */ if (mmap_page->mapping != mapping) { + WARN_ON(mmap_page->mapping); unlock_page(mmap_page); - /* - * Sanity check - the locking in - * ocfs2_pagemkwrite() should ensure - * that this code doesn't trigger. - */ - ret = -EINVAL; - mlog_errno(ret); + ret = -EAGAIN; goto out; } page_cache_get(mmap_page); wc->w_pages[i] = mmap_page; + wc->w_target_locked = true; } else { wc->w_pages[i] = find_or_create_page(mapping, index, GFP_NOFS); @@ -1160,6 +1199,8 @@ static int ocfs2_grab_pages_for_write(struct address_space *mapping, wc->w_target_page = wc->w_pages[i]; } out: + if (ret) + wc->w_target_locked = false; return ret; } @@ -1817,11 +1858,23 @@ try_again: */ ret = ocfs2_grab_pages_for_write(mapping, wc, wc->w_cpos, pos, len, cluster_of_pages, mmap_page); - if (ret) { + if (ret && ret != -EAGAIN) { mlog_errno(ret); goto out_quota; } + /* + * ocfs2_grab_pages_for_write() returns -EAGAIN if it could not lock + * the target page. In this case, we exit with no error and no target + * page. This will trigger the caller, page_mkwrite(), to re-try + * the operation. + */ + if (ret == -EAGAIN) { + BUG_ON(wc->w_target_page); + ret = 0; + goto out_quota; + } + ret = ocfs2_write_cluster_by_desc(mapping, data_ac, meta_ac, wc, pos, len); if (ret) { diff --git a/fs/ocfs2/aops.h b/fs/ocfs2/aops.h index 75cf3ad987a6..ffb2da370a99 100644 --- a/fs/ocfs2/aops.h +++ b/fs/ocfs2/aops.h @@ -78,6 +78,7 @@ enum ocfs2_iocb_lock_bits { OCFS2_IOCB_RW_LOCK = 0, OCFS2_IOCB_RW_LOCK_LEVEL, OCFS2_IOCB_SEM, + OCFS2_IOCB_UNALIGNED_IO, OCFS2_IOCB_NUM_LOCKS }; @@ -91,4 +92,17 @@ enum ocfs2_iocb_lock_bits { clear_bit(OCFS2_IOCB_SEM, (unsigned long *)&iocb->private) #define ocfs2_iocb_is_sem_locked(iocb) \ test_bit(OCFS2_IOCB_SEM, (unsigned long *)&iocb->private) + +#define ocfs2_iocb_set_unaligned_aio(iocb) \ + set_bit(OCFS2_IOCB_UNALIGNED_IO, (unsigned long *)&iocb->private) +#define ocfs2_iocb_clear_unaligned_aio(iocb) \ + clear_bit(OCFS2_IOCB_UNALIGNED_IO, (unsigned long *)&iocb->private) +#define ocfs2_iocb_is_unaligned_aio(iocb) \ + test_bit(OCFS2_IOCB_UNALIGNED_IO, (unsigned long *)&iocb->private) + +#define OCFS2_IOEND_WQ_HASH_SZ 37 +#define ocfs2_ioend_wq(v) (&ocfs2__ioend_wq[((unsigned long)(v)) %\ + OCFS2_IOEND_WQ_HASH_SZ]) +extern wait_queue_head_t ocfs2__ioend_wq[OCFS2_IOEND_WQ_HASH_SZ]; + #endif /* OCFS2_FILE_H */ diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index 9a3e6bbff27b..a4e855e3690e 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -216,6 +216,7 @@ struct o2hb_region { struct list_head hr_all_item; unsigned hr_unclean_stop:1, + hr_aborted_start:1, hr_item_pinned:1, hr_item_dropped:1; @@ -254,6 +255,10 @@ struct o2hb_region { * a more complete api that doesn't lead to this sort of fragility. */ atomic_t hr_steady_iterations; + /* terminate o2hb thread if it does not reach steady state + * (hr_steady_iterations == 0) within hr_unsteady_iterations */ + atomic_t hr_unsteady_iterations; + char hr_dev_name[BDEVNAME_SIZE]; unsigned int hr_timeout_ms; @@ -324,6 +329,10 @@ static void o2hb_write_timeout(struct work_struct *work) static void o2hb_arm_write_timeout(struct o2hb_region *reg) { + /* Arm writeout only after thread reaches steady state */ + if (atomic_read(®->hr_steady_iterations) != 0) + return; + mlog(ML_HEARTBEAT, "Queue write timeout for %u ms\n", O2HB_MAX_WRITE_TIMEOUT_MS); @@ -537,9 +546,14 @@ static int o2hb_verify_crc(struct o2hb_region *reg, return read == computed; } -/* We want to make sure that nobody is heartbeating on top of us -- - * this will help detect an invalid configuration. */ -static void o2hb_check_last_timestamp(struct o2hb_region *reg) +/* + * Compare the slot data with what we wrote in the last iteration. + * If the match fails, print an appropriate error message. This is to + * detect errors like... another node hearting on the same slot, + * flaky device that is losing writes, etc. + * Returns 1 if check succeeds, 0 otherwise. + */ +static int o2hb_check_own_slot(struct o2hb_region *reg) { struct o2hb_disk_slot *slot; struct o2hb_disk_heartbeat_block *hb_block; @@ -548,13 +562,13 @@ static void o2hb_check_last_timestamp(struct o2hb_region *reg) slot = ®->hr_slots[o2nm_this_node()]; /* Don't check on our 1st timestamp */ if (!slot->ds_last_time) - return; + return 0; hb_block = slot->ds_raw_block; if (le64_to_cpu(hb_block->hb_seq) == slot->ds_last_time && le64_to_cpu(hb_block->hb_generation) == slot->ds_last_generation && hb_block->hb_node == slot->ds_node_num) - return; + return 1; #define ERRSTR1 "Another node is heartbeating on device" #define ERRSTR2 "Heartbeat generation mismatch on device" @@ -574,6 +588,8 @@ static void o2hb_check_last_timestamp(struct o2hb_region *reg) (unsigned long long)slot->ds_last_time, hb_block->hb_node, (unsigned long long)le64_to_cpu(hb_block->hb_generation), (unsigned long long)le64_to_cpu(hb_block->hb_seq)); + + return 0; } static inline void o2hb_prepare_block(struct o2hb_region *reg, @@ -719,17 +735,24 @@ static void o2hb_shutdown_slot(struct o2hb_disk_slot *slot) o2nm_node_put(node); } -static void o2hb_set_quorum_device(struct o2hb_region *reg, - struct o2hb_disk_slot *slot) +static void o2hb_set_quorum_device(struct o2hb_region *reg) { - assert_spin_locked(&o2hb_live_lock); - if (!o2hb_global_heartbeat_active()) return; - if (test_bit(reg->hr_region_num, o2hb_quorum_region_bitmap)) + /* Prevent race with o2hb_heartbeat_group_drop_item() */ + if (kthread_should_stop()) + return; + + /* Tag region as quorum only after thread reaches steady state */ + if (atomic_read(®->hr_steady_iterations) != 0) return; + spin_lock(&o2hb_live_lock); + + if (test_bit(reg->hr_region_num, o2hb_quorum_region_bitmap)) + goto unlock; + /* * A region can be added to the quorum only when it sees all * live nodes heartbeat on it. In other words, the region has been @@ -737,13 +760,10 @@ static void o2hb_set_quorum_device(struct o2hb_region *reg, */ if (memcmp(reg->hr_live_node_bitmap, o2hb_live_node_bitmap, sizeof(o2hb_live_node_bitmap))) - return; - - if (slot->ds_changed_samples < O2HB_LIVE_THRESHOLD) - return; + goto unlock; - printk(KERN_NOTICE "o2hb: Region %s is now a quorum device\n", - config_item_name(®->hr_item)); + printk(KERN_NOTICE "o2hb: Region %s (%s) is now a quorum device\n", + config_item_name(®->hr_item), reg->hr_dev_name); set_bit(reg->hr_region_num, o2hb_quorum_region_bitmap); @@ -754,6 +774,8 @@ static void o2hb_set_quorum_device(struct o2hb_region *reg, if (o2hb_pop_count(&o2hb_quorum_region_bitmap, O2NM_MAX_REGIONS) > O2HB_PIN_CUT_OFF) o2hb_region_unpin(NULL); +unlock: + spin_unlock(&o2hb_live_lock); } static int o2hb_check_slot(struct o2hb_region *reg, @@ -925,8 +947,6 @@ fire_callbacks: slot->ds_equal_samples = 0; } out: - o2hb_set_quorum_device(reg, slot); - spin_unlock(&o2hb_live_lock); o2hb_run_event_list(&event); @@ -957,7 +977,8 @@ static int o2hb_highest_node(unsigned long *nodes, static int o2hb_do_disk_heartbeat(struct o2hb_region *reg) { - int i, ret, highest_node, change = 0; + int i, ret, highest_node; + int membership_change = 0, own_slot_ok = 0; unsigned long configured_nodes[BITS_TO_LONGS(O2NM_MAX_NODES)]; unsigned long live_node_bitmap[BITS_TO_LONGS(O2NM_MAX_NODES)]; struct o2hb_bio_wait_ctxt write_wc; @@ -966,7 +987,7 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg) sizeof(configured_nodes)); if (ret) { mlog_errno(ret); - return ret; + goto bail; } /* @@ -982,8 +1003,9 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg) highest_node = o2hb_highest_node(configured_nodes, O2NM_MAX_NODES); if (highest_node >= O2NM_MAX_NODES) { - mlog(ML_NOTICE, "ocfs2_heartbeat: no configured nodes found!\n"); - return -EINVAL; + mlog(ML_NOTICE, "o2hb: No configured nodes found!\n"); + ret = -EINVAL; + goto bail; } /* No sense in reading the slots of nodes that don't exist @@ -993,29 +1015,27 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg) ret = o2hb_read_slots(reg, highest_node + 1); if (ret < 0) { mlog_errno(ret); - return ret; + goto bail; } /* With an up to date view of the slots, we can check that no * other node has been improperly configured to heartbeat in * our slot. */ - o2hb_check_last_timestamp(reg); + own_slot_ok = o2hb_check_own_slot(reg); /* fill in the proper info for our next heartbeat */ o2hb_prepare_block(reg, reg->hr_generation); - /* And fire off the write. Note that we don't wait on this I/O - * until later. */ ret = o2hb_issue_node_write(reg, &write_wc); if (ret < 0) { mlog_errno(ret); - return ret; + goto bail; } i = -1; while((i = find_next_bit(configured_nodes, O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) { - change |= o2hb_check_slot(reg, ®->hr_slots[i]); + membership_change |= o2hb_check_slot(reg, ®->hr_slots[i]); } /* @@ -1030,18 +1050,39 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg) * disk */ mlog(ML_ERROR, "Write error %d on device \"%s\"\n", write_wc.wc_error, reg->hr_dev_name); - return write_wc.wc_error; + ret = write_wc.wc_error; + goto bail; } - o2hb_arm_write_timeout(reg); + /* Skip disarming the timeout if own slot has stale/bad data */ + if (own_slot_ok) { + o2hb_set_quorum_device(reg); + o2hb_arm_write_timeout(reg); + } +bail: /* let the person who launched us know when things are steady */ - if (!change && (atomic_read(®->hr_steady_iterations) != 0)) { - if (atomic_dec_and_test(®->hr_steady_iterations)) + if (atomic_read(®->hr_steady_iterations) != 0) { + if (!ret && own_slot_ok && !membership_change) { + if (atomic_dec_and_test(®->hr_steady_iterations)) + wake_up(&o2hb_steady_queue); + } + } + + if (atomic_read(®->hr_steady_iterations) != 0) { + if (atomic_dec_and_test(®->hr_unsteady_iterations)) { + printk(KERN_NOTICE "o2hb: Unable to stabilize " + "heartbeart on region %s (%s)\n", + config_item_name(®->hr_item), + reg->hr_dev_name); + atomic_set(®->hr_steady_iterations, 0); + reg->hr_aborted_start = 1; wake_up(&o2hb_steady_queue); + ret = -EIO; + } } - return 0; + return ret; } /* Subtract b from a, storing the result in a. a *must* have a larger @@ -1095,7 +1136,8 @@ static int o2hb_thread(void *data) /* Pin node */ o2nm_depend_this_node(); - while (!kthread_should_stop() && !reg->hr_unclean_stop) { + while (!kthread_should_stop() && + !reg->hr_unclean_stop && !reg->hr_aborted_start) { /* We track the time spent inside * o2hb_do_disk_heartbeat so that we avoid more than * hr_timeout_ms between disk writes. On busy systems @@ -1103,10 +1145,7 @@ static int o2hb_thread(void *data) * likely to time itself out. */ do_gettimeofday(&before_hb); - i = 0; - do { - ret = o2hb_do_disk_heartbeat(reg); - } while (ret && ++i < 2); + ret = o2hb_do_disk_heartbeat(reg); do_gettimeofday(&after_hb); elapsed_msec = o2hb_elapsed_msecs(&before_hb, &after_hb); @@ -1117,7 +1156,8 @@ static int o2hb_thread(void *data) after_hb.tv_sec, (unsigned long) after_hb.tv_usec, elapsed_msec); - if (elapsed_msec < reg->hr_timeout_ms) { + if (!kthread_should_stop() && + elapsed_msec < reg->hr_timeout_ms) { /* the kthread api has blocked signals for us so no * need to record the return value. */ msleep_interruptible(reg->hr_timeout_ms - elapsed_msec); @@ -1134,20 +1174,20 @@ static int o2hb_thread(void *data) * to timeout on this region when we could just as easily * write a clear generation - thus indicating to them that * this node has left this region. - * - * XXX: Should we skip this on unclean_stop? */ - o2hb_prepare_block(reg, 0); - ret = o2hb_issue_node_write(reg, &write_wc); - if (ret == 0) { - o2hb_wait_on_io(reg, &write_wc); - } else { - mlog_errno(ret); + */ + if (!reg->hr_unclean_stop && !reg->hr_aborted_start) { + o2hb_prepare_block(reg, 0); + ret = o2hb_issue_node_write(reg, &write_wc); + if (ret == 0) + o2hb_wait_on_io(reg, &write_wc); + else + mlog_errno(ret); } /* Unpin node */ o2nm_undepend_this_node(); - mlog(ML_HEARTBEAT|ML_KTHREAD, "hb thread exiting\n"); + mlog(ML_HEARTBEAT|ML_KTHREAD, "o2hb thread exiting\n"); return 0; } @@ -1158,6 +1198,7 @@ static int o2hb_debug_open(struct inode *inode, struct file *file) struct o2hb_debug_buf *db = inode->i_private; struct o2hb_region *reg; unsigned long map[BITS_TO_LONGS(O2NM_MAX_NODES)]; + unsigned long lts; char *buf = NULL; int i = -1; int out = 0; @@ -1194,9 +1235,11 @@ static int o2hb_debug_open(struct inode *inode, struct file *file) case O2HB_DB_TYPE_REGION_ELAPSED_TIME: reg = (struct o2hb_region *)db->db_data; - out += snprintf(buf + out, PAGE_SIZE - out, "%u\n", - jiffies_to_msecs(jiffies - - reg->hr_last_timeout_start)); + lts = reg->hr_last_timeout_start; + /* If 0, it has never been set before */ + if (lts) + lts = jiffies_to_msecs(jiffies - lts); + out += snprintf(buf + out, PAGE_SIZE - out, "%lu\n", lts); goto done; case O2HB_DB_TYPE_REGION_PINNED: @@ -1426,6 +1469,8 @@ static void o2hb_region_release(struct config_item *item) struct page *page; struct o2hb_region *reg = to_o2hb_region(item); + mlog(ML_HEARTBEAT, "hb region release (%s)\n", reg->hr_dev_name); + if (reg->hr_tmp_block) kfree(reg->hr_tmp_block); @@ -1792,7 +1837,10 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, live_threshold <<= 1; spin_unlock(&o2hb_live_lock); } - atomic_set(®->hr_steady_iterations, live_threshold + 1); + ++live_threshold; + atomic_set(®->hr_steady_iterations, live_threshold); + /* unsteady_iterations is double the steady_iterations */ + atomic_set(®->hr_unsteady_iterations, (live_threshold << 1)); hb_task = kthread_run(o2hb_thread, reg, "o2hb-%s", reg->hr_item.ci_name); @@ -1809,14 +1857,12 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, ret = wait_event_interruptible(o2hb_steady_queue, atomic_read(®->hr_steady_iterations) == 0); if (ret) { - /* We got interrupted (hello ptrace!). Clean up */ - spin_lock(&o2hb_live_lock); - hb_task = reg->hr_task; - reg->hr_task = NULL; - spin_unlock(&o2hb_live_lock); + atomic_set(®->hr_steady_iterations, 0); + reg->hr_aborted_start = 1; + } - if (hb_task) - kthread_stop(hb_task); + if (reg->hr_aborted_start) { + ret = -EIO; goto out; } @@ -1833,8 +1879,8 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, ret = -EIO; if (hb_task && o2hb_global_heartbeat_active()) - printk(KERN_NOTICE "o2hb: Heartbeat started on region %s\n", - config_item_name(®->hr_item)); + printk(KERN_NOTICE "o2hb: Heartbeat started on region %s (%s)\n", + config_item_name(®->hr_item), reg->hr_dev_name); out: if (filp) @@ -2092,13 +2138,6 @@ static void o2hb_heartbeat_group_drop_item(struct config_group *group, /* stop the thread when the user removes the region dir */ spin_lock(&o2hb_live_lock); - if (o2hb_global_heartbeat_active()) { - clear_bit(reg->hr_region_num, o2hb_region_bitmap); - clear_bit(reg->hr_region_num, o2hb_live_region_bitmap); - if (test_bit(reg->hr_region_num, o2hb_quorum_region_bitmap)) - quorum_region = 1; - clear_bit(reg->hr_region_num, o2hb_quorum_region_bitmap); - } hb_task = reg->hr_task; reg->hr_task = NULL; reg->hr_item_dropped = 1; @@ -2107,19 +2146,30 @@ static void o2hb_heartbeat_group_drop_item(struct config_group *group, if (hb_task) kthread_stop(hb_task); + if (o2hb_global_heartbeat_active()) { + spin_lock(&o2hb_live_lock); + clear_bit(reg->hr_region_num, o2hb_region_bitmap); + clear_bit(reg->hr_region_num, o2hb_live_region_bitmap); + if (test_bit(reg->hr_region_num, o2hb_quorum_region_bitmap)) + quorum_region = 1; + clear_bit(reg->hr_region_num, o2hb_quorum_region_bitmap); + spin_unlock(&o2hb_live_lock); + printk(KERN_NOTICE "o2hb: Heartbeat %s on region %s (%s)\n", + ((atomic_read(®->hr_steady_iterations) == 0) ? + "stopped" : "start aborted"), config_item_name(item), + reg->hr_dev_name); + } + /* * If we're racing a dev_write(), we need to wake them. They will * check reg->hr_task */ if (atomic_read(®->hr_steady_iterations) != 0) { + reg->hr_aborted_start = 1; atomic_set(®->hr_steady_iterations, 0); wake_up(&o2hb_steady_queue); } - if (o2hb_global_heartbeat_active()) - printk(KERN_NOTICE "o2hb: Heartbeat stopped on region %s\n", - config_item_name(®->hr_item)); - config_item_put(item); if (!o2hb_global_heartbeat_active() || !quorum_region) diff --git a/fs/ocfs2/cluster/netdebug.c b/fs/ocfs2/cluster/netdebug.c index 3a5835904b3d..73ba81928bce 100644 --- a/fs/ocfs2/cluster/netdebug.c +++ b/fs/ocfs2/cluster/netdebug.c @@ -47,6 +47,7 @@ #define SC_DEBUG_NAME "sock_containers" #define NST_DEBUG_NAME "send_tracking" #define STATS_DEBUG_NAME "stats" +#define NODES_DEBUG_NAME "connected_nodes" #define SHOW_SOCK_CONTAINERS 0 #define SHOW_SOCK_STATS 1 @@ -55,6 +56,7 @@ static struct dentry *o2net_dentry; static struct dentry *sc_dentry; static struct dentry *nst_dentry; static struct dentry *stats_dentry; +static struct dentry *nodes_dentry; static DEFINE_SPINLOCK(o2net_debug_lock); @@ -491,53 +493,87 @@ static const struct file_operations sc_seq_fops = { .release = sc_fop_release, }; -int o2net_debugfs_init(void) +static int o2net_fill_bitmap(char *buf, int len) { - o2net_dentry = debugfs_create_dir(O2NET_DEBUG_DIR, NULL); - if (!o2net_dentry) { - mlog_errno(-ENOMEM); - goto bail; - } + unsigned long map[BITS_TO_LONGS(O2NM_MAX_NODES)]; + int i = -1, out = 0; - nst_dentry = debugfs_create_file(NST_DEBUG_NAME, S_IFREG|S_IRUSR, - o2net_dentry, NULL, - &nst_seq_fops); - if (!nst_dentry) { - mlog_errno(-ENOMEM); - goto bail; - } + o2net_fill_node_map(map, sizeof(map)); - sc_dentry = debugfs_create_file(SC_DEBUG_NAME, S_IFREG|S_IRUSR, - o2net_dentry, NULL, - &sc_seq_fops); - if (!sc_dentry) { - mlog_errno(-ENOMEM); - goto bail; - } + while ((i = find_next_bit(map, O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) + out += snprintf(buf + out, PAGE_SIZE - out, "%d ", i); + out += snprintf(buf + out, PAGE_SIZE - out, "\n"); - stats_dentry = debugfs_create_file(STATS_DEBUG_NAME, S_IFREG|S_IRUSR, - o2net_dentry, NULL, - &stats_seq_fops); - if (!stats_dentry) { - mlog_errno(-ENOMEM); - goto bail; - } + return out; +} + +static int nodes_fop_open(struct inode *inode, struct file *file) +{ + char *buf; + + buf = kmalloc(PAGE_SIZE, GFP_KERNEL); + if (!buf) + return -ENOMEM; + + i_size_write(inode, o2net_fill_bitmap(buf, PAGE_SIZE)); + + file->private_data = buf; return 0; -bail: - debugfs_remove(stats_dentry); - debugfs_remove(sc_dentry); - debugfs_remove(nst_dentry); - debugfs_remove(o2net_dentry); - return -ENOMEM; } +static int o2net_debug_release(struct inode *inode, struct file *file) +{ + kfree(file->private_data); + return 0; +} + +static ssize_t o2net_debug_read(struct file *file, char __user *buf, + size_t nbytes, loff_t *ppos) +{ + return simple_read_from_buffer(buf, nbytes, ppos, file->private_data, + i_size_read(file->f_mapping->host)); +} + +static const struct file_operations nodes_fops = { + .open = nodes_fop_open, + .release = o2net_debug_release, + .read = o2net_debug_read, + .llseek = generic_file_llseek, +}; + void o2net_debugfs_exit(void) { + debugfs_remove(nodes_dentry); debugfs_remove(stats_dentry); debugfs_remove(sc_dentry); debugfs_remove(nst_dentry); debugfs_remove(o2net_dentry); } +int o2net_debugfs_init(void) +{ + umode_t mode = S_IFREG|S_IRUSR; + + o2net_dentry = debugfs_create_dir(O2NET_DEBUG_DIR, NULL); + if (o2net_dentry) + nst_dentry = debugfs_create_file(NST_DEBUG_NAME, mode, + o2net_dentry, NULL, &nst_seq_fops); + if (nst_dentry) + sc_dentry = debugfs_create_file(SC_DEBUG_NAME, mode, + o2net_dentry, NULL, &sc_seq_fops); + if (sc_dentry) + stats_dentry = debugfs_create_file(STATS_DEBUG_NAME, mode, + o2net_dentry, NULL, &stats_seq_fops); + if (stats_dentry) + nodes_dentry = debugfs_create_file(NODES_DEBUG_NAME, mode, + o2net_dentry, NULL, &nodes_fops); + if (nodes_dentry) + return 0; + + o2net_debugfs_exit(); + mlog_errno(-ENOMEM); + return -ENOMEM; +} + #endif /* CONFIG_DEBUG_FS */ diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c index ad7d0c155de4..044e7b58d31c 100644 --- a/fs/ocfs2/cluster/tcp.c +++ b/fs/ocfs2/cluster/tcp.c @@ -546,7 +546,7 @@ static void o2net_set_nn_state(struct o2net_node *nn, } if (was_valid && !valid) { - printk(KERN_NOTICE "o2net: no longer connected to " + printk(KERN_NOTICE "o2net: No longer connected to " SC_NODEF_FMT "\n", SC_NODEF_ARGS(old_sc)); o2net_complete_nodes_nsw(nn); } @@ -556,7 +556,7 @@ static void o2net_set_nn_state(struct o2net_node *nn, cancel_delayed_work(&nn->nn_connect_expired); printk(KERN_NOTICE "o2net: %s " SC_NODEF_FMT "\n", o2nm_this_node() > sc->sc_node->nd_num ? - "connected to" : "accepted connection from", + "Connected to" : "Accepted connection from", SC_NODEF_ARGS(sc)); } @@ -644,7 +644,7 @@ static void o2net_state_change(struct sock *sk) o2net_sc_queue_work(sc, &sc->sc_connect_work); break; default: - printk(KERN_INFO "o2net: connection to " SC_NODEF_FMT + printk(KERN_INFO "o2net: Connection to " SC_NODEF_FMT " shutdown, state %d\n", SC_NODEF_ARGS(sc), sk->sk_state); o2net_sc_queue_work(sc, &sc->sc_shutdown_work); @@ -1035,6 +1035,25 @@ static int o2net_tx_can_proceed(struct o2net_node *nn, return ret; } +/* Get a map of all nodes to which this node is currently connected to */ +void o2net_fill_node_map(unsigned long *map, unsigned bytes) +{ + struct o2net_sock_container *sc; + int node, ret; + + BUG_ON(bytes < (BITS_TO_LONGS(O2NM_MAX_NODES) * sizeof(unsigned long))); + + memset(map, 0, bytes); + for (node = 0; node < O2NM_MAX_NODES; ++node) { + o2net_tx_can_proceed(o2net_nn_from_num(node), &sc, &ret); + if (!ret) { + set_bit(node, map); + sc_put(sc); + } + } +} +EXPORT_SYMBOL_GPL(o2net_fill_node_map); + int o2net_send_message_vec(u32 msg_type, u32 key, struct kvec *caller_vec, size_t caller_veclen, u8 target_node, int *status) { @@ -1285,11 +1304,11 @@ static int o2net_check_handshake(struct o2net_sock_container *sc) struct o2net_node *nn = o2net_nn_from_num(sc->sc_node->nd_num); if (hand->protocol_version != cpu_to_be64(O2NET_PROTOCOL_VERSION)) { - mlog(ML_NOTICE, SC_NODEF_FMT " advertised net protocol " - "version %llu but %llu is required, disconnecting\n", - SC_NODEF_ARGS(sc), - (unsigned long long)be64_to_cpu(hand->protocol_version), - O2NET_PROTOCOL_VERSION); + printk(KERN_NOTICE "o2net: " SC_NODEF_FMT " Advertised net " + "protocol version %llu but %llu is required. " + "Disconnecting.\n", SC_NODEF_ARGS(sc), + (unsigned long long)be64_to_cpu(hand->protocol_version), + O2NET_PROTOCOL_VERSION); /* don't bother reconnecting if its the wrong version. */ o2net_ensure_shutdown(nn, sc, -ENOTCONN); @@ -1303,33 +1322,33 @@ static int o2net_check_handshake(struct o2net_sock_container *sc) */ if (be32_to_cpu(hand->o2net_idle_timeout_ms) != o2net_idle_timeout()) { - mlog(ML_NOTICE, SC_NODEF_FMT " uses a network idle timeout of " - "%u ms, but we use %u ms locally. disconnecting\n", - SC_NODEF_ARGS(sc), - be32_to_cpu(hand->o2net_idle_timeout_ms), - o2net_idle_timeout()); + printk(KERN_NOTICE "o2net: " SC_NODEF_FMT " uses a network " + "idle timeout of %u ms, but we use %u ms locally. " + "Disconnecting.\n", SC_NODEF_ARGS(sc), + be32_to_cpu(hand->o2net_idle_timeout_ms), + o2net_idle_timeout()); o2net_ensure_shutdown(nn, sc, -ENOTCONN); return -1; } if (be32_to_cpu(hand->o2net_keepalive_delay_ms) != o2net_keepalive_delay()) { - mlog(ML_NOTICE, SC_NODEF_FMT " uses a keepalive delay of " - "%u ms, but we use %u ms locally. disconnecting\n", - SC_NODEF_ARGS(sc), - be32_to_cpu(hand->o2net_keepalive_delay_ms), - o2net_keepalive_delay()); + printk(KERN_NOTICE "o2net: " SC_NODEF_FMT " uses a keepalive " + "delay of %u ms, but we use %u ms locally. " + "Disconnecting.\n", SC_NODEF_ARGS(sc), + be32_to_cpu(hand->o2net_keepalive_delay_ms), + o2net_keepalive_delay()); o2net_ensure_shutdown(nn, sc, -ENOTCONN); return -1; } if (be32_to_cpu(hand->o2hb_heartbeat_timeout_ms) != O2HB_MAX_WRITE_TIMEOUT_MS) { - mlog(ML_NOTICE, SC_NODEF_FMT " uses a heartbeat timeout of " - "%u ms, but we use %u ms locally. disconnecting\n", - SC_NODEF_ARGS(sc), - be32_to_cpu(hand->o2hb_heartbeat_timeout_ms), - O2HB_MAX_WRITE_TIMEOUT_MS); + printk(KERN_NOTICE "o2net: " SC_NODEF_FMT " uses a heartbeat " + "timeout of %u ms, but we use %u ms locally. " + "Disconnecting.\n", SC_NODEF_ARGS(sc), + be32_to_cpu(hand->o2hb_heartbeat_timeout_ms), + O2HB_MAX_WRITE_TIMEOUT_MS); o2net_ensure_shutdown(nn, sc, -ENOTCONN); return -1; } @@ -1540,28 +1559,16 @@ static void o2net_idle_timer(unsigned long data) { struct o2net_sock_container *sc = (struct o2net_sock_container *)data; struct o2net_node *nn = o2net_nn_from_num(sc->sc_node->nd_num); - #ifdef CONFIG_DEBUG_FS - ktime_t now = ktime_get(); + unsigned long msecs = ktime_to_ms(ktime_get()) - + ktime_to_ms(sc->sc_tv_timer); +#else + unsigned long msecs = o2net_idle_timeout(); #endif - printk(KERN_NOTICE "o2net: connection to " SC_NODEF_FMT " has been idle for %u.%u " - "seconds, shutting it down.\n", SC_NODEF_ARGS(sc), - o2net_idle_timeout() / 1000, - o2net_idle_timeout() % 1000); - -#ifdef CONFIG_DEBUG_FS - mlog(ML_NOTICE, "Here are some times that might help debug the " - "situation: (Timer: %lld, Now %lld, DataReady %lld, Advance %lld-%lld, " - "Key 0x%08x, Func %u, FuncTime %lld-%lld)\n", - (long long)ktime_to_us(sc->sc_tv_timer), (long long)ktime_to_us(now), - (long long)ktime_to_us(sc->sc_tv_data_ready), - (long long)ktime_to_us(sc->sc_tv_advance_start), - (long long)ktime_to_us(sc->sc_tv_advance_stop), - sc->sc_msg_key, sc->sc_msg_type, - (long long)ktime_to_us(sc->sc_tv_func_start), - (long long)ktime_to_us(sc->sc_tv_func_stop)); -#endif + printk(KERN_NOTICE "o2net: Connection to " SC_NODEF_FMT " has been " + "idle for %lu.%lu secs, shutting it down.\n", SC_NODEF_ARGS(sc), + msecs / 1000, msecs % 1000); /* * Initialize the nn_timeout so that the next connection attempt @@ -1694,8 +1701,8 @@ static void o2net_start_connect(struct work_struct *work) out: if (ret) { - mlog(ML_NOTICE, "connect attempt to " SC_NODEF_FMT " failed " - "with errno %d\n", SC_NODEF_ARGS(sc), ret); + printk(KERN_NOTICE "o2net: Connect attempt to " SC_NODEF_FMT + " failed with errno %d\n", SC_NODEF_ARGS(sc), ret); /* 0 err so that another will be queued and attempted * from set_nn_state */ if (sc) @@ -1718,8 +1725,8 @@ static void o2net_connect_expired(struct work_struct *work) spin_lock(&nn->nn_lock); if (!nn->nn_sc_valid) { - mlog(ML_ERROR, "no connection established with node %u after " - "%u.%u seconds, giving up and returning errors.\n", + printk(KERN_NOTICE "o2net: No connection established with " + "node %u after %u.%u seconds, giving up.\n", o2net_num_from_nn(nn), o2net_idle_timeout() / 1000, o2net_idle_timeout() % 1000); @@ -1862,21 +1869,21 @@ static int o2net_accept_one(struct socket *sock) node = o2nm_get_node_by_ip(sin.sin_addr.s_addr); if (node == NULL) { - mlog(ML_NOTICE, "attempt to connect from unknown node at %pI4:%d\n", - &sin.sin_addr.s_addr, ntohs(sin.sin_port)); + printk(KERN_NOTICE "o2net: Attempt to connect from unknown " + "node at %pI4:%d\n", &sin.sin_addr.s_addr, + ntohs(sin.sin_port)); ret = -EINVAL; goto out; } if (o2nm_this_node() >= node->nd_num) { local_node = o2nm_get_node_by_num(o2nm_this_node()); - mlog(ML_NOTICE, "unexpected connect attempt seen at node '%s' (" - "%u, %pI4:%d) from node '%s' (%u, %pI4:%d)\n", - local_node->nd_name, local_node->nd_num, - &(local_node->nd_ipv4_address), - ntohs(local_node->nd_ipv4_port), - node->nd_name, node->nd_num, &sin.sin_addr.s_addr, - ntohs(sin.sin_port)); + printk(KERN_NOTICE "o2net: Unexpected connect attempt seen " + "at node '%s' (%u, %pI4:%d) from node '%s' (%u, " + "%pI4:%d)\n", local_node->nd_name, local_node->nd_num, + &(local_node->nd_ipv4_address), + ntohs(local_node->nd_ipv4_port), node->nd_name, + node->nd_num, &sin.sin_addr.s_addr, ntohs(sin.sin_port)); ret = -EINVAL; goto out; } @@ -1901,10 +1908,10 @@ static int o2net_accept_one(struct socket *sock) ret = 0; spin_unlock(&nn->nn_lock); if (ret) { - mlog(ML_NOTICE, "attempt to connect from node '%s' at " - "%pI4:%d but it already has an open connection\n", - node->nd_name, &sin.sin_addr.s_addr, - ntohs(sin.sin_port)); + printk(KERN_NOTICE "o2net: Attempt to connect from node '%s' " + "at %pI4:%d but it already has an open connection\n", + node->nd_name, &sin.sin_addr.s_addr, + ntohs(sin.sin_port)); goto out; } @@ -1984,7 +1991,7 @@ static int o2net_open_listening_sock(__be32 addr, __be16 port) ret = sock_create(PF_INET, SOCK_STREAM, IPPROTO_TCP, &sock); if (ret < 0) { - mlog(ML_ERROR, "unable to create socket, ret=%d\n", ret); + printk(KERN_ERR "o2net: Error %d while creating socket\n", ret); goto out; } @@ -2001,16 +2008,15 @@ static int o2net_open_listening_sock(__be32 addr, __be16 port) sock->sk->sk_reuse = 1; ret = sock->ops->bind(sock, (struct sockaddr *)&sin, sizeof(sin)); if (ret < 0) { - mlog(ML_ERROR, "unable to bind socket at %pI4:%u, " - "ret=%d\n", &addr, ntohs(port), ret); + printk(KERN_ERR "o2net: Error %d while binding socket at " + "%pI4:%u\n", ret, &addr, ntohs(port)); goto out; } ret = sock->ops->listen(sock, 64); - if (ret < 0) { - mlog(ML_ERROR, "unable to listen on %pI4:%u, ret=%d\n", - &addr, ntohs(port), ret); - } + if (ret < 0) + printk(KERN_ERR "o2net: Error %d while listening on %pI4:%u\n", + ret, &addr, ntohs(port)); out: if (ret) { diff --git a/fs/ocfs2/cluster/tcp.h b/fs/ocfs2/cluster/tcp.h index fd6179eb26d4..5bada2a69b50 100644 --- a/fs/ocfs2/cluster/tcp.h +++ b/fs/ocfs2/cluster/tcp.h @@ -106,6 +106,8 @@ int o2net_register_handler(u32 msg_type, u32 key, u32 max_len, struct list_head *unreg_list); void o2net_unregister_handler_list(struct list_head *list); +void o2net_fill_node_map(unsigned long *map, unsigned bytes); + struct o2nm_node; int o2net_register_hb_callbacks(void); void o2net_unregister_hb_callbacks(void); diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c index e2878b5895fb..8fe4e2892ab9 100644 --- a/fs/ocfs2/dir.c +++ b/fs/ocfs2/dir.c @@ -1184,8 +1184,7 @@ static int __ocfs2_delete_entry(handle_t *handle, struct inode *dir, if (pde) le16_add_cpu(&pde->rec_len, le16_to_cpu(de->rec_len)); - else - de->inode = 0; + de->inode = 0; dir->i_version++; ocfs2_journal_dirty(handle, bh); goto bail; diff --git a/fs/ocfs2/dlm/dlmcommon.h b/fs/ocfs2/dlm/dlmcommon.h index d602abb51b61..a5952ceecba5 100644 --- a/fs/ocfs2/dlm/dlmcommon.h +++ b/fs/ocfs2/dlm/dlmcommon.h @@ -859,8 +859,8 @@ void dlm_complete_recovery_thread(struct dlm_ctxt *dlm); void dlm_wait_for_recovery(struct dlm_ctxt *dlm); void dlm_kick_recovery_thread(struct dlm_ctxt *dlm); int dlm_is_node_dead(struct dlm_ctxt *dlm, u8 node); -int dlm_wait_for_node_death(struct dlm_ctxt *dlm, u8 node, int timeout); -int dlm_wait_for_node_recovery(struct dlm_ctxt *dlm, u8 node, int timeout); +void dlm_wait_for_node_death(struct dlm_ctxt *dlm, u8 node, int timeout); +void dlm_wait_for_node_recovery(struct dlm_ctxt *dlm, u8 node, int timeout); void dlm_put(struct dlm_ctxt *dlm); struct dlm_ctxt *dlm_grab(struct dlm_ctxt *dlm); @@ -877,9 +877,8 @@ static inline void dlm_lockres_get(struct dlm_lock_resource *res) kref_get(&res->refs); } void dlm_lockres_put(struct dlm_lock_resource *res); -void __dlm_unhash_lockres(struct dlm_lock_resource *res); -void __dlm_insert_lockres(struct dlm_ctxt *dlm, - struct dlm_lock_resource *res); +void __dlm_unhash_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res); +void __dlm_insert_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res); struct dlm_lock_resource * __dlm_lookup_lockres_full(struct dlm_ctxt *dlm, const char *name, unsigned int len, @@ -902,46 +901,15 @@ struct dlm_lock_resource *dlm_new_lockres(struct dlm_ctxt *dlm, const char *name, unsigned int namelen); -#define dlm_lockres_set_refmap_bit(bit,res) \ - __dlm_lockres_set_refmap_bit(bit,res,__FILE__,__LINE__) -#define dlm_lockres_clear_refmap_bit(bit,res) \ - __dlm_lockres_clear_refmap_bit(bit,res,__FILE__,__LINE__) +void dlm_lockres_set_refmap_bit(struct dlm_ctxt *dlm, + struct dlm_lock_resource *res, int bit); +void dlm_lockres_clear_refmap_bit(struct dlm_ctxt *dlm, + struct dlm_lock_resource *res, int bit); -static inline void __dlm_lockres_set_refmap_bit(int bit, - struct dlm_lock_resource *res, - const char *file, - int line) -{ - //printk("%s:%d:%.*s: setting bit %d\n", file, line, - // res->lockname.len, res->lockname.name, bit); - set_bit(bit, res->refmap); -} - -static inline void __dlm_lockres_clear_refmap_bit(int bit, - struct dlm_lock_resource *res, - const char *file, - int line) -{ - //printk("%s:%d:%.*s: clearing bit %d\n", file, line, - // res->lockname.len, res->lockname.name, bit); - clear_bit(bit, res->refmap); -} - -void __dlm_lockres_drop_inflight_ref(struct dlm_ctxt *dlm, - struct dlm_lock_resource *res, - const char *file, - int line); -void __dlm_lockres_grab_inflight_ref(struct dlm_ctxt *dlm, - struct dlm_lock_resource *res, - int new_lockres, - const char *file, - int line); -#define dlm_lockres_drop_inflight_ref(d,r) \ - __dlm_lockres_drop_inflight_ref(d,r,__FILE__,__LINE__) -#define dlm_lockres_grab_inflight_ref(d,r) \ - __dlm_lockres_grab_inflight_ref(d,r,0,__FILE__,__LINE__) -#define dlm_lockres_grab_inflight_ref_new(d,r) \ - __dlm_lockres_grab_inflight_ref(d,r,1,__FILE__,__LINE__) +void dlm_lockres_drop_inflight_ref(struct dlm_ctxt *dlm, + struct dlm_lock_resource *res); +void dlm_lockres_grab_inflight_ref(struct dlm_ctxt *dlm, + struct dlm_lock_resource *res); void dlm_queue_ast(struct dlm_ctxt *dlm, struct dlm_lock *lock); void dlm_queue_bast(struct dlm_ctxt *dlm, struct dlm_lock *lock); diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c index 6ed6b95dcf93..92f2ead0fab6 100644 --- a/fs/ocfs2/dlm/dlmdomain.c +++ b/fs/ocfs2/dlm/dlmdomain.c @@ -157,16 +157,18 @@ static int dlm_protocol_compare(struct dlm_protocol_version *existing, static void dlm_unregister_domain_handlers(struct dlm_ctxt *dlm); -void __dlm_unhash_lockres(struct dlm_lock_resource *lockres) +void __dlm_unhash_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) { - if (!hlist_unhashed(&lockres->hash_node)) { - hlist_del_init(&lockres->hash_node); - dlm_lockres_put(lockres); - } + if (hlist_unhashed(&res->hash_node)) + return; + + mlog(0, "%s: Unhash res %.*s\n", dlm->name, res->lockname.len, + res->lockname.name); + hlist_del_init(&res->hash_node); + dlm_lockres_put(res); } -void __dlm_insert_lockres(struct dlm_ctxt *dlm, - struct dlm_lock_resource *res) +void __dlm_insert_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) { struct hlist_head *bucket; struct qstr *q; @@ -180,6 +182,9 @@ void __dlm_insert_lockres(struct dlm_ctxt *dlm, dlm_lockres_get(res); hlist_add_head(&res->hash_node, bucket); + + mlog(0, "%s: Hash res %.*s\n", dlm->name, res->lockname.len, + res->lockname.name); } struct dlm_lock_resource * __dlm_lookup_lockres_full(struct dlm_ctxt *dlm, @@ -539,17 +544,17 @@ again: static void __dlm_print_nodes(struct dlm_ctxt *dlm) { - int node = -1; + int node = -1, num = 0; assert_spin_locked(&dlm->spinlock); - printk(KERN_NOTICE "o2dlm: Nodes in domain %s: ", dlm->name); - + printk("( "); while ((node = find_next_bit(dlm->domain_map, O2NM_MAX_NODES, node + 1)) < O2NM_MAX_NODES) { printk("%d ", node); + ++num; } - printk("\n"); + printk(") %u nodes\n", num); } static int dlm_exit_domain_handler(struct o2net_msg *msg, u32 len, void *data, @@ -566,11 +571,10 @@ static int dlm_exit_domain_handler(struct o2net_msg *msg, u32 len, void *data, node = exit_msg->node_idx; - printk(KERN_NOTICE "o2dlm: Node %u leaves domain %s\n", node, dlm->name); - spin_lock(&dlm->spinlock); clear_bit(node, dlm->domain_map); clear_bit(node, dlm->exit_domain_map); + printk(KERN_NOTICE "o2dlm: Node %u leaves domain %s ", node, dlm->name); __dlm_print_nodes(dlm); /* notify anything attached to the heartbeat events */ @@ -755,6 +759,7 @@ void dlm_unregister_domain(struct dlm_ctxt *dlm) dlm_mark_domain_leaving(dlm); dlm_leave_domain(dlm); + printk(KERN_NOTICE "o2dlm: Leaving domain %s\n", dlm->name); dlm_force_free_mles(dlm); dlm_complete_dlm_shutdown(dlm); } @@ -970,7 +975,7 @@ static int dlm_assert_joined_handler(struct o2net_msg *msg, u32 len, void *data, clear_bit(assert->node_idx, dlm->exit_domain_map); __dlm_set_joining_node(dlm, DLM_LOCK_RES_OWNER_UNKNOWN); - printk(KERN_NOTICE "o2dlm: Node %u joins domain %s\n", + printk(KERN_NOTICE "o2dlm: Node %u joins domain %s ", assert->node_idx, dlm->name); __dlm_print_nodes(dlm); @@ -1701,8 +1706,10 @@ static int dlm_try_to_join_domain(struct dlm_ctxt *dlm) bail: spin_lock(&dlm->spinlock); __dlm_set_joining_node(dlm, DLM_LOCK_RES_OWNER_UNKNOWN); - if (!status) + if (!status) { + printk(KERN_NOTICE "o2dlm: Joining domain %s ", dlm->name); __dlm_print_nodes(dlm); + } spin_unlock(&dlm->spinlock); if (ctxt) { @@ -2131,13 +2138,6 @@ struct dlm_ctxt * dlm_register_domain(const char *domain, goto leave; } - if (!o2hb_check_local_node_heartbeating()) { - mlog(ML_ERROR, "the local node has not been configured, or is " - "not heartbeating\n"); - ret = -EPROTO; - goto leave; - } - mlog(0, "register called for domain \"%s\"\n", domain); retry: diff --git a/fs/ocfs2/dlm/dlmlock.c b/fs/ocfs2/dlm/dlmlock.c index 8d39e0fd66f7..975810b98492 100644 --- a/fs/ocfs2/dlm/dlmlock.c +++ b/fs/ocfs2/dlm/dlmlock.c @@ -183,10 +183,6 @@ static enum dlm_status dlmlock_master(struct dlm_ctxt *dlm, kick_thread = 1; } } - /* reduce the inflight count, this may result in the lockres - * being purged below during calc_usage */ - if (lock->ml.node == dlm->node_num) - dlm_lockres_drop_inflight_ref(dlm, res); spin_unlock(&res->spinlock); wake_up(&res->wq); @@ -231,10 +227,16 @@ static enum dlm_status dlmlock_remote(struct dlm_ctxt *dlm, lock->ml.type, res->lockname.len, res->lockname.name, flags); + /* + * Wait if resource is getting recovered, remastered, etc. + * If the resource was remastered and new owner is self, then exit. + */ spin_lock(&res->spinlock); - - /* will exit this call with spinlock held */ __dlm_wait_on_lockres(res); + if (res->owner == dlm->node_num) { + spin_unlock(&res->spinlock); + return DLM_RECOVERING; + } res->state |= DLM_LOCK_RES_IN_PROGRESS; /* add lock to local (secondary) queue */ @@ -319,27 +321,23 @@ static enum dlm_status dlm_send_remote_lock_request(struct dlm_ctxt *dlm, tmpret = o2net_send_message(DLM_CREATE_LOCK_MSG, dlm->key, &create, sizeof(create), res->owner, &status); if (tmpret >= 0) { - // successfully sent and received - ret = status; // this is already a dlm_status + ret = status; if (ret == DLM_REJECTED) { - mlog(ML_ERROR, "%s:%.*s: BUG. this is a stale lockres " - "no longer owned by %u. that node is coming back " - "up currently.\n", dlm->name, create.namelen, + mlog(ML_ERROR, "%s: res %.*s, Stale lockres no longer " + "owned by node %u. That node is coming back up " + "currently.\n", dlm->name, create.namelen, create.name, res->owner); dlm_print_one_lock_resource(res); BUG(); } } else { - mlog(ML_ERROR, "Error %d when sending message %u (key 0x%x) to " - "node %u\n", tmpret, DLM_CREATE_LOCK_MSG, dlm->key, - res->owner); - if (dlm_is_host_down(tmpret)) { + mlog(ML_ERROR, "%s: res %.*s, Error %d send CREATE LOCK to " + "node %u\n", dlm->name, create.namelen, create.name, + tmpret, res->owner); + if (dlm_is_host_down(tmpret)) ret = DLM_RECOVERING; - mlog(0, "node %u died so returning DLM_RECOVERING " - "from lock message!\n", res->owner); - } else { + else ret = dlm_err_to_dlm_status(tmpret); - } } return ret; @@ -440,7 +438,7 @@ struct dlm_lock * dlm_new_lock(int type, u8 node, u64 cookie, /* zero memory only if kernel-allocated */ lksb = kzalloc(sizeof(*lksb), GFP_NOFS); if (!lksb) { - kfree(lock); + kmem_cache_free(dlm_lock_cache, lock); return NULL; } kernel_allocated = 1; @@ -718,18 +716,10 @@ retry_lock: if (status == DLM_RECOVERING || status == DLM_MIGRATING || status == DLM_FORWARD) { - mlog(0, "retrying lock with migration/" - "recovery/in progress\n"); msleep(100); - /* no waiting for dlm_reco_thread */ if (recovery) { if (status != DLM_RECOVERING) goto retry_lock; - - mlog(0, "%s: got RECOVERING " - "for $RECOVERY lock, master " - "was %u\n", dlm->name, - res->owner); /* wait to see the node go down, then * drop down and allow the lockres to * get cleaned up. need to remaster. */ @@ -741,6 +731,14 @@ retry_lock: } } + /* Inflight taken in dlm_get_lock_resource() is dropped here */ + spin_lock(&res->spinlock); + dlm_lockres_drop_inflight_ref(dlm, res); + spin_unlock(&res->spinlock); + + dlm_lockres_calc_usage(dlm, res); + dlm_kick_thread(dlm, res); + if (status != DLM_NORMAL) { lock->lksb->flags &= ~DLM_LKSB_GET_LVB; if (status != DLM_NOTQUEUED) diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index 11eefb8c12e9..005261c333b0 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -631,39 +631,54 @@ error: return NULL; } -void __dlm_lockres_grab_inflight_ref(struct dlm_ctxt *dlm, - struct dlm_lock_resource *res, - int new_lockres, - const char *file, - int line) +void dlm_lockres_set_refmap_bit(struct dlm_ctxt *dlm, + struct dlm_lock_resource *res, int bit) { - if (!new_lockres) - assert_spin_locked(&res->spinlock); + assert_spin_locked(&res->spinlock); + + mlog(0, "res %.*s, set node %u, %ps()\n", res->lockname.len, + res->lockname.name, bit, __builtin_return_address(0)); + + set_bit(bit, res->refmap); +} + +void dlm_lockres_clear_refmap_bit(struct dlm_ctxt *dlm, + struct dlm_lock_resource *res, int bit) +{ + assert_spin_locked(&res->spinlock); + + mlog(0, "res %.*s, clr node %u, %ps()\n", res->lockname.len, + res->lockname.name, bit, __builtin_return_address(0)); + + clear_bit(bit, res->refmap); +} + + +void dlm_lockres_grab_inflight_ref(struct dlm_ctxt *dlm, + struct dlm_lock_resource *res) +{ + assert_spin_locked(&res->spinlock); - if (!test_bit(dlm->node_num, res->refmap)) { - BUG_ON(res->inflight_locks != 0); - dlm_lockres_set_refmap_bit(dlm->node_num, res); - } res->inflight_locks++; - mlog(0, "%s:%.*s: inflight++: now %u\n", - dlm->name, res->lockname.len, res->lockname.name, - res->inflight_locks); + + mlog(0, "%s: res %.*s, inflight++: now %u, %ps()\n", dlm->name, + res->lockname.len, res->lockname.name, res->inflight_locks, + __builtin_return_address(0)); } -void __dlm_lockres_drop_inflight_ref(struct dlm_ctxt *dlm, - struct dlm_lock_resource *res, - const char *file, - int line) +void dlm_lockres_drop_inflight_ref(struct dlm_ctxt *dlm, + struct dlm_lock_resource *res) { assert_spin_locked(&res->spinlock); BUG_ON(res->inflight_locks == 0); + res->inflight_locks--; - mlog(0, "%s:%.*s: inflight--: now %u\n", - dlm->name, res->lockname.len, res->lockname.name, - res->inflight_locks); - if (res->inflight_locks == 0) - dlm_lockres_clear_refmap_bit(dlm->node_num, res); + + mlog(0, "%s: res %.*s, inflight--: now %u, %ps()\n", dlm->name, + res->lockname.len, res->lockname.name, res->inflight_locks, + __builtin_return_address(0)); + wake_up(&res->wq); } @@ -697,7 +712,6 @@ struct dlm_lock_resource * dlm_get_lock_resource(struct dlm_ctxt *dlm, unsigned int hash; int tries = 0; int bit, wait_on_recovery = 0; - int drop_inflight_if_nonlocal = 0; BUG_ON(!lockid); @@ -709,36 +723,33 @@ lookup: spin_lock(&dlm->spinlock); tmpres = __dlm_lookup_lockres_full(dlm, lockid, namelen, hash); if (tmpres) { - int dropping_ref = 0; - spin_unlock(&dlm->spinlock); - spin_lock(&tmpres->spinlock); - /* We wait for the other thread that is mastering the resource */ + /* Wait on the thread that is mastering the resource */ if (tmpres->owner == DLM_LOCK_RES_OWNER_UNKNOWN) { __dlm_wait_on_lockres(tmpres); BUG_ON(tmpres->owner == DLM_LOCK_RES_OWNER_UNKNOWN); + spin_unlock(&tmpres->spinlock); + dlm_lockres_put(tmpres); + tmpres = NULL; + goto lookup; } - if (tmpres->owner == dlm->node_num) { - BUG_ON(tmpres->state & DLM_LOCK_RES_DROPPING_REF); - dlm_lockres_grab_inflight_ref(dlm, tmpres); - } else if (tmpres->state & DLM_LOCK_RES_DROPPING_REF) - dropping_ref = 1; - spin_unlock(&tmpres->spinlock); - - /* wait until done messaging the master, drop our ref to allow - * the lockres to be purged, start over. */ - if (dropping_ref) { - spin_lock(&tmpres->spinlock); - __dlm_wait_on_lockres_flags(tmpres, DLM_LOCK_RES_DROPPING_REF); + /* Wait on the resource purge to complete before continuing */ + if (tmpres->state & DLM_LOCK_RES_DROPPING_REF) { + BUG_ON(tmpres->owner == dlm->node_num); + __dlm_wait_on_lockres_flags(tmpres, + DLM_LOCK_RES_DROPPING_REF); spin_unlock(&tmpres->spinlock); dlm_lockres_put(tmpres); tmpres = NULL; goto lookup; } - mlog(0, "found in hash!\n"); + /* Grab inflight ref to pin the resource */ + dlm_lockres_grab_inflight_ref(dlm, tmpres); + + spin_unlock(&tmpres->spinlock); if (res) dlm_lockres_put(res); res = tmpres; @@ -829,8 +840,8 @@ lookup: * but they might own this lockres. wait on them. */ bit = find_next_bit(dlm->recovery_map, O2NM_MAX_NODES, 0); if (bit < O2NM_MAX_NODES) { - mlog(ML_NOTICE, "%s:%.*s: at least one node (%d) to " - "recover before lock mastery can begin\n", + mlog(0, "%s: res %.*s, At least one node (%d) " + "to recover before lock mastery can begin\n", dlm->name, namelen, (char *)lockid, bit); wait_on_recovery = 1; } @@ -843,12 +854,11 @@ lookup: /* finally add the lockres to its hash bucket */ __dlm_insert_lockres(dlm, res); - /* since this lockres is new it doesn't not require the spinlock */ - dlm_lockres_grab_inflight_ref_new(dlm, res); - /* if this node does not become the master make sure to drop - * this inflight reference below */ - drop_inflight_if_nonlocal = 1; + /* Grab inflight ref to pin the resource */ + spin_lock(&res->spinlock); + dlm_lockres_grab_inflight_ref(dlm, res); + spin_unlock(&res->spinlock); /* get an extra ref on the mle in case this is a BLOCK * if so, the creator of the BLOCK may try to put the last @@ -864,8 +874,8 @@ redo_request: * dlm spinlock would be detectable be a change on the mle, * so we only need to clear out the recovery map once. */ if (dlm_is_recovery_lock(lockid, namelen)) { - mlog(ML_NOTICE, "%s: recovery map is not empty, but " - "must master $RECOVERY lock now\n", dlm->name); + mlog(0, "%s: Recovery map is not empty, but must " + "master $RECOVERY lock now\n", dlm->name); if (!dlm_pre_master_reco_lockres(dlm, res)) wait_on_recovery = 0; else { @@ -883,8 +893,8 @@ redo_request: spin_lock(&dlm->spinlock); bit = find_next_bit(dlm->recovery_map, O2NM_MAX_NODES, 0); if (bit < O2NM_MAX_NODES) { - mlog(ML_NOTICE, "%s:%.*s: at least one node (%d) to " - "recover before lock mastery can begin\n", + mlog(0, "%s: res %.*s, At least one node (%d) " + "to recover before lock mastery can begin\n", dlm->name, namelen, (char *)lockid, bit); wait_on_recovery = 1; } else @@ -913,8 +923,8 @@ redo_request: * yet, keep going until it does. this is how the * master will know that asserts are needed back to * the lower nodes. */ - mlog(0, "%s:%.*s: requests only up to %u but master " - "is %u, keep going\n", dlm->name, namelen, + mlog(0, "%s: res %.*s, Requests only up to %u but " + "master is %u, keep going\n", dlm->name, namelen, lockid, nodenum, mle->master); } } @@ -924,13 +934,12 @@ wait: ret = dlm_wait_for_lock_mastery(dlm, res, mle, &blocked); if (ret < 0) { wait_on_recovery = 1; - mlog(0, "%s:%.*s: node map changed, redo the " - "master request now, blocked=%d\n", - dlm->name, res->lockname.len, + mlog(0, "%s: res %.*s, Node map changed, redo the master " + "request now, blocked=%d\n", dlm->name, res->lockname.len, res->lockname.name, blocked); if (++tries > 20) { - mlog(ML_ERROR, "%s:%.*s: spinning on " - "dlm_wait_for_lock_mastery, blocked=%d\n", + mlog(ML_ERROR, "%s: res %.*s, Spinning on " + "dlm_wait_for_lock_mastery, blocked = %d\n", dlm->name, res->lockname.len, res->lockname.name, blocked); dlm_print_one_lock_resource(res); @@ -940,7 +949,8 @@ wait: goto redo_request; } - mlog(0, "lockres mastered by %u\n", res->owner); + mlog(0, "%s: res %.*s, Mastered by %u\n", dlm->name, res->lockname.len, + res->lockname.name, res->owner); /* make sure we never continue without this */ BUG_ON(res->owner == O2NM_MAX_NODES); @@ -952,8 +962,6 @@ wait: wake_waiters: spin_lock(&res->spinlock); - if (res->owner != dlm->node_num && drop_inflight_if_nonlocal) - dlm_lockres_drop_inflight_ref(dlm, res); res->state &= ~DLM_LOCK_RES_IN_PROGRESS; spin_unlock(&res->spinlock); wake_up(&res->wq); @@ -1426,9 +1434,7 @@ way_up_top: } if (res->owner == dlm->node_num) { - mlog(0, "%s:%.*s: setting bit %u in refmap\n", - dlm->name, namelen, name, request->node_idx); - dlm_lockres_set_refmap_bit(request->node_idx, res); + dlm_lockres_set_refmap_bit(dlm, res, request->node_idx); spin_unlock(&res->spinlock); response = DLM_MASTER_RESP_YES; if (mle) @@ -1493,10 +1499,8 @@ way_up_top: * go back and clean the mles on any * other nodes */ dispatch_assert = 1; - dlm_lockres_set_refmap_bit(request->node_idx, res); - mlog(0, "%s:%.*s: setting bit %u in refmap\n", - dlm->name, namelen, name, - request->node_idx); + dlm_lockres_set_refmap_bit(dlm, res, + request->node_idx); } else response = DLM_MASTER_RESP_NO; } else { @@ -1702,7 +1706,7 @@ again: "lockres, set the bit in the refmap\n", namelen, lockname, to); spin_lock(&res->spinlock); - dlm_lockres_set_refmap_bit(to, res); + dlm_lockres_set_refmap_bit(dlm, res, to); spin_unlock(&res->spinlock); } } @@ -2187,8 +2191,6 @@ int dlm_drop_lockres_ref(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) namelen = res->lockname.len; BUG_ON(namelen > O2NM_MAX_NAME_LEN); - mlog(0, "%s:%.*s: sending deref to %d\n", - dlm->name, namelen, lockname, res->owner); memset(&deref, 0, sizeof(deref)); deref.node_idx = dlm->node_num; deref.namelen = namelen; @@ -2197,14 +2199,12 @@ int dlm_drop_lockres_ref(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) ret = o2net_send_message(DLM_DEREF_LOCKRES_MSG, dlm->key, &deref, sizeof(deref), res->owner, &r); if (ret < 0) - mlog(ML_ERROR, "Error %d when sending message %u (key 0x%x) to " - "node %u\n", ret, DLM_DEREF_LOCKRES_MSG, dlm->key, - res->owner); + mlog(ML_ERROR, "%s: res %.*s, error %d send DEREF to node %u\n", + dlm->name, namelen, lockname, ret, res->owner); else if (r < 0) { /* BAD. other node says I did not have a ref. */ - mlog(ML_ERROR,"while dropping ref on %s:%.*s " - "(master=%u) got %d.\n", dlm->name, namelen, - lockname, res->owner, r); + mlog(ML_ERROR, "%s: res %.*s, DEREF to node %u got %d\n", + dlm->name, namelen, lockname, res->owner, r); dlm_print_one_lock_resource(res); BUG(); } @@ -2260,7 +2260,7 @@ int dlm_deref_lockres_handler(struct o2net_msg *msg, u32 len, void *data, else { BUG_ON(res->state & DLM_LOCK_RES_DROPPING_REF); if (test_bit(node, res->refmap)) { - dlm_lockres_clear_refmap_bit(node, res); + dlm_lockres_clear_refmap_bit(dlm, res, node); cleared = 1; } } @@ -2320,7 +2320,7 @@ static void dlm_deref_lockres_worker(struct dlm_work_item *item, void *data) BUG_ON(res->state & DLM_LOCK_RES_DROPPING_REF); if (test_bit(node, res->refmap)) { __dlm_wait_on_lockres_flags(res, DLM_LOCK_RES_SETREF_INPROG); - dlm_lockres_clear_refmap_bit(node, res); + dlm_lockres_clear_refmap_bit(dlm, res, node); cleared = 1; } spin_unlock(&res->spinlock); @@ -2802,7 +2802,8 @@ static void dlm_remove_nonlocal_locks(struct dlm_ctxt *dlm, BUG_ON(!list_empty(&lock->bast_list)); BUG_ON(lock->ast_pending); BUG_ON(lock->bast_pending); - dlm_lockres_clear_refmap_bit(lock->ml.node, res); + dlm_lockres_clear_refmap_bit(dlm, res, + lock->ml.node); list_del_init(&lock->list); dlm_lock_put(lock); /* In a normal unlock, we would have added a @@ -2823,7 +2824,7 @@ static void dlm_remove_nonlocal_locks(struct dlm_ctxt *dlm, mlog(0, "%s:%.*s: node %u had a ref to this " "migrating lockres, clearing\n", dlm->name, res->lockname.len, res->lockname.name, bit); - dlm_lockres_clear_refmap_bit(bit, res); + dlm_lockres_clear_refmap_bit(dlm, res, bit); } bit++; } @@ -2916,9 +2917,9 @@ static int dlm_do_migrate_request(struct dlm_ctxt *dlm, &migrate, sizeof(migrate), nodenum, &status); if (ret < 0) { - mlog(ML_ERROR, "Error %d when sending message %u (key " - "0x%x) to node %u\n", ret, DLM_MIGRATE_REQUEST_MSG, - dlm->key, nodenum); + mlog(ML_ERROR, "%s: res %.*s, Error %d send " + "MIGRATE_REQUEST to node %u\n", dlm->name, + migrate.namelen, migrate.name, ret, nodenum); if (!dlm_is_host_down(ret)) { mlog(ML_ERROR, "unhandled error=%d!\n", ret); BUG(); @@ -2937,7 +2938,7 @@ static int dlm_do_migrate_request(struct dlm_ctxt *dlm, dlm->name, res->lockname.len, res->lockname.name, nodenum); spin_lock(&res->spinlock); - dlm_lockres_set_refmap_bit(nodenum, res); + dlm_lockres_set_refmap_bit(dlm, res, nodenum); spin_unlock(&res->spinlock); } } @@ -3271,7 +3272,7 @@ int dlm_finish_migration(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, * mastery reference here since old_master will briefly have * a reference after the migration completes */ spin_lock(&res->spinlock); - dlm_lockres_set_refmap_bit(old_master, res); + dlm_lockres_set_refmap_bit(dlm, res, old_master); spin_unlock(&res->spinlock); mlog(0, "now time to do a migrate request to other nodes\n"); diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index 7efab6d28a21..01ebfd0bdad7 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -362,40 +362,38 @@ static int dlm_is_node_recovered(struct dlm_ctxt *dlm, u8 node) } -int dlm_wait_for_node_death(struct dlm_ctxt *dlm, u8 node, int timeout) +void dlm_wait_for_node_death(struct dlm_ctxt *dlm, u8 node, int timeout) { - if (timeout) { - mlog(ML_NOTICE, "%s: waiting %dms for notification of " - "death of node %u\n", dlm->name, timeout, node); + if (dlm_is_node_dead(dlm, node)) + return; + + printk(KERN_NOTICE "o2dlm: Waiting on the death of node %u in " + "domain %s\n", node, dlm->name); + + if (timeout) wait_event_timeout(dlm->dlm_reco_thread_wq, - dlm_is_node_dead(dlm, node), - msecs_to_jiffies(timeout)); - } else { - mlog(ML_NOTICE, "%s: waiting indefinitely for notification " - "of death of node %u\n", dlm->name, node); + dlm_is_node_dead(dlm, node), + msecs_to_jiffies(timeout)); + else wait_event(dlm->dlm_reco_thread_wq, dlm_is_node_dead(dlm, node)); - } - /* for now, return 0 */ - return 0; } -int dlm_wait_for_node_recovery(struct dlm_ctxt *dlm, u8 node, int timeout) +void dlm_wait_for_node_recovery(struct dlm_ctxt *dlm, u8 node, int timeout) { - if (timeout) { - mlog(0, "%s: waiting %dms for notification of " - "recovery of node %u\n", dlm->name, timeout, node); + if (dlm_is_node_recovered(dlm, node)) + return; + + printk(KERN_NOTICE "o2dlm: Waiting on the recovery of node %u in " + "domain %s\n", node, dlm->name); + + if (timeout) wait_event_timeout(dlm->dlm_reco_thread_wq, - dlm_is_node_recovered(dlm, node), - msecs_to_jiffies(timeout)); - } else { - mlog(0, "%s: waiting indefinitely for notification " - "of recovery of node %u\n", dlm->name, node); + dlm_is_node_recovered(dlm, node), + msecs_to_jiffies(timeout)); + else wait_event(dlm->dlm_reco_thread_wq, dlm_is_node_recovered(dlm, node)); - } - /* for now, return 0 */ - return 0; } /* callers of the top-level api calls (dlmlock/dlmunlock) should @@ -430,6 +428,8 @@ static void dlm_begin_recovery(struct dlm_ctxt *dlm) { spin_lock(&dlm->spinlock); BUG_ON(dlm->reco.state & DLM_RECO_STATE_ACTIVE); + printk(KERN_NOTICE "o2dlm: Begin recovery on domain %s for node %u\n", + dlm->name, dlm->reco.dead_node); dlm->reco.state |= DLM_RECO_STATE_ACTIVE; spin_unlock(&dlm->spinlock); } @@ -440,9 +440,18 @@ static void dlm_end_recovery(struct dlm_ctxt *dlm) BUG_ON(!(dlm->reco.state & DLM_RECO_STATE_ACTIVE)); dlm->reco.state &= ~DLM_RECO_STATE_ACTIVE; spin_unlock(&dlm->spinlock); + printk(KERN_NOTICE "o2dlm: End recovery on domain %s\n", dlm->name); wake_up(&dlm->reco.event); } +static void dlm_print_recovery_master(struct dlm_ctxt *dlm) +{ + printk(KERN_NOTICE "o2dlm: Node %u (%s) is the Recovery Master for the " + "dead node %u in domain %s\n", dlm->reco.new_master, + (dlm->node_num == dlm->reco.new_master ? "me" : "he"), + dlm->reco.dead_node, dlm->name); +} + static int dlm_do_recovery(struct dlm_ctxt *dlm) { int status = 0; @@ -505,9 +514,8 @@ static int dlm_do_recovery(struct dlm_ctxt *dlm) } mlog(0, "another node will master this recovery session.\n"); } - mlog(0, "dlm=%s (%d), new_master=%u, this node=%u, dead_node=%u\n", - dlm->name, task_pid_nr(dlm->dlm_reco_thread_task), dlm->reco.new_master, - dlm->node_num, dlm->reco.dead_node); + + dlm_print_recovery_master(dlm); /* it is safe to start everything back up here * because all of the dead node's lock resources @@ -518,15 +526,13 @@ static int dlm_do_recovery(struct dlm_ctxt *dlm) return 0; master_here: - mlog(ML_NOTICE, "(%d) Node %u is the Recovery Master for the Dead Node " - "%u for Domain %s\n", task_pid_nr(dlm->dlm_reco_thread_task), - dlm->node_num, dlm->reco.dead_node, dlm->name); + dlm_print_recovery_master(dlm); status = dlm_remaster_locks(dlm, dlm->reco.dead_node); if (status < 0) { /* we should never hit this anymore */ - mlog(ML_ERROR, "error %d remastering locks for node %u, " - "retrying.\n", status, dlm->reco.dead_node); + mlog(ML_ERROR, "%s: Error %d remastering locks for node %u, " + "retrying.\n", dlm->name, status, dlm->reco.dead_node); /* yield a bit to allow any final network messages * to get handled on remaining nodes */ msleep(100); @@ -567,7 +573,7 @@ static int dlm_remaster_locks(struct dlm_ctxt *dlm, u8 dead_node) BUG_ON(ndata->state != DLM_RECO_NODE_DATA_INIT); ndata->state = DLM_RECO_NODE_DATA_REQUESTING; - mlog(0, "requesting lock info from node %u\n", + mlog(0, "%s: Requesting lock info from node %u\n", dlm->name, ndata->node_num); if (ndata->node_num == dlm->node_num) { @@ -640,7 +646,7 @@ static int dlm_remaster_locks(struct dlm_ctxt *dlm, u8 dead_node) spin_unlock(&dlm_reco_state_lock); } - mlog(0, "done requesting all lock info\n"); + mlog(0, "%s: Done requesting all lock info\n", dlm->name); /* nodes should be sending reco data now * just need to wait */ @@ -802,10 +808,9 @@ static int dlm_request_all_locks(struct dlm_ctxt *dlm, u8 request_from, /* negative status is handled by caller */ if (ret < 0) - mlog(ML_ERROR, "Error %d when sending message %u (key " - "0x%x) to node %u\n", ret, DLM_LOCK_REQUEST_MSG, - dlm->key, request_from); - + mlog(ML_ERROR, "%s: Error %d send LOCK_REQUEST to node %u " + "to recover dead node %u\n", dlm->name, ret, + request_from, dead_node); // return from here, then // sleep until all received or error return ret; @@ -956,9 +961,9 @@ static int dlm_send_all_done_msg(struct dlm_ctxt *dlm, u8 dead_node, u8 send_to) ret = o2net_send_message(DLM_RECO_DATA_DONE_MSG, dlm->key, &done_msg, sizeof(done_msg), send_to, &tmpret); if (ret < 0) { - mlog(ML_ERROR, "Error %d when sending message %u (key " - "0x%x) to node %u\n", ret, DLM_RECO_DATA_DONE_MSG, - dlm->key, send_to); + mlog(ML_ERROR, "%s: Error %d send RECO_DATA_DONE to node %u " + "to recover dead node %u\n", dlm->name, ret, send_to, + dead_node); if (!dlm_is_host_down(ret)) { BUG(); } @@ -1127,9 +1132,11 @@ static int dlm_send_mig_lockres_msg(struct dlm_ctxt *dlm, if (ret < 0) { /* XXX: negative status is not handled. * this will end up killing this node. */ - mlog(ML_ERROR, "Error %d when sending message %u (key " - "0x%x) to node %u\n", ret, DLM_MIG_LOCKRES_MSG, - dlm->key, send_to); + mlog(ML_ERROR, "%s: res %.*s, Error %d send MIG_LOCKRES to " + "node %u (%s)\n", dlm->name, mres->lockname_len, + mres->lockname, ret, send_to, + (orig_flags & DLM_MRES_MIGRATION ? + "migration" : "recovery")); } else { /* might get an -ENOMEM back here */ ret = status; @@ -1767,7 +1774,7 @@ static int dlm_process_recovery_data(struct dlm_ctxt *dlm, dlm->name, mres->lockname_len, mres->lockname, from); spin_lock(&res->spinlock); - dlm_lockres_set_refmap_bit(from, res); + dlm_lockres_set_refmap_bit(dlm, res, from); spin_unlock(&res->spinlock); added++; break; @@ -1965,7 +1972,7 @@ skip_lvb: mlog(0, "%s:%.*s: added lock for node %u, " "setting refmap bit\n", dlm->name, res->lockname.len, res->lockname.name, ml->node); - dlm_lockres_set_refmap_bit(ml->node, res); + dlm_lockres_set_refmap_bit(dlm, res, ml->node); added++; } spin_unlock(&res->spinlock); @@ -2084,6 +2091,9 @@ static void dlm_finish_local_lockres_recovery(struct dlm_ctxt *dlm, list_for_each_entry_safe(res, next, &dlm->reco.resources, recovering) { if (res->owner == dead_node) { + mlog(0, "%s: res %.*s, Changing owner from %u to %u\n", + dlm->name, res->lockname.len, res->lockname.name, + res->owner, new_master); list_del_init(&res->recovering); spin_lock(&res->spinlock); /* new_master has our reference from @@ -2105,40 +2115,30 @@ static void dlm_finish_local_lockres_recovery(struct dlm_ctxt *dlm, for (i = 0; i < DLM_HASH_BUCKETS; i++) { bucket = dlm_lockres_hash(dlm, i); hlist_for_each_entry(res, hash_iter, bucket, hash_node) { - if (res->state & DLM_LOCK_RES_RECOVERING) { - if (res->owner == dead_node) { - mlog(0, "(this=%u) res %.*s owner=%u " - "was not on recovering list, but " - "clearing state anyway\n", - dlm->node_num, res->lockname.len, - res->lockname.name, new_master); - } else if (res->owner == dlm->node_num) { - mlog(0, "(this=%u) res %.*s owner=%u " - "was not on recovering list, " - "owner is THIS node, clearing\n", - dlm->node_num, res->lockname.len, - res->lockname.name, new_master); - } else - continue; + if (!(res->state & DLM_LOCK_RES_RECOVERING)) + continue; - if (!list_empty(&res->recovering)) { - mlog(0, "%s:%.*s: lockres was " - "marked RECOVERING, owner=%u\n", - dlm->name, res->lockname.len, - res->lockname.name, res->owner); - list_del_init(&res->recovering); - dlm_lockres_put(res); - } - spin_lock(&res->spinlock); - /* new_master has our reference from - * the lock state sent during recovery */ - dlm_change_lockres_owner(dlm, res, new_master); - res->state &= ~DLM_LOCK_RES_RECOVERING; - if (__dlm_lockres_has_locks(res)) - __dlm_dirty_lockres(dlm, res); - spin_unlock(&res->spinlock); - wake_up(&res->wq); + if (res->owner != dead_node && + res->owner != dlm->node_num) + continue; + + if (!list_empty(&res->recovering)) { + list_del_init(&res->recovering); + dlm_lockres_put(res); } + + /* new_master has our reference from + * the lock state sent during recovery */ + mlog(0, "%s: res %.*s, Changing owner from %u to %u\n", + dlm->name, res->lockname.len, res->lockname.name, + res->owner, new_master); + spin_lock(&res->spinlock); + dlm_change_lockres_owner(dlm, res, new_master); + res->state &= ~DLM_LOCK_RES_RECOVERING; + if (__dlm_lockres_has_locks(res)) + __dlm_dirty_lockres(dlm, res); + spin_unlock(&res->spinlock); + wake_up(&res->wq); } } } @@ -2252,12 +2252,12 @@ static void dlm_free_dead_locks(struct dlm_ctxt *dlm, res->lockname.len, res->lockname.name, freed, dead_node); __dlm_print_one_lock_resource(res); } - dlm_lockres_clear_refmap_bit(dead_node, res); + dlm_lockres_clear_refmap_bit(dlm, res, dead_node); } else if (test_bit(dead_node, res->refmap)) { mlog(0, "%s:%.*s: dead node %u had a ref, but had " "no locks and had not purged before dying\n", dlm->name, res->lockname.len, res->lockname.name, dead_node); - dlm_lockres_clear_refmap_bit(dead_node, res); + dlm_lockres_clear_refmap_bit(dlm, res, dead_node); } /* do not kick thread yet */ @@ -2324,9 +2324,9 @@ static void dlm_do_local_recovery_cleanup(struct dlm_ctxt *dlm, u8 dead_node) dlm_revalidate_lvb(dlm, res, dead_node); if (res->owner == dead_node) { if (res->state & DLM_LOCK_RES_DROPPING_REF) { - mlog(ML_NOTICE, "Ignore %.*s for " + mlog(ML_NOTICE, "%s: res %.*s, Skip " "recovery as it is being freed\n", - res->lockname.len, + dlm->name, res->lockname.len, res->lockname.name); } else dlm_move_lockres_to_recovery_list(dlm, diff --git a/fs/ocfs2/dlm/dlmthread.c b/fs/ocfs2/dlm/dlmthread.c index 1d6d1d22c471..e73c833fc2a1 100644 --- a/fs/ocfs2/dlm/dlmthread.c +++ b/fs/ocfs2/dlm/dlmthread.c @@ -94,24 +94,26 @@ int __dlm_lockres_unused(struct dlm_lock_resource *res) { int bit; + assert_spin_locked(&res->spinlock); + if (__dlm_lockres_has_locks(res)) return 0; + /* Locks are in the process of being created */ + if (res->inflight_locks) + return 0; + if (!list_empty(&res->dirty) || res->state & DLM_LOCK_RES_DIRTY) return 0; if (res->state & DLM_LOCK_RES_RECOVERING) return 0; + /* Another node has this resource with this node as the master */ bit = find_next_bit(res->refmap, O2NM_MAX_NODES, 0); if (bit < O2NM_MAX_NODES) return 0; - /* - * since the bit for dlm->node_num is not set, inflight_locks better - * be zero - */ - BUG_ON(res->inflight_locks != 0); return 1; } @@ -185,8 +187,6 @@ static void dlm_purge_lockres(struct dlm_ctxt *dlm, /* clear our bit from the master's refmap, ignore errors */ ret = dlm_drop_lockres_ref(dlm, res); if (ret < 0) { - mlog(ML_ERROR, "%s: deref %.*s failed %d\n", dlm->name, - res->lockname.len, res->lockname.name, ret); if (!dlm_is_host_down(ret)) BUG(); } @@ -209,7 +209,7 @@ static void dlm_purge_lockres(struct dlm_ctxt *dlm, BUG(); } - __dlm_unhash_lockres(res); + __dlm_unhash_lockres(dlm, res); /* lockres is not in the hash now. drop the flag and wake up * any processes waiting in dlm_get_lock_resource. */ diff --git a/fs/ocfs2/dlmfs/dlmfs.c b/fs/ocfs2/dlmfs/dlmfs.c index b42076797049..abfac0d7ae9c 100644 --- a/fs/ocfs2/dlmfs/dlmfs.c +++ b/fs/ocfs2/dlmfs/dlmfs.c @@ -354,7 +354,6 @@ static struct inode *dlmfs_alloc_inode(struct super_block *sb) static void dlmfs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(dlmfs_inode_cache, DLMFS_I(inode)); } @@ -401,16 +400,14 @@ static struct backing_dev_info dlmfs_backing_dev_info = { static struct inode *dlmfs_get_root_inode(struct super_block *sb) { struct inode *inode = new_inode(sb); - int mode = S_IFDIR | 0755; + umode_t mode = S_IFDIR | 0755; struct dlmfs_inode_private *ip; if (inode) { ip = DLMFS_I(inode); inode->i_ino = get_next_ino(); - inode->i_mode = mode; - inode->i_uid = current_fsuid(); - inode->i_gid = current_fsgid(); + inode_init_owner(inode, NULL, mode); inode->i_mapping->backing_dev_info = &dlmfs_backing_dev_info; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; inc_nlink(inode); @@ -424,7 +421,7 @@ static struct inode *dlmfs_get_root_inode(struct super_block *sb) static struct inode *dlmfs_get_inode(struct inode *parent, struct dentry *dentry, - int mode) + umode_t mode) { struct super_block *sb = parent->i_sb; struct inode * inode = new_inode(sb); @@ -434,9 +431,7 @@ static struct inode *dlmfs_get_inode(struct inode *parent, return NULL; inode->i_ino = get_next_ino(); - inode->i_mode = mode; - inode->i_uid = current_fsuid(); - inode->i_gid = current_fsgid(); + inode_init_owner(inode, parent, mode); inode->i_mapping->backing_dev_info = &dlmfs_backing_dev_info; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; @@ -473,13 +468,6 @@ static struct inode *dlmfs_get_inode(struct inode *parent, inc_nlink(inode); break; } - - if (parent->i_mode & S_ISGID) { - inode->i_gid = parent->i_gid; - if (S_ISDIR(mode)) - inode->i_mode |= S_ISGID; - } - return inode; } @@ -489,7 +477,7 @@ static struct inode *dlmfs_get_inode(struct inode *parent, /* SMP-safe */ static int dlmfs_mkdir(struct inode * dir, struct dentry * dentry, - int mode) + umode_t mode) { int status; struct inode *inode = NULL; @@ -537,7 +525,7 @@ bail: static int dlmfs_create(struct inode *dir, struct dentry *dentry, - int mode, + umode_t mode, struct nameidata *nd) { int status = 0; diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c index e1ed5e502ff2..81a4cd22f80b 100644 --- a/fs/ocfs2/dlmglue.c +++ b/fs/ocfs2/dlmglue.c @@ -1692,7 +1692,7 @@ int ocfs2_open_lock(struct inode *inode) mlog(0, "inode %llu take PRMODE open lock\n", (unsigned long long)OCFS2_I(inode)->ip_blkno); - if (ocfs2_mount_local(osb)) + if (ocfs2_is_hard_readonly(osb) || ocfs2_mount_local(osb)) goto out; lockres = &OCFS2_I(inode)->ip_open_lockres; @@ -1718,6 +1718,12 @@ int ocfs2_try_open_lock(struct inode *inode, int write) (unsigned long long)OCFS2_I(inode)->ip_blkno, write ? "EXMODE" : "PRMODE"); + if (ocfs2_is_hard_readonly(osb)) { + if (write) + status = -EROFS; + goto out; + } + if (ocfs2_mount_local(osb)) goto out; @@ -2298,7 +2304,7 @@ int ocfs2_inode_lock_full_nested(struct inode *inode, if (ocfs2_is_hard_readonly(osb)) { if (ex) status = -EROFS; - goto bail; + goto getbh; } if (ocfs2_mount_local(osb)) @@ -2356,7 +2362,7 @@ local: mlog_errno(status); goto bail; } - +getbh: if (ret_bh) { status = ocfs2_assign_bh(inode, ret_bh, local_bh); if (status < 0) { @@ -2628,8 +2634,11 @@ int ocfs2_dentry_lock(struct dentry *dentry, int ex) BUG_ON(!dl); - if (ocfs2_is_hard_readonly(osb)) - return -EROFS; + if (ocfs2_is_hard_readonly(osb)) { + if (ex) + return -EROFS; + return 0; + } if (ocfs2_mount_local(osb)) return 0; @@ -2647,7 +2656,7 @@ void ocfs2_dentry_unlock(struct dentry *dentry, int ex) struct ocfs2_dentry_lock *dl = dentry->d_fsdata; struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb); - if (!ocfs2_mount_local(osb)) + if (!ocfs2_is_hard_readonly(osb) && !ocfs2_mount_local(osb)) ocfs2_cluster_unlock(osb, &dl->dl_lockres, level); } diff --git a/fs/ocfs2/extent_map.c b/fs/ocfs2/extent_map.c index 23457b491e8c..2f5b92ef0e53 100644 --- a/fs/ocfs2/extent_map.c +++ b/fs/ocfs2/extent_map.c @@ -832,6 +832,102 @@ out: return ret; } +int ocfs2_seek_data_hole_offset(struct file *file, loff_t *offset, int origin) +{ + struct inode *inode = file->f_mapping->host; + int ret; + unsigned int is_last = 0, is_data = 0; + u16 cs_bits = OCFS2_SB(inode->i_sb)->s_clustersize_bits; + u32 cpos, cend, clen, hole_size; + u64 extoff, extlen; + struct buffer_head *di_bh = NULL; + struct ocfs2_extent_rec rec; + + BUG_ON(origin != SEEK_DATA && origin != SEEK_HOLE); + + ret = ocfs2_inode_lock(inode, &di_bh, 0); + if (ret) { + mlog_errno(ret); + goto out; + } + + down_read(&OCFS2_I(inode)->ip_alloc_sem); + + if (*offset >= inode->i_size) { + ret = -ENXIO; + goto out_unlock; + } + + if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) { + if (origin == SEEK_HOLE) + *offset = inode->i_size; + goto out_unlock; + } + + clen = 0; + cpos = *offset >> cs_bits; + cend = ocfs2_clusters_for_bytes(inode->i_sb, inode->i_size); + + while (cpos < cend && !is_last) { + ret = ocfs2_get_clusters_nocache(inode, di_bh, cpos, &hole_size, + &rec, &is_last); + if (ret) { + mlog_errno(ret); + goto out_unlock; + } + + extoff = cpos; + extoff <<= cs_bits; + + if (rec.e_blkno == 0ULL) { + clen = hole_size; + is_data = 0; + } else { + clen = le16_to_cpu(rec.e_leaf_clusters) - + (cpos - le32_to_cpu(rec.e_cpos)); + is_data = (rec.e_flags & OCFS2_EXT_UNWRITTEN) ? 0 : 1; + } + + if ((!is_data && origin == SEEK_HOLE) || + (is_data && origin == SEEK_DATA)) { + if (extoff > *offset) + *offset = extoff; + goto out_unlock; + } + + if (!is_last) + cpos += clen; + } + + if (origin == SEEK_HOLE) { + extoff = cpos; + extoff <<= cs_bits; + extlen = clen; + extlen <<= cs_bits; + + if ((extoff + extlen) > inode->i_size) + extlen = inode->i_size - extoff; + extoff += extlen; + if (extoff > *offset) + *offset = extoff; + goto out_unlock; + } + + ret = -ENXIO; + +out_unlock: + + brelse(di_bh); + + up_read(&OCFS2_I(inode)->ip_alloc_sem); + + ocfs2_inode_unlock(inode, 0); +out: + if (ret && ret != -ENXIO) + ret = -ENXIO; + return ret; +} + int ocfs2_read_virt_blocks(struct inode *inode, u64 v_block, int nr, struct buffer_head *bhs[], int flags, int (*validate)(struct super_block *sb, diff --git a/fs/ocfs2/extent_map.h b/fs/ocfs2/extent_map.h index e79d41c2c909..67ea57d2fd59 100644 --- a/fs/ocfs2/extent_map.h +++ b/fs/ocfs2/extent_map.h @@ -53,6 +53,8 @@ int ocfs2_extent_map_get_blocks(struct inode *inode, u64 v_blkno, u64 *p_blkno, int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, u64 map_start, u64 map_len); +int ocfs2_seek_data_hole_offset(struct file *file, loff_t *offset, int origin); + int ocfs2_xattr_get_clusters(struct inode *inode, u32 v_cluster, u32 *p_cluster, u32 *num_clusters, struct ocfs2_extent_list *el, diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index de4ea1af041b..061591a3ab08 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -1950,6 +1950,9 @@ static int __ocfs2_change_file_space(struct file *file, struct inode *inode, if (ret < 0) mlog_errno(ret); + if (file->f_flags & O_SYNC) + handle->h_sync = 1; + ocfs2_commit_trans(osb, handle); out_inode_unlock: @@ -2052,6 +2055,23 @@ out: return ret; } +static void ocfs2_aiodio_wait(struct inode *inode) +{ + wait_queue_head_t *wq = ocfs2_ioend_wq(inode); + + wait_event(*wq, (atomic_read(&OCFS2_I(inode)->ip_unaligned_aio) == 0)); +} + +static int ocfs2_is_io_unaligned(struct inode *inode, size_t count, loff_t pos) +{ + int blockmask = inode->i_sb->s_blocksize - 1; + loff_t final_size = pos + count; + + if ((pos & blockmask) || (final_size & blockmask)) + return 1; + return 0; +} + static int ocfs2_prepare_inode_for_refcount(struct inode *inode, struct file *file, loff_t pos, size_t count, @@ -2108,7 +2128,7 @@ static int ocfs2_prepare_inode_for_write(struct file *file, * remove_suid() calls ->setattr without any hint that * we may have already done our cluster locking. Since * ocfs2_setattr() *must* take cluster locks to - * proceeed, this will lead us to recursively lock the + * proceed, this will lead us to recursively lock the * inode. There's also the dinode i_size state which * can be lost via setattr during extending writes (we * set inode->i_size at the end of a write. */ @@ -2230,6 +2250,7 @@ static ssize_t ocfs2_file_aio_write(struct kiocb *iocb, struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); int full_coherency = !(osb->s_mount_opt & OCFS2_MOUNT_COHERENCY_BUFFERED); + int unaligned_dio = 0; trace_ocfs2_file_aio_write(inode, file, file->f_path.dentry, (unsigned long long)OCFS2_I(inode)->ip_blkno, @@ -2297,6 +2318,10 @@ relock: goto out; } + if (direct_io && !is_sync_kiocb(iocb)) + unaligned_dio = ocfs2_is_io_unaligned(inode, iocb->ki_left, + *ppos); + /* * We can't complete the direct I/O as requested, fall back to * buffered I/O. @@ -2311,6 +2336,18 @@ relock: goto relock; } + if (unaligned_dio) { + /* + * Wait on previous unaligned aio to complete before + * proceeding. + */ + ocfs2_aiodio_wait(inode); + + /* Mark the iocb as needing a decrement in ocfs2_dio_end_io */ + atomic_inc(&OCFS2_I(inode)->ip_unaligned_aio); + ocfs2_iocb_set_unaligned_aio(iocb); + } + /* * To later detect whether a journal commit for sync writes is * necessary, we sample i_size, and cluster count here. @@ -2382,8 +2419,12 @@ out_dio: if ((ret == -EIOCBQUEUED) || (!ocfs2_iocb_is_rw_locked(iocb))) { rw_level = -1; have_alloc_sem = 0; + unaligned_dio = 0; } + if (unaligned_dio) + atomic_dec(&OCFS2_I(inode)->ip_unaligned_aio); + out: if (rw_level != -1) ocfs2_rw_unlock(inode, rw_level); @@ -2591,6 +2632,57 @@ bail: return ret; } +/* Refer generic_file_llseek_unlocked() */ +static loff_t ocfs2_file_llseek(struct file *file, loff_t offset, int origin) +{ + struct inode *inode = file->f_mapping->host; + int ret = 0; + + mutex_lock(&inode->i_mutex); + + switch (origin) { + case SEEK_SET: + break; + case SEEK_END: + offset += inode->i_size; + break; + case SEEK_CUR: + if (offset == 0) { + offset = file->f_pos; + goto out; + } + offset += file->f_pos; + break; + case SEEK_DATA: + case SEEK_HOLE: + ret = ocfs2_seek_data_hole_offset(file, &offset, origin); + if (ret) + goto out; + break; + default: + ret = -EINVAL; + goto out; + } + + if (offset < 0 && !(file->f_mode & FMODE_UNSIGNED_OFFSET)) + ret = -EINVAL; + if (!ret && offset > inode->i_sb->s_maxbytes) + ret = -EINVAL; + if (ret) + goto out; + + if (offset != file->f_pos) { + file->f_pos = offset; + file->f_version = 0; + } + +out: + mutex_unlock(&inode->i_mutex); + if (ret) + return ret; + return offset; +} + const struct inode_operations ocfs2_file_iops = { .setattr = ocfs2_setattr, .getattr = ocfs2_getattr, @@ -2615,7 +2707,7 @@ const struct inode_operations ocfs2_special_file_iops = { * ocfs2_fops_no_plocks and ocfs2_dops_no_plocks! */ const struct file_operations ocfs2_fops = { - .llseek = generic_file_llseek, + .llseek = ocfs2_file_llseek, .read = do_sync_read, .write = do_sync_write, .mmap = ocfs2_mmap, @@ -2663,7 +2755,7 @@ const struct file_operations ocfs2_dops = { * the cluster. */ const struct file_operations ocfs2_fops_no_plocks = { - .llseek = generic_file_llseek, + .llseek = ocfs2_file_llseek, .read = do_sync_read, .write = do_sync_write, .mmap = ocfs2_mmap, diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c index a22d2c098890..17454a904d7b 100644 --- a/fs/ocfs2/inode.c +++ b/fs/ocfs2/inode.c @@ -951,7 +951,7 @@ static void ocfs2_cleanup_delete_inode(struct inode *inode, trace_ocfs2_cleanup_delete_inode( (unsigned long long)OCFS2_I(inode)->ip_blkno, sync_data); if (sync_data) - write_inode_now(inode, 1); + filemap_write_and_wait(inode->i_mapping); truncate_inode_pages(&inode->i_data, 0); } diff --git a/fs/ocfs2/inode.h b/fs/ocfs2/inode.h index 1c508b149b3a..88924a3133fa 100644 --- a/fs/ocfs2/inode.h +++ b/fs/ocfs2/inode.h @@ -43,6 +43,9 @@ struct ocfs2_inode_info /* protects extended attribute changes on this inode */ struct rw_semaphore ip_xattr_sem; + /* Number of outstanding AIO's which are not page aligned */ + atomic_t ip_unaligned_aio; + /* These fields are protected by ip_lock */ spinlock_t ip_lock; u32 ip_open_count; diff --git a/fs/ocfs2/ioctl.c b/fs/ocfs2/ioctl.c index bc91072b7219..a6fda3c188aa 100644 --- a/fs/ocfs2/ioctl.c +++ b/fs/ocfs2/ioctl.c @@ -122,7 +122,7 @@ static int ocfs2_set_inode_attr(struct inode *inode, unsigned flags, if ((oldflags & OCFS2_IMMUTABLE_FL) || ((flags ^ oldflags) & (OCFS2_APPEND_FL | OCFS2_IMMUTABLE_FL))) { if (!capable(CAP_LINUX_IMMUTABLE)) - goto bail_unlock; + goto bail_commit; } ocfs2_inode->ip_attr = flags; @@ -132,6 +132,7 @@ static int ocfs2_set_inode_attr(struct inode *inode, unsigned flags, if (status < 0) mlog_errno(status); +bail_commit: ocfs2_commit_trans(osb, handle); bail_unlock: ocfs2_inode_unlock(inode, 1); @@ -381,7 +382,7 @@ int ocfs2_info_handle_freeinode(struct inode *inode, if (!oifi) { status = -ENOMEM; mlog_errno(status); - goto bail; + goto out_err; } if (o2info_from_user(*oifi, req)) @@ -431,7 +432,7 @@ bail: o2info_set_request_error(&oifi->ifi_req, req); kfree(oifi); - +out_err: return status; } @@ -666,7 +667,7 @@ int ocfs2_info_handle_freefrag(struct inode *inode, if (!oiff) { status = -ENOMEM; mlog_errno(status); - goto bail; + goto out_err; } if (o2info_from_user(*oiff, req)) @@ -716,7 +717,7 @@ bail: o2info_set_request_error(&oiff->iff_req, req); kfree(oiff); - +out_err: return status; } @@ -905,12 +906,12 @@ long ocfs2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) if (get_user(flags, (int __user *) arg)) return -EFAULT; - status = mnt_want_write(filp->f_path.mnt); + status = mnt_want_write_file(filp); if (status) return status; status = ocfs2_set_inode_attr(inode, flags, OCFS2_FL_MODIFIABLE); - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return status; case OCFS2_IOC_RESVSP: case OCFS2_IOC_RESVSP64: diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index 295d56454e8b..0a42ae96dca7 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -1544,9 +1544,9 @@ static int ocfs2_replay_journal(struct ocfs2_super *osb, /* we need to run complete recovery for offline orphan slots */ ocfs2_replay_map_set_state(osb, REPLAY_NEEDED); - mlog(ML_NOTICE, "Recovering node %d from slot %d on device (%u,%u)\n", - node_num, slot_num, - MAJOR(osb->sb->s_dev), MINOR(osb->sb->s_dev)); + printk(KERN_NOTICE "ocfs2: Begin replay journal (node %d, slot %d) on "\ + "device (%u,%u)\n", node_num, slot_num, MAJOR(osb->sb->s_dev), + MINOR(osb->sb->s_dev)); OCFS2_I(inode)->ip_clusters = le32_to_cpu(fe->i_clusters); @@ -1601,6 +1601,9 @@ static int ocfs2_replay_journal(struct ocfs2_super *osb, jbd2_journal_destroy(journal); + printk(KERN_NOTICE "ocfs2: End replay journal (node %d, slot %d) on "\ + "device (%u,%u)\n", node_num, slot_num, MAJOR(osb->sb->s_dev), + MINOR(osb->sb->s_dev)); done: /* drop the lock on this nodes journal */ if (got_lock) @@ -1808,6 +1811,20 @@ static inline unsigned long ocfs2_orphan_scan_timeout(void) * every slot, queuing a recovery of the slot on the ocfs2_wq thread. This * is done to catch any orphans that are left over in orphan directories. * + * It scans all slots, even ones that are in use. It does so to handle the + * case described below: + * + * Node 1 has an inode it was using. The dentry went away due to memory + * pressure. Node 1 closes the inode, but it's on the free list. The node + * has the open lock. + * Node 2 unlinks the inode. It grabs the dentry lock to notify others, + * but node 1 has no dentry and doesn't get the message. It trylocks the + * open lock, sees that another node has a PR, and does nothing. + * Later node 2 runs its orphan dir. It igets the inode, trylocks the + * open lock, sees the PR still, and does nothing. + * Basically, we have to trigger an orphan iput on node 1. The only way + * for this to happen is if node 1 runs node 2's orphan dir. + * * ocfs2_queue_orphan_scan gets called every ORPHAN_SCAN_SCHEDULE_TIMEOUT * seconds. It gets an EX lock on os_lockres and checks sequence number * stored in LVB. If the sequence number has changed, it means some other diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h index 68cf2f6d3c6a..a3385b63ff5e 100644 --- a/fs/ocfs2/journal.h +++ b/fs/ocfs2/journal.h @@ -441,10 +441,11 @@ static inline int ocfs2_mknod_credits(struct super_block *sb, int is_dir, #define OCFS2_SIMPLE_DIR_EXTEND_CREDITS (2) /* file update (nlink, etc) + directory mtime/ctime + dir entry block + quota - * update on dir + index leaf + dx root update for free list */ + * update on dir + index leaf + dx root update for free list + + * previous dirblock update in the free list */ static inline int ocfs2_link_credits(struct super_block *sb) { - return 2*OCFS2_INODE_UPDATE_CREDITS + 3 + + return 2*OCFS2_INODE_UPDATE_CREDITS + 4 + ocfs2_quota_trans_credits(sb); } diff --git a/fs/ocfs2/mmap.c b/fs/ocfs2/mmap.c index 3e9393ca39eb..9cd41083e991 100644 --- a/fs/ocfs2/mmap.c +++ b/fs/ocfs2/mmap.c @@ -61,7 +61,7 @@ static int ocfs2_fault(struct vm_area_struct *area, struct vm_fault *vmf) static int __ocfs2_page_mkwrite(struct file *file, struct buffer_head *di_bh, struct page *page) { - int ret; + int ret = VM_FAULT_NOPAGE; struct inode *inode = file->f_path.dentry->d_inode; struct address_space *mapping = inode->i_mapping; loff_t pos = page_offset(page); @@ -71,32 +71,25 @@ static int __ocfs2_page_mkwrite(struct file *file, struct buffer_head *di_bh, void *fsdata; loff_t size = i_size_read(inode); - /* - * Another node might have truncated while we were waiting on - * cluster locks. - * We don't check size == 0 before the shift. This is borrowed - * from do_generic_file_read. - */ last_index = (size - 1) >> PAGE_CACHE_SHIFT; - if (unlikely(!size || page->index > last_index)) { - ret = -EINVAL; - goto out; - } /* - * The i_size check above doesn't catch the case where nodes - * truncated and then re-extended the file. We'll re-check the - * page mapping after taking the page lock inside of - * ocfs2_write_begin_nolock(). + * There are cases that lead to the page no longer bebongs to the + * mapping. + * 1) pagecache truncates locally due to memory pressure. + * 2) pagecache truncates when another is taking EX lock against + * inode lock. see ocfs2_data_convert_worker. + * + * The i_size check doesn't catch the case where nodes truncated and + * then re-extended the file. We'll re-check the page mapping after + * taking the page lock inside of ocfs2_write_begin_nolock(). + * + * Let VM retry with these cases. */ - if (!PageUptodate(page) || page->mapping != inode->i_mapping) { - /* - * the page has been umapped in ocfs2_data_downconvert_worker. - * So return 0 here and let VFS retry. - */ - ret = 0; + if ((page->mapping != inode->i_mapping) || + (!PageUptodate(page)) || + (page_offset(page) >= size)) goto out; - } /* * Call ocfs2_write_begin() and ocfs2_write_end() to take @@ -116,17 +109,21 @@ static int __ocfs2_page_mkwrite(struct file *file, struct buffer_head *di_bh, if (ret) { if (ret != -ENOSPC) mlog_errno(ret); + if (ret == -ENOMEM) + ret = VM_FAULT_OOM; + else + ret = VM_FAULT_SIGBUS; goto out; } - ret = ocfs2_write_end_nolock(mapping, pos, len, len, locked_page, - fsdata); - if (ret < 0) { - mlog_errno(ret); + if (!locked_page) { + ret = VM_FAULT_NOPAGE; goto out; } + ret = ocfs2_write_end_nolock(mapping, pos, len, len, locked_page, + fsdata); BUG_ON(ret != len); - ret = 0; + ret = VM_FAULT_LOCKED; out: return ret; } @@ -168,8 +165,6 @@ static int ocfs2_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) out: ocfs2_unblock_signals(&oldset); - if (ret) - ret = VM_FAULT_SIGBUS; return ret; } diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c index d53cb706f14c..b1e3fce72ea4 100644 --- a/fs/ocfs2/move_extents.c +++ b/fs/ocfs2/move_extents.c @@ -745,7 +745,7 @@ static int ocfs2_move_extent(struct ocfs2_move_extents_context *context, */ ocfs2_probe_alloc_group(inode, gd_bh, &goal_bit, len, move_max_hop, new_phys_cpos); - if (!new_phys_cpos) { + if (!*new_phys_cpos) { ret = -ENOSPC; goto out_commit; } @@ -1059,7 +1059,7 @@ int ocfs2_ioctl_move_extents(struct file *filp, void __user *argp) struct ocfs2_move_extents range; struct ocfs2_move_extents_context *context = NULL; - status = mnt_want_write(filp->f_path.mnt); + status = mnt_want_write_file(filp); if (status) return status; @@ -1145,7 +1145,7 @@ out: kfree(context); - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return status; } diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index a8b2bfea574e..be244692550d 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -185,7 +185,7 @@ bail: return ret; } -static struct inode *ocfs2_get_init_inode(struct inode *dir, int mode) +static struct inode *ocfs2_get_init_inode(struct inode *dir, umode_t mode) { struct inode *inode; @@ -207,7 +207,7 @@ static struct inode *ocfs2_get_init_inode(struct inode *dir, int mode) static int ocfs2_mknod(struct inode *dir, struct dentry *dentry, - int mode, + umode_t mode, dev_t dev) { int status = 0; @@ -602,7 +602,7 @@ static int ocfs2_mknod_locked(struct ocfs2_super *osb, static int ocfs2_mkdir(struct inode *dir, struct dentry *dentry, - int mode) + umode_t mode) { int ret; @@ -617,7 +617,7 @@ static int ocfs2_mkdir(struct inode *dir, static int ocfs2_create(struct inode *dir, struct dentry *dentry, - int mode, + umode_t mode, struct nameidata *nd) { int ret; diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index 409285854f64..d355e6e36b36 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -836,18 +836,65 @@ static inline unsigned int ocfs2_clusters_to_megabytes(struct super_block *sb, static inline void _ocfs2_set_bit(unsigned int bit, unsigned long *bitmap) { - __test_and_set_bit_le(bit, bitmap); + __set_bit_le(bit, bitmap); } #define ocfs2_set_bit(bit, addr) _ocfs2_set_bit((bit), (unsigned long *)(addr)) static inline void _ocfs2_clear_bit(unsigned int bit, unsigned long *bitmap) { - __test_and_clear_bit_le(bit, bitmap); + __clear_bit_le(bit, bitmap); } #define ocfs2_clear_bit(bit, addr) _ocfs2_clear_bit((bit), (unsigned long *)(addr)) #define ocfs2_test_bit test_bit_le #define ocfs2_find_next_zero_bit find_next_zero_bit_le #define ocfs2_find_next_bit find_next_bit_le + +static inline void *correct_addr_and_bit_unaligned(int *bit, void *addr) +{ +#if BITS_PER_LONG == 64 + *bit += ((unsigned long) addr & 7UL) << 3; + addr = (void *) ((unsigned long) addr & ~7UL); +#elif BITS_PER_LONG == 32 + *bit += ((unsigned long) addr & 3UL) << 3; + addr = (void *) ((unsigned long) addr & ~3UL); +#else +#error "how many bits you are?!" +#endif + return addr; +} + +static inline void ocfs2_set_bit_unaligned(int bit, void *bitmap) +{ + bitmap = correct_addr_and_bit_unaligned(&bit, bitmap); + ocfs2_set_bit(bit, bitmap); +} + +static inline void ocfs2_clear_bit_unaligned(int bit, void *bitmap) +{ + bitmap = correct_addr_and_bit_unaligned(&bit, bitmap); + ocfs2_clear_bit(bit, bitmap); +} + +static inline int ocfs2_test_bit_unaligned(int bit, void *bitmap) +{ + bitmap = correct_addr_and_bit_unaligned(&bit, bitmap); + return ocfs2_test_bit(bit, bitmap); +} + +static inline int ocfs2_find_next_zero_bit_unaligned(void *bitmap, int max, + int start) +{ + int fix = 0, ret, tmpmax; + bitmap = correct_addr_and_bit_unaligned(&fix, bitmap); + tmpmax = max + fix; + start += fix; + + ret = ocfs2_find_next_zero_bit(bitmap, tmpmax, start) - fix; + if (ret > max) + return max; + return ret; +} + #endif /* OCFS2_H */ diff --git a/fs/ocfs2/quota_local.c b/fs/ocfs2/quota_local.c index dc8007fc9247..f100bf70a906 100644 --- a/fs/ocfs2/quota_local.c +++ b/fs/ocfs2/quota_local.c @@ -404,7 +404,9 @@ struct ocfs2_quota_recovery *ocfs2_begin_quota_recovery( int status = 0; struct ocfs2_quota_recovery *rec; - mlog(ML_NOTICE, "Beginning quota recovery in slot %u\n", slot_num); + printk(KERN_NOTICE "ocfs2: Beginning quota recovery on device (%s) for " + "slot %u\n", osb->dev_str, slot_num); + rec = ocfs2_alloc_quota_recovery(); if (!rec) return ERR_PTR(-ENOMEM); @@ -549,8 +551,8 @@ static int ocfs2_recover_local_quota_file(struct inode *lqinode, goto out_commit; } lock_buffer(qbh); - WARN_ON(!ocfs2_test_bit(bit, dchunk->dqc_bitmap)); - ocfs2_clear_bit(bit, dchunk->dqc_bitmap); + WARN_ON(!ocfs2_test_bit_unaligned(bit, dchunk->dqc_bitmap)); + ocfs2_clear_bit_unaligned(bit, dchunk->dqc_bitmap); le32_add_cpu(&dchunk->dqc_free, 1); unlock_buffer(qbh); ocfs2_journal_dirty(handle, qbh); @@ -596,7 +598,9 @@ int ocfs2_finish_quota_recovery(struct ocfs2_super *osb, struct inode *lqinode; unsigned int flags; - mlog(ML_NOTICE, "Finishing quota recovery in slot %u\n", slot_num); + printk(KERN_NOTICE "ocfs2: Finishing quota recovery on device (%s) for " + "slot %u\n", osb->dev_str, slot_num); + mutex_lock(&sb_dqopt(sb)->dqonoff_mutex); for (type = 0; type < MAXQUOTAS; type++) { if (list_empty(&(rec->r_list[type]))) @@ -612,8 +616,9 @@ int ocfs2_finish_quota_recovery(struct ocfs2_super *osb, /* Someone else is holding the lock? Then he must be * doing the recovery. Just skip the file... */ if (status == -EAGAIN) { - mlog(ML_NOTICE, "skipping quota recovery for slot %d " - "because quota file is locked.\n", slot_num); + printk(KERN_NOTICE "ocfs2: Skipping quota recovery on " + "device (%s) for slot %d because quota file is " + "locked.\n", osb->dev_str, slot_num); status = 0; goto out_put; } else if (status < 0) { @@ -944,7 +949,7 @@ static struct ocfs2_quota_chunk *ocfs2_find_free_entry(struct super_block *sb, * ol_quota_entries_per_block(sb); } - found = ocfs2_find_next_zero_bit(dchunk->dqc_bitmap, len, 0); + found = ocfs2_find_next_zero_bit_unaligned(dchunk->dqc_bitmap, len, 0); /* We failed? */ if (found == len) { mlog(ML_ERROR, "Did not find empty entry in chunk %d with %u" @@ -1208,7 +1213,7 @@ static void olq_alloc_dquot(struct buffer_head *bh, void *private) struct ocfs2_local_disk_chunk *dchunk; dchunk = (struct ocfs2_local_disk_chunk *)bh->b_data; - ocfs2_set_bit(*offset, dchunk->dqc_bitmap); + ocfs2_set_bit_unaligned(*offset, dchunk->dqc_bitmap); le32_add_cpu(&dchunk->dqc_free, -1); } @@ -1289,7 +1294,7 @@ int ocfs2_local_release_dquot(handle_t *handle, struct dquot *dquot) (od->dq_chunk->qc_headerbh->b_data); /* Mark structure as freed */ lock_buffer(od->dq_chunk->qc_headerbh); - ocfs2_clear_bit(offset, dchunk->dqc_bitmap); + ocfs2_clear_bit_unaligned(offset, dchunk->dqc_bitmap); le32_add_cpu(&dchunk->dqc_free, 1); unlock_buffer(od->dq_chunk->qc_headerbh); ocfs2_journal_dirty(handle, od->dq_chunk->qc_headerbh); diff --git a/fs/ocfs2/slot_map.c b/fs/ocfs2/slot_map.c index 26fc0014d509..1424c151cccc 100644 --- a/fs/ocfs2/slot_map.c +++ b/fs/ocfs2/slot_map.c @@ -493,8 +493,8 @@ int ocfs2_find_slot(struct ocfs2_super *osb) goto bail; } } else - mlog(ML_NOTICE, "slot %d is already allocated to this node!\n", - slot); + printk(KERN_INFO "ocfs2: Slot %d on device (%s) was already " + "allocated to this node!\n", slot, osb->dev_str); ocfs2_set_slot(si, slot, osb->node_num); osb->slot_num = slot; diff --git a/fs/ocfs2/stack_o2cb.c b/fs/ocfs2/stack_o2cb.c index 19965b00c43c..94368017edb3 100644 --- a/fs/ocfs2/stack_o2cb.c +++ b/fs/ocfs2/stack_o2cb.c @@ -28,6 +28,7 @@ #include "cluster/masklog.h" #include "cluster/nodemanager.h" #include "cluster/heartbeat.h" +#include "cluster/tcp.h" #include "stackglue.h" @@ -256,6 +257,61 @@ static void o2cb_dump_lksb(struct ocfs2_dlm_lksb *lksb) } /* + * Check if this node is heartbeating and is connected to all other + * heartbeating nodes. + */ +static int o2cb_cluster_check(void) +{ + u8 node_num; + int i; + unsigned long hbmap[BITS_TO_LONGS(O2NM_MAX_NODES)]; + unsigned long netmap[BITS_TO_LONGS(O2NM_MAX_NODES)]; + + node_num = o2nm_this_node(); + if (node_num == O2NM_MAX_NODES) { + printk(KERN_ERR "o2cb: This node has not been configured.\n"); + return -EINVAL; + } + + /* + * o2dlm expects o2net sockets to be created. If not, then + * dlm_join_domain() fails with a stack of errors which are both cryptic + * and incomplete. The idea here is to detect upfront whether we have + * managed to connect to all nodes or not. If not, then list the nodes + * to allow the user to check the configuration (incorrect IP, firewall, + * etc.) Yes, this is racy. But its not the end of the world. + */ +#define O2CB_MAP_STABILIZE_COUNT 60 + for (i = 0; i < O2CB_MAP_STABILIZE_COUNT; ++i) { + o2hb_fill_node_map(hbmap, sizeof(hbmap)); + if (!test_bit(node_num, hbmap)) { + printk(KERN_ERR "o2cb: %s heartbeat has not been " + "started.\n", (o2hb_global_heartbeat_active() ? + "Global" : "Local")); + return -EINVAL; + } + o2net_fill_node_map(netmap, sizeof(netmap)); + /* Force set the current node to allow easy compare */ + set_bit(node_num, netmap); + if (!memcmp(hbmap, netmap, sizeof(hbmap))) + return 0; + if (i < O2CB_MAP_STABILIZE_COUNT) + msleep(1000); + } + + printk(KERN_ERR "o2cb: This node could not connect to nodes:"); + i = -1; + while ((i = find_next_bit(hbmap, O2NM_MAX_NODES, + i + 1)) < O2NM_MAX_NODES) { + if (!test_bit(i, netmap)) + printk(" %u", i); + } + printk(".\n"); + + return -ENOTCONN; +} + +/* * Called from the dlm when it's about to evict a node. This is how the * classic stack signals node death. */ @@ -263,8 +319,8 @@ static void o2dlm_eviction_cb(int node_num, void *data) { struct ocfs2_cluster_connection *conn = data; - mlog(ML_NOTICE, "o2dlm has evicted node %d from group %.*s\n", - node_num, conn->cc_namelen, conn->cc_name); + printk(KERN_NOTICE "o2cb: o2dlm has evicted node %d from domain %.*s\n", + node_num, conn->cc_namelen, conn->cc_name); conn->cc_recovery_handler(node_num, conn->cc_recovery_data); } @@ -280,12 +336,11 @@ static int o2cb_cluster_connect(struct ocfs2_cluster_connection *conn) BUG_ON(conn == NULL); BUG_ON(conn->cc_proto == NULL); - /* for now we only have one cluster/node, make sure we see it - * in the heartbeat universe */ - if (!o2hb_check_local_node_heartbeating()) { - if (o2hb_global_heartbeat_active()) - mlog(ML_ERROR, "Global heartbeat not started\n"); - rc = -EINVAL; + /* Ensure cluster stack is up and all nodes are connected */ + rc = o2cb_cluster_check(); + if (rc) { + printk(KERN_ERR "o2cb: Cluster check failed. Fix errors " + "before retrying.\n"); goto out; } diff --git a/fs/ocfs2/stack_user.c b/fs/ocfs2/stack_user.c index a5ebe421195f..286edf1e231f 100644 --- a/fs/ocfs2/stack_user.c +++ b/fs/ocfs2/stack_user.c @@ -827,8 +827,8 @@ static int user_cluster_connect(struct ocfs2_cluster_connection *conn) goto out; } - rc = dlm_new_lockspace(conn->cc_name, strlen(conn->cc_name), - &fsdlm, DLM_LSFL_FS, DLM_LVB_LEN); + rc = dlm_new_lockspace(conn->cc_name, NULL, DLM_LSFL_FS, DLM_LVB_LEN, + NULL, NULL, NULL, &fsdlm); if (rc) { ocfs2_live_connection_drop(control); goto out; diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 56f61027236b..604e12c4e979 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -54,6 +54,7 @@ #include "ocfs1_fs_compat.h" #include "alloc.h" +#include "aops.h" #include "blockcheck.h" #include "dlmglue.h" #include "export.h" @@ -107,7 +108,7 @@ static int ocfs2_parse_options(struct super_block *sb, char *options, int is_remount); static int ocfs2_check_set_options(struct super_block *sb, struct mount_options *options); -static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt); +static int ocfs2_show_options(struct seq_file *s, struct dentry *root); static void ocfs2_put_super(struct super_block *sb); static int ocfs2_mount_volume(struct super_block *sb); static int ocfs2_remount(struct super_block *sb, int *flags, char *data); @@ -568,7 +569,6 @@ static struct inode *ocfs2_alloc_inode(struct super_block *sb) static void ocfs2_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(ocfs2_inode_cachep, OCFS2_I(inode)); } @@ -1107,9 +1107,9 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent) ocfs2_set_ro_flag(osb, 1); - printk(KERN_NOTICE "Readonly device detected. No cluster " - "services will be utilized for this mount. Recovery " - "will be skipped.\n"); + printk(KERN_NOTICE "ocfs2: Readonly device (%s) detected. " + "Cluster services will not be used for this mount. " + "Recovery will be skipped.\n", osb->dev_str); } if (!ocfs2_is_hard_readonly(osb)) { @@ -1533,9 +1533,9 @@ bail: return status; } -static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt) +static int ocfs2_show_options(struct seq_file *s, struct dentry *root) { - struct ocfs2_super *osb = OCFS2_SB(mnt->mnt_sb); + struct ocfs2_super *osb = OCFS2_SB(root->d_sb); unsigned long opts = osb->s_mount_opt; unsigned int local_alloc_megs; @@ -1567,8 +1567,7 @@ static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt) if (osb->preferred_slot != OCFS2_INVALID_SLOT) seq_printf(s, ",preferred_slot=%d", osb->preferred_slot); - if (!(mnt->mnt_flags & MNT_NOATIME) && !(mnt->mnt_flags & MNT_RELATIME)) - seq_printf(s, ",atime_quantum=%u", osb->s_atime_quantum); + seq_printf(s, ",atime_quantum=%u", osb->s_atime_quantum); if (osb->osb_commit_interval) seq_printf(s, ",commit=%u", @@ -1616,12 +1615,17 @@ static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt) return 0; } +wait_queue_head_t ocfs2__ioend_wq[OCFS2_IOEND_WQ_HASH_SZ]; + static int __init ocfs2_init(void) { - int status; + int status, i; ocfs2_print_version(); + for (i = 0; i < OCFS2_IOEND_WQ_HASH_SZ; i++) + init_waitqueue_head(&ocfs2__ioend_wq[i]); + status = init_ocfs2_uptodate_cache(); if (status < 0) { mlog_errno(status); @@ -1760,7 +1764,7 @@ static void ocfs2_inode_init_once(void *data) ocfs2_extent_map_init(&oi->vfs_inode); INIT_LIST_HEAD(&oi->ip_io_markers); oi->ip_dir_start_lookup = 0; - + atomic_set(&oi->ip_unaligned_aio, 0); init_rwsem(&oi->ip_alloc_sem); init_rwsem(&oi->ip_xattr_sem); mutex_init(&oi->ip_io_mutex); @@ -1974,7 +1978,8 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err) * If we failed before we got a uuid_str yet, we can't stop * heartbeat. Otherwise, do it. */ - if (!mnt_err && !ocfs2_mount_local(osb) && osb->uuid_str) + if (!mnt_err && !ocfs2_mount_local(osb) && osb->uuid_str && + !ocfs2_is_hard_readonly(osb)) hangup_needed = 1; if (osb->cconn) @@ -2353,7 +2358,7 @@ static int ocfs2_initialize_super(struct super_block *sb, mlog_errno(status); goto bail; } - cleancache_init_shared_fs((char *)&uuid_net_key, sb); + cleancache_init_shared_fs((char *)&di->id2.i_super.s_uuid, sb); bail: return status; @@ -2462,8 +2467,8 @@ static int ocfs2_check_volume(struct ocfs2_super *osb) goto finally; } } else { - mlog(ML_NOTICE, "File system was not unmounted cleanly, " - "recovering volume.\n"); + printk(KERN_NOTICE "ocfs2: File system on device (%s) was not " + "unmounted cleanly, recovering it.\n", osb->dev_str); } local = ocfs2_mount_local(osb); diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c index 194fb22ef79d..0ba9ea1e7961 100644 --- a/fs/ocfs2/xattr.c +++ b/fs/ocfs2/xattr.c @@ -623,7 +623,7 @@ int ocfs2_calc_security_init(struct inode *dir, int ocfs2_calc_xattr_init(struct inode *dir, struct buffer_head *dir_bh, - int mode, + umode_t mode, struct ocfs2_security_xattr_info *si, int *want_clusters, int *xattr_credits, @@ -2376,16 +2376,18 @@ static int ocfs2_remove_value_outside(struct inode*inode, } ret = ocfs2_xattr_value_truncate(inode, vb, 0, &ctxt); - if (ret < 0) { - mlog_errno(ret); - break; - } ocfs2_commit_trans(osb, ctxt.handle); if (ctxt.meta_ac) { ocfs2_free_alloc_context(ctxt.meta_ac); ctxt.meta_ac = NULL; } + + if (ret < 0) { + mlog_errno(ret); + break; + } + } if (ctxt.meta_ac) diff --git a/fs/ocfs2/xattr.h b/fs/ocfs2/xattr.h index d63cfb72316b..e5c7f15465b4 100644 --- a/fs/ocfs2/xattr.h +++ b/fs/ocfs2/xattr.h @@ -68,7 +68,7 @@ int ocfs2_calc_security_init(struct inode *, struct ocfs2_security_xattr_info *, int *, int *, struct ocfs2_alloc_context **); int ocfs2_calc_xattr_init(struct inode *, struct buffer_head *, - int, struct ocfs2_security_xattr_info *, + umode_t, struct ocfs2_security_xattr_info *, int *, int *, int *); /* diff --git a/fs/omfs/dir.c b/fs/omfs/dir.c index 98e544274390..f00576ec320f 100644 --- a/fs/omfs/dir.c +++ b/fs/omfs/dir.c @@ -255,7 +255,7 @@ static int omfs_remove(struct inode *dir, struct dentry *dentry) return 0; } -static int omfs_add_node(struct inode *dir, struct dentry *dentry, int mode) +static int omfs_add_node(struct inode *dir, struct dentry *dentry, umode_t mode) { int err; struct inode *inode = omfs_new_inode(dir, mode); @@ -279,12 +279,12 @@ out_free_inode: return err; } -static int omfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int omfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { return omfs_add_node(dir, dentry, mode | S_IFDIR); } -static int omfs_create(struct inode *dir, struct dentry *dentry, int mode, +static int omfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { return omfs_add_node(dir, dentry, mode | S_IFREG); diff --git a/fs/omfs/inode.c b/fs/omfs/inode.c index e043c4cb9a97..6065bb0ba207 100644 --- a/fs/omfs/inode.c +++ b/fs/omfs/inode.c @@ -28,7 +28,7 @@ struct buffer_head *omfs_bread(struct super_block *sb, sector_t block) return sb_bread(sb, clus_to_blk(sbi, block)); } -struct inode *omfs_new_inode(struct inode *dir, int mode) +struct inode *omfs_new_inode(struct inode *dir, umode_t mode) { struct inode *inode; u64 new_block; diff --git a/fs/omfs/omfs.h b/fs/omfs/omfs.h index 7d414fef501a..8941f12c6b01 100644 --- a/fs/omfs/omfs.h +++ b/fs/omfs/omfs.h @@ -60,7 +60,7 @@ extern int omfs_shrink_inode(struct inode *inode); /* inode.c */ extern struct buffer_head *omfs_bread(struct super_block *sb, sector_t block); extern struct inode *omfs_iget(struct super_block *sb, ino_t inode); -extern struct inode *omfs_new_inode(struct inode *dir, int mode); +extern struct inode *omfs_new_inode(struct inode *dir, umode_t mode); extern int omfs_reserve_block(struct super_block *sb, sector_t block); extern int omfs_find_empty_block(struct super_block *sb, int mode, ino_t *ino); extern int omfs_sync_inode(struct inode *inode); diff --git a/fs/open.c b/fs/open.c index 22c41b543f2d..77becc041149 100644 --- a/fs/open.c +++ b/fs/open.c @@ -456,7 +456,7 @@ static int chmod_common(struct path *path, umode_t mode) if (error) return error; mutex_lock(&inode->i_mutex); - error = security_path_chmod(path->dentry, path->mnt, mode); + error = security_path_chmod(path, mode); if (error) goto out_unlock; newattrs.ia_mode = (mode & S_IALLUGO) | (inode->i_mode & ~S_IALLUGO); @@ -468,7 +468,7 @@ out_unlock: return error; } -SYSCALL_DEFINE2(fchmod, unsigned int, fd, mode_t, mode) +SYSCALL_DEFINE2(fchmod, unsigned int, fd, umode_t, mode) { struct file * file; int err = -EBADF; @@ -482,7 +482,7 @@ SYSCALL_DEFINE2(fchmod, unsigned int, fd, mode_t, mode) return err; } -SYSCALL_DEFINE3(fchmodat, int, dfd, const char __user *, filename, mode_t, mode) +SYSCALL_DEFINE3(fchmodat, int, dfd, const char __user *, filename, umode_t, mode) { struct path path; int error; @@ -495,7 +495,7 @@ SYSCALL_DEFINE3(fchmodat, int, dfd, const char __user *, filename, mode_t, mode) return error; } -SYSCALL_DEFINE2(chmod, const char __user *, filename, mode_t, mode) +SYSCALL_DEFINE2(chmod, const char __user *, filename, umode_t, mode) { return sys_fchmodat(AT_FDCWD, filename, mode); } @@ -608,7 +608,7 @@ SYSCALL_DEFINE3(fchown, unsigned int, fd, uid_t, user, gid_t, group) dentry = file->f_path.dentry; audit_inode(NULL, dentry); error = chown_common(&file->f_path, user, group); - mnt_drop_write(file->f_path.mnt); + mnt_drop_write_file(file); out_fput: fput(file); out: @@ -877,7 +877,7 @@ void fd_install(unsigned int fd, struct file *file) EXPORT_SYMBOL(fd_install); -static inline int build_open_flags(int flags, int mode, struct open_flags *op) +static inline int build_open_flags(int flags, umode_t mode, struct open_flags *op) { int lookup_flags = 0; int acc_mode; @@ -948,7 +948,7 @@ static inline int build_open_flags(int flags, int mode, struct open_flags *op) * have to. But in generally you should not do this, so please move * along, nothing to see here.. */ -struct file *filp_open(const char *filename, int flags, int mode) +struct file *filp_open(const char *filename, int flags, umode_t mode) { struct open_flags op; int lookup = build_open_flags(flags, mode, &op); @@ -970,7 +970,7 @@ struct file *file_open_root(struct dentry *dentry, struct vfsmount *mnt, } EXPORT_SYMBOL(file_open_root); -long do_sys_open(int dfd, const char __user *filename, int flags, int mode) +long do_sys_open(int dfd, const char __user *filename, int flags, umode_t mode) { struct open_flags op; int lookup = build_open_flags(flags, mode, &op); @@ -994,7 +994,7 @@ long do_sys_open(int dfd, const char __user *filename, int flags, int mode) return fd; } -SYSCALL_DEFINE3(open, const char __user *, filename, int, flags, int, mode) +SYSCALL_DEFINE3(open, const char __user *, filename, int, flags, umode_t, mode) { long ret; @@ -1008,7 +1008,7 @@ SYSCALL_DEFINE3(open, const char __user *, filename, int, flags, int, mode) } SYSCALL_DEFINE4(openat, int, dfd, const char __user *, filename, int, flags, - int, mode) + umode_t, mode) { long ret; @@ -1027,7 +1027,7 @@ SYSCALL_DEFINE4(openat, int, dfd, const char __user *, filename, int, flags, * For backward compatibility? Maybe this should be moved * into arch/i386 instead? */ -SYSCALL_DEFINE2(creat, const char __user *, pathname, int, mode) +SYSCALL_DEFINE2(creat, const char __user *, pathname, umode_t, mode) { return sys_open(pathname, O_CREAT | O_WRONLY | O_TRUNC, mode); } diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index e4e0ff7962e2..a88c03bc749d 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -346,7 +346,6 @@ static struct inode *openprom_alloc_inode(struct super_block *sb) static void openprom_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(op_inode_cachep, OP_I(inode)); } diff --git a/fs/partitions/Kconfig b/fs/partitions/Kconfig deleted file mode 100644 index cb5f0a3f1b03..000000000000 --- a/fs/partitions/Kconfig +++ /dev/null @@ -1,251 +0,0 @@ -# -# Partition configuration -# -config PARTITION_ADVANCED - bool "Advanced partition selection" - help - Say Y here if you would like to use hard disks under Linux which - were partitioned under an operating system running on a different - architecture than your Linux system. - - Note that the answer to this question won't directly affect the - kernel: saying N will just cause the configurator to skip all - the questions about foreign partitioning schemes. - - If unsure, say N. - -config ACORN_PARTITION - bool "Acorn partition support" if PARTITION_ADVANCED - default y if ARCH_ACORN - help - Support hard disks partitioned under Acorn operating systems. - -config ACORN_PARTITION_CUMANA - bool "Cumana partition support" if PARTITION_ADVANCED - default y if ARCH_ACORN - depends on ACORN_PARTITION - help - Say Y here if you would like to use hard disks under Linux which - were partitioned using the Cumana interface on Acorn machines. - -config ACORN_PARTITION_EESOX - bool "EESOX partition support" if PARTITION_ADVANCED - default y if ARCH_ACORN - depends on ACORN_PARTITION - -config ACORN_PARTITION_ICS - bool "ICS partition support" if PARTITION_ADVANCED - default y if ARCH_ACORN - depends on ACORN_PARTITION - help - Say Y here if you would like to use hard disks under Linux which - were partitioned using the ICS interface on Acorn machines. - -config ACORN_PARTITION_ADFS - bool "Native filecore partition support" if PARTITION_ADVANCED - default y if ARCH_ACORN - depends on ACORN_PARTITION - help - The Acorn Disc Filing System is the standard file system of the - RiscOS operating system which runs on Acorn's ARM-based Risc PC - systems and the Acorn Archimedes range of machines. If you say - `Y' here, Linux will support disk partitions created under ADFS. - -config ACORN_PARTITION_POWERTEC - bool "PowerTec partition support" if PARTITION_ADVANCED - default y if ARCH_ACORN - depends on ACORN_PARTITION - help - Support reading partition tables created on Acorn machines using - the PowerTec SCSI drive. - -config ACORN_PARTITION_RISCIX - bool "RISCiX partition support" if PARTITION_ADVANCED - default y if ARCH_ACORN - depends on ACORN_PARTITION - help - Once upon a time, there was a native Unix port for the Acorn series - of machines called RISCiX. If you say 'Y' here, Linux will be able - to read disks partitioned under RISCiX. - -config OSF_PARTITION - bool "Alpha OSF partition support" if PARTITION_ADVANCED - default y if ALPHA - help - Say Y here if you would like to use hard disks under Linux which - were partitioned on an Alpha machine. - -config AMIGA_PARTITION - bool "Amiga partition table support" if PARTITION_ADVANCED - default y if (AMIGA || AFFS_FS=y) - help - Say Y here if you would like to use hard disks under Linux which - were partitioned under AmigaOS. - -config ATARI_PARTITION - bool "Atari partition table support" if PARTITION_ADVANCED - default y if ATARI - help - Say Y here if you would like to use hard disks under Linux which - were partitioned under the Atari OS. - -config IBM_PARTITION - bool "IBM disk label and partition support" - depends on PARTITION_ADVANCED && S390 - help - Say Y here if you would like to be able to read the hard disk - partition table format used by IBM DASD disks operating under CMS. - Otherwise, say N. - -config MAC_PARTITION - bool "Macintosh partition map support" if PARTITION_ADVANCED - default y if (MAC || PPC_PMAC) - help - Say Y here if you would like to use hard disks under Linux which - were partitioned on a Macintosh. - -config MSDOS_PARTITION - bool "PC BIOS (MSDOS partition tables) support" if PARTITION_ADVANCED - default y - help - Say Y here. - -config BSD_DISKLABEL - bool "BSD disklabel (FreeBSD partition tables) support" - depends on PARTITION_ADVANCED && MSDOS_PARTITION - help - FreeBSD uses its own hard disk partition scheme on your PC. It - requires only one entry in the primary partition table of your disk - and manages it similarly to DOS extended partitions, putting in its - first sector a new partition table in BSD disklabel format. Saying Y - here allows you to read these disklabels and further mount FreeBSD - partitions from within Linux if you have also said Y to "UFS - file system support", above. If you don't know what all this is - about, say N. - -config MINIX_SUBPARTITION - bool "Minix subpartition support" - depends on PARTITION_ADVANCED && MSDOS_PARTITION - help - Minix 2.0.0/2.0.2 subpartition table support for Linux. - Say Y here if you want to mount and use Minix 2.0.0/2.0.2 - subpartitions. - -config SOLARIS_X86_PARTITION - bool "Solaris (x86) partition table support" - depends on PARTITION_ADVANCED && MSDOS_PARTITION - help - Like most systems, Solaris x86 uses its own hard disk partition - table format, incompatible with all others. Saying Y here allows you - to read these partition tables and further mount Solaris x86 - partitions from within Linux if you have also said Y to "UFS - file system support", above. - -config UNIXWARE_DISKLABEL - bool "Unixware slices support" - depends on PARTITION_ADVANCED && MSDOS_PARTITION - ---help--- - Like some systems, UnixWare uses its own slice table inside a - partition (VTOC - Virtual Table of Contents). Its format is - incompatible with all other OSes. Saying Y here allows you to read - VTOC and further mount UnixWare partitions read-only from within - Linux if you have also said Y to "UFS file system support" or - "System V and Coherent file system support", above. - - This is mainly used to carry data from a UnixWare box to your - Linux box via a removable medium like magneto-optical, ZIP or - removable IDE drives. Note, however, that a good portable way to - transport files and directories between unixes (and even other - operating systems) is given by the tar program ("man tar" or - preferably "info tar"). - - If you don't know what all this is about, say N. - -config LDM_PARTITION - bool "Windows Logical Disk Manager (Dynamic Disk) support" - depends on PARTITION_ADVANCED - ---help--- - Say Y here if you would like to use hard disks under Linux which - were partitioned using Windows 2000's/XP's or Vista's Logical Disk - Manager. They are also known as "Dynamic Disks". - - Note this driver only supports Dynamic Disks with a protective MBR - label, i.e. DOS partition table. It does not support GPT labelled - Dynamic Disks yet as can be created with Vista. - - Windows 2000 introduced the concept of Dynamic Disks to get around - the limitations of the PC's partitioning scheme. The Logical Disk - Manager allows the user to repartition a disk and create spanned, - mirrored, striped or RAID volumes, all without the need for - rebooting. - - Normal partitions are now called Basic Disks under Windows 2000, XP, - and Vista. - - For a fuller description read <file:Documentation/ldm.txt>. - - If unsure, say N. - -config LDM_DEBUG - bool "Windows LDM extra logging" - depends on LDM_PARTITION - help - Say Y here if you would like LDM to log verbosely. This could be - helpful if the driver doesn't work as expected and you'd like to - report a bug. - - If unsure, say N. - -config SGI_PARTITION - bool "SGI partition support" if PARTITION_ADVANCED - default y if DEFAULT_SGI_PARTITION - help - Say Y here if you would like to be able to read the hard disk - partition table format used by SGI machines. - -config ULTRIX_PARTITION - bool "Ultrix partition table support" if PARTITION_ADVANCED - default y if MACH_DECSTATION - help - Say Y here if you would like to be able to read the hard disk - partition table format used by DEC (now Compaq) Ultrix machines. - Otherwise, say N. - -config SUN_PARTITION - bool "Sun partition tables support" if PARTITION_ADVANCED - default y if (SPARC || SUN3 || SUN3X) - ---help--- - Like most systems, SunOS uses its own hard disk partition table - format, incompatible with all others. Saying Y here allows you to - read these partition tables and further mount SunOS partitions from - within Linux if you have also said Y to "UFS file system support", - above. This is mainly used to carry data from a SPARC under SunOS to - your Linux box via a removable medium like magneto-optical or ZIP - drives; note however that a good portable way to transport files and - directories between unixes (and even other operating systems) is - given by the tar program ("man tar" or preferably "info tar"). If - you don't know what all this is about, say N. - -config KARMA_PARTITION - bool "Karma Partition support" - depends on PARTITION_ADVANCED - help - Say Y here if you would like to mount the Rio Karma MP3 player, as it - uses a proprietary partition table. - -config EFI_PARTITION - bool "EFI GUID Partition support" - depends on PARTITION_ADVANCED - select CRC32 - help - Say Y here if you would like to use hard disks under Linux which - were partitioned using EFI GPT. - -config SYSV68_PARTITION - bool "SYSV68 partition table support" if PARTITION_ADVANCED - default y if VME - help - Say Y here if you would like to be able to read the hard disk - partition table format used by Motorola Delta machines (using - sysv68). - Otherwise, say N. diff --git a/fs/partitions/Makefile b/fs/partitions/Makefile deleted file mode 100644 index 03af8eac51da..000000000000 --- a/fs/partitions/Makefile +++ /dev/null @@ -1,20 +0,0 @@ -# -# Makefile for the linux kernel. -# - -obj-$(CONFIG_BLOCK) := check.o - -obj-$(CONFIG_ACORN_PARTITION) += acorn.o -obj-$(CONFIG_AMIGA_PARTITION) += amiga.o -obj-$(CONFIG_ATARI_PARTITION) += atari.o -obj-$(CONFIG_MAC_PARTITION) += mac.o -obj-$(CONFIG_LDM_PARTITION) += ldm.o -obj-$(CONFIG_MSDOS_PARTITION) += msdos.o -obj-$(CONFIG_OSF_PARTITION) += osf.o -obj-$(CONFIG_SGI_PARTITION) += sgi.o -obj-$(CONFIG_SUN_PARTITION) += sun.o -obj-$(CONFIG_ULTRIX_PARTITION) += ultrix.o -obj-$(CONFIG_IBM_PARTITION) += ibm.o -obj-$(CONFIG_EFI_PARTITION) += efi.o -obj-$(CONFIG_KARMA_PARTITION) += karma.o -obj-$(CONFIG_SYSV68_PARTITION) += sysv68.o diff --git a/fs/partitions/acorn.c b/fs/partitions/acorn.c deleted file mode 100644 index fbeb697374d5..000000000000 --- a/fs/partitions/acorn.c +++ /dev/null @@ -1,556 +0,0 @@ -/* - * linux/fs/partitions/acorn.c - * - * Copyright (c) 1996-2000 Russell King. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - * - * Scan ADFS partitions on hard disk drives. Unfortunately, there - * isn't a standard for partitioning drives on Acorn machines, so - * every single manufacturer of SCSI and IDE cards created their own - * method. - */ -#include <linux/buffer_head.h> -#include <linux/adfs_fs.h> - -#include "check.h" -#include "acorn.h" - -/* - * Partition types. (Oh for reusability) - */ -#define PARTITION_RISCIX_MFM 1 -#define PARTITION_RISCIX_SCSI 2 -#define PARTITION_LINUX 9 - -#if defined(CONFIG_ACORN_PARTITION_CUMANA) || \ - defined(CONFIG_ACORN_PARTITION_ADFS) -static struct adfs_discrecord * -adfs_partition(struct parsed_partitions *state, char *name, char *data, - unsigned long first_sector, int slot) -{ - struct adfs_discrecord *dr; - unsigned int nr_sects; - - if (adfs_checkbblk(data)) - return NULL; - - dr = (struct adfs_discrecord *)(data + 0x1c0); - - if (dr->disc_size == 0 && dr->disc_size_high == 0) - return NULL; - - nr_sects = (le32_to_cpu(dr->disc_size_high) << 23) | - (le32_to_cpu(dr->disc_size) >> 9); - - if (name) { - strlcat(state->pp_buf, " [", PAGE_SIZE); - strlcat(state->pp_buf, name, PAGE_SIZE); - strlcat(state->pp_buf, "]", PAGE_SIZE); - } - put_partition(state, slot, first_sector, nr_sects); - return dr; -} -#endif - -#ifdef CONFIG_ACORN_PARTITION_RISCIX - -struct riscix_part { - __le32 start; - __le32 length; - __le32 one; - char name[16]; -}; - -struct riscix_record { - __le32 magic; -#define RISCIX_MAGIC cpu_to_le32(0x4a657320) - __le32 date; - struct riscix_part part[8]; -}; - -#if defined(CONFIG_ACORN_PARTITION_CUMANA) || \ - defined(CONFIG_ACORN_PARTITION_ADFS) -static int riscix_partition(struct parsed_partitions *state, - unsigned long first_sect, int slot, - unsigned long nr_sects) -{ - Sector sect; - struct riscix_record *rr; - - rr = read_part_sector(state, first_sect, §); - if (!rr) - return -1; - - strlcat(state->pp_buf, " [RISCiX]", PAGE_SIZE); - - - if (rr->magic == RISCIX_MAGIC) { - unsigned long size = nr_sects > 2 ? 2 : nr_sects; - int part; - - strlcat(state->pp_buf, " <", PAGE_SIZE); - - put_partition(state, slot++, first_sect, size); - for (part = 0; part < 8; part++) { - if (rr->part[part].one && - memcmp(rr->part[part].name, "All\0", 4)) { - put_partition(state, slot++, - le32_to_cpu(rr->part[part].start), - le32_to_cpu(rr->part[part].length)); - strlcat(state->pp_buf, "(", PAGE_SIZE); - strlcat(state->pp_buf, rr->part[part].name, PAGE_SIZE); - strlcat(state->pp_buf, ")", PAGE_SIZE); - } - } - - strlcat(state->pp_buf, " >\n", PAGE_SIZE); - } else { - put_partition(state, slot++, first_sect, nr_sects); - } - - put_dev_sector(sect); - return slot; -} -#endif -#endif - -#define LINUX_NATIVE_MAGIC 0xdeafa1de -#define LINUX_SWAP_MAGIC 0xdeafab1e - -struct linux_part { - __le32 magic; - __le32 start_sect; - __le32 nr_sects; -}; - -#if defined(CONFIG_ACORN_PARTITION_CUMANA) || \ - defined(CONFIG_ACORN_PARTITION_ADFS) -static int linux_partition(struct parsed_partitions *state, - unsigned long first_sect, int slot, - unsigned long nr_sects) -{ - Sector sect; - struct linux_part *linuxp; - unsigned long size = nr_sects > 2 ? 2 : nr_sects; - - strlcat(state->pp_buf, " [Linux]", PAGE_SIZE); - - put_partition(state, slot++, first_sect, size); - - linuxp = read_part_sector(state, first_sect, §); - if (!linuxp) - return -1; - - strlcat(state->pp_buf, " <", PAGE_SIZE); - while (linuxp->magic == cpu_to_le32(LINUX_NATIVE_MAGIC) || - linuxp->magic == cpu_to_le32(LINUX_SWAP_MAGIC)) { - if (slot == state->limit) - break; - put_partition(state, slot++, first_sect + - le32_to_cpu(linuxp->start_sect), - le32_to_cpu(linuxp->nr_sects)); - linuxp ++; - } - strlcat(state->pp_buf, " >", PAGE_SIZE); - - put_dev_sector(sect); - return slot; -} -#endif - -#ifdef CONFIG_ACORN_PARTITION_CUMANA -int adfspart_check_CUMANA(struct parsed_partitions *state) -{ - unsigned long first_sector = 0; - unsigned int start_blk = 0; - Sector sect; - unsigned char *data; - char *name = "CUMANA/ADFS"; - int first = 1; - int slot = 1; - - /* - * Try Cumana style partitions - sector 6 contains ADFS boot block - * with pointer to next 'drive'. - * - * There are unknowns in this code - is the 'cylinder number' of the - * next partition relative to the start of this one - I'm assuming - * it is. - * - * Also, which ID did Cumana use? - * - * This is totally unfinished, and will require more work to get it - * going. Hence it is totally untested. - */ - do { - struct adfs_discrecord *dr; - unsigned int nr_sects; - - data = read_part_sector(state, start_blk * 2 + 6, §); - if (!data) - return -1; - - if (slot == state->limit) - break; - - dr = adfs_partition(state, name, data, first_sector, slot++); - if (!dr) - break; - - name = NULL; - - nr_sects = (data[0x1fd] + (data[0x1fe] << 8)) * - (dr->heads + (dr->lowsector & 0x40 ? 1 : 0)) * - dr->secspertrack; - - if (!nr_sects) - break; - - first = 0; - first_sector += nr_sects; - start_blk += nr_sects >> (BLOCK_SIZE_BITS - 9); - nr_sects = 0; /* hmm - should be partition size */ - - switch (data[0x1fc] & 15) { - case 0: /* No partition / ADFS? */ - break; - -#ifdef CONFIG_ACORN_PARTITION_RISCIX - case PARTITION_RISCIX_SCSI: - /* RISCiX - we don't know how to find the next one. */ - slot = riscix_partition(state, first_sector, slot, - nr_sects); - break; -#endif - - case PARTITION_LINUX: - slot = linux_partition(state, first_sector, slot, - nr_sects); - break; - } - put_dev_sector(sect); - if (slot == -1) - return -1; - } while (1); - put_dev_sector(sect); - return first ? 0 : 1; -} -#endif - -#ifdef CONFIG_ACORN_PARTITION_ADFS -/* - * Purpose: allocate ADFS partitions. - * - * Params : hd - pointer to gendisk structure to store partition info. - * dev - device number to access. - * - * Returns: -1 on error, 0 for no ADFS boot sector, 1 for ok. - * - * Alloc : hda = whole drive - * hda1 = ADFS partition on first drive. - * hda2 = non-ADFS partition. - */ -int adfspart_check_ADFS(struct parsed_partitions *state) -{ - unsigned long start_sect, nr_sects, sectscyl, heads; - Sector sect; - unsigned char *data; - struct adfs_discrecord *dr; - unsigned char id; - int slot = 1; - - data = read_part_sector(state, 6, §); - if (!data) - return -1; - - dr = adfs_partition(state, "ADFS", data, 0, slot++); - if (!dr) { - put_dev_sector(sect); - return 0; - } - - heads = dr->heads + ((dr->lowsector >> 6) & 1); - sectscyl = dr->secspertrack * heads; - start_sect = ((data[0x1fe] << 8) + data[0x1fd]) * sectscyl; - id = data[0x1fc] & 15; - put_dev_sector(sect); - - /* - * Work out start of non-adfs partition. - */ - nr_sects = (state->bdev->bd_inode->i_size >> 9) - start_sect; - - if (start_sect) { - switch (id) { -#ifdef CONFIG_ACORN_PARTITION_RISCIX - case PARTITION_RISCIX_SCSI: - case PARTITION_RISCIX_MFM: - slot = riscix_partition(state, start_sect, slot, - nr_sects); - break; -#endif - - case PARTITION_LINUX: - slot = linux_partition(state, start_sect, slot, - nr_sects); - break; - } - } - strlcat(state->pp_buf, "\n", PAGE_SIZE); - return 1; -} -#endif - -#ifdef CONFIG_ACORN_PARTITION_ICS - -struct ics_part { - __le32 start; - __le32 size; -}; - -static int adfspart_check_ICSLinux(struct parsed_partitions *state, - unsigned long block) -{ - Sector sect; - unsigned char *data = read_part_sector(state, block, §); - int result = 0; - - if (data) { - if (memcmp(data, "LinuxPart", 9) == 0) - result = 1; - put_dev_sector(sect); - } - - return result; -} - -/* - * Check for a valid ICS partition using the checksum. - */ -static inline int valid_ics_sector(const unsigned char *data) -{ - unsigned long sum; - int i; - - for (i = 0, sum = 0x50617274; i < 508; i++) - sum += data[i]; - - sum -= le32_to_cpu(*(__le32 *)(&data[508])); - - return sum == 0; -} - -/* - * Purpose: allocate ICS partitions. - * Params : hd - pointer to gendisk structure to store partition info. - * dev - device number to access. - * Returns: -1 on error, 0 for no ICS table, 1 for partitions ok. - * Alloc : hda = whole drive - * hda1 = ADFS partition 0 on first drive. - * hda2 = ADFS partition 1 on first drive. - * ..etc.. - */ -int adfspart_check_ICS(struct parsed_partitions *state) -{ - const unsigned char *data; - const struct ics_part *p; - int slot; - Sector sect; - - /* - * Try ICS style partitions - sector 0 contains partition info. - */ - data = read_part_sector(state, 0, §); - if (!data) - return -1; - - if (!valid_ics_sector(data)) { - put_dev_sector(sect); - return 0; - } - - strlcat(state->pp_buf, " [ICS]", PAGE_SIZE); - - for (slot = 1, p = (const struct ics_part *)data; p->size; p++) { - u32 start = le32_to_cpu(p->start); - s32 size = le32_to_cpu(p->size); /* yes, it's signed. */ - - if (slot == state->limit) - break; - - /* - * Negative sizes tell the RISC OS ICS driver to ignore - * this partition - in effect it says that this does not - * contain an ADFS filesystem. - */ - if (size < 0) { - size = -size; - - /* - * Our own extension - We use the first sector - * of the partition to identify what type this - * partition is. We must not make this visible - * to the filesystem. - */ - if (size > 1 && adfspart_check_ICSLinux(state, start)) { - start += 1; - size -= 1; - } - } - - if (size) - put_partition(state, slot++, start, size); - } - - put_dev_sector(sect); - strlcat(state->pp_buf, "\n", PAGE_SIZE); - return 1; -} -#endif - -#ifdef CONFIG_ACORN_PARTITION_POWERTEC -struct ptec_part { - __le32 unused1; - __le32 unused2; - __le32 start; - __le32 size; - __le32 unused5; - char type[8]; -}; - -static inline int valid_ptec_sector(const unsigned char *data) -{ - unsigned char checksum = 0x2a; - int i; - - /* - * If it looks like a PC/BIOS partition, then it - * probably isn't PowerTec. - */ - if (data[510] == 0x55 && data[511] == 0xaa) - return 0; - - for (i = 0; i < 511; i++) - checksum += data[i]; - - return checksum == data[511]; -} - -/* - * Purpose: allocate ICS partitions. - * Params : hd - pointer to gendisk structure to store partition info. - * dev - device number to access. - * Returns: -1 on error, 0 for no ICS table, 1 for partitions ok. - * Alloc : hda = whole drive - * hda1 = ADFS partition 0 on first drive. - * hda2 = ADFS partition 1 on first drive. - * ..etc.. - */ -int adfspart_check_POWERTEC(struct parsed_partitions *state) -{ - Sector sect; - const unsigned char *data; - const struct ptec_part *p; - int slot = 1; - int i; - - data = read_part_sector(state, 0, §); - if (!data) - return -1; - - if (!valid_ptec_sector(data)) { - put_dev_sector(sect); - return 0; - } - - strlcat(state->pp_buf, " [POWERTEC]", PAGE_SIZE); - - for (i = 0, p = (const struct ptec_part *)data; i < 12; i++, p++) { - u32 start = le32_to_cpu(p->start); - u32 size = le32_to_cpu(p->size); - - if (size) - put_partition(state, slot++, start, size); - } - - put_dev_sector(sect); - strlcat(state->pp_buf, "\n", PAGE_SIZE); - return 1; -} -#endif - -#ifdef CONFIG_ACORN_PARTITION_EESOX -struct eesox_part { - char magic[6]; - char name[10]; - __le32 start; - __le32 unused6; - __le32 unused7; - __le32 unused8; -}; - -/* - * Guess who created this format? - */ -static const char eesox_name[] = { - 'N', 'e', 'i', 'l', ' ', - 'C', 'r', 'i', 't', 'c', 'h', 'e', 'l', 'l', ' ', ' ' -}; - -/* - * EESOX SCSI partition format. - * - * This is a goddamned awful partition format. We don't seem to store - * the size of the partition in this table, only the start addresses. - * - * There are two possibilities where the size comes from: - * 1. The individual ADFS boot block entries that are placed on the disk. - * 2. The start address of the next entry. - */ -int adfspart_check_EESOX(struct parsed_partitions *state) -{ - Sector sect; - const unsigned char *data; - unsigned char buffer[256]; - struct eesox_part *p; - sector_t start = 0; - int i, slot = 1; - - data = read_part_sector(state, 7, §); - if (!data) - return -1; - - /* - * "Decrypt" the partition table. God knows why... - */ - for (i = 0; i < 256; i++) - buffer[i] = data[i] ^ eesox_name[i & 15]; - - put_dev_sector(sect); - - for (i = 0, p = (struct eesox_part *)buffer; i < 8; i++, p++) { - sector_t next; - - if (memcmp(p->magic, "Eesox", 6)) - break; - - next = le32_to_cpu(p->start); - if (i) - put_partition(state, slot++, start, next - start); - start = next; - } - - if (i != 0) { - sector_t size; - - size = get_capacity(state->bdev->bd_disk); - put_partition(state, slot++, start, size - start); - strlcat(state->pp_buf, "\n", PAGE_SIZE); - } - - return i ? 1 : 0; -} -#endif diff --git a/fs/partitions/acorn.h b/fs/partitions/acorn.h deleted file mode 100644 index ede828529692..000000000000 --- a/fs/partitions/acorn.h +++ /dev/null @@ -1,14 +0,0 @@ -/* - * linux/fs/partitions/acorn.h - * - * Copyright (C) 1996-2001 Russell King. - * - * I _hate_ this partitioning mess - why can't we have one defined - * format, and everyone stick to it? - */ - -int adfspart_check_CUMANA(struct parsed_partitions *state); -int adfspart_check_ADFS(struct parsed_partitions *state); -int adfspart_check_ICS(struct parsed_partitions *state); -int adfspart_check_POWERTEC(struct parsed_partitions *state); -int adfspart_check_EESOX(struct parsed_partitions *state); diff --git a/fs/partitions/amiga.c b/fs/partitions/amiga.c deleted file mode 100644 index 70cbf44a1560..000000000000 --- a/fs/partitions/amiga.c +++ /dev/null @@ -1,139 +0,0 @@ -/* - * fs/partitions/amiga.c - * - * Code extracted from drivers/block/genhd.c - * - * Copyright (C) 1991-1998 Linus Torvalds - * Re-organised Feb 1998 Russell King - */ - -#include <linux/types.h> -#include <linux/affs_hardblocks.h> - -#include "check.h" -#include "amiga.h" - -static __inline__ u32 -checksum_block(__be32 *m, int size) -{ - u32 sum = 0; - - while (size--) - sum += be32_to_cpu(*m++); - return sum; -} - -int amiga_partition(struct parsed_partitions *state) -{ - Sector sect; - unsigned char *data; - struct RigidDiskBlock *rdb; - struct PartitionBlock *pb; - int start_sect, nr_sects, blk, part, res = 0; - int blksize = 1; /* Multiplier for disk block size */ - int slot = 1; - char b[BDEVNAME_SIZE]; - - for (blk = 0; ; blk++, put_dev_sector(sect)) { - if (blk == RDB_ALLOCATION_LIMIT) - goto rdb_done; - data = read_part_sector(state, blk, §); - if (!data) { - if (warn_no_part) - printk("Dev %s: unable to read RDB block %d\n", - bdevname(state->bdev, b), blk); - res = -1; - goto rdb_done; - } - if (*(__be32 *)data != cpu_to_be32(IDNAME_RIGIDDISK)) - continue; - - rdb = (struct RigidDiskBlock *)data; - if (checksum_block((__be32 *)data, be32_to_cpu(rdb->rdb_SummedLongs) & 0x7F) == 0) - break; - /* Try again with 0xdc..0xdf zeroed, Windows might have - * trashed it. - */ - *(__be32 *)(data+0xdc) = 0; - if (checksum_block((__be32 *)data, - be32_to_cpu(rdb->rdb_SummedLongs) & 0x7F)==0) { - printk("Warning: Trashed word at 0xd0 in block %d " - "ignored in checksum calculation\n",blk); - break; - } - - printk("Dev %s: RDB in block %d has bad checksum\n", - bdevname(state->bdev, b), blk); - } - - /* blksize is blocks per 512 byte standard block */ - blksize = be32_to_cpu( rdb->rdb_BlockBytes ) / 512; - - { - char tmp[7 + 10 + 1 + 1]; - - /* Be more informative */ - snprintf(tmp, sizeof(tmp), " RDSK (%d)", blksize * 512); - strlcat(state->pp_buf, tmp, PAGE_SIZE); - } - blk = be32_to_cpu(rdb->rdb_PartitionList); - put_dev_sector(sect); - for (part = 1; blk>0 && part<=16; part++, put_dev_sector(sect)) { - blk *= blksize; /* Read in terms partition table understands */ - data = read_part_sector(state, blk, §); - if (!data) { - if (warn_no_part) - printk("Dev %s: unable to read partition block %d\n", - bdevname(state->bdev, b), blk); - res = -1; - goto rdb_done; - } - pb = (struct PartitionBlock *)data; - blk = be32_to_cpu(pb->pb_Next); - if (pb->pb_ID != cpu_to_be32(IDNAME_PARTITION)) - continue; - if (checksum_block((__be32 *)pb, be32_to_cpu(pb->pb_SummedLongs) & 0x7F) != 0 ) - continue; - - /* Tell Kernel about it */ - - nr_sects = (be32_to_cpu(pb->pb_Environment[10]) + 1 - - be32_to_cpu(pb->pb_Environment[9])) * - be32_to_cpu(pb->pb_Environment[3]) * - be32_to_cpu(pb->pb_Environment[5]) * - blksize; - if (!nr_sects) - continue; - start_sect = be32_to_cpu(pb->pb_Environment[9]) * - be32_to_cpu(pb->pb_Environment[3]) * - be32_to_cpu(pb->pb_Environment[5]) * - blksize; - put_partition(state,slot++,start_sect,nr_sects); - { - /* Be even more informative to aid mounting */ - char dostype[4]; - char tmp[42]; - - __be32 *dt = (__be32 *)dostype; - *dt = pb->pb_Environment[16]; - if (dostype[3] < ' ') - snprintf(tmp, sizeof(tmp), " (%c%c%c^%c)", - dostype[0], dostype[1], - dostype[2], dostype[3] + '@' ); - else - snprintf(tmp, sizeof(tmp), " (%c%c%c%c)", - dostype[0], dostype[1], - dostype[2], dostype[3]); - strlcat(state->pp_buf, tmp, PAGE_SIZE); - snprintf(tmp, sizeof(tmp), "(res %d spb %d)", - be32_to_cpu(pb->pb_Environment[6]), - be32_to_cpu(pb->pb_Environment[4])); - strlcat(state->pp_buf, tmp, PAGE_SIZE); - } - res = 1; - } - strlcat(state->pp_buf, "\n", PAGE_SIZE); - -rdb_done: - return res; -} diff --git a/fs/partitions/amiga.h b/fs/partitions/amiga.h deleted file mode 100644 index d094585cadaa..000000000000 --- a/fs/partitions/amiga.h +++ /dev/null @@ -1,6 +0,0 @@ -/* - * fs/partitions/amiga.h - */ - -int amiga_partition(struct parsed_partitions *state); - diff --git a/fs/partitions/atari.c b/fs/partitions/atari.c deleted file mode 100644 index 9875b05e80a2..000000000000 --- a/fs/partitions/atari.c +++ /dev/null @@ -1,149 +0,0 @@ -/* - * fs/partitions/atari.c - * - * Code extracted from drivers/block/genhd.c - * - * Copyright (C) 1991-1998 Linus Torvalds - * Re-organised Feb 1998 Russell King - */ - -#include <linux/ctype.h> -#include "check.h" -#include "atari.h" - -/* ++guenther: this should be settable by the user ("make config")?. - */ -#define ICD_PARTS - -/* check if a partition entry looks valid -- Atari format is assumed if at - least one of the primary entries is ok this way */ -#define VALID_PARTITION(pi,hdsiz) \ - (((pi)->flg & 1) && \ - isalnum((pi)->id[0]) && isalnum((pi)->id[1]) && isalnum((pi)->id[2]) && \ - be32_to_cpu((pi)->st) <= (hdsiz) && \ - be32_to_cpu((pi)->st) + be32_to_cpu((pi)->siz) <= (hdsiz)) - -static inline int OK_id(char *s) -{ - return memcmp (s, "GEM", 3) == 0 || memcmp (s, "BGM", 3) == 0 || - memcmp (s, "LNX", 3) == 0 || memcmp (s, "SWP", 3) == 0 || - memcmp (s, "RAW", 3) == 0 ; -} - -int atari_partition(struct parsed_partitions *state) -{ - Sector sect; - struct rootsector *rs; - struct partition_info *pi; - u32 extensect; - u32 hd_size; - int slot; -#ifdef ICD_PARTS - int part_fmt = 0; /* 0:unknown, 1:AHDI, 2:ICD/Supra */ -#endif - - rs = read_part_sector(state, 0, §); - if (!rs) - return -1; - - /* Verify this is an Atari rootsector: */ - hd_size = state->bdev->bd_inode->i_size >> 9; - if (!VALID_PARTITION(&rs->part[0], hd_size) && - !VALID_PARTITION(&rs->part[1], hd_size) && - !VALID_PARTITION(&rs->part[2], hd_size) && - !VALID_PARTITION(&rs->part[3], hd_size)) { - /* - * if there's no valid primary partition, assume that no Atari - * format partition table (there's no reliable magic or the like - * :-() - */ - put_dev_sector(sect); - return 0; - } - - pi = &rs->part[0]; - strlcat(state->pp_buf, " AHDI", PAGE_SIZE); - for (slot = 1; pi < &rs->part[4] && slot < state->limit; slot++, pi++) { - struct rootsector *xrs; - Sector sect2; - ulong partsect; - - if ( !(pi->flg & 1) ) - continue; - /* active partition */ - if (memcmp (pi->id, "XGM", 3) != 0) { - /* we don't care about other id's */ - put_partition (state, slot, be32_to_cpu(pi->st), - be32_to_cpu(pi->siz)); - continue; - } - /* extension partition */ -#ifdef ICD_PARTS - part_fmt = 1; -#endif - strlcat(state->pp_buf, " XGM<", PAGE_SIZE); - partsect = extensect = be32_to_cpu(pi->st); - while (1) { - xrs = read_part_sector(state, partsect, §2); - if (!xrs) { - printk (" block %ld read failed\n", partsect); - put_dev_sector(sect); - return -1; - } - - /* ++roman: sanity check: bit 0 of flg field must be set */ - if (!(xrs->part[0].flg & 1)) { - printk( "\nFirst sub-partition in extended partition is not valid!\n" ); - put_dev_sector(sect2); - break; - } - - put_partition(state, slot, - partsect + be32_to_cpu(xrs->part[0].st), - be32_to_cpu(xrs->part[0].siz)); - - if (!(xrs->part[1].flg & 1)) { - /* end of linked partition list */ - put_dev_sector(sect2); - break; - } - if (memcmp( xrs->part[1].id, "XGM", 3 ) != 0) { - printk("\nID of extended partition is not XGM!\n"); - put_dev_sector(sect2); - break; - } - - partsect = be32_to_cpu(xrs->part[1].st) + extensect; - put_dev_sector(sect2); - if (++slot == state->limit) { - printk( "\nMaximum number of partitions reached!\n" ); - break; - } - } - strlcat(state->pp_buf, " >", PAGE_SIZE); - } -#ifdef ICD_PARTS - if ( part_fmt!=1 ) { /* no extended partitions -> test ICD-format */ - pi = &rs->icdpart[0]; - /* sanity check: no ICD format if first partition invalid */ - if (OK_id(pi->id)) { - strlcat(state->pp_buf, " ICD<", PAGE_SIZE); - for (; pi < &rs->icdpart[8] && slot < state->limit; slot++, pi++) { - /* accept only GEM,BGM,RAW,LNX,SWP partitions */ - if (!((pi->flg & 1) && OK_id(pi->id))) - continue; - part_fmt = 2; - put_partition (state, slot, - be32_to_cpu(pi->st), - be32_to_cpu(pi->siz)); - } - strlcat(state->pp_buf, " >", PAGE_SIZE); - } - } -#endif - put_dev_sector(sect); - - strlcat(state->pp_buf, "\n", PAGE_SIZE); - - return 1; -} diff --git a/fs/partitions/atari.h b/fs/partitions/atari.h deleted file mode 100644 index fe2d32a89f36..000000000000 --- a/fs/partitions/atari.h +++ /dev/null @@ -1,34 +0,0 @@ -/* - * fs/partitions/atari.h - * Moved by Russell King from: - * - * linux/include/linux/atari_rootsec.h - * definitions for Atari Rootsector layout - * by Andreas Schwab (schwab@ls5.informatik.uni-dortmund.de) - * - * modified for ICD/Supra partitioning scheme restricted to at most 12 - * partitions - * by Guenther Kelleter (guenther@pool.informatik.rwth-aachen.de) - */ - -struct partition_info -{ - u8 flg; /* bit 0: active; bit 7: bootable */ - char id[3]; /* "GEM", "BGM", "XGM", or other */ - __be32 st; /* start of partition */ - __be32 siz; /* length of partition */ -}; - -struct rootsector -{ - char unused[0x156]; /* room for boot code */ - struct partition_info icdpart[8]; /* info for ICD-partitions 5..12 */ - char unused2[0xc]; - u32 hd_siz; /* size of disk in blocks */ - struct partition_info part[4]; - u32 bsl_st; /* start of bad sector list */ - u32 bsl_cnt; /* length of bad sector list */ - u16 checksum; /* checksum for bootable disks */ -} __attribute__((__packed__)); - -int atari_partition(struct parsed_partitions *state); diff --git a/fs/partitions/check.c b/fs/partitions/check.c deleted file mode 100644 index e3c63d1c5e13..000000000000 --- a/fs/partitions/check.c +++ /dev/null @@ -1,687 +0,0 @@ -/* - * fs/partitions/check.c - * - * Code extracted from drivers/block/genhd.c - * Copyright (C) 1991-1998 Linus Torvalds - * Re-organised Feb 1998 Russell King - * - * We now have independent partition support from the - * block drivers, which allows all the partition code to - * be grouped in one location, and it to be mostly self - * contained. - * - * Added needed MAJORS for new pairs, {hdi,hdj}, {hdk,hdl} - */ - -#include <linux/init.h> -#include <linux/module.h> -#include <linux/fs.h> -#include <linux/slab.h> -#include <linux/kmod.h> -#include <linux/ctype.h> -#include <linux/genhd.h> -#include <linux/blktrace_api.h> - -#include "check.h" - -#include "acorn.h" -#include "amiga.h" -#include "atari.h" -#include "ldm.h" -#include "mac.h" -#include "msdos.h" -#include "osf.h" -#include "sgi.h" -#include "sun.h" -#include "ibm.h" -#include "ultrix.h" -#include "efi.h" -#include "karma.h" -#include "sysv68.h" - -#ifdef CONFIG_BLK_DEV_MD -extern void md_autodetect_dev(dev_t dev); -#endif - -int warn_no_part = 1; /*This is ugly: should make genhd removable media aware*/ - -static int (*check_part[])(struct parsed_partitions *) = { - /* - * Probe partition formats with tables at disk address 0 - * that also have an ADFS boot block at 0xdc0. - */ -#ifdef CONFIG_ACORN_PARTITION_ICS - adfspart_check_ICS, -#endif -#ifdef CONFIG_ACORN_PARTITION_POWERTEC - adfspart_check_POWERTEC, -#endif -#ifdef CONFIG_ACORN_PARTITION_EESOX - adfspart_check_EESOX, -#endif - - /* - * Now move on to formats that only have partition info at - * disk address 0xdc0. Since these may also have stale - * PC/BIOS partition tables, they need to come before - * the msdos entry. - */ -#ifdef CONFIG_ACORN_PARTITION_CUMANA - adfspart_check_CUMANA, -#endif -#ifdef CONFIG_ACORN_PARTITION_ADFS - adfspart_check_ADFS, -#endif - -#ifdef CONFIG_EFI_PARTITION - efi_partition, /* this must come before msdos */ -#endif -#ifdef CONFIG_SGI_PARTITION - sgi_partition, -#endif -#ifdef CONFIG_LDM_PARTITION - ldm_partition, /* this must come before msdos */ -#endif -#ifdef CONFIG_MSDOS_PARTITION - msdos_partition, -#endif -#ifdef CONFIG_OSF_PARTITION - osf_partition, -#endif -#ifdef CONFIG_SUN_PARTITION - sun_partition, -#endif -#ifdef CONFIG_AMIGA_PARTITION - amiga_partition, -#endif -#ifdef CONFIG_ATARI_PARTITION - atari_partition, -#endif -#ifdef CONFIG_MAC_PARTITION - mac_partition, -#endif -#ifdef CONFIG_ULTRIX_PARTITION - ultrix_partition, -#endif -#ifdef CONFIG_IBM_PARTITION - ibm_partition, -#endif -#ifdef CONFIG_KARMA_PARTITION - karma_partition, -#endif -#ifdef CONFIG_SYSV68_PARTITION - sysv68_partition, -#endif - NULL -}; - -/* - * disk_name() is used by partition check code and the genhd driver. - * It formats the devicename of the indicated disk into - * the supplied buffer (of size at least 32), and returns - * a pointer to that same buffer (for convenience). - */ - -char *disk_name(struct gendisk *hd, int partno, char *buf) -{ - if (!partno) - snprintf(buf, BDEVNAME_SIZE, "%s", hd->disk_name); - else if (isdigit(hd->disk_name[strlen(hd->disk_name)-1])) - snprintf(buf, BDEVNAME_SIZE, "%sp%d", hd->disk_name, partno); - else - snprintf(buf, BDEVNAME_SIZE, "%s%d", hd->disk_name, partno); - - return buf; -} - -const char *bdevname(struct block_device *bdev, char *buf) -{ - return disk_name(bdev->bd_disk, bdev->bd_part->partno, buf); -} - -EXPORT_SYMBOL(bdevname); - -/* - * There's very little reason to use this, you should really - * have a struct block_device just about everywhere and use - * bdevname() instead. - */ -const char *__bdevname(dev_t dev, char *buffer) -{ - scnprintf(buffer, BDEVNAME_SIZE, "unknown-block(%u,%u)", - MAJOR(dev), MINOR(dev)); - return buffer; -} - -EXPORT_SYMBOL(__bdevname); - -static struct parsed_partitions * -check_partition(struct gendisk *hd, struct block_device *bdev) -{ - struct parsed_partitions *state; - int i, res, err; - - state = kzalloc(sizeof(struct parsed_partitions), GFP_KERNEL); - if (!state) - return NULL; - state->pp_buf = (char *)__get_free_page(GFP_KERNEL); - if (!state->pp_buf) { - kfree(state); - return NULL; - } - state->pp_buf[0] = '\0'; - - state->bdev = bdev; - disk_name(hd, 0, state->name); - snprintf(state->pp_buf, PAGE_SIZE, " %s:", state->name); - if (isdigit(state->name[strlen(state->name)-1])) - sprintf(state->name, "p"); - - state->limit = disk_max_parts(hd); - i = res = err = 0; - while (!res && check_part[i]) { - memset(&state->parts, 0, sizeof(state->parts)); - res = check_part[i++](state); - if (res < 0) { - /* We have hit an I/O error which we don't report now. - * But record it, and let the others do their job. - */ - err = res; - res = 0; - } - - } - if (res > 0) { - printk(KERN_INFO "%s", state->pp_buf); - - free_page((unsigned long)state->pp_buf); - return state; - } - if (state->access_beyond_eod) - err = -ENOSPC; - if (err) - /* The partition is unrecognized. So report I/O errors if there were any */ - res = err; - if (!res) - strlcat(state->pp_buf, " unknown partition table\n", PAGE_SIZE); - else if (warn_no_part) - strlcat(state->pp_buf, " unable to read partition table\n", PAGE_SIZE); - - printk(KERN_INFO "%s", state->pp_buf); - - free_page((unsigned long)state->pp_buf); - kfree(state); - return ERR_PTR(res); -} - -static ssize_t part_partition_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct hd_struct *p = dev_to_part(dev); - - return sprintf(buf, "%d\n", p->partno); -} - -static ssize_t part_start_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct hd_struct *p = dev_to_part(dev); - - return sprintf(buf, "%llu\n",(unsigned long long)p->start_sect); -} - -ssize_t part_size_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct hd_struct *p = dev_to_part(dev); - return sprintf(buf, "%llu\n",(unsigned long long)p->nr_sects); -} - -static ssize_t part_ro_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct hd_struct *p = dev_to_part(dev); - return sprintf(buf, "%d\n", p->policy ? 1 : 0); -} - -static ssize_t part_alignment_offset_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct hd_struct *p = dev_to_part(dev); - return sprintf(buf, "%llu\n", (unsigned long long)p->alignment_offset); -} - -static ssize_t part_discard_alignment_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct hd_struct *p = dev_to_part(dev); - return sprintf(buf, "%u\n", p->discard_alignment); -} - -ssize_t part_stat_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct hd_struct *p = dev_to_part(dev); - int cpu; - - cpu = part_stat_lock(); - part_round_stats(cpu, p); - part_stat_unlock(); - return sprintf(buf, - "%8lu %8lu %8llu %8u " - "%8lu %8lu %8llu %8u " - "%8u %8u %8u" - "\n", - part_stat_read(p, ios[READ]), - part_stat_read(p, merges[READ]), - (unsigned long long)part_stat_read(p, sectors[READ]), - jiffies_to_msecs(part_stat_read(p, ticks[READ])), - part_stat_read(p, ios[WRITE]), - part_stat_read(p, merges[WRITE]), - (unsigned long long)part_stat_read(p, sectors[WRITE]), - jiffies_to_msecs(part_stat_read(p, ticks[WRITE])), - part_in_flight(p), - jiffies_to_msecs(part_stat_read(p, io_ticks)), - jiffies_to_msecs(part_stat_read(p, time_in_queue))); -} - -ssize_t part_inflight_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct hd_struct *p = dev_to_part(dev); - - return sprintf(buf, "%8u %8u\n", atomic_read(&p->in_flight[0]), - atomic_read(&p->in_flight[1])); -} - -#ifdef CONFIG_FAIL_MAKE_REQUEST -ssize_t part_fail_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct hd_struct *p = dev_to_part(dev); - - return sprintf(buf, "%d\n", p->make_it_fail); -} - -ssize_t part_fail_store(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct hd_struct *p = dev_to_part(dev); - int i; - - if (count > 0 && sscanf(buf, "%d", &i) > 0) - p->make_it_fail = (i == 0) ? 0 : 1; - - return count; -} -#endif - -static DEVICE_ATTR(partition, S_IRUGO, part_partition_show, NULL); -static DEVICE_ATTR(start, S_IRUGO, part_start_show, NULL); -static DEVICE_ATTR(size, S_IRUGO, part_size_show, NULL); -static DEVICE_ATTR(ro, S_IRUGO, part_ro_show, NULL); -static DEVICE_ATTR(alignment_offset, S_IRUGO, part_alignment_offset_show, NULL); -static DEVICE_ATTR(discard_alignment, S_IRUGO, part_discard_alignment_show, - NULL); -static DEVICE_ATTR(stat, S_IRUGO, part_stat_show, NULL); -static DEVICE_ATTR(inflight, S_IRUGO, part_inflight_show, NULL); -#ifdef CONFIG_FAIL_MAKE_REQUEST -static struct device_attribute dev_attr_fail = - __ATTR(make-it-fail, S_IRUGO|S_IWUSR, part_fail_show, part_fail_store); -#endif - -static struct attribute *part_attrs[] = { - &dev_attr_partition.attr, - &dev_attr_start.attr, - &dev_attr_size.attr, - &dev_attr_ro.attr, - &dev_attr_alignment_offset.attr, - &dev_attr_discard_alignment.attr, - &dev_attr_stat.attr, - &dev_attr_inflight.attr, -#ifdef CONFIG_FAIL_MAKE_REQUEST - &dev_attr_fail.attr, -#endif - NULL -}; - -static struct attribute_group part_attr_group = { - .attrs = part_attrs, -}; - -static const struct attribute_group *part_attr_groups[] = { - &part_attr_group, -#ifdef CONFIG_BLK_DEV_IO_TRACE - &blk_trace_attr_group, -#endif - NULL -}; - -static void part_release(struct device *dev) -{ - struct hd_struct *p = dev_to_part(dev); - free_part_stats(p); - free_part_info(p); - kfree(p); -} - -struct device_type part_type = { - .name = "partition", - .groups = part_attr_groups, - .release = part_release, -}; - -static void delete_partition_rcu_cb(struct rcu_head *head) -{ - struct hd_struct *part = container_of(head, struct hd_struct, rcu_head); - - part->start_sect = 0; - part->nr_sects = 0; - part_stat_set_all(part, 0); - put_device(part_to_dev(part)); -} - -void __delete_partition(struct hd_struct *part) -{ - call_rcu(&part->rcu_head, delete_partition_rcu_cb); -} - -void delete_partition(struct gendisk *disk, int partno) -{ - struct disk_part_tbl *ptbl = disk->part_tbl; - struct hd_struct *part; - - if (partno >= ptbl->len) - return; - - part = ptbl->part[partno]; - if (!part) - return; - - blk_free_devt(part_devt(part)); - rcu_assign_pointer(ptbl->part[partno], NULL); - rcu_assign_pointer(ptbl->last_lookup, NULL); - kobject_put(part->holder_dir); - device_del(part_to_dev(part)); - - hd_struct_put(part); -} - -static ssize_t whole_disk_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - return 0; -} -static DEVICE_ATTR(whole_disk, S_IRUSR | S_IRGRP | S_IROTH, - whole_disk_show, NULL); - -struct hd_struct *add_partition(struct gendisk *disk, int partno, - sector_t start, sector_t len, int flags, - struct partition_meta_info *info) -{ - struct hd_struct *p; - dev_t devt = MKDEV(0, 0); - struct device *ddev = disk_to_dev(disk); - struct device *pdev; - struct disk_part_tbl *ptbl; - const char *dname; - int err; - - err = disk_expand_part_tbl(disk, partno); - if (err) - return ERR_PTR(err); - ptbl = disk->part_tbl; - - if (ptbl->part[partno]) - return ERR_PTR(-EBUSY); - - p = kzalloc(sizeof(*p), GFP_KERNEL); - if (!p) - return ERR_PTR(-EBUSY); - - if (!init_part_stats(p)) { - err = -ENOMEM; - goto out_free; - } - pdev = part_to_dev(p); - - p->start_sect = start; - p->alignment_offset = - queue_limit_alignment_offset(&disk->queue->limits, start); - p->discard_alignment = - queue_limit_discard_alignment(&disk->queue->limits, start); - p->nr_sects = len; - p->partno = partno; - p->policy = get_disk_ro(disk); - - if (info) { - struct partition_meta_info *pinfo = alloc_part_info(disk); - if (!pinfo) - goto out_free_stats; - memcpy(pinfo, info, sizeof(*info)); - p->info = pinfo; - } - - dname = dev_name(ddev); - if (isdigit(dname[strlen(dname) - 1])) - dev_set_name(pdev, "%sp%d", dname, partno); - else - dev_set_name(pdev, "%s%d", dname, partno); - - device_initialize(pdev); - pdev->class = &block_class; - pdev->type = &part_type; - pdev->parent = ddev; - - err = blk_alloc_devt(p, &devt); - if (err) - goto out_free_info; - pdev->devt = devt; - - /* delay uevent until 'holders' subdir is created */ - dev_set_uevent_suppress(pdev, 1); - err = device_add(pdev); - if (err) - goto out_put; - - err = -ENOMEM; - p->holder_dir = kobject_create_and_add("holders", &pdev->kobj); - if (!p->holder_dir) - goto out_del; - - dev_set_uevent_suppress(pdev, 0); - if (flags & ADDPART_FLAG_WHOLEDISK) { - err = device_create_file(pdev, &dev_attr_whole_disk); - if (err) - goto out_del; - } - - /* everything is up and running, commence */ - rcu_assign_pointer(ptbl->part[partno], p); - - /* suppress uevent if the disk suppresses it */ - if (!dev_get_uevent_suppress(ddev)) - kobject_uevent(&pdev->kobj, KOBJ_ADD); - - hd_ref_init(p); - return p; - -out_free_info: - free_part_info(p); -out_free_stats: - free_part_stats(p); -out_free: - kfree(p); - return ERR_PTR(err); -out_del: - kobject_put(p->holder_dir); - device_del(pdev); -out_put: - put_device(pdev); - blk_free_devt(devt); - return ERR_PTR(err); -} - -static bool disk_unlock_native_capacity(struct gendisk *disk) -{ - const struct block_device_operations *bdops = disk->fops; - - if (bdops->unlock_native_capacity && - !(disk->flags & GENHD_FL_NATIVE_CAPACITY)) { - printk(KERN_CONT "enabling native capacity\n"); - bdops->unlock_native_capacity(disk); - disk->flags |= GENHD_FL_NATIVE_CAPACITY; - return true; - } else { - printk(KERN_CONT "truncated\n"); - return false; - } -} - -int rescan_partitions(struct gendisk *disk, struct block_device *bdev) -{ - struct parsed_partitions *state = NULL; - struct disk_part_iter piter; - struct hd_struct *part; - int p, highest, res; -rescan: - if (state && !IS_ERR(state)) { - kfree(state); - state = NULL; - } - - if (bdev->bd_part_count) - return -EBUSY; - res = invalidate_partition(disk, 0); - if (res) - return res; - - disk_part_iter_init(&piter, disk, DISK_PITER_INCL_EMPTY); - while ((part = disk_part_iter_next(&piter))) - delete_partition(disk, part->partno); - disk_part_iter_exit(&piter); - - if (disk->fops->revalidate_disk) - disk->fops->revalidate_disk(disk); - check_disk_size_change(disk, bdev); - bdev->bd_invalidated = 0; - if (!get_capacity(disk) || !(state = check_partition(disk, bdev))) - return 0; - if (IS_ERR(state)) { - /* - * I/O error reading the partition table. If any - * partition code tried to read beyond EOD, retry - * after unlocking native capacity. - */ - if (PTR_ERR(state) == -ENOSPC) { - printk(KERN_WARNING "%s: partition table beyond EOD, ", - disk->disk_name); - if (disk_unlock_native_capacity(disk)) - goto rescan; - } - return -EIO; - } - /* - * If any partition code tried to read beyond EOD, try - * unlocking native capacity even if partition table is - * successfully read as we could be missing some partitions. - */ - if (state->access_beyond_eod) { - printk(KERN_WARNING - "%s: partition table partially beyond EOD, ", - disk->disk_name); - if (disk_unlock_native_capacity(disk)) - goto rescan; - } - - /* tell userspace that the media / partition table may have changed */ - kobject_uevent(&disk_to_dev(disk)->kobj, KOBJ_CHANGE); - - /* Detect the highest partition number and preallocate - * disk->part_tbl. This is an optimization and not strictly - * necessary. - */ - for (p = 1, highest = 0; p < state->limit; p++) - if (state->parts[p].size) - highest = p; - - disk_expand_part_tbl(disk, highest); - - /* add partitions */ - for (p = 1; p < state->limit; p++) { - sector_t size, from; - struct partition_meta_info *info = NULL; - - size = state->parts[p].size; - if (!size) - continue; - - from = state->parts[p].from; - if (from >= get_capacity(disk)) { - printk(KERN_WARNING - "%s: p%d start %llu is beyond EOD, ", - disk->disk_name, p, (unsigned long long) from); - if (disk_unlock_native_capacity(disk)) - goto rescan; - continue; - } - - if (from + size > get_capacity(disk)) { - printk(KERN_WARNING - "%s: p%d size %llu extends beyond EOD, ", - disk->disk_name, p, (unsigned long long) size); - - if (disk_unlock_native_capacity(disk)) { - /* free state and restart */ - goto rescan; - } else { - /* - * we can not ignore partitions of broken tables - * created by for example camera firmware, but - * we limit them to the end of the disk to avoid - * creating invalid block devices - */ - size = get_capacity(disk) - from; - } - } - - if (state->parts[p].has_info) - info = &state->parts[p].info; - part = add_partition(disk, p, from, size, - state->parts[p].flags, - &state->parts[p].info); - if (IS_ERR(part)) { - printk(KERN_ERR " %s: p%d could not be added: %ld\n", - disk->disk_name, p, -PTR_ERR(part)); - continue; - } -#ifdef CONFIG_BLK_DEV_MD - if (state->parts[p].flags & ADDPART_FLAG_RAID) - md_autodetect_dev(part_to_dev(part)->devt); -#endif - } - kfree(state); - return 0; -} - -unsigned char *read_dev_sector(struct block_device *bdev, sector_t n, Sector *p) -{ - struct address_space *mapping = bdev->bd_inode->i_mapping; - struct page *page; - - page = read_mapping_page(mapping, (pgoff_t)(n >> (PAGE_CACHE_SHIFT-9)), - NULL); - if (!IS_ERR(page)) { - if (PageError(page)) - goto fail; - p->v = page; - return (unsigned char *)page_address(page) + ((n & ((1 << (PAGE_CACHE_SHIFT - 9)) - 1)) << 9); -fail: - page_cache_release(page); - } - p->v = NULL; - return NULL; -} - -EXPORT_SYMBOL(read_dev_sector); diff --git a/fs/partitions/check.h b/fs/partitions/check.h deleted file mode 100644 index d68bf4dc3bc2..000000000000 --- a/fs/partitions/check.h +++ /dev/null @@ -1,49 +0,0 @@ -#include <linux/pagemap.h> -#include <linux/blkdev.h> -#include <linux/genhd.h> - -/* - * add_gd_partition adds a partitions details to the devices partition - * description. - */ -struct parsed_partitions { - struct block_device *bdev; - char name[BDEVNAME_SIZE]; - struct { - sector_t from; - sector_t size; - int flags; - bool has_info; - struct partition_meta_info info; - } parts[DISK_MAX_PARTS]; - int next; - int limit; - bool access_beyond_eod; - char *pp_buf; -}; - -static inline void *read_part_sector(struct parsed_partitions *state, - sector_t n, Sector *p) -{ - if (n >= get_capacity(state->bdev->bd_disk)) { - state->access_beyond_eod = true; - return NULL; - } - return read_dev_sector(state->bdev, n, p); -} - -static inline void -put_partition(struct parsed_partitions *p, int n, sector_t from, sector_t size) -{ - if (n < p->limit) { - char tmp[1 + BDEVNAME_SIZE + 10 + 1]; - - p->parts[n].from = from; - p->parts[n].size = size; - snprintf(tmp, sizeof(tmp), " %s%d", p->name, n); - strlcat(p->pp_buf, tmp, PAGE_SIZE); - } -} - -extern int warn_no_part; - diff --git a/fs/partitions/efi.c b/fs/partitions/efi.c deleted file mode 100644 index 6296b403c67a..000000000000 --- a/fs/partitions/efi.c +++ /dev/null @@ -1,675 +0,0 @@ -/************************************************************ - * EFI GUID Partition Table handling - * - * http://www.uefi.org/specs/ - * http://www.intel.com/technology/efi/ - * - * efi.[ch] by Matt Domsch <Matt_Domsch@dell.com> - * Copyright 2000,2001,2002,2004 Dell Inc. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - * - * - * TODO: - * - * Changelog: - * Mon Nov 09 2004 Matt Domsch <Matt_Domsch@dell.com> - * - test for valid PMBR and valid PGPT before ever reading - * AGPT, allow override with 'gpt' kernel command line option. - * - check for first/last_usable_lba outside of size of disk - * - * Tue Mar 26 2002 Matt Domsch <Matt_Domsch@dell.com> - * - Ported to 2.5.7-pre1 and 2.5.7-dj2 - * - Applied patch to avoid fault in alternate header handling - * - cleaned up find_valid_gpt - * - On-disk structure and copy in memory is *always* LE now - - * swab fields as needed - * - remove print_gpt_header() - * - only use first max_p partition entries, to keep the kernel minor number - * and partition numbers tied. - * - * Mon Feb 04 2002 Matt Domsch <Matt_Domsch@dell.com> - * - Removed __PRIPTR_PREFIX - not being used - * - * Mon Jan 14 2002 Matt Domsch <Matt_Domsch@dell.com> - * - Ported to 2.5.2-pre11 + library crc32 patch Linus applied - * - * Thu Dec 6 2001 Matt Domsch <Matt_Domsch@dell.com> - * - Added compare_gpts(). - * - moved le_efi_guid_to_cpus() back into this file. GPT is the only - * thing that keeps EFI GUIDs on disk. - * - Changed gpt structure names and members to be simpler and more Linux-like. - * - * Wed Oct 17 2001 Matt Domsch <Matt_Domsch@dell.com> - * - Removed CONFIG_DEVFS_VOLUMES_UUID code entirely per Martin Wilck - * - * Wed Oct 10 2001 Matt Domsch <Matt_Domsch@dell.com> - * - Changed function comments to DocBook style per Andreas Dilger suggestion. - * - * Mon Oct 08 2001 Matt Domsch <Matt_Domsch@dell.com> - * - Change read_lba() to use the page cache per Al Viro's work. - * - print u64s properly on all architectures - * - fixed debug_printk(), now Dprintk() - * - * Mon Oct 01 2001 Matt Domsch <Matt_Domsch@dell.com> - * - Style cleanups - * - made most functions static - * - Endianness addition - * - remove test for second alternate header, as it's not per spec, - * and is unnecessary. There's now a method to read/write the last - * sector of an odd-sized disk from user space. No tools have ever - * been released which used this code, so it's effectively dead. - * - Per Asit Mallick of Intel, added a test for a valid PMBR. - * - Added kernel command line option 'gpt' to override valid PMBR test. - * - * Wed Jun 6 2001 Martin Wilck <Martin.Wilck@Fujitsu-Siemens.com> - * - added devfs volume UUID support (/dev/volumes/uuids) for - * mounting file systems by the partition GUID. - * - * Tue Dec 5 2000 Matt Domsch <Matt_Domsch@dell.com> - * - Moved crc32() to linux/lib, added efi_crc32(). - * - * Thu Nov 30 2000 Matt Domsch <Matt_Domsch@dell.com> - * - Replaced Intel's CRC32 function with an equivalent - * non-license-restricted version. - * - * Wed Oct 25 2000 Matt Domsch <Matt_Domsch@dell.com> - * - Fixed the last_lba() call to return the proper last block - * - * Thu Oct 12 2000 Matt Domsch <Matt_Domsch@dell.com> - * - Thanks to Andries Brouwer for his debugging assistance. - * - Code works, detects all the partitions. - * - ************************************************************/ -#include <linux/crc32.h> -#include <linux/ctype.h> -#include <linux/math64.h> -#include <linux/slab.h> -#include "check.h" -#include "efi.h" - -/* This allows a kernel command line option 'gpt' to override - * the test for invalid PMBR. Not __initdata because reloading - * the partition tables happens after init too. - */ -static int force_gpt; -static int __init -force_gpt_fn(char *str) -{ - force_gpt = 1; - return 1; -} -__setup("gpt", force_gpt_fn); - - -/** - * efi_crc32() - EFI version of crc32 function - * @buf: buffer to calculate crc32 of - * @len - length of buf - * - * Description: Returns EFI-style CRC32 value for @buf - * - * This function uses the little endian Ethernet polynomial - * but seeds the function with ~0, and xor's with ~0 at the end. - * Note, the EFI Specification, v1.02, has a reference to - * Dr. Dobbs Journal, May 1994 (actually it's in May 1992). - */ -static inline u32 -efi_crc32(const void *buf, unsigned long len) -{ - return (crc32(~0L, buf, len) ^ ~0L); -} - -/** - * last_lba(): return number of last logical block of device - * @bdev: block device - * - * Description: Returns last LBA value on success, 0 on error. - * This is stored (by sd and ide-geometry) in - * the part[0] entry for this disk, and is the number of - * physical sectors available on the disk. - */ -static u64 last_lba(struct block_device *bdev) -{ - if (!bdev || !bdev->bd_inode) - return 0; - return div_u64(bdev->bd_inode->i_size, - bdev_logical_block_size(bdev)) - 1ULL; -} - -static inline int -pmbr_part_valid(struct partition *part) -{ - if (part->sys_ind == EFI_PMBR_OSTYPE_EFI_GPT && - le32_to_cpu(part->start_sect) == 1UL) - return 1; - return 0; -} - -/** - * is_pmbr_valid(): test Protective MBR for validity - * @mbr: pointer to a legacy mbr structure - * - * Description: Returns 1 if PMBR is valid, 0 otherwise. - * Validity depends on two things: - * 1) MSDOS signature is in the last two bytes of the MBR - * 2) One partition of type 0xEE is found - */ -static int -is_pmbr_valid(legacy_mbr *mbr) -{ - int i; - if (!mbr || le16_to_cpu(mbr->signature) != MSDOS_MBR_SIGNATURE) - return 0; - for (i = 0; i < 4; i++) - if (pmbr_part_valid(&mbr->partition_record[i])) - return 1; - return 0; -} - -/** - * read_lba(): Read bytes from disk, starting at given LBA - * @state - * @lba - * @buffer - * @size_t - * - * Description: Reads @count bytes from @state->bdev into @buffer. - * Returns number of bytes read on success, 0 on error. - */ -static size_t read_lba(struct parsed_partitions *state, - u64 lba, u8 *buffer, size_t count) -{ - size_t totalreadcount = 0; - struct block_device *bdev = state->bdev; - sector_t n = lba * (bdev_logical_block_size(bdev) / 512); - - if (!buffer || lba > last_lba(bdev)) - return 0; - - while (count) { - int copied = 512; - Sector sect; - unsigned char *data = read_part_sector(state, n++, §); - if (!data) - break; - if (copied > count) - copied = count; - memcpy(buffer, data, copied); - put_dev_sector(sect); - buffer += copied; - totalreadcount +=copied; - count -= copied; - } - return totalreadcount; -} - -/** - * alloc_read_gpt_entries(): reads partition entries from disk - * @state - * @gpt - GPT header - * - * Description: Returns ptes on success, NULL on error. - * Allocates space for PTEs based on information found in @gpt. - * Notes: remember to free pte when you're done! - */ -static gpt_entry *alloc_read_gpt_entries(struct parsed_partitions *state, - gpt_header *gpt) -{ - size_t count; - gpt_entry *pte; - - if (!gpt) - return NULL; - - count = le32_to_cpu(gpt->num_partition_entries) * - le32_to_cpu(gpt->sizeof_partition_entry); - if (!count) - return NULL; - pte = kzalloc(count, GFP_KERNEL); - if (!pte) - return NULL; - - if (read_lba(state, le64_to_cpu(gpt->partition_entry_lba), - (u8 *) pte, - count) < count) { - kfree(pte); - pte=NULL; - return NULL; - } - return pte; -} - -/** - * alloc_read_gpt_header(): Allocates GPT header, reads into it from disk - * @state - * @lba is the Logical Block Address of the partition table - * - * Description: returns GPT header on success, NULL on error. Allocates - * and fills a GPT header starting at @ from @state->bdev. - * Note: remember to free gpt when finished with it. - */ -static gpt_header *alloc_read_gpt_header(struct parsed_partitions *state, - u64 lba) -{ - gpt_header *gpt; - unsigned ssz = bdev_logical_block_size(state->bdev); - - gpt = kzalloc(ssz, GFP_KERNEL); - if (!gpt) - return NULL; - - if (read_lba(state, lba, (u8 *) gpt, ssz) < ssz) { - kfree(gpt); - gpt=NULL; - return NULL; - } - - return gpt; -} - -/** - * is_gpt_valid() - tests one GPT header and PTEs for validity - * @state - * @lba is the logical block address of the GPT header to test - * @gpt is a GPT header ptr, filled on return. - * @ptes is a PTEs ptr, filled on return. - * - * Description: returns 1 if valid, 0 on error. - * If valid, returns pointers to newly allocated GPT header and PTEs. - */ -static int is_gpt_valid(struct parsed_partitions *state, u64 lba, - gpt_header **gpt, gpt_entry **ptes) -{ - u32 crc, origcrc; - u64 lastlba; - - if (!ptes) - return 0; - if (!(*gpt = alloc_read_gpt_header(state, lba))) - return 0; - - /* Check the GUID Partition Table signature */ - if (le64_to_cpu((*gpt)->signature) != GPT_HEADER_SIGNATURE) { - pr_debug("GUID Partition Table Header signature is wrong:" - "%lld != %lld\n", - (unsigned long long)le64_to_cpu((*gpt)->signature), - (unsigned long long)GPT_HEADER_SIGNATURE); - goto fail; - } - - /* Check the GUID Partition Table header size */ - if (le32_to_cpu((*gpt)->header_size) > - bdev_logical_block_size(state->bdev)) { - pr_debug("GUID Partition Table Header size is wrong: %u > %u\n", - le32_to_cpu((*gpt)->header_size), - bdev_logical_block_size(state->bdev)); - goto fail; - } - - /* Check the GUID Partition Table CRC */ - origcrc = le32_to_cpu((*gpt)->header_crc32); - (*gpt)->header_crc32 = 0; - crc = efi_crc32((const unsigned char *) (*gpt), le32_to_cpu((*gpt)->header_size)); - - if (crc != origcrc) { - pr_debug("GUID Partition Table Header CRC is wrong: %x != %x\n", - crc, origcrc); - goto fail; - } - (*gpt)->header_crc32 = cpu_to_le32(origcrc); - - /* Check that the my_lba entry points to the LBA that contains - * the GUID Partition Table */ - if (le64_to_cpu((*gpt)->my_lba) != lba) { - pr_debug("GPT my_lba incorrect: %lld != %lld\n", - (unsigned long long)le64_to_cpu((*gpt)->my_lba), - (unsigned long long)lba); - goto fail; - } - - /* Check the first_usable_lba and last_usable_lba are - * within the disk. - */ - lastlba = last_lba(state->bdev); - if (le64_to_cpu((*gpt)->first_usable_lba) > lastlba) { - pr_debug("GPT: first_usable_lba incorrect: %lld > %lld\n", - (unsigned long long)le64_to_cpu((*gpt)->first_usable_lba), - (unsigned long long)lastlba); - goto fail; - } - if (le64_to_cpu((*gpt)->last_usable_lba) > lastlba) { - pr_debug("GPT: last_usable_lba incorrect: %lld > %lld\n", - (unsigned long long)le64_to_cpu((*gpt)->last_usable_lba), - (unsigned long long)lastlba); - goto fail; - } - - /* Check that sizeof_partition_entry has the correct value */ - if (le32_to_cpu((*gpt)->sizeof_partition_entry) != sizeof(gpt_entry)) { - pr_debug("GUID Partitition Entry Size check failed.\n"); - goto fail; - } - - if (!(*ptes = alloc_read_gpt_entries(state, *gpt))) - goto fail; - - /* Check the GUID Partition Entry Array CRC */ - crc = efi_crc32((const unsigned char *) (*ptes), - le32_to_cpu((*gpt)->num_partition_entries) * - le32_to_cpu((*gpt)->sizeof_partition_entry)); - - if (crc != le32_to_cpu((*gpt)->partition_entry_array_crc32)) { - pr_debug("GUID Partitition Entry Array CRC check failed.\n"); - goto fail_ptes; - } - - /* We're done, all's well */ - return 1; - - fail_ptes: - kfree(*ptes); - *ptes = NULL; - fail: - kfree(*gpt); - *gpt = NULL; - return 0; -} - -/** - * is_pte_valid() - tests one PTE for validity - * @pte is the pte to check - * @lastlba is last lba of the disk - * - * Description: returns 1 if valid, 0 on error. - */ -static inline int -is_pte_valid(const gpt_entry *pte, const u64 lastlba) -{ - if ((!efi_guidcmp(pte->partition_type_guid, NULL_GUID)) || - le64_to_cpu(pte->starting_lba) > lastlba || - le64_to_cpu(pte->ending_lba) > lastlba) - return 0; - return 1; -} - -/** - * compare_gpts() - Search disk for valid GPT headers and PTEs - * @pgpt is the primary GPT header - * @agpt is the alternate GPT header - * @lastlba is the last LBA number - * Description: Returns nothing. Sanity checks pgpt and agpt fields - * and prints warnings on discrepancies. - * - */ -static void -compare_gpts(gpt_header *pgpt, gpt_header *agpt, u64 lastlba) -{ - int error_found = 0; - if (!pgpt || !agpt) - return; - if (le64_to_cpu(pgpt->my_lba) != le64_to_cpu(agpt->alternate_lba)) { - printk(KERN_WARNING - "GPT:Primary header LBA != Alt. header alternate_lba\n"); - printk(KERN_WARNING "GPT:%lld != %lld\n", - (unsigned long long)le64_to_cpu(pgpt->my_lba), - (unsigned long long)le64_to_cpu(agpt->alternate_lba)); - error_found++; - } - if (le64_to_cpu(pgpt->alternate_lba) != le64_to_cpu(agpt->my_lba)) { - printk(KERN_WARNING - "GPT:Primary header alternate_lba != Alt. header my_lba\n"); - printk(KERN_WARNING "GPT:%lld != %lld\n", - (unsigned long long)le64_to_cpu(pgpt->alternate_lba), - (unsigned long long)le64_to_cpu(agpt->my_lba)); - error_found++; - } - if (le64_to_cpu(pgpt->first_usable_lba) != - le64_to_cpu(agpt->first_usable_lba)) { - printk(KERN_WARNING "GPT:first_usable_lbas don't match.\n"); - printk(KERN_WARNING "GPT:%lld != %lld\n", - (unsigned long long)le64_to_cpu(pgpt->first_usable_lba), - (unsigned long long)le64_to_cpu(agpt->first_usable_lba)); - error_found++; - } - if (le64_to_cpu(pgpt->last_usable_lba) != - le64_to_cpu(agpt->last_usable_lba)) { - printk(KERN_WARNING "GPT:last_usable_lbas don't match.\n"); - printk(KERN_WARNING "GPT:%lld != %lld\n", - (unsigned long long)le64_to_cpu(pgpt->last_usable_lba), - (unsigned long long)le64_to_cpu(agpt->last_usable_lba)); - error_found++; - } - if (efi_guidcmp(pgpt->disk_guid, agpt->disk_guid)) { - printk(KERN_WARNING "GPT:disk_guids don't match.\n"); - error_found++; - } - if (le32_to_cpu(pgpt->num_partition_entries) != - le32_to_cpu(agpt->num_partition_entries)) { - printk(KERN_WARNING "GPT:num_partition_entries don't match: " - "0x%x != 0x%x\n", - le32_to_cpu(pgpt->num_partition_entries), - le32_to_cpu(agpt->num_partition_entries)); - error_found++; - } - if (le32_to_cpu(pgpt->sizeof_partition_entry) != - le32_to_cpu(agpt->sizeof_partition_entry)) { - printk(KERN_WARNING - "GPT:sizeof_partition_entry values don't match: " - "0x%x != 0x%x\n", - le32_to_cpu(pgpt->sizeof_partition_entry), - le32_to_cpu(agpt->sizeof_partition_entry)); - error_found++; - } - if (le32_to_cpu(pgpt->partition_entry_array_crc32) != - le32_to_cpu(agpt->partition_entry_array_crc32)) { - printk(KERN_WARNING - "GPT:partition_entry_array_crc32 values don't match: " - "0x%x != 0x%x\n", - le32_to_cpu(pgpt->partition_entry_array_crc32), - le32_to_cpu(agpt->partition_entry_array_crc32)); - error_found++; - } - if (le64_to_cpu(pgpt->alternate_lba) != lastlba) { - printk(KERN_WARNING - "GPT:Primary header thinks Alt. header is not at the end of the disk.\n"); - printk(KERN_WARNING "GPT:%lld != %lld\n", - (unsigned long long)le64_to_cpu(pgpt->alternate_lba), - (unsigned long long)lastlba); - error_found++; - } - - if (le64_to_cpu(agpt->my_lba) != lastlba) { - printk(KERN_WARNING - "GPT:Alternate GPT header not at the end of the disk.\n"); - printk(KERN_WARNING "GPT:%lld != %lld\n", - (unsigned long long)le64_to_cpu(agpt->my_lba), - (unsigned long long)lastlba); - error_found++; - } - - if (error_found) - printk(KERN_WARNING - "GPT: Use GNU Parted to correct GPT errors.\n"); - return; -} - -/** - * find_valid_gpt() - Search disk for valid GPT headers and PTEs - * @state - * @gpt is a GPT header ptr, filled on return. - * @ptes is a PTEs ptr, filled on return. - * Description: Returns 1 if valid, 0 on error. - * If valid, returns pointers to newly allocated GPT header and PTEs. - * Validity depends on PMBR being valid (or being overridden by the - * 'gpt' kernel command line option) and finding either the Primary - * GPT header and PTEs valid, or the Alternate GPT header and PTEs - * valid. If the Primary GPT header is not valid, the Alternate GPT header - * is not checked unless the 'gpt' kernel command line option is passed. - * This protects against devices which misreport their size, and forces - * the user to decide to use the Alternate GPT. - */ -static int find_valid_gpt(struct parsed_partitions *state, gpt_header **gpt, - gpt_entry **ptes) -{ - int good_pgpt = 0, good_agpt = 0, good_pmbr = 0; - gpt_header *pgpt = NULL, *agpt = NULL; - gpt_entry *pptes = NULL, *aptes = NULL; - legacy_mbr *legacymbr; - u64 lastlba; - - if (!ptes) - return 0; - - lastlba = last_lba(state->bdev); - if (!force_gpt) { - /* This will be added to the EFI Spec. per Intel after v1.02. */ - legacymbr = kzalloc(sizeof (*legacymbr), GFP_KERNEL); - if (legacymbr) { - read_lba(state, 0, (u8 *) legacymbr, - sizeof (*legacymbr)); - good_pmbr = is_pmbr_valid(legacymbr); - kfree(legacymbr); - } - if (!good_pmbr) - goto fail; - } - - good_pgpt = is_gpt_valid(state, GPT_PRIMARY_PARTITION_TABLE_LBA, - &pgpt, &pptes); - if (good_pgpt) - good_agpt = is_gpt_valid(state, - le64_to_cpu(pgpt->alternate_lba), - &agpt, &aptes); - if (!good_agpt && force_gpt) - good_agpt = is_gpt_valid(state, lastlba, &agpt, &aptes); - - /* The obviously unsuccessful case */ - if (!good_pgpt && !good_agpt) - goto fail; - - compare_gpts(pgpt, agpt, lastlba); - - /* The good cases */ - if (good_pgpt) { - *gpt = pgpt; - *ptes = pptes; - kfree(agpt); - kfree(aptes); - if (!good_agpt) { - printk(KERN_WARNING - "Alternate GPT is invalid, " - "using primary GPT.\n"); - } - return 1; - } - else if (good_agpt) { - *gpt = agpt; - *ptes = aptes; - kfree(pgpt); - kfree(pptes); - printk(KERN_WARNING - "Primary GPT is invalid, using alternate GPT.\n"); - return 1; - } - - fail: - kfree(pgpt); - kfree(agpt); - kfree(pptes); - kfree(aptes); - *gpt = NULL; - *ptes = NULL; - return 0; -} - -/** - * efi_partition(struct parsed_partitions *state) - * @state - * - * Description: called from check.c, if the disk contains GPT - * partitions, sets up partition entries in the kernel. - * - * If the first block on the disk is a legacy MBR, - * it will get handled by msdos_partition(). - * If it's a Protective MBR, we'll handle it here. - * - * We do not create a Linux partition for GPT, but - * only for the actual data partitions. - * Returns: - * -1 if unable to read the partition table - * 0 if this isn't our partition table - * 1 if successful - * - */ -int efi_partition(struct parsed_partitions *state) -{ - gpt_header *gpt = NULL; - gpt_entry *ptes = NULL; - u32 i; - unsigned ssz = bdev_logical_block_size(state->bdev) / 512; - u8 unparsed_guid[37]; - - if (!find_valid_gpt(state, &gpt, &ptes) || !gpt || !ptes) { - kfree(gpt); - kfree(ptes); - return 0; - } - - pr_debug("GUID Partition Table is valid! Yea!\n"); - - for (i = 0; i < le32_to_cpu(gpt->num_partition_entries) && i < state->limit-1; i++) { - struct partition_meta_info *info; - unsigned label_count = 0; - unsigned label_max; - u64 start = le64_to_cpu(ptes[i].starting_lba); - u64 size = le64_to_cpu(ptes[i].ending_lba) - - le64_to_cpu(ptes[i].starting_lba) + 1ULL; - - if (!is_pte_valid(&ptes[i], last_lba(state->bdev))) - continue; - - put_partition(state, i+1, start * ssz, size * ssz); - - /* If this is a RAID volume, tell md */ - if (!efi_guidcmp(ptes[i].partition_type_guid, - PARTITION_LINUX_RAID_GUID)) - state->parts[i + 1].flags = ADDPART_FLAG_RAID; - - info = &state->parts[i + 1].info; - /* Instead of doing a manual swap to big endian, reuse the - * common ASCII hex format as the interim. - */ - efi_guid_unparse(&ptes[i].unique_partition_guid, unparsed_guid); - part_pack_uuid(unparsed_guid, info->uuid); - - /* Naively convert UTF16-LE to 7 bits. */ - label_max = min(sizeof(info->volname) - 1, - sizeof(ptes[i].partition_name)); - info->volname[label_max] = 0; - while (label_count < label_max) { - u8 c = ptes[i].partition_name[label_count] & 0xff; - if (c && !isprint(c)) - c = '!'; - info->volname[label_count] = c; - label_count++; - } - state->parts[i + 1].has_info = true; - } - kfree(ptes); - kfree(gpt); - strlcat(state->pp_buf, "\n", PAGE_SIZE); - return 1; -} diff --git a/fs/partitions/efi.h b/fs/partitions/efi.h deleted file mode 100644 index b69ab729558f..000000000000 --- a/fs/partitions/efi.h +++ /dev/null @@ -1,134 +0,0 @@ -/************************************************************ - * EFI GUID Partition Table - * Per Intel EFI Specification v1.02 - * http://developer.intel.com/technology/efi/efi.htm - * - * By Matt Domsch <Matt_Domsch@dell.com> Fri Sep 22 22:15:56 CDT 2000 - * Copyright 2000,2001 Dell Inc. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - * - ************************************************************/ - -#ifndef FS_PART_EFI_H_INCLUDED -#define FS_PART_EFI_H_INCLUDED - -#include <linux/types.h> -#include <linux/fs.h> -#include <linux/genhd.h> -#include <linux/kernel.h> -#include <linux/major.h> -#include <linux/string.h> -#include <linux/efi.h> - -#define MSDOS_MBR_SIGNATURE 0xaa55 -#define EFI_PMBR_OSTYPE_EFI 0xEF -#define EFI_PMBR_OSTYPE_EFI_GPT 0xEE - -#define GPT_HEADER_SIGNATURE 0x5452415020494645ULL -#define GPT_HEADER_REVISION_V1 0x00010000 -#define GPT_PRIMARY_PARTITION_TABLE_LBA 1 - -#define PARTITION_SYSTEM_GUID \ - EFI_GUID( 0xC12A7328, 0xF81F, 0x11d2, \ - 0xBA, 0x4B, 0x00, 0xA0, 0xC9, 0x3E, 0xC9, 0x3B) -#define LEGACY_MBR_PARTITION_GUID \ - EFI_GUID( 0x024DEE41, 0x33E7, 0x11d3, \ - 0x9D, 0x69, 0x00, 0x08, 0xC7, 0x81, 0xF3, 0x9F) -#define PARTITION_MSFT_RESERVED_GUID \ - EFI_GUID( 0xE3C9E316, 0x0B5C, 0x4DB8, \ - 0x81, 0x7D, 0xF9, 0x2D, 0xF0, 0x02, 0x15, 0xAE) -#define PARTITION_BASIC_DATA_GUID \ - EFI_GUID( 0xEBD0A0A2, 0xB9E5, 0x4433, \ - 0x87, 0xC0, 0x68, 0xB6, 0xB7, 0x26, 0x99, 0xC7) -#define PARTITION_LINUX_RAID_GUID \ - EFI_GUID( 0xa19d880f, 0x05fc, 0x4d3b, \ - 0xa0, 0x06, 0x74, 0x3f, 0x0f, 0x84, 0x91, 0x1e) -#define PARTITION_LINUX_SWAP_GUID \ - EFI_GUID( 0x0657fd6d, 0xa4ab, 0x43c4, \ - 0x84, 0xe5, 0x09, 0x33, 0xc8, 0x4b, 0x4f, 0x4f) -#define PARTITION_LINUX_LVM_GUID \ - EFI_GUID( 0xe6d6d379, 0xf507, 0x44c2, \ - 0xa2, 0x3c, 0x23, 0x8f, 0x2a, 0x3d, 0xf9, 0x28) - -typedef struct _gpt_header { - __le64 signature; - __le32 revision; - __le32 header_size; - __le32 header_crc32; - __le32 reserved1; - __le64 my_lba; - __le64 alternate_lba; - __le64 first_usable_lba; - __le64 last_usable_lba; - efi_guid_t disk_guid; - __le64 partition_entry_lba; - __le32 num_partition_entries; - __le32 sizeof_partition_entry; - __le32 partition_entry_array_crc32; - - /* The rest of the logical block is reserved by UEFI and must be zero. - * EFI standard handles this by: - * - * uint8_t reserved2[ BlockSize - 92 ]; - */ -} __attribute__ ((packed)) gpt_header; - -typedef struct _gpt_entry_attributes { - u64 required_to_function:1; - u64 reserved:47; - u64 type_guid_specific:16; -} __attribute__ ((packed)) gpt_entry_attributes; - -typedef struct _gpt_entry { - efi_guid_t partition_type_guid; - efi_guid_t unique_partition_guid; - __le64 starting_lba; - __le64 ending_lba; - gpt_entry_attributes attributes; - efi_char16_t partition_name[72 / sizeof (efi_char16_t)]; -} __attribute__ ((packed)) gpt_entry; - -typedef struct _legacy_mbr { - u8 boot_code[440]; - __le32 unique_mbr_signature; - __le16 unknown; - struct partition partition_record[4]; - __le16 signature; -} __attribute__ ((packed)) legacy_mbr; - -/* Functions */ -extern int efi_partition(struct parsed_partitions *state); - -#endif - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * -------------------------------------------------------------------------- - * Local variables: - * c-indent-level: 4 - * c-brace-imaginary-offset: 0 - * c-brace-offset: -4 - * c-argdecl-indent: 4 - * c-label-offset: -4 - * c-continued-statement-offset: 4 - * c-continued-brace-offset: 0 - * indent-tabs-mode: nil - * tab-width: 8 - * End: - */ diff --git a/fs/partitions/ibm.c b/fs/partitions/ibm.c deleted file mode 100644 index d513a07f44bb..000000000000 --- a/fs/partitions/ibm.c +++ /dev/null @@ -1,275 +0,0 @@ -/* - * File...........: linux/fs/partitions/ibm.c - * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com> - * Volker Sameske <sameske@de.ibm.com> - * Bugreports.to..: <Linux390@de.ibm.com> - * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 1999,2000 - */ - -#include <linux/buffer_head.h> -#include <linux/hdreg.h> -#include <linux/slab.h> -#include <asm/dasd.h> -#include <asm/ebcdic.h> -#include <asm/uaccess.h> -#include <asm/vtoc.h> - -#include "check.h" -#include "ibm.h" - -/* - * compute the block number from a - * cyl-cyl-head-head structure - */ -static sector_t -cchh2blk (struct vtoc_cchh *ptr, struct hd_geometry *geo) { - - sector_t cyl; - __u16 head; - - /*decode cylinder and heads for large volumes */ - cyl = ptr->hh & 0xFFF0; - cyl <<= 12; - cyl |= ptr->cc; - head = ptr->hh & 0x000F; - return cyl * geo->heads * geo->sectors + - head * geo->sectors; -} - -/* - * compute the block number from a - * cyl-cyl-head-head-block structure - */ -static sector_t -cchhb2blk (struct vtoc_cchhb *ptr, struct hd_geometry *geo) { - - sector_t cyl; - __u16 head; - - /*decode cylinder and heads for large volumes */ - cyl = ptr->hh & 0xFFF0; - cyl <<= 12; - cyl |= ptr->cc; - head = ptr->hh & 0x000F; - return cyl * geo->heads * geo->sectors + - head * geo->sectors + - ptr->b; -} - -/* - */ -int ibm_partition(struct parsed_partitions *state) -{ - struct block_device *bdev = state->bdev; - int blocksize, res; - loff_t i_size, offset, size, fmt_size; - dasd_information2_t *info; - struct hd_geometry *geo; - char type[5] = {0,}; - char name[7] = {0,}; - union label_t { - struct vtoc_volume_label_cdl vol; - struct vtoc_volume_label_ldl lnx; - struct vtoc_cms_label cms; - } *label; - unsigned char *data; - Sector sect; - sector_t labelsect; - char tmp[64]; - - res = 0; - blocksize = bdev_logical_block_size(bdev); - if (blocksize <= 0) - goto out_exit; - i_size = i_size_read(bdev->bd_inode); - if (i_size == 0) - goto out_exit; - - info = kmalloc(sizeof(dasd_information2_t), GFP_KERNEL); - if (info == NULL) - goto out_exit; - geo = kmalloc(sizeof(struct hd_geometry), GFP_KERNEL); - if (geo == NULL) - goto out_nogeo; - label = kmalloc(sizeof(union label_t), GFP_KERNEL); - if (label == NULL) - goto out_nolab; - - if (ioctl_by_bdev(bdev, BIODASDINFO2, (unsigned long)info) != 0 || - ioctl_by_bdev(bdev, HDIO_GETGEO, (unsigned long)geo) != 0) - goto out_freeall; - - /* - * Special case for FBA disks: label sector does not depend on - * blocksize. - */ - if ((info->cu_type == 0x6310 && info->dev_type == 0x9336) || - (info->cu_type == 0x3880 && info->dev_type == 0x3370)) - labelsect = info->label_block; - else - labelsect = info->label_block * (blocksize >> 9); - - /* - * Get volume label, extract name and type. - */ - data = read_part_sector(state, labelsect, §); - if (data == NULL) - goto out_readerr; - - memcpy(label, data, sizeof(union label_t)); - put_dev_sector(sect); - - if ((!info->FBA_layout) && (!strcmp(info->type, "ECKD"))) { - strncpy(type, label->vol.vollbl, 4); - strncpy(name, label->vol.volid, 6); - } else { - strncpy(type, label->lnx.vollbl, 4); - strncpy(name, label->lnx.volid, 6); - } - EBCASC(type, 4); - EBCASC(name, 6); - - res = 1; - - /* - * Three different formats: LDL, CDL and unformated disk - * - * identified by info->format - * - * unformated disks we do not have to care about - */ - if (info->format == DASD_FORMAT_LDL) { - if (strncmp(type, "CMS1", 4) == 0) { - /* - * VM style CMS1 labeled disk - */ - blocksize = label->cms.block_size; - if (label->cms.disk_offset != 0) { - snprintf(tmp, sizeof(tmp), "CMS1/%8s(MDSK):", name); - strlcat(state->pp_buf, tmp, PAGE_SIZE); - /* disk is reserved minidisk */ - offset = label->cms.disk_offset; - size = (label->cms.block_count - 1) - * (blocksize >> 9); - } else { - snprintf(tmp, sizeof(tmp), "CMS1/%8s:", name); - strlcat(state->pp_buf, tmp, PAGE_SIZE); - offset = (info->label_block + 1); - size = label->cms.block_count - * (blocksize >> 9); - } - put_partition(state, 1, offset*(blocksize >> 9), - size-offset*(blocksize >> 9)); - } else { - if (strncmp(type, "LNX1", 4) == 0) { - snprintf(tmp, sizeof(tmp), "LNX1/%8s:", name); - strlcat(state->pp_buf, tmp, PAGE_SIZE); - if (label->lnx.ldl_version == 0xf2) { - fmt_size = label->lnx.formatted_blocks - * (blocksize >> 9); - } else if (!strcmp(info->type, "ECKD")) { - /* formated w/o large volume support */ - fmt_size = geo->cylinders * geo->heads - * geo->sectors * (blocksize >> 9); - } else { - /* old label and no usable disk geometry - * (e.g. DIAG) */ - fmt_size = i_size >> 9; - } - size = i_size >> 9; - if (fmt_size < size) - size = fmt_size; - offset = (info->label_block + 1); - } else { - /* unlabeled disk */ - strlcat(state->pp_buf, "(nonl)", PAGE_SIZE); - size = i_size >> 9; - offset = (info->label_block + 1); - } - put_partition(state, 1, offset*(blocksize >> 9), - size-offset*(blocksize >> 9)); - } - } else if (info->format == DASD_FORMAT_CDL) { - /* - * New style CDL formatted disk - */ - sector_t blk; - int counter; - - /* - * check if VOL1 label is available - * if not, something is wrong, skipping partition detection - */ - if (strncmp(type, "VOL1", 4) == 0) { - snprintf(tmp, sizeof(tmp), "VOL1/%8s:", name); - strlcat(state->pp_buf, tmp, PAGE_SIZE); - /* - * get block number and read then go through format1 - * labels - */ - blk = cchhb2blk(&label->vol.vtoc, geo) + 1; - counter = 0; - data = read_part_sector(state, blk * (blocksize/512), - §); - while (data != NULL) { - struct vtoc_format1_label f1; - - memcpy(&f1, data, - sizeof(struct vtoc_format1_label)); - put_dev_sector(sect); - - /* skip FMT4 / FMT5 / FMT7 labels */ - if (f1.DS1FMTID == _ascebc['4'] - || f1.DS1FMTID == _ascebc['5'] - || f1.DS1FMTID == _ascebc['7'] - || f1.DS1FMTID == _ascebc['9']) { - blk++; - data = read_part_sector(state, - blk * (blocksize/512), §); - continue; - } - - /* only FMT1 and 8 labels valid at this point */ - if (f1.DS1FMTID != _ascebc['1'] && - f1.DS1FMTID != _ascebc['8']) - break; - - /* OK, we got valid partition data */ - offset = cchh2blk(&f1.DS1EXT1.llimit, geo); - size = cchh2blk(&f1.DS1EXT1.ulimit, geo) - - offset + geo->sectors; - if (counter >= state->limit) - break; - put_partition(state, counter + 1, - offset * (blocksize >> 9), - size * (blocksize >> 9)); - counter++; - blk++; - data = read_part_sector(state, - blk * (blocksize/512), §); - } - - if (!data) - /* Are we not supposed to report this ? */ - goto out_readerr; - } else - printk(KERN_WARNING "Warning, expected Label VOL1 not " - "found, treating as CDL formated Disk"); - - } - - strlcat(state->pp_buf, "\n", PAGE_SIZE); - goto out_freeall; - - -out_readerr: - res = -1; -out_freeall: - kfree(label); -out_nolab: - kfree(geo); -out_nogeo: - kfree(info); -out_exit: - return res; -} diff --git a/fs/partitions/ibm.h b/fs/partitions/ibm.h deleted file mode 100644 index 08fb0804a812..000000000000 --- a/fs/partitions/ibm.h +++ /dev/null @@ -1 +0,0 @@ -int ibm_partition(struct parsed_partitions *); diff --git a/fs/partitions/karma.c b/fs/partitions/karma.c deleted file mode 100644 index 0ea19312706b..000000000000 --- a/fs/partitions/karma.c +++ /dev/null @@ -1,57 +0,0 @@ -/* - * fs/partitions/karma.c - * Rio Karma partition info. - * - * Copyright (C) 2006 Bob Copeland (me@bobcopeland.com) - * based on osf.c - */ - -#include "check.h" -#include "karma.h" - -int karma_partition(struct parsed_partitions *state) -{ - int i; - int slot = 1; - Sector sect; - unsigned char *data; - struct disklabel { - u8 d_reserved[270]; - struct d_partition { - __le32 p_res; - u8 p_fstype; - u8 p_res2[3]; - __le32 p_offset; - __le32 p_size; - } d_partitions[2]; - u8 d_blank[208]; - __le16 d_magic; - } __attribute__((packed)) *label; - struct d_partition *p; - - data = read_part_sector(state, 0, §); - if (!data) - return -1; - - label = (struct disklabel *)data; - if (le16_to_cpu(label->d_magic) != KARMA_LABEL_MAGIC) { - put_dev_sector(sect); - return 0; - } - - p = label->d_partitions; - for (i = 0 ; i < 2; i++, p++) { - if (slot == state->limit) - break; - - if (p->p_fstype == 0x4d && le32_to_cpu(p->p_size)) { - put_partition(state, slot, le32_to_cpu(p->p_offset), - le32_to_cpu(p->p_size)); - } - slot++; - } - strlcat(state->pp_buf, "\n", PAGE_SIZE); - put_dev_sector(sect); - return 1; -} - diff --git a/fs/partitions/karma.h b/fs/partitions/karma.h deleted file mode 100644 index c764b2e9df21..000000000000 --- a/fs/partitions/karma.h +++ /dev/null @@ -1,8 +0,0 @@ -/* - * fs/partitions/karma.h - */ - -#define KARMA_LABEL_MAGIC 0xAB56 - -int karma_partition(struct parsed_partitions *state); - diff --git a/fs/partitions/ldm.c b/fs/partitions/ldm.c deleted file mode 100644 index bd8ae788f689..000000000000 --- a/fs/partitions/ldm.c +++ /dev/null @@ -1,1570 +0,0 @@ -/** - * ldm - Support for Windows Logical Disk Manager (Dynamic Disks) - * - * Copyright (C) 2001,2002 Richard Russon <ldm@flatcap.org> - * Copyright (c) 2001-2007 Anton Altaparmakov - * Copyright (C) 2001,2002 Jakob Kemi <jakob.kemi@telia.com> - * - * Documentation is available at http://www.linux-ntfs.org/doku.php?id=downloads - * - * This program is free software; you can redistribute it and/or modify it under - * the terms of the GNU General Public License as published by the Free Software - * Foundation; either version 2 of the License, or (at your option) any later - * version. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS - * FOR A PARTICULAR PURPOSE. See the GNU General Public License for more - * details. - * - * You should have received a copy of the GNU General Public License along with - * this program (in the main directory of the source in the file COPYING); if - * not, write to the Free Software Foundation, Inc., 59 Temple Place, Suite 330, - * Boston, MA 02111-1307 USA - */ - -#include <linux/slab.h> -#include <linux/pagemap.h> -#include <linux/stringify.h> -#include <linux/kernel.h> -#include "ldm.h" -#include "check.h" -#include "msdos.h" - -/** - * ldm_debug/info/error/crit - Output an error message - * @f: A printf format string containing the message - * @...: Variables to substitute into @f - * - * ldm_debug() writes a DEBUG level message to the syslog but only if the - * driver was compiled with debug enabled. Otherwise, the call turns into a NOP. - */ -#ifndef CONFIG_LDM_DEBUG -#define ldm_debug(...) do {} while (0) -#else -#define ldm_debug(f, a...) _ldm_printk (KERN_DEBUG, __func__, f, ##a) -#endif - -#define ldm_crit(f, a...) _ldm_printk (KERN_CRIT, __func__, f, ##a) -#define ldm_error(f, a...) _ldm_printk (KERN_ERR, __func__, f, ##a) -#define ldm_info(f, a...) _ldm_printk (KERN_INFO, __func__, f, ##a) - -static __printf(3, 4) -void _ldm_printk(const char *level, const char *function, const char *fmt, ...) -{ - struct va_format vaf; - va_list args; - - va_start (args, fmt); - - vaf.fmt = fmt; - vaf.va = &args; - - printk("%s%s(): %pV\n", level, function, &vaf); - - va_end(args); -} - -/** - * ldm_parse_hexbyte - Convert a ASCII hex number to a byte - * @src: Pointer to at least 2 characters to convert. - * - * Convert a two character ASCII hex string to a number. - * - * Return: 0-255 Success, the byte was parsed correctly - * -1 Error, an invalid character was supplied - */ -static int ldm_parse_hexbyte (const u8 *src) -{ - unsigned int x; /* For correct wrapping */ - int h; - - /* high part */ - x = h = hex_to_bin(src[0]); - if (h < 0) - return -1; - - /* low part */ - h = hex_to_bin(src[1]); - if (h < 0) - return -1; - - return (x << 4) + h; -} - -/** - * ldm_parse_guid - Convert GUID from ASCII to binary - * @src: 36 char string of the form fa50ff2b-f2e8-45de-83fa-65417f2f49ba - * @dest: Memory block to hold binary GUID (16 bytes) - * - * N.B. The GUID need not be NULL terminated. - * - * Return: 'true' @dest contains binary GUID - * 'false' @dest contents are undefined - */ -static bool ldm_parse_guid (const u8 *src, u8 *dest) -{ - static const int size[] = { 4, 2, 2, 2, 6 }; - int i, j, v; - - if (src[8] != '-' || src[13] != '-' || - src[18] != '-' || src[23] != '-') - return false; - - for (j = 0; j < 5; j++, src++) - for (i = 0; i < size[j]; i++, src+=2, *dest++ = v) - if ((v = ldm_parse_hexbyte (src)) < 0) - return false; - - return true; -} - -/** - * ldm_parse_privhead - Read the LDM Database PRIVHEAD structure - * @data: Raw database PRIVHEAD structure loaded from the device - * @ph: In-memory privhead structure in which to return parsed information - * - * This parses the LDM database PRIVHEAD structure supplied in @data and - * sets up the in-memory privhead structure @ph with the obtained information. - * - * Return: 'true' @ph contains the PRIVHEAD data - * 'false' @ph contents are undefined - */ -static bool ldm_parse_privhead(const u8 *data, struct privhead *ph) -{ - bool is_vista = false; - - BUG_ON(!data || !ph); - if (MAGIC_PRIVHEAD != get_unaligned_be64(data)) { - ldm_error("Cannot find PRIVHEAD structure. LDM database is" - " corrupt. Aborting."); - return false; - } - ph->ver_major = get_unaligned_be16(data + 0x000C); - ph->ver_minor = get_unaligned_be16(data + 0x000E); - ph->logical_disk_start = get_unaligned_be64(data + 0x011B); - ph->logical_disk_size = get_unaligned_be64(data + 0x0123); - ph->config_start = get_unaligned_be64(data + 0x012B); - ph->config_size = get_unaligned_be64(data + 0x0133); - /* Version 2.11 is Win2k/XP and version 2.12 is Vista. */ - if (ph->ver_major == 2 && ph->ver_minor == 12) - is_vista = true; - if (!is_vista && (ph->ver_major != 2 || ph->ver_minor != 11)) { - ldm_error("Expected PRIVHEAD version 2.11 or 2.12, got %d.%d." - " Aborting.", ph->ver_major, ph->ver_minor); - return false; - } - ldm_debug("PRIVHEAD version %d.%d (Windows %s).", ph->ver_major, - ph->ver_minor, is_vista ? "Vista" : "2000/XP"); - if (ph->config_size != LDM_DB_SIZE) { /* 1 MiB in sectors. */ - /* Warn the user and continue, carefully. */ - ldm_info("Database is normally %u bytes, it claims to " - "be %llu bytes.", LDM_DB_SIZE, - (unsigned long long)ph->config_size); - } - if ((ph->logical_disk_size == 0) || (ph->logical_disk_start + - ph->logical_disk_size > ph->config_start)) { - ldm_error("PRIVHEAD disk size doesn't match real disk size"); - return false; - } - if (!ldm_parse_guid(data + 0x0030, ph->disk_id)) { - ldm_error("PRIVHEAD contains an invalid GUID."); - return false; - } - ldm_debug("Parsed PRIVHEAD successfully."); - return true; -} - -/** - * ldm_parse_tocblock - Read the LDM Database TOCBLOCK structure - * @data: Raw database TOCBLOCK structure loaded from the device - * @toc: In-memory toc structure in which to return parsed information - * - * This parses the LDM Database TOCBLOCK (table of contents) structure supplied - * in @data and sets up the in-memory tocblock structure @toc with the obtained - * information. - * - * N.B. The *_start and *_size values returned in @toc are not range-checked. - * - * Return: 'true' @toc contains the TOCBLOCK data - * 'false' @toc contents are undefined - */ -static bool ldm_parse_tocblock (const u8 *data, struct tocblock *toc) -{ - BUG_ON (!data || !toc); - - if (MAGIC_TOCBLOCK != get_unaligned_be64(data)) { - ldm_crit ("Cannot find TOCBLOCK, database may be corrupt."); - return false; - } - strncpy (toc->bitmap1_name, data + 0x24, sizeof (toc->bitmap1_name)); - toc->bitmap1_name[sizeof (toc->bitmap1_name) - 1] = 0; - toc->bitmap1_start = get_unaligned_be64(data + 0x2E); - toc->bitmap1_size = get_unaligned_be64(data + 0x36); - - if (strncmp (toc->bitmap1_name, TOC_BITMAP1, - sizeof (toc->bitmap1_name)) != 0) { - ldm_crit ("TOCBLOCK's first bitmap is '%s', should be '%s'.", - TOC_BITMAP1, toc->bitmap1_name); - return false; - } - strncpy (toc->bitmap2_name, data + 0x46, sizeof (toc->bitmap2_name)); - toc->bitmap2_name[sizeof (toc->bitmap2_name) - 1] = 0; - toc->bitmap2_start = get_unaligned_be64(data + 0x50); - toc->bitmap2_size = get_unaligned_be64(data + 0x58); - if (strncmp (toc->bitmap2_name, TOC_BITMAP2, - sizeof (toc->bitmap2_name)) != 0) { - ldm_crit ("TOCBLOCK's second bitmap is '%s', should be '%s'.", - TOC_BITMAP2, toc->bitmap2_name); - return false; - } - ldm_debug ("Parsed TOCBLOCK successfully."); - return true; -} - -/** - * ldm_parse_vmdb - Read the LDM Database VMDB structure - * @data: Raw database VMDB structure loaded from the device - * @vm: In-memory vmdb structure in which to return parsed information - * - * This parses the LDM Database VMDB structure supplied in @data and sets up - * the in-memory vmdb structure @vm with the obtained information. - * - * N.B. The *_start, *_size and *_seq values will be range-checked later. - * - * Return: 'true' @vm contains VMDB info - * 'false' @vm contents are undefined - */ -static bool ldm_parse_vmdb (const u8 *data, struct vmdb *vm) -{ - BUG_ON (!data || !vm); - - if (MAGIC_VMDB != get_unaligned_be32(data)) { - ldm_crit ("Cannot find the VMDB, database may be corrupt."); - return false; - } - - vm->ver_major = get_unaligned_be16(data + 0x12); - vm->ver_minor = get_unaligned_be16(data + 0x14); - if ((vm->ver_major != 4) || (vm->ver_minor != 10)) { - ldm_error ("Expected VMDB version %d.%d, got %d.%d. " - "Aborting.", 4, 10, vm->ver_major, vm->ver_minor); - return false; - } - - vm->vblk_size = get_unaligned_be32(data + 0x08); - if (vm->vblk_size == 0) { - ldm_error ("Illegal VBLK size"); - return false; - } - - vm->vblk_offset = get_unaligned_be32(data + 0x0C); - vm->last_vblk_seq = get_unaligned_be32(data + 0x04); - - ldm_debug ("Parsed VMDB successfully."); - return true; -} - -/** - * ldm_compare_privheads - Compare two privhead objects - * @ph1: First privhead - * @ph2: Second privhead - * - * This compares the two privhead structures @ph1 and @ph2. - * - * Return: 'true' Identical - * 'false' Different - */ -static bool ldm_compare_privheads (const struct privhead *ph1, - const struct privhead *ph2) -{ - BUG_ON (!ph1 || !ph2); - - return ((ph1->ver_major == ph2->ver_major) && - (ph1->ver_minor == ph2->ver_minor) && - (ph1->logical_disk_start == ph2->logical_disk_start) && - (ph1->logical_disk_size == ph2->logical_disk_size) && - (ph1->config_start == ph2->config_start) && - (ph1->config_size == ph2->config_size) && - !memcmp (ph1->disk_id, ph2->disk_id, GUID_SIZE)); -} - -/** - * ldm_compare_tocblocks - Compare two tocblock objects - * @toc1: First toc - * @toc2: Second toc - * - * This compares the two tocblock structures @toc1 and @toc2. - * - * Return: 'true' Identical - * 'false' Different - */ -static bool ldm_compare_tocblocks (const struct tocblock *toc1, - const struct tocblock *toc2) -{ - BUG_ON (!toc1 || !toc2); - - return ((toc1->bitmap1_start == toc2->bitmap1_start) && - (toc1->bitmap1_size == toc2->bitmap1_size) && - (toc1->bitmap2_start == toc2->bitmap2_start) && - (toc1->bitmap2_size == toc2->bitmap2_size) && - !strncmp (toc1->bitmap1_name, toc2->bitmap1_name, - sizeof (toc1->bitmap1_name)) && - !strncmp (toc1->bitmap2_name, toc2->bitmap2_name, - sizeof (toc1->bitmap2_name))); -} - -/** - * ldm_validate_privheads - Compare the primary privhead with its backups - * @state: Partition check state including device holding the LDM Database - * @ph1: Memory struct to fill with ph contents - * - * Read and compare all three privheads from disk. - * - * The privheads on disk show the size and location of the main disk area and - * the configuration area (the database). The values are range-checked against - * @hd, which contains the real size of the disk. - * - * Return: 'true' Success - * 'false' Error - */ -static bool ldm_validate_privheads(struct parsed_partitions *state, - struct privhead *ph1) -{ - static const int off[3] = { OFF_PRIV1, OFF_PRIV2, OFF_PRIV3 }; - struct privhead *ph[3] = { ph1 }; - Sector sect; - u8 *data; - bool result = false; - long num_sects; - int i; - - BUG_ON (!state || !ph1); - - ph[1] = kmalloc (sizeof (*ph[1]), GFP_KERNEL); - ph[2] = kmalloc (sizeof (*ph[2]), GFP_KERNEL); - if (!ph[1] || !ph[2]) { - ldm_crit ("Out of memory."); - goto out; - } - - /* off[1 & 2] are relative to ph[0]->config_start */ - ph[0]->config_start = 0; - - /* Read and parse privheads */ - for (i = 0; i < 3; i++) { - data = read_part_sector(state, ph[0]->config_start + off[i], - §); - if (!data) { - ldm_crit ("Disk read failed."); - goto out; - } - result = ldm_parse_privhead (data, ph[i]); - put_dev_sector (sect); - if (!result) { - ldm_error ("Cannot find PRIVHEAD %d.", i+1); /* Log again */ - if (i < 2) - goto out; /* Already logged */ - else - break; /* FIXME ignore for now, 3rd PH can fail on odd-sized disks */ - } - } - - num_sects = state->bdev->bd_inode->i_size >> 9; - - if ((ph[0]->config_start > num_sects) || - ((ph[0]->config_start + ph[0]->config_size) > num_sects)) { - ldm_crit ("Database extends beyond the end of the disk."); - goto out; - } - - if ((ph[0]->logical_disk_start > ph[0]->config_start) || - ((ph[0]->logical_disk_start + ph[0]->logical_disk_size) - > ph[0]->config_start)) { - ldm_crit ("Disk and database overlap."); - goto out; - } - - if (!ldm_compare_privheads (ph[0], ph[1])) { - ldm_crit ("Primary and backup PRIVHEADs don't match."); - goto out; - } - /* FIXME ignore this for now - if (!ldm_compare_privheads (ph[0], ph[2])) { - ldm_crit ("Primary and backup PRIVHEADs don't match."); - goto out; - }*/ - ldm_debug ("Validated PRIVHEADs successfully."); - result = true; -out: - kfree (ph[1]); - kfree (ph[2]); - return result; -} - -/** - * ldm_validate_tocblocks - Validate the table of contents and its backups - * @state: Partition check state including device holding the LDM Database - * @base: Offset, into @state->bdev, of the database - * @ldb: Cache of the database structures - * - * Find and compare the four tables of contents of the LDM Database stored on - * @state->bdev and return the parsed information into @toc1. - * - * The offsets and sizes of the configs are range-checked against a privhead. - * - * Return: 'true' @toc1 contains validated TOCBLOCK info - * 'false' @toc1 contents are undefined - */ -static bool ldm_validate_tocblocks(struct parsed_partitions *state, - unsigned long base, struct ldmdb *ldb) -{ - static const int off[4] = { OFF_TOCB1, OFF_TOCB2, OFF_TOCB3, OFF_TOCB4}; - struct tocblock *tb[4]; - struct privhead *ph; - Sector sect; - u8 *data; - int i, nr_tbs; - bool result = false; - - BUG_ON(!state || !ldb); - ph = &ldb->ph; - tb[0] = &ldb->toc; - tb[1] = kmalloc(sizeof(*tb[1]) * 3, GFP_KERNEL); - if (!tb[1]) { - ldm_crit("Out of memory."); - goto err; - } - tb[2] = (struct tocblock*)((u8*)tb[1] + sizeof(*tb[1])); - tb[3] = (struct tocblock*)((u8*)tb[2] + sizeof(*tb[2])); - /* - * Try to read and parse all four TOCBLOCKs. - * - * Windows Vista LDM v2.12 does not always have all four TOCBLOCKs so - * skip any that fail as long as we get at least one valid TOCBLOCK. - */ - for (nr_tbs = i = 0; i < 4; i++) { - data = read_part_sector(state, base + off[i], §); - if (!data) { - ldm_error("Disk read failed for TOCBLOCK %d.", i); - continue; - } - if (ldm_parse_tocblock(data, tb[nr_tbs])) - nr_tbs++; - put_dev_sector(sect); - } - if (!nr_tbs) { - ldm_crit("Failed to find a valid TOCBLOCK."); - goto err; - } - /* Range check the TOCBLOCK against a privhead. */ - if (((tb[0]->bitmap1_start + tb[0]->bitmap1_size) > ph->config_size) || - ((tb[0]->bitmap2_start + tb[0]->bitmap2_size) > - ph->config_size)) { - ldm_crit("The bitmaps are out of range. Giving up."); - goto err; - } - /* Compare all loaded TOCBLOCKs. */ - for (i = 1; i < nr_tbs; i++) { - if (!ldm_compare_tocblocks(tb[0], tb[i])) { - ldm_crit("TOCBLOCKs 0 and %d do not match.", i); - goto err; - } - } - ldm_debug("Validated %d TOCBLOCKs successfully.", nr_tbs); - result = true; -err: - kfree(tb[1]); - return result; -} - -/** - * ldm_validate_vmdb - Read the VMDB and validate it - * @state: Partition check state including device holding the LDM Database - * @base: Offset, into @bdev, of the database - * @ldb: Cache of the database structures - * - * Find the vmdb of the LDM Database stored on @bdev and return the parsed - * information in @ldb. - * - * Return: 'true' @ldb contains validated VBDB info - * 'false' @ldb contents are undefined - */ -static bool ldm_validate_vmdb(struct parsed_partitions *state, - unsigned long base, struct ldmdb *ldb) -{ - Sector sect; - u8 *data; - bool result = false; - struct vmdb *vm; - struct tocblock *toc; - - BUG_ON (!state || !ldb); - - vm = &ldb->vm; - toc = &ldb->toc; - - data = read_part_sector(state, base + OFF_VMDB, §); - if (!data) { - ldm_crit ("Disk read failed."); - return false; - } - - if (!ldm_parse_vmdb (data, vm)) - goto out; /* Already logged */ - - /* Are there uncommitted transactions? */ - if (get_unaligned_be16(data + 0x10) != 0x01) { - ldm_crit ("Database is not in a consistent state. Aborting."); - goto out; - } - - if (vm->vblk_offset != 512) - ldm_info ("VBLKs start at offset 0x%04x.", vm->vblk_offset); - - /* - * The last_vblkd_seq can be before the end of the vmdb, just make sure - * it is not out of bounds. - */ - if ((vm->vblk_size * vm->last_vblk_seq) > (toc->bitmap1_size << 9)) { - ldm_crit ("VMDB exceeds allowed size specified by TOCBLOCK. " - "Database is corrupt. Aborting."); - goto out; - } - - result = true; -out: - put_dev_sector (sect); - return result; -} - - -/** - * ldm_validate_partition_table - Determine whether bdev might be a dynamic disk - * @state: Partition check state including device holding the LDM Database - * - * This function provides a weak test to decide whether the device is a dynamic - * disk or not. It looks for an MS-DOS-style partition table containing at - * least one partition of type 0x42 (formerly SFS, now used by Windows for - * dynamic disks). - * - * N.B. The only possible error can come from the read_part_sector and that is - * only likely to happen if the underlying device is strange. If that IS - * the case we should return zero to let someone else try. - * - * Return: 'true' @state->bdev is a dynamic disk - * 'false' @state->bdev is not a dynamic disk, or an error occurred - */ -static bool ldm_validate_partition_table(struct parsed_partitions *state) -{ - Sector sect; - u8 *data; - struct partition *p; - int i; - bool result = false; - - BUG_ON(!state); - - data = read_part_sector(state, 0, §); - if (!data) { - ldm_info ("Disk read failed."); - return false; - } - - if (*(__le16*) (data + 0x01FE) != cpu_to_le16 (MSDOS_LABEL_MAGIC)) - goto out; - - p = (struct partition*)(data + 0x01BE); - for (i = 0; i < 4; i++, p++) - if (SYS_IND (p) == LDM_PARTITION) { - result = true; - break; - } - - if (result) - ldm_debug ("Found W2K dynamic disk partition type."); - -out: - put_dev_sector (sect); - return result; -} - -/** - * ldm_get_disk_objid - Search a linked list of vblk's for a given Disk Id - * @ldb: Cache of the database structures - * - * The LDM Database contains a list of all partitions on all dynamic disks. - * The primary PRIVHEAD, at the beginning of the physical disk, tells us - * the GUID of this disk. This function searches for the GUID in a linked - * list of vblk's. - * - * Return: Pointer, A matching vblk was found - * NULL, No match, or an error - */ -static struct vblk * ldm_get_disk_objid (const struct ldmdb *ldb) -{ - struct list_head *item; - - BUG_ON (!ldb); - - list_for_each (item, &ldb->v_disk) { - struct vblk *v = list_entry (item, struct vblk, list); - if (!memcmp (v->vblk.disk.disk_id, ldb->ph.disk_id, GUID_SIZE)) - return v; - } - - return NULL; -} - -/** - * ldm_create_data_partitions - Create data partitions for this device - * @pp: List of the partitions parsed so far - * @ldb: Cache of the database structures - * - * The database contains ALL the partitions for ALL disk groups, so we need to - * filter out this specific disk. Using the disk's object id, we can find all - * the partitions in the database that belong to this disk. - * - * Add each partition in our database, to the parsed_partitions structure. - * - * N.B. This function creates the partitions in the order it finds partition - * objects in the linked list. - * - * Return: 'true' Partition created - * 'false' Error, probably a range checking problem - */ -static bool ldm_create_data_partitions (struct parsed_partitions *pp, - const struct ldmdb *ldb) -{ - struct list_head *item; - struct vblk *vb; - struct vblk *disk; - struct vblk_part *part; - int part_num = 1; - - BUG_ON (!pp || !ldb); - - disk = ldm_get_disk_objid (ldb); - if (!disk) { - ldm_crit ("Can't find the ID of this disk in the database."); - return false; - } - - strlcat(pp->pp_buf, " [LDM]", PAGE_SIZE); - - /* Create the data partitions */ - list_for_each (item, &ldb->v_part) { - vb = list_entry (item, struct vblk, list); - part = &vb->vblk.part; - - if (part->disk_id != disk->obj_id) - continue; - - put_partition (pp, part_num, ldb->ph.logical_disk_start + - part->start, part->size); - part_num++; - } - - strlcat(pp->pp_buf, "\n", PAGE_SIZE); - return true; -} - - -/** - * ldm_relative - Calculate the next relative offset - * @buffer: Block of data being worked on - * @buflen: Size of the block of data - * @base: Size of the previous fixed width fields - * @offset: Cumulative size of the previous variable-width fields - * - * Because many of the VBLK fields are variable-width, it's necessary - * to calculate each offset based on the previous one and the length - * of the field it pointed to. - * - * Return: -1 Error, the calculated offset exceeded the size of the buffer - * n OK, a range-checked offset into buffer - */ -static int ldm_relative(const u8 *buffer, int buflen, int base, int offset) -{ - - base += offset; - if (!buffer || offset < 0 || base > buflen) { - if (!buffer) - ldm_error("!buffer"); - if (offset < 0) - ldm_error("offset (%d) < 0", offset); - if (base > buflen) - ldm_error("base (%d) > buflen (%d)", base, buflen); - return -1; - } - if (base + buffer[base] >= buflen) { - ldm_error("base (%d) + buffer[base] (%d) >= buflen (%d)", base, - buffer[base], buflen); - return -1; - } - return buffer[base] + offset + 1; -} - -/** - * ldm_get_vnum - Convert a variable-width, big endian number, into cpu order - * @block: Pointer to the variable-width number to convert - * - * Large numbers in the LDM Database are often stored in a packed format. Each - * number is prefixed by a one byte width marker. All numbers in the database - * are stored in big-endian byte order. This function reads one of these - * numbers and returns the result - * - * N.B. This function DOES NOT perform any range checking, though the most - * it will read is eight bytes. - * - * Return: n A number - * 0 Zero, or an error occurred - */ -static u64 ldm_get_vnum (const u8 *block) -{ - u64 tmp = 0; - u8 length; - - BUG_ON (!block); - - length = *block++; - - if (length && length <= 8) - while (length--) - tmp = (tmp << 8) | *block++; - else - ldm_error ("Illegal length %d.", length); - - return tmp; -} - -/** - * ldm_get_vstr - Read a length-prefixed string into a buffer - * @block: Pointer to the length marker - * @buffer: Location to copy string to - * @buflen: Size of the output buffer - * - * Many of the strings in the LDM Database are not NULL terminated. Instead - * they are prefixed by a one byte length marker. This function copies one of - * these strings into a buffer. - * - * N.B. This function DOES NOT perform any range checking on the input. - * If the buffer is too small, the output will be truncated. - * - * Return: 0, Error and @buffer contents are undefined - * n, String length in characters (excluding NULL) - * buflen-1, String was truncated. - */ -static int ldm_get_vstr (const u8 *block, u8 *buffer, int buflen) -{ - int length; - - BUG_ON (!block || !buffer); - - length = block[0]; - if (length >= buflen) { - ldm_error ("Truncating string %d -> %d.", length, buflen); - length = buflen - 1; - } - memcpy (buffer, block + 1, length); - buffer[length] = 0; - return length; -} - - -/** - * ldm_parse_cmp3 - Read a raw VBLK Component object into a vblk structure - * @buffer: Block of data being worked on - * @buflen: Size of the block of data - * @vb: In-memory vblk in which to return information - * - * Read a raw VBLK Component object (version 3) into a vblk structure. - * - * Return: 'true' @vb contains a Component VBLK - * 'false' @vb contents are not defined - */ -static bool ldm_parse_cmp3 (const u8 *buffer, int buflen, struct vblk *vb) -{ - int r_objid, r_name, r_vstate, r_child, r_parent, r_stripe, r_cols, len; - struct vblk_comp *comp; - - BUG_ON (!buffer || !vb); - - r_objid = ldm_relative (buffer, buflen, 0x18, 0); - r_name = ldm_relative (buffer, buflen, 0x18, r_objid); - r_vstate = ldm_relative (buffer, buflen, 0x18, r_name); - r_child = ldm_relative (buffer, buflen, 0x1D, r_vstate); - r_parent = ldm_relative (buffer, buflen, 0x2D, r_child); - - if (buffer[0x12] & VBLK_FLAG_COMP_STRIPE) { - r_stripe = ldm_relative (buffer, buflen, 0x2E, r_parent); - r_cols = ldm_relative (buffer, buflen, 0x2E, r_stripe); - len = r_cols; - } else { - r_stripe = 0; - r_cols = 0; - len = r_parent; - } - if (len < 0) - return false; - - len += VBLK_SIZE_CMP3; - if (len != get_unaligned_be32(buffer + 0x14)) - return false; - - comp = &vb->vblk.comp; - ldm_get_vstr (buffer + 0x18 + r_name, comp->state, - sizeof (comp->state)); - comp->type = buffer[0x18 + r_vstate]; - comp->children = ldm_get_vnum (buffer + 0x1D + r_vstate); - comp->parent_id = ldm_get_vnum (buffer + 0x2D + r_child); - comp->chunksize = r_stripe ? ldm_get_vnum (buffer+r_parent+0x2E) : 0; - - return true; -} - -/** - * ldm_parse_dgr3 - Read a raw VBLK Disk Group object into a vblk structure - * @buffer: Block of data being worked on - * @buflen: Size of the block of data - * @vb: In-memory vblk in which to return information - * - * Read a raw VBLK Disk Group object (version 3) into a vblk structure. - * - * Return: 'true' @vb contains a Disk Group VBLK - * 'false' @vb contents are not defined - */ -static int ldm_parse_dgr3 (const u8 *buffer, int buflen, struct vblk *vb) -{ - int r_objid, r_name, r_diskid, r_id1, r_id2, len; - struct vblk_dgrp *dgrp; - - BUG_ON (!buffer || !vb); - - r_objid = ldm_relative (buffer, buflen, 0x18, 0); - r_name = ldm_relative (buffer, buflen, 0x18, r_objid); - r_diskid = ldm_relative (buffer, buflen, 0x18, r_name); - - if (buffer[0x12] & VBLK_FLAG_DGR3_IDS) { - r_id1 = ldm_relative (buffer, buflen, 0x24, r_diskid); - r_id2 = ldm_relative (buffer, buflen, 0x24, r_id1); - len = r_id2; - } else { - r_id1 = 0; - r_id2 = 0; - len = r_diskid; - } - if (len < 0) - return false; - - len += VBLK_SIZE_DGR3; - if (len != get_unaligned_be32(buffer + 0x14)) - return false; - - dgrp = &vb->vblk.dgrp; - ldm_get_vstr (buffer + 0x18 + r_name, dgrp->disk_id, - sizeof (dgrp->disk_id)); - return true; -} - -/** - * ldm_parse_dgr4 - Read a raw VBLK Disk Group object into a vblk structure - * @buffer: Block of data being worked on - * @buflen: Size of the block of data - * @vb: In-memory vblk in which to return information - * - * Read a raw VBLK Disk Group object (version 4) into a vblk structure. - * - * Return: 'true' @vb contains a Disk Group VBLK - * 'false' @vb contents are not defined - */ -static bool ldm_parse_dgr4 (const u8 *buffer, int buflen, struct vblk *vb) -{ - char buf[64]; - int r_objid, r_name, r_id1, r_id2, len; - struct vblk_dgrp *dgrp; - - BUG_ON (!buffer || !vb); - - r_objid = ldm_relative (buffer, buflen, 0x18, 0); - r_name = ldm_relative (buffer, buflen, 0x18, r_objid); - - if (buffer[0x12] & VBLK_FLAG_DGR4_IDS) { - r_id1 = ldm_relative (buffer, buflen, 0x44, r_name); - r_id2 = ldm_relative (buffer, buflen, 0x44, r_id1); - len = r_id2; - } else { - r_id1 = 0; - r_id2 = 0; - len = r_name; - } - if (len < 0) - return false; - - len += VBLK_SIZE_DGR4; - if (len != get_unaligned_be32(buffer + 0x14)) - return false; - - dgrp = &vb->vblk.dgrp; - - ldm_get_vstr (buffer + 0x18 + r_objid, buf, sizeof (buf)); - return true; -} - -/** - * ldm_parse_dsk3 - Read a raw VBLK Disk object into a vblk structure - * @buffer: Block of data being worked on - * @buflen: Size of the block of data - * @vb: In-memory vblk in which to return information - * - * Read a raw VBLK Disk object (version 3) into a vblk structure. - * - * Return: 'true' @vb contains a Disk VBLK - * 'false' @vb contents are not defined - */ -static bool ldm_parse_dsk3 (const u8 *buffer, int buflen, struct vblk *vb) -{ - int r_objid, r_name, r_diskid, r_altname, len; - struct vblk_disk *disk; - - BUG_ON (!buffer || !vb); - - r_objid = ldm_relative (buffer, buflen, 0x18, 0); - r_name = ldm_relative (buffer, buflen, 0x18, r_objid); - r_diskid = ldm_relative (buffer, buflen, 0x18, r_name); - r_altname = ldm_relative (buffer, buflen, 0x18, r_diskid); - len = r_altname; - if (len < 0) - return false; - - len += VBLK_SIZE_DSK3; - if (len != get_unaligned_be32(buffer + 0x14)) - return false; - - disk = &vb->vblk.disk; - ldm_get_vstr (buffer + 0x18 + r_diskid, disk->alt_name, - sizeof (disk->alt_name)); - if (!ldm_parse_guid (buffer + 0x19 + r_name, disk->disk_id)) - return false; - - return true; -} - -/** - * ldm_parse_dsk4 - Read a raw VBLK Disk object into a vblk structure - * @buffer: Block of data being worked on - * @buflen: Size of the block of data - * @vb: In-memory vblk in which to return information - * - * Read a raw VBLK Disk object (version 4) into a vblk structure. - * - * Return: 'true' @vb contains a Disk VBLK - * 'false' @vb contents are not defined - */ -static bool ldm_parse_dsk4 (const u8 *buffer, int buflen, struct vblk *vb) -{ - int r_objid, r_name, len; - struct vblk_disk *disk; - - BUG_ON (!buffer || !vb); - - r_objid = ldm_relative (buffer, buflen, 0x18, 0); - r_name = ldm_relative (buffer, buflen, 0x18, r_objid); - len = r_name; - if (len < 0) - return false; - - len += VBLK_SIZE_DSK4; - if (len != get_unaligned_be32(buffer + 0x14)) - return false; - - disk = &vb->vblk.disk; - memcpy (disk->disk_id, buffer + 0x18 + r_name, GUID_SIZE); - return true; -} - -/** - * ldm_parse_prt3 - Read a raw VBLK Partition object into a vblk structure - * @buffer: Block of data being worked on - * @buflen: Size of the block of data - * @vb: In-memory vblk in which to return information - * - * Read a raw VBLK Partition object (version 3) into a vblk structure. - * - * Return: 'true' @vb contains a Partition VBLK - * 'false' @vb contents are not defined - */ -static bool ldm_parse_prt3(const u8 *buffer, int buflen, struct vblk *vb) -{ - int r_objid, r_name, r_size, r_parent, r_diskid, r_index, len; - struct vblk_part *part; - - BUG_ON(!buffer || !vb); - r_objid = ldm_relative(buffer, buflen, 0x18, 0); - if (r_objid < 0) { - ldm_error("r_objid %d < 0", r_objid); - return false; - } - r_name = ldm_relative(buffer, buflen, 0x18, r_objid); - if (r_name < 0) { - ldm_error("r_name %d < 0", r_name); - return false; - } - r_size = ldm_relative(buffer, buflen, 0x34, r_name); - if (r_size < 0) { - ldm_error("r_size %d < 0", r_size); - return false; - } - r_parent = ldm_relative(buffer, buflen, 0x34, r_size); - if (r_parent < 0) { - ldm_error("r_parent %d < 0", r_parent); - return false; - } - r_diskid = ldm_relative(buffer, buflen, 0x34, r_parent); - if (r_diskid < 0) { - ldm_error("r_diskid %d < 0", r_diskid); - return false; - } - if (buffer[0x12] & VBLK_FLAG_PART_INDEX) { - r_index = ldm_relative(buffer, buflen, 0x34, r_diskid); - if (r_index < 0) { - ldm_error("r_index %d < 0", r_index); - return false; - } - len = r_index; - } else { - r_index = 0; - len = r_diskid; - } - if (len < 0) { - ldm_error("len %d < 0", len); - return false; - } - len += VBLK_SIZE_PRT3; - if (len > get_unaligned_be32(buffer + 0x14)) { - ldm_error("len %d > BE32(buffer + 0x14) %d", len, - get_unaligned_be32(buffer + 0x14)); - return false; - } - part = &vb->vblk.part; - part->start = get_unaligned_be64(buffer + 0x24 + r_name); - part->volume_offset = get_unaligned_be64(buffer + 0x2C + r_name); - part->size = ldm_get_vnum(buffer + 0x34 + r_name); - part->parent_id = ldm_get_vnum(buffer + 0x34 + r_size); - part->disk_id = ldm_get_vnum(buffer + 0x34 + r_parent); - if (vb->flags & VBLK_FLAG_PART_INDEX) - part->partnum = buffer[0x35 + r_diskid]; - else - part->partnum = 0; - return true; -} - -/** - * ldm_parse_vol5 - Read a raw VBLK Volume object into a vblk structure - * @buffer: Block of data being worked on - * @buflen: Size of the block of data - * @vb: In-memory vblk in which to return information - * - * Read a raw VBLK Volume object (version 5) into a vblk structure. - * - * Return: 'true' @vb contains a Volume VBLK - * 'false' @vb contents are not defined - */ -static bool ldm_parse_vol5(const u8 *buffer, int buflen, struct vblk *vb) -{ - int r_objid, r_name, r_vtype, r_disable_drive_letter, r_child, r_size; - int r_id1, r_id2, r_size2, r_drive, len; - struct vblk_volu *volu; - - BUG_ON(!buffer || !vb); - r_objid = ldm_relative(buffer, buflen, 0x18, 0); - if (r_objid < 0) { - ldm_error("r_objid %d < 0", r_objid); - return false; - } - r_name = ldm_relative(buffer, buflen, 0x18, r_objid); - if (r_name < 0) { - ldm_error("r_name %d < 0", r_name); - return false; - } - r_vtype = ldm_relative(buffer, buflen, 0x18, r_name); - if (r_vtype < 0) { - ldm_error("r_vtype %d < 0", r_vtype); - return false; - } - r_disable_drive_letter = ldm_relative(buffer, buflen, 0x18, r_vtype); - if (r_disable_drive_letter < 0) { - ldm_error("r_disable_drive_letter %d < 0", - r_disable_drive_letter); - return false; - } - r_child = ldm_relative(buffer, buflen, 0x2D, r_disable_drive_letter); - if (r_child < 0) { - ldm_error("r_child %d < 0", r_child); - return false; - } - r_size = ldm_relative(buffer, buflen, 0x3D, r_child); - if (r_size < 0) { - ldm_error("r_size %d < 0", r_size); - return false; - } - if (buffer[0x12] & VBLK_FLAG_VOLU_ID1) { - r_id1 = ldm_relative(buffer, buflen, 0x52, r_size); - if (r_id1 < 0) { - ldm_error("r_id1 %d < 0", r_id1); - return false; - } - } else - r_id1 = r_size; - if (buffer[0x12] & VBLK_FLAG_VOLU_ID2) { - r_id2 = ldm_relative(buffer, buflen, 0x52, r_id1); - if (r_id2 < 0) { - ldm_error("r_id2 %d < 0", r_id2); - return false; - } - } else - r_id2 = r_id1; - if (buffer[0x12] & VBLK_FLAG_VOLU_SIZE) { - r_size2 = ldm_relative(buffer, buflen, 0x52, r_id2); - if (r_size2 < 0) { - ldm_error("r_size2 %d < 0", r_size2); - return false; - } - } else - r_size2 = r_id2; - if (buffer[0x12] & VBLK_FLAG_VOLU_DRIVE) { - r_drive = ldm_relative(buffer, buflen, 0x52, r_size2); - if (r_drive < 0) { - ldm_error("r_drive %d < 0", r_drive); - return false; - } - } else - r_drive = r_size2; - len = r_drive; - if (len < 0) { - ldm_error("len %d < 0", len); - return false; - } - len += VBLK_SIZE_VOL5; - if (len > get_unaligned_be32(buffer + 0x14)) { - ldm_error("len %d > BE32(buffer + 0x14) %d", len, - get_unaligned_be32(buffer + 0x14)); - return false; - } - volu = &vb->vblk.volu; - ldm_get_vstr(buffer + 0x18 + r_name, volu->volume_type, - sizeof(volu->volume_type)); - memcpy(volu->volume_state, buffer + 0x18 + r_disable_drive_letter, - sizeof(volu->volume_state)); - volu->size = ldm_get_vnum(buffer + 0x3D + r_child); - volu->partition_type = buffer[0x41 + r_size]; - memcpy(volu->guid, buffer + 0x42 + r_size, sizeof(volu->guid)); - if (buffer[0x12] & VBLK_FLAG_VOLU_DRIVE) { - ldm_get_vstr(buffer + 0x52 + r_size, volu->drive_hint, - sizeof(volu->drive_hint)); - } - return true; -} - -/** - * ldm_parse_vblk - Read a raw VBLK object into a vblk structure - * @buf: Block of data being worked on - * @len: Size of the block of data - * @vb: In-memory vblk in which to return information - * - * Read a raw VBLK object into a vblk structure. This function just reads the - * information common to all VBLK types, then delegates the rest of the work to - * helper functions: ldm_parse_*. - * - * Return: 'true' @vb contains a VBLK - * 'false' @vb contents are not defined - */ -static bool ldm_parse_vblk (const u8 *buf, int len, struct vblk *vb) -{ - bool result = false; - int r_objid; - - BUG_ON (!buf || !vb); - - r_objid = ldm_relative (buf, len, 0x18, 0); - if (r_objid < 0) { - ldm_error ("VBLK header is corrupt."); - return false; - } - - vb->flags = buf[0x12]; - vb->type = buf[0x13]; - vb->obj_id = ldm_get_vnum (buf + 0x18); - ldm_get_vstr (buf+0x18+r_objid, vb->name, sizeof (vb->name)); - - switch (vb->type) { - case VBLK_CMP3: result = ldm_parse_cmp3 (buf, len, vb); break; - case VBLK_DSK3: result = ldm_parse_dsk3 (buf, len, vb); break; - case VBLK_DSK4: result = ldm_parse_dsk4 (buf, len, vb); break; - case VBLK_DGR3: result = ldm_parse_dgr3 (buf, len, vb); break; - case VBLK_DGR4: result = ldm_parse_dgr4 (buf, len, vb); break; - case VBLK_PRT3: result = ldm_parse_prt3 (buf, len, vb); break; - case VBLK_VOL5: result = ldm_parse_vol5 (buf, len, vb); break; - } - - if (result) - ldm_debug ("Parsed VBLK 0x%llx (type: 0x%02x) ok.", - (unsigned long long) vb->obj_id, vb->type); - else - ldm_error ("Failed to parse VBLK 0x%llx (type: 0x%02x).", - (unsigned long long) vb->obj_id, vb->type); - - return result; -} - - -/** - * ldm_ldmdb_add - Adds a raw VBLK entry to the ldmdb database - * @data: Raw VBLK to add to the database - * @len: Size of the raw VBLK - * @ldb: Cache of the database structures - * - * The VBLKs are sorted into categories. Partitions are also sorted by offset. - * - * N.B. This function does not check the validity of the VBLKs. - * - * Return: 'true' The VBLK was added - * 'false' An error occurred - */ -static bool ldm_ldmdb_add (u8 *data, int len, struct ldmdb *ldb) -{ - struct vblk *vb; - struct list_head *item; - - BUG_ON (!data || !ldb); - - vb = kmalloc (sizeof (*vb), GFP_KERNEL); - if (!vb) { - ldm_crit ("Out of memory."); - return false; - } - - if (!ldm_parse_vblk (data, len, vb)) { - kfree(vb); - return false; /* Already logged */ - } - - /* Put vblk into the correct list. */ - switch (vb->type) { - case VBLK_DGR3: - case VBLK_DGR4: - list_add (&vb->list, &ldb->v_dgrp); - break; - case VBLK_DSK3: - case VBLK_DSK4: - list_add (&vb->list, &ldb->v_disk); - break; - case VBLK_VOL5: - list_add (&vb->list, &ldb->v_volu); - break; - case VBLK_CMP3: - list_add (&vb->list, &ldb->v_comp); - break; - case VBLK_PRT3: - /* Sort by the partition's start sector. */ - list_for_each (item, &ldb->v_part) { - struct vblk *v = list_entry (item, struct vblk, list); - if ((v->vblk.part.disk_id == vb->vblk.part.disk_id) && - (v->vblk.part.start > vb->vblk.part.start)) { - list_add_tail (&vb->list, &v->list); - return true; - } - } - list_add_tail (&vb->list, &ldb->v_part); - break; - } - return true; -} - -/** - * ldm_frag_add - Add a VBLK fragment to a list - * @data: Raw fragment to be added to the list - * @size: Size of the raw fragment - * @frags: Linked list of VBLK fragments - * - * Fragmented VBLKs may not be consecutive in the database, so they are placed - * in a list so they can be pieced together later. - * - * Return: 'true' Success, the VBLK was added to the list - * 'false' Error, a problem occurred - */ -static bool ldm_frag_add (const u8 *data, int size, struct list_head *frags) -{ - struct frag *f; - struct list_head *item; - int rec, num, group; - - BUG_ON (!data || !frags); - - if (size < 2 * VBLK_SIZE_HEAD) { - ldm_error("Value of size is to small."); - return false; - } - - group = get_unaligned_be32(data + 0x08); - rec = get_unaligned_be16(data + 0x0C); - num = get_unaligned_be16(data + 0x0E); - if ((num < 1) || (num > 4)) { - ldm_error ("A VBLK claims to have %d parts.", num); - return false; - } - if (rec >= num) { - ldm_error("REC value (%d) exceeds NUM value (%d)", rec, num); - return false; - } - - list_for_each (item, frags) { - f = list_entry (item, struct frag, list); - if (f->group == group) - goto found; - } - - f = kmalloc (sizeof (*f) + size*num, GFP_KERNEL); - if (!f) { - ldm_crit ("Out of memory."); - return false; - } - - f->group = group; - f->num = num; - f->rec = rec; - f->map = 0xFF << num; - - list_add_tail (&f->list, frags); -found: - if (rec >= f->num) { - ldm_error("REC value (%d) exceeds NUM value (%d)", rec, f->num); - return false; - } - - if (f->map & (1 << rec)) { - ldm_error ("Duplicate VBLK, part %d.", rec); - f->map &= 0x7F; /* Mark the group as broken */ - return false; - } - - f->map |= (1 << rec); - - data += VBLK_SIZE_HEAD; - size -= VBLK_SIZE_HEAD; - - memcpy (f->data+rec*(size-VBLK_SIZE_HEAD)+VBLK_SIZE_HEAD, data, size); - - return true; -} - -/** - * ldm_frag_free - Free a linked list of VBLK fragments - * @list: Linked list of fragments - * - * Free a linked list of VBLK fragments - * - * Return: none - */ -static void ldm_frag_free (struct list_head *list) -{ - struct list_head *item, *tmp; - - BUG_ON (!list); - - list_for_each_safe (item, tmp, list) - kfree (list_entry (item, struct frag, list)); -} - -/** - * ldm_frag_commit - Validate fragmented VBLKs and add them to the database - * @frags: Linked list of VBLK fragments - * @ldb: Cache of the database structures - * - * Now that all the fragmented VBLKs have been collected, they must be added to - * the database for later use. - * - * Return: 'true' All the fragments we added successfully - * 'false' One or more of the fragments we invalid - */ -static bool ldm_frag_commit (struct list_head *frags, struct ldmdb *ldb) -{ - struct frag *f; - struct list_head *item; - - BUG_ON (!frags || !ldb); - - list_for_each (item, frags) { - f = list_entry (item, struct frag, list); - - if (f->map != 0xFF) { - ldm_error ("VBLK group %d is incomplete (0x%02x).", - f->group, f->map); - return false; - } - - if (!ldm_ldmdb_add (f->data, f->num*ldb->vm.vblk_size, ldb)) - return false; /* Already logged */ - } - return true; -} - -/** - * ldm_get_vblks - Read the on-disk database of VBLKs into memory - * @state: Partition check state including device holding the LDM Database - * @base: Offset, into @state->bdev, of the database - * @ldb: Cache of the database structures - * - * To use the information from the VBLKs, they need to be read from the disk, - * unpacked and validated. We cache them in @ldb according to their type. - * - * Return: 'true' All the VBLKs were read successfully - * 'false' An error occurred - */ -static bool ldm_get_vblks(struct parsed_partitions *state, unsigned long base, - struct ldmdb *ldb) -{ - int size, perbuf, skip, finish, s, v, recs; - u8 *data = NULL; - Sector sect; - bool result = false; - LIST_HEAD (frags); - - BUG_ON(!state || !ldb); - - size = ldb->vm.vblk_size; - perbuf = 512 / size; - skip = ldb->vm.vblk_offset >> 9; /* Bytes to sectors */ - finish = (size * ldb->vm.last_vblk_seq) >> 9; - - for (s = skip; s < finish; s++) { /* For each sector */ - data = read_part_sector(state, base + OFF_VMDB + s, §); - if (!data) { - ldm_crit ("Disk read failed."); - goto out; - } - - for (v = 0; v < perbuf; v++, data+=size) { /* For each vblk */ - if (MAGIC_VBLK != get_unaligned_be32(data)) { - ldm_error ("Expected to find a VBLK."); - goto out; - } - - recs = get_unaligned_be16(data + 0x0E); /* Number of records */ - if (recs == 1) { - if (!ldm_ldmdb_add (data, size, ldb)) - goto out; /* Already logged */ - } else if (recs > 1) { - if (!ldm_frag_add (data, size, &frags)) - goto out; /* Already logged */ - } - /* else Record is not in use, ignore it. */ - } - put_dev_sector (sect); - data = NULL; - } - - result = ldm_frag_commit (&frags, ldb); /* Failures, already logged */ -out: - if (data) - put_dev_sector (sect); - ldm_frag_free (&frags); - - return result; -} - -/** - * ldm_free_vblks - Free a linked list of vblk's - * @lh: Head of a linked list of struct vblk - * - * Free a list of vblk's and free the memory used to maintain the list. - * - * Return: none - */ -static void ldm_free_vblks (struct list_head *lh) -{ - struct list_head *item, *tmp; - - BUG_ON (!lh); - - list_for_each_safe (item, tmp, lh) - kfree (list_entry (item, struct vblk, list)); -} - - -/** - * ldm_partition - Find out whether a device is a dynamic disk and handle it - * @state: Partition check state including device holding the LDM Database - * - * This determines whether the device @bdev is a dynamic disk and if so creates - * the partitions necessary in the gendisk structure pointed to by @hd. - * - * We create a dummy device 1, which contains the LDM database, and then create - * each partition described by the LDM database in sequence as devices 2+. For - * example, if the device is hda, we would have: hda1: LDM database, hda2, hda3, - * and so on: the actual data containing partitions. - * - * Return: 1 Success, @state->bdev is a dynamic disk and we handled it - * 0 Success, @state->bdev is not a dynamic disk - * -1 An error occurred before enough information had been read - * Or @state->bdev is a dynamic disk, but it may be corrupted - */ -int ldm_partition(struct parsed_partitions *state) -{ - struct ldmdb *ldb; - unsigned long base; - int result = -1; - - BUG_ON(!state); - - /* Look for signs of a Dynamic Disk */ - if (!ldm_validate_partition_table(state)) - return 0; - - ldb = kmalloc (sizeof (*ldb), GFP_KERNEL); - if (!ldb) { - ldm_crit ("Out of memory."); - goto out; - } - - /* Parse and check privheads. */ - if (!ldm_validate_privheads(state, &ldb->ph)) - goto out; /* Already logged */ - - /* All further references are relative to base (database start). */ - base = ldb->ph.config_start; - - /* Parse and check tocs and vmdb. */ - if (!ldm_validate_tocblocks(state, base, ldb) || - !ldm_validate_vmdb(state, base, ldb)) - goto out; /* Already logged */ - - /* Initialize vblk lists in ldmdb struct */ - INIT_LIST_HEAD (&ldb->v_dgrp); - INIT_LIST_HEAD (&ldb->v_disk); - INIT_LIST_HEAD (&ldb->v_volu); - INIT_LIST_HEAD (&ldb->v_comp); - INIT_LIST_HEAD (&ldb->v_part); - - if (!ldm_get_vblks(state, base, ldb)) { - ldm_crit ("Failed to read the VBLKs from the database."); - goto cleanup; - } - - /* Finally, create the data partition devices. */ - if (ldm_create_data_partitions(state, ldb)) { - ldm_debug ("Parsed LDM database successfully."); - result = 1; - } - /* else Already logged */ - -cleanup: - ldm_free_vblks (&ldb->v_dgrp); - ldm_free_vblks (&ldb->v_disk); - ldm_free_vblks (&ldb->v_volu); - ldm_free_vblks (&ldb->v_comp); - ldm_free_vblks (&ldb->v_part); -out: - kfree (ldb); - return result; -} diff --git a/fs/partitions/ldm.h b/fs/partitions/ldm.h deleted file mode 100644 index 374242c0971a..000000000000 --- a/fs/partitions/ldm.h +++ /dev/null @@ -1,215 +0,0 @@ -/** - * ldm - Part of the Linux-NTFS project. - * - * Copyright (C) 2001,2002 Richard Russon <ldm@flatcap.org> - * Copyright (c) 2001-2007 Anton Altaparmakov - * Copyright (C) 2001,2002 Jakob Kemi <jakob.kemi@telia.com> - * - * Documentation is available at http://www.linux-ntfs.org/doku.php?id=downloads - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License as published by the Free - * Software Foundation; either version 2 of the License, or (at your option) - * any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program (in the main directory of the Linux-NTFS source - * in the file COPYING); if not, write to the Free Software Foundation, - * Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - */ - -#ifndef _FS_PT_LDM_H_ -#define _FS_PT_LDM_H_ - -#include <linux/types.h> -#include <linux/list.h> -#include <linux/genhd.h> -#include <linux/fs.h> -#include <asm/unaligned.h> -#include <asm/byteorder.h> - -struct parsed_partitions; - -/* Magic numbers in CPU format. */ -#define MAGIC_VMDB 0x564D4442 /* VMDB */ -#define MAGIC_VBLK 0x56424C4B /* VBLK */ -#define MAGIC_PRIVHEAD 0x5052495648454144ULL /* PRIVHEAD */ -#define MAGIC_TOCBLOCK 0x544F43424C4F434BULL /* TOCBLOCK */ - -/* The defined vblk types. */ -#define VBLK_VOL5 0x51 /* Volume, version 5 */ -#define VBLK_CMP3 0x32 /* Component, version 3 */ -#define VBLK_PRT3 0x33 /* Partition, version 3 */ -#define VBLK_DSK3 0x34 /* Disk, version 3 */ -#define VBLK_DSK4 0x44 /* Disk, version 4 */ -#define VBLK_DGR3 0x35 /* Disk Group, version 3 */ -#define VBLK_DGR4 0x45 /* Disk Group, version 4 */ - -/* vblk flags indicating extra information will be present */ -#define VBLK_FLAG_COMP_STRIPE 0x10 -#define VBLK_FLAG_PART_INDEX 0x08 -#define VBLK_FLAG_DGR3_IDS 0x08 -#define VBLK_FLAG_DGR4_IDS 0x08 -#define VBLK_FLAG_VOLU_ID1 0x08 -#define VBLK_FLAG_VOLU_ID2 0x20 -#define VBLK_FLAG_VOLU_SIZE 0x80 -#define VBLK_FLAG_VOLU_DRIVE 0x02 - -/* size of a vblk's static parts */ -#define VBLK_SIZE_HEAD 16 -#define VBLK_SIZE_CMP3 22 /* Name and version */ -#define VBLK_SIZE_DGR3 12 -#define VBLK_SIZE_DGR4 44 -#define VBLK_SIZE_DSK3 12 -#define VBLK_SIZE_DSK4 45 -#define VBLK_SIZE_PRT3 28 -#define VBLK_SIZE_VOL5 58 - -/* component types */ -#define COMP_STRIPE 0x01 /* Stripe-set */ -#define COMP_BASIC 0x02 /* Basic disk */ -#define COMP_RAID 0x03 /* Raid-set */ - -/* Other constants. */ -#define LDM_DB_SIZE 2048 /* Size in sectors (= 1MiB). */ - -#define OFF_PRIV1 6 /* Offset of the first privhead - relative to the start of the - device in sectors */ - -/* Offsets to structures within the LDM Database in sectors. */ -#define OFF_PRIV2 1856 /* Backup private headers. */ -#define OFF_PRIV3 2047 - -#define OFF_TOCB1 1 /* Tables of contents. */ -#define OFF_TOCB2 2 -#define OFF_TOCB3 2045 -#define OFF_TOCB4 2046 - -#define OFF_VMDB 17 /* List of partitions. */ - -#define LDM_PARTITION 0x42 /* Formerly SFS (Landis). */ - -#define TOC_BITMAP1 "config" /* Names of the two defined */ -#define TOC_BITMAP2 "log" /* bitmaps in the TOCBLOCK. */ - -/* Borrowed from msdos.c */ -#define SYS_IND(p) (get_unaligned(&(p)->sys_ind)) - -struct frag { /* VBLK Fragment handling */ - struct list_head list; - u32 group; - u8 num; /* Total number of records */ - u8 rec; /* This is record number n */ - u8 map; /* Which portions are in use */ - u8 data[0]; -}; - -/* In memory LDM database structures. */ - -#define GUID_SIZE 16 - -struct privhead { /* Offsets and sizes are in sectors. */ - u16 ver_major; - u16 ver_minor; - u64 logical_disk_start; - u64 logical_disk_size; - u64 config_start; - u64 config_size; - u8 disk_id[GUID_SIZE]; -}; - -struct tocblock { /* We have exactly two bitmaps. */ - u8 bitmap1_name[16]; - u64 bitmap1_start; - u64 bitmap1_size; - u8 bitmap2_name[16]; - u64 bitmap2_start; - u64 bitmap2_size; -}; - -struct vmdb { /* VMDB: The database header */ - u16 ver_major; - u16 ver_minor; - u32 vblk_size; - u32 vblk_offset; - u32 last_vblk_seq; -}; - -struct vblk_comp { /* VBLK Component */ - u8 state[16]; - u64 parent_id; - u8 type; - u8 children; - u16 chunksize; -}; - -struct vblk_dgrp { /* VBLK Disk Group */ - u8 disk_id[64]; -}; - -struct vblk_disk { /* VBLK Disk */ - u8 disk_id[GUID_SIZE]; - u8 alt_name[128]; -}; - -struct vblk_part { /* VBLK Partition */ - u64 start; - u64 size; /* start, size and vol_off in sectors */ - u64 volume_offset; - u64 parent_id; - u64 disk_id; - u8 partnum; -}; - -struct vblk_volu { /* VBLK Volume */ - u8 volume_type[16]; - u8 volume_state[16]; - u8 guid[16]; - u8 drive_hint[4]; - u64 size; - u8 partition_type; -}; - -struct vblk_head { /* VBLK standard header */ - u32 group; - u16 rec; - u16 nrec; -}; - -struct vblk { /* Generalised VBLK */ - u8 name[64]; - u64 obj_id; - u32 sequence; - u8 flags; - u8 type; - union { - struct vblk_comp comp; - struct vblk_dgrp dgrp; - struct vblk_disk disk; - struct vblk_part part; - struct vblk_volu volu; - } vblk; - struct list_head list; -}; - -struct ldmdb { /* Cache of the database */ - struct privhead ph; - struct tocblock toc; - struct vmdb vm; - struct list_head v_dgrp; - struct list_head v_disk; - struct list_head v_volu; - struct list_head v_comp; - struct list_head v_part; -}; - -int ldm_partition(struct parsed_partitions *state); - -#endif /* _FS_PT_LDM_H_ */ - diff --git a/fs/partitions/mac.c b/fs/partitions/mac.c deleted file mode 100644 index 11f688bd76c5..000000000000 --- a/fs/partitions/mac.c +++ /dev/null @@ -1,134 +0,0 @@ -/* - * fs/partitions/mac.c - * - * Code extracted from drivers/block/genhd.c - * Copyright (C) 1991-1998 Linus Torvalds - * Re-organised Feb 1998 Russell King - */ - -#include <linux/ctype.h> -#include "check.h" -#include "mac.h" - -#ifdef CONFIG_PPC_PMAC -#include <asm/machdep.h> -extern void note_bootable_part(dev_t dev, int part, int goodness); -#endif - -/* - * Code to understand MacOS partition tables. - */ - -static inline void mac_fix_string(char *stg, int len) -{ - int i; - - for (i = len - 1; i >= 0 && stg[i] == ' '; i--) - stg[i] = 0; -} - -int mac_partition(struct parsed_partitions *state) -{ - Sector sect; - unsigned char *data; - int slot, blocks_in_map; - unsigned secsize; -#ifdef CONFIG_PPC_PMAC - int found_root = 0; - int found_root_goodness = 0; -#endif - struct mac_partition *part; - struct mac_driver_desc *md; - - /* Get 0th block and look at the first partition map entry. */ - md = read_part_sector(state, 0, §); - if (!md) - return -1; - if (be16_to_cpu(md->signature) != MAC_DRIVER_MAGIC) { - put_dev_sector(sect); - return 0; - } - secsize = be16_to_cpu(md->block_size); - put_dev_sector(sect); - data = read_part_sector(state, secsize/512, §); - if (!data) - return -1; - part = (struct mac_partition *) (data + secsize%512); - if (be16_to_cpu(part->signature) != MAC_PARTITION_MAGIC) { - put_dev_sector(sect); - return 0; /* not a MacOS disk */ - } - blocks_in_map = be32_to_cpu(part->map_count); - if (blocks_in_map < 0 || blocks_in_map >= DISK_MAX_PARTS) { - put_dev_sector(sect); - return 0; - } - strlcat(state->pp_buf, " [mac]", PAGE_SIZE); - for (slot = 1; slot <= blocks_in_map; ++slot) { - int pos = slot * secsize; - put_dev_sector(sect); - data = read_part_sector(state, pos/512, §); - if (!data) - return -1; - part = (struct mac_partition *) (data + pos%512); - if (be16_to_cpu(part->signature) != MAC_PARTITION_MAGIC) - break; - put_partition(state, slot, - be32_to_cpu(part->start_block) * (secsize/512), - be32_to_cpu(part->block_count) * (secsize/512)); - - if (!strnicmp(part->type, "Linux_RAID", 10)) - state->parts[slot].flags = ADDPART_FLAG_RAID; -#ifdef CONFIG_PPC_PMAC - /* - * If this is the first bootable partition, tell the - * setup code, in case it wants to make this the root. - */ - if (machine_is(powermac)) { - int goodness = 0; - - mac_fix_string(part->processor, 16); - mac_fix_string(part->name, 32); - mac_fix_string(part->type, 32); - - if ((be32_to_cpu(part->status) & MAC_STATUS_BOOTABLE) - && strcasecmp(part->processor, "powerpc") == 0) - goodness++; - - if (strcasecmp(part->type, "Apple_UNIX_SVR2") == 0 - || (strnicmp(part->type, "Linux", 5) == 0 - && strcasecmp(part->type, "Linux_swap") != 0)) { - int i, l; - - goodness++; - l = strlen(part->name); - if (strcmp(part->name, "/") == 0) - goodness++; - for (i = 0; i <= l - 4; ++i) { - if (strnicmp(part->name + i, "root", - 4) == 0) { - goodness += 2; - break; - } - } - if (strnicmp(part->name, "swap", 4) == 0) - goodness--; - } - - if (goodness > found_root_goodness) { - found_root = slot; - found_root_goodness = goodness; - } - } -#endif /* CONFIG_PPC_PMAC */ - } -#ifdef CONFIG_PPC_PMAC - if (found_root_goodness) - note_bootable_part(state->bdev->bd_dev, found_root, - found_root_goodness); -#endif - - put_dev_sector(sect); - strlcat(state->pp_buf, "\n", PAGE_SIZE); - return 1; -} diff --git a/fs/partitions/mac.h b/fs/partitions/mac.h deleted file mode 100644 index 3c7d98436380..000000000000 --- a/fs/partitions/mac.h +++ /dev/null @@ -1,44 +0,0 @@ -/* - * fs/partitions/mac.h - */ - -#define MAC_PARTITION_MAGIC 0x504d - -/* type field value for A/UX or other Unix partitions */ -#define APPLE_AUX_TYPE "Apple_UNIX_SVR2" - -struct mac_partition { - __be16 signature; /* expected to be MAC_PARTITION_MAGIC */ - __be16 res1; - __be32 map_count; /* # blocks in partition map */ - __be32 start_block; /* absolute starting block # of partition */ - __be32 block_count; /* number of blocks in partition */ - char name[32]; /* partition name */ - char type[32]; /* string type description */ - __be32 data_start; /* rel block # of first data block */ - __be32 data_count; /* number of data blocks */ - __be32 status; /* partition status bits */ - __be32 boot_start; - __be32 boot_size; - __be32 boot_load; - __be32 boot_load2; - __be32 boot_entry; - __be32 boot_entry2; - __be32 boot_cksum; - char processor[16]; /* identifies ISA of boot */ - /* there is more stuff after this that we don't need */ -}; - -#define MAC_STATUS_BOOTABLE 8 /* partition is bootable */ - -#define MAC_DRIVER_MAGIC 0x4552 - -/* Driver descriptor structure, in block 0 */ -struct mac_driver_desc { - __be16 signature; /* expected to be MAC_DRIVER_MAGIC */ - __be16 block_size; - __be32 block_count; - /* ... more stuff */ -}; - -int mac_partition(struct parsed_partitions *state); diff --git a/fs/partitions/msdos.c b/fs/partitions/msdos.c deleted file mode 100644 index 5f79a6677c69..000000000000 --- a/fs/partitions/msdos.c +++ /dev/null @@ -1,552 +0,0 @@ -/* - * fs/partitions/msdos.c - * - * Code extracted from drivers/block/genhd.c - * Copyright (C) 1991-1998 Linus Torvalds - * - * Thanks to Branko Lankester, lankeste@fwi.uva.nl, who found a bug - * in the early extended-partition checks and added DM partitions - * - * Support for DiskManager v6.0x added by Mark Lord, - * with information provided by OnTrack. This now works for linux fdisk - * and LILO, as well as loadlin and bootln. Note that disks other than - * /dev/hda *must* have a "DOS" type 0x51 partition in the first slot (hda1). - * - * More flexible handling of extended partitions - aeb, 950831 - * - * Check partition table on IDE disks for common CHS translations - * - * Re-organised Feb 1998 Russell King - */ -#include <linux/msdos_fs.h> - -#include "check.h" -#include "msdos.h" -#include "efi.h" - -/* - * Many architectures don't like unaligned accesses, while - * the nr_sects and start_sect partition table entries are - * at a 2 (mod 4) address. - */ -#include <asm/unaligned.h> - -#define SYS_IND(p) get_unaligned(&p->sys_ind) - -static inline sector_t nr_sects(struct partition *p) -{ - return (sector_t)get_unaligned_le32(&p->nr_sects); -} - -static inline sector_t start_sect(struct partition *p) -{ - return (sector_t)get_unaligned_le32(&p->start_sect); -} - -static inline int is_extended_partition(struct partition *p) -{ - return (SYS_IND(p) == DOS_EXTENDED_PARTITION || - SYS_IND(p) == WIN98_EXTENDED_PARTITION || - SYS_IND(p) == LINUX_EXTENDED_PARTITION); -} - -#define MSDOS_LABEL_MAGIC1 0x55 -#define MSDOS_LABEL_MAGIC2 0xAA - -static inline int -msdos_magic_present(unsigned char *p) -{ - return (p[0] == MSDOS_LABEL_MAGIC1 && p[1] == MSDOS_LABEL_MAGIC2); -} - -/* Value is EBCDIC 'IBMA' */ -#define AIX_LABEL_MAGIC1 0xC9 -#define AIX_LABEL_MAGIC2 0xC2 -#define AIX_LABEL_MAGIC3 0xD4 -#define AIX_LABEL_MAGIC4 0xC1 -static int aix_magic_present(struct parsed_partitions *state, unsigned char *p) -{ - struct partition *pt = (struct partition *) (p + 0x1be); - Sector sect; - unsigned char *d; - int slot, ret = 0; - - if (!(p[0] == AIX_LABEL_MAGIC1 && - p[1] == AIX_LABEL_MAGIC2 && - p[2] == AIX_LABEL_MAGIC3 && - p[3] == AIX_LABEL_MAGIC4)) - return 0; - /* Assume the partition table is valid if Linux partitions exists */ - for (slot = 1; slot <= 4; slot++, pt++) { - if (pt->sys_ind == LINUX_SWAP_PARTITION || - pt->sys_ind == LINUX_RAID_PARTITION || - pt->sys_ind == LINUX_DATA_PARTITION || - pt->sys_ind == LINUX_LVM_PARTITION || - is_extended_partition(pt)) - return 0; - } - d = read_part_sector(state, 7, §); - if (d) { - if (d[0] == '_' && d[1] == 'L' && d[2] == 'V' && d[3] == 'M') - ret = 1; - put_dev_sector(sect); - }; - return ret; -} - -/* - * Create devices for each logical partition in an extended partition. - * The logical partitions form a linked list, with each entry being - * a partition table with two entries. The first entry - * is the real data partition (with a start relative to the partition - * table start). The second is a pointer to the next logical partition - * (with a start relative to the entire extended partition). - * We do not create a Linux partition for the partition tables, but - * only for the actual data partitions. - */ - -static void parse_extended(struct parsed_partitions *state, - sector_t first_sector, sector_t first_size) -{ - struct partition *p; - Sector sect; - unsigned char *data; - sector_t this_sector, this_size; - sector_t sector_size = bdev_logical_block_size(state->bdev) / 512; - int loopct = 0; /* number of links followed - without finding a data partition */ - int i; - - this_sector = first_sector; - this_size = first_size; - - while (1) { - if (++loopct > 100) - return; - if (state->next == state->limit) - return; - data = read_part_sector(state, this_sector, §); - if (!data) - return; - - if (!msdos_magic_present(data + 510)) - goto done; - - p = (struct partition *) (data + 0x1be); - - /* - * Usually, the first entry is the real data partition, - * the 2nd entry is the next extended partition, or empty, - * and the 3rd and 4th entries are unused. - * However, DRDOS sometimes has the extended partition as - * the first entry (when the data partition is empty), - * and OS/2 seems to use all four entries. - */ - - /* - * First process the data partition(s) - */ - for (i=0; i<4; i++, p++) { - sector_t offs, size, next; - if (!nr_sects(p) || is_extended_partition(p)) - continue; - - /* Check the 3rd and 4th entries - - these sometimes contain random garbage */ - offs = start_sect(p)*sector_size; - size = nr_sects(p)*sector_size; - next = this_sector + offs; - if (i >= 2) { - if (offs + size > this_size) - continue; - if (next < first_sector) - continue; - if (next + size > first_sector + first_size) - continue; - } - - put_partition(state, state->next, next, size); - if (SYS_IND(p) == LINUX_RAID_PARTITION) - state->parts[state->next].flags = ADDPART_FLAG_RAID; - loopct = 0; - if (++state->next == state->limit) - goto done; - } - /* - * Next, process the (first) extended partition, if present. - * (So far, there seems to be no reason to make - * parse_extended() recursive and allow a tree - * of extended partitions.) - * It should be a link to the next logical partition. - */ - p -= 4; - for (i=0; i<4; i++, p++) - if (nr_sects(p) && is_extended_partition(p)) - break; - if (i == 4) - goto done; /* nothing left to do */ - - this_sector = first_sector + start_sect(p) * sector_size; - this_size = nr_sects(p) * sector_size; - put_dev_sector(sect); - } -done: - put_dev_sector(sect); -} - -/* james@bpgc.com: Solaris has a nasty indicator: 0x82 which also - indicates linux swap. Be careful before believing this is Solaris. */ - -static void parse_solaris_x86(struct parsed_partitions *state, - sector_t offset, sector_t size, int origin) -{ -#ifdef CONFIG_SOLARIS_X86_PARTITION - Sector sect; - struct solaris_x86_vtoc *v; - int i; - short max_nparts; - - v = read_part_sector(state, offset + 1, §); - if (!v) - return; - if (le32_to_cpu(v->v_sanity) != SOLARIS_X86_VTOC_SANE) { - put_dev_sector(sect); - return; - } - { - char tmp[1 + BDEVNAME_SIZE + 10 + 11 + 1]; - - snprintf(tmp, sizeof(tmp), " %s%d: <solaris:", state->name, origin); - strlcat(state->pp_buf, tmp, PAGE_SIZE); - } - if (le32_to_cpu(v->v_version) != 1) { - char tmp[64]; - - snprintf(tmp, sizeof(tmp), " cannot handle version %d vtoc>\n", - le32_to_cpu(v->v_version)); - strlcat(state->pp_buf, tmp, PAGE_SIZE); - put_dev_sector(sect); - return; - } - /* Ensure we can handle previous case of VTOC with 8 entries gracefully */ - max_nparts = le16_to_cpu (v->v_nparts) > 8 ? SOLARIS_X86_NUMSLICE : 8; - for (i=0; i<max_nparts && state->next<state->limit; i++) { - struct solaris_x86_slice *s = &v->v_slice[i]; - char tmp[3 + 10 + 1 + 1]; - - if (s->s_size == 0) - continue; - snprintf(tmp, sizeof(tmp), " [s%d]", i); - strlcat(state->pp_buf, tmp, PAGE_SIZE); - /* solaris partitions are relative to current MS-DOS - * one; must add the offset of the current partition */ - put_partition(state, state->next++, - le32_to_cpu(s->s_start)+offset, - le32_to_cpu(s->s_size)); - } - put_dev_sector(sect); - strlcat(state->pp_buf, " >\n", PAGE_SIZE); -#endif -} - -#if defined(CONFIG_BSD_DISKLABEL) -/* - * Create devices for BSD partitions listed in a disklabel, under a - * dos-like partition. See parse_extended() for more information. - */ -static void parse_bsd(struct parsed_partitions *state, - sector_t offset, sector_t size, int origin, char *flavour, - int max_partitions) -{ - Sector sect; - struct bsd_disklabel *l; - struct bsd_partition *p; - char tmp[64]; - - l = read_part_sector(state, offset + 1, §); - if (!l) - return; - if (le32_to_cpu(l->d_magic) != BSD_DISKMAGIC) { - put_dev_sector(sect); - return; - } - - snprintf(tmp, sizeof(tmp), " %s%d: <%s:", state->name, origin, flavour); - strlcat(state->pp_buf, tmp, PAGE_SIZE); - - if (le16_to_cpu(l->d_npartitions) < max_partitions) - max_partitions = le16_to_cpu(l->d_npartitions); - for (p = l->d_partitions; p - l->d_partitions < max_partitions; p++) { - sector_t bsd_start, bsd_size; - - if (state->next == state->limit) - break; - if (p->p_fstype == BSD_FS_UNUSED) - continue; - bsd_start = le32_to_cpu(p->p_offset); - bsd_size = le32_to_cpu(p->p_size); - if (offset == bsd_start && size == bsd_size) - /* full parent partition, we have it already */ - continue; - if (offset > bsd_start || offset+size < bsd_start+bsd_size) { - strlcat(state->pp_buf, "bad subpartition - ignored\n", PAGE_SIZE); - continue; - } - put_partition(state, state->next++, bsd_start, bsd_size); - } - put_dev_sector(sect); - if (le16_to_cpu(l->d_npartitions) > max_partitions) { - snprintf(tmp, sizeof(tmp), " (ignored %d more)", - le16_to_cpu(l->d_npartitions) - max_partitions); - strlcat(state->pp_buf, tmp, PAGE_SIZE); - } - strlcat(state->pp_buf, " >\n", PAGE_SIZE); -} -#endif - -static void parse_freebsd(struct parsed_partitions *state, - sector_t offset, sector_t size, int origin) -{ -#ifdef CONFIG_BSD_DISKLABEL - parse_bsd(state, offset, size, origin, "bsd", BSD_MAXPARTITIONS); -#endif -} - -static void parse_netbsd(struct parsed_partitions *state, - sector_t offset, sector_t size, int origin) -{ -#ifdef CONFIG_BSD_DISKLABEL - parse_bsd(state, offset, size, origin, "netbsd", BSD_MAXPARTITIONS); -#endif -} - -static void parse_openbsd(struct parsed_partitions *state, - sector_t offset, sector_t size, int origin) -{ -#ifdef CONFIG_BSD_DISKLABEL - parse_bsd(state, offset, size, origin, "openbsd", - OPENBSD_MAXPARTITIONS); -#endif -} - -/* - * Create devices for Unixware partitions listed in a disklabel, under a - * dos-like partition. See parse_extended() for more information. - */ -static void parse_unixware(struct parsed_partitions *state, - sector_t offset, sector_t size, int origin) -{ -#ifdef CONFIG_UNIXWARE_DISKLABEL - Sector sect; - struct unixware_disklabel *l; - struct unixware_slice *p; - - l = read_part_sector(state, offset + 29, §); - if (!l) - return; - if (le32_to_cpu(l->d_magic) != UNIXWARE_DISKMAGIC || - le32_to_cpu(l->vtoc.v_magic) != UNIXWARE_DISKMAGIC2) { - put_dev_sector(sect); - return; - } - { - char tmp[1 + BDEVNAME_SIZE + 10 + 12 + 1]; - - snprintf(tmp, sizeof(tmp), " %s%d: <unixware:", state->name, origin); - strlcat(state->pp_buf, tmp, PAGE_SIZE); - } - p = &l->vtoc.v_slice[1]; - /* I omit the 0th slice as it is the same as whole disk. */ - while (p - &l->vtoc.v_slice[0] < UNIXWARE_NUMSLICE) { - if (state->next == state->limit) - break; - - if (p->s_label != UNIXWARE_FS_UNUSED) - put_partition(state, state->next++, - le32_to_cpu(p->start_sect), - le32_to_cpu(p->nr_sects)); - p++; - } - put_dev_sector(sect); - strlcat(state->pp_buf, " >\n", PAGE_SIZE); -#endif -} - -/* - * Minix 2.0.0/2.0.2 subpartition support. - * Anand Krishnamurthy <anandk@wiproge.med.ge.com> - * Rajeev V. Pillai <rajeevvp@yahoo.com> - */ -static void parse_minix(struct parsed_partitions *state, - sector_t offset, sector_t size, int origin) -{ -#ifdef CONFIG_MINIX_SUBPARTITION - Sector sect; - unsigned char *data; - struct partition *p; - int i; - - data = read_part_sector(state, offset, §); - if (!data) - return; - - p = (struct partition *)(data + 0x1be); - - /* The first sector of a Minix partition can have either - * a secondary MBR describing its subpartitions, or - * the normal boot sector. */ - if (msdos_magic_present (data + 510) && - SYS_IND(p) == MINIX_PARTITION) { /* subpartition table present */ - char tmp[1 + BDEVNAME_SIZE + 10 + 9 + 1]; - - snprintf(tmp, sizeof(tmp), " %s%d: <minix:", state->name, origin); - strlcat(state->pp_buf, tmp, PAGE_SIZE); - for (i = 0; i < MINIX_NR_SUBPARTITIONS; i++, p++) { - if (state->next == state->limit) - break; - /* add each partition in use */ - if (SYS_IND(p) == MINIX_PARTITION) - put_partition(state, state->next++, - start_sect(p), nr_sects(p)); - } - strlcat(state->pp_buf, " >\n", PAGE_SIZE); - } - put_dev_sector(sect); -#endif /* CONFIG_MINIX_SUBPARTITION */ -} - -static struct { - unsigned char id; - void (*parse)(struct parsed_partitions *, sector_t, sector_t, int); -} subtypes[] = { - {FREEBSD_PARTITION, parse_freebsd}, - {NETBSD_PARTITION, parse_netbsd}, - {OPENBSD_PARTITION, parse_openbsd}, - {MINIX_PARTITION, parse_minix}, - {UNIXWARE_PARTITION, parse_unixware}, - {SOLARIS_X86_PARTITION, parse_solaris_x86}, - {NEW_SOLARIS_X86_PARTITION, parse_solaris_x86}, - {0, NULL}, -}; - -int msdos_partition(struct parsed_partitions *state) -{ - sector_t sector_size = bdev_logical_block_size(state->bdev) / 512; - Sector sect; - unsigned char *data; - struct partition *p; - struct fat_boot_sector *fb; - int slot; - - data = read_part_sector(state, 0, §); - if (!data) - return -1; - if (!msdos_magic_present(data + 510)) { - put_dev_sector(sect); - return 0; - } - - if (aix_magic_present(state, data)) { - put_dev_sector(sect); - strlcat(state->pp_buf, " [AIX]", PAGE_SIZE); - return 0; - } - - /* - * Now that the 55aa signature is present, this is probably - * either the boot sector of a FAT filesystem or a DOS-type - * partition table. Reject this in case the boot indicator - * is not 0 or 0x80. - */ - p = (struct partition *) (data + 0x1be); - for (slot = 1; slot <= 4; slot++, p++) { - if (p->boot_ind != 0 && p->boot_ind != 0x80) { - /* - * Even without a valid boot inidicator value - * its still possible this is valid FAT filesystem - * without a partition table. - */ - fb = (struct fat_boot_sector *) data; - if (slot == 1 && fb->reserved && fb->fats - && fat_valid_media(fb->media)) { - strlcat(state->pp_buf, "\n", PAGE_SIZE); - put_dev_sector(sect); - return 1; - } else { - put_dev_sector(sect); - return 0; - } - } - } - -#ifdef CONFIG_EFI_PARTITION - p = (struct partition *) (data + 0x1be); - for (slot = 1 ; slot <= 4 ; slot++, p++) { - /* If this is an EFI GPT disk, msdos should ignore it. */ - if (SYS_IND(p) == EFI_PMBR_OSTYPE_EFI_GPT) { - put_dev_sector(sect); - return 0; - } - } -#endif - p = (struct partition *) (data + 0x1be); - - /* - * Look for partitions in two passes: - * First find the primary and DOS-type extended partitions. - * On the second pass look inside *BSD, Unixware and Solaris partitions. - */ - - state->next = 5; - for (slot = 1 ; slot <= 4 ; slot++, p++) { - sector_t start = start_sect(p)*sector_size; - sector_t size = nr_sects(p)*sector_size; - if (!size) - continue; - if (is_extended_partition(p)) { - /* - * prevent someone doing mkfs or mkswap on an - * extended partition, but leave room for LILO - * FIXME: this uses one logical sector for > 512b - * sector, although it may not be enough/proper. - */ - sector_t n = 2; - n = min(size, max(sector_size, n)); - put_partition(state, slot, start, n); - - strlcat(state->pp_buf, " <", PAGE_SIZE); - parse_extended(state, start, size); - strlcat(state->pp_buf, " >", PAGE_SIZE); - continue; - } - put_partition(state, slot, start, size); - if (SYS_IND(p) == LINUX_RAID_PARTITION) - state->parts[slot].flags = ADDPART_FLAG_RAID; - if (SYS_IND(p) == DM6_PARTITION) - strlcat(state->pp_buf, "[DM]", PAGE_SIZE); - if (SYS_IND(p) == EZD_PARTITION) - strlcat(state->pp_buf, "[EZD]", PAGE_SIZE); - } - - strlcat(state->pp_buf, "\n", PAGE_SIZE); - - /* second pass - output for each on a separate line */ - p = (struct partition *) (0x1be + data); - for (slot = 1 ; slot <= 4 ; slot++, p++) { - unsigned char id = SYS_IND(p); - int n; - - if (!nr_sects(p)) - continue; - - for (n = 0; subtypes[n].parse && id != subtypes[n].id; n++) - ; - - if (!subtypes[n].parse) - continue; - subtypes[n].parse(state, start_sect(p) * sector_size, - nr_sects(p) * sector_size, slot); - } - put_dev_sector(sect); - return 1; -} diff --git a/fs/partitions/msdos.h b/fs/partitions/msdos.h deleted file mode 100644 index 38c781c490b3..000000000000 --- a/fs/partitions/msdos.h +++ /dev/null @@ -1,8 +0,0 @@ -/* - * fs/partitions/msdos.h - */ - -#define MSDOS_LABEL_MAGIC 0xAA55 - -int msdos_partition(struct parsed_partitions *state); - diff --git a/fs/partitions/osf.c b/fs/partitions/osf.c deleted file mode 100644 index 764b86a01965..000000000000 --- a/fs/partitions/osf.c +++ /dev/null @@ -1,86 +0,0 @@ -/* - * fs/partitions/osf.c - * - * Code extracted from drivers/block/genhd.c - * - * Copyright (C) 1991-1998 Linus Torvalds - * Re-organised Feb 1998 Russell King - */ - -#include "check.h" -#include "osf.h" - -#define MAX_OSF_PARTITIONS 18 - -int osf_partition(struct parsed_partitions *state) -{ - int i; - int slot = 1; - unsigned int npartitions; - Sector sect; - unsigned char *data; - struct disklabel { - __le32 d_magic; - __le16 d_type,d_subtype; - u8 d_typename[16]; - u8 d_packname[16]; - __le32 d_secsize; - __le32 d_nsectors; - __le32 d_ntracks; - __le32 d_ncylinders; - __le32 d_secpercyl; - __le32 d_secprtunit; - __le16 d_sparespertrack; - __le16 d_sparespercyl; - __le32 d_acylinders; - __le16 d_rpm, d_interleave, d_trackskew, d_cylskew; - __le32 d_headswitch, d_trkseek, d_flags; - __le32 d_drivedata[5]; - __le32 d_spare[5]; - __le32 d_magic2; - __le16 d_checksum; - __le16 d_npartitions; - __le32 d_bbsize, d_sbsize; - struct d_partition { - __le32 p_size; - __le32 p_offset; - __le32 p_fsize; - u8 p_fstype; - u8 p_frag; - __le16 p_cpg; - } d_partitions[MAX_OSF_PARTITIONS]; - } * label; - struct d_partition * partition; - - data = read_part_sector(state, 0, §); - if (!data) - return -1; - - label = (struct disklabel *) (data+64); - partition = label->d_partitions; - if (le32_to_cpu(label->d_magic) != DISKLABELMAGIC) { - put_dev_sector(sect); - return 0; - } - if (le32_to_cpu(label->d_magic2) != DISKLABELMAGIC) { - put_dev_sector(sect); - return 0; - } - npartitions = le16_to_cpu(label->d_npartitions); - if (npartitions > MAX_OSF_PARTITIONS) { - put_dev_sector(sect); - return 0; - } - for (i = 0 ; i < npartitions; i++, partition++) { - if (slot == state->limit) - break; - if (le32_to_cpu(partition->p_size)) - put_partition(state, slot, - le32_to_cpu(partition->p_offset), - le32_to_cpu(partition->p_size)); - slot++; - } - strlcat(state->pp_buf, "\n", PAGE_SIZE); - put_dev_sector(sect); - return 1; -} diff --git a/fs/partitions/osf.h b/fs/partitions/osf.h deleted file mode 100644 index 20ed2315ec16..000000000000 --- a/fs/partitions/osf.h +++ /dev/null @@ -1,7 +0,0 @@ -/* - * fs/partitions/osf.h - */ - -#define DISKLABELMAGIC (0x82564557UL) - -int osf_partition(struct parsed_partitions *state); diff --git a/fs/partitions/sgi.c b/fs/partitions/sgi.c deleted file mode 100644 index ea8a86dceaf4..000000000000 --- a/fs/partitions/sgi.c +++ /dev/null @@ -1,82 +0,0 @@ -/* - * fs/partitions/sgi.c - * - * Code extracted from drivers/block/genhd.c - */ - -#include "check.h" -#include "sgi.h" - -struct sgi_disklabel { - __be32 magic_mushroom; /* Big fat spliff... */ - __be16 root_part_num; /* Root partition number */ - __be16 swap_part_num; /* Swap partition number */ - s8 boot_file[16]; /* Name of boot file for ARCS */ - u8 _unused0[48]; /* Device parameter useless crapola.. */ - struct sgi_volume { - s8 name[8]; /* Name of volume */ - __be32 block_num; /* Logical block number */ - __be32 num_bytes; /* How big, in bytes */ - } volume[15]; - struct sgi_partition { - __be32 num_blocks; /* Size in logical blocks */ - __be32 first_block; /* First logical block */ - __be32 type; /* Type of this partition */ - } partitions[16]; - __be32 csum; /* Disk label checksum */ - __be32 _unused1; /* Padding */ -}; - -int sgi_partition(struct parsed_partitions *state) -{ - int i, csum; - __be32 magic; - int slot = 1; - unsigned int start, blocks; - __be32 *ui, cs; - Sector sect; - struct sgi_disklabel *label; - struct sgi_partition *p; - char b[BDEVNAME_SIZE]; - - label = read_part_sector(state, 0, §); - if (!label) - return -1; - p = &label->partitions[0]; - magic = label->magic_mushroom; - if(be32_to_cpu(magic) != SGI_LABEL_MAGIC) { - /*printk("Dev %s SGI disklabel: bad magic %08x\n", - bdevname(bdev, b), be32_to_cpu(magic));*/ - put_dev_sector(sect); - return 0; - } - ui = ((__be32 *) (label + 1)) - 1; - for(csum = 0; ui >= ((__be32 *) label);) { - cs = *ui--; - csum += be32_to_cpu(cs); - } - if(csum) { - printk(KERN_WARNING "Dev %s SGI disklabel: csum bad, label corrupted\n", - bdevname(state->bdev, b)); - put_dev_sector(sect); - return 0; - } - /* All SGI disk labels have 16 partitions, disks under Linux only - * have 15 minor's. Luckily there are always a few zero length - * partitions which we don't care about so we never overflow the - * current_minor. - */ - for(i = 0; i < 16; i++, p++) { - blocks = be32_to_cpu(p->num_blocks); - start = be32_to_cpu(p->first_block); - if (blocks) { - put_partition(state, slot, start, blocks); - if (be32_to_cpu(p->type) == LINUX_RAID_PARTITION) - state->parts[slot].flags = ADDPART_FLAG_RAID; - } - slot++; - } - strlcat(state->pp_buf, "\n", PAGE_SIZE); - put_dev_sector(sect); - return 1; -} diff --git a/fs/partitions/sgi.h b/fs/partitions/sgi.h deleted file mode 100644 index b9553ebdd5a9..000000000000 --- a/fs/partitions/sgi.h +++ /dev/null @@ -1,8 +0,0 @@ -/* - * fs/partitions/sgi.h - */ - -extern int sgi_partition(struct parsed_partitions *state); - -#define SGI_LABEL_MAGIC 0x0be5a941 - diff --git a/fs/partitions/sun.c b/fs/partitions/sun.c deleted file mode 100644 index b5b6fcfb3d36..000000000000 --- a/fs/partitions/sun.c +++ /dev/null @@ -1,122 +0,0 @@ -/* - * fs/partitions/sun.c - * - * Code extracted from drivers/block/genhd.c - * - * Copyright (C) 1991-1998 Linus Torvalds - * Re-organised Feb 1998 Russell King - */ - -#include "check.h" -#include "sun.h" - -int sun_partition(struct parsed_partitions *state) -{ - int i; - __be16 csum; - int slot = 1; - __be16 *ush; - Sector sect; - struct sun_disklabel { - unsigned char info[128]; /* Informative text string */ - struct sun_vtoc { - __be32 version; /* Layout version */ - char volume[8]; /* Volume name */ - __be16 nparts; /* Number of partitions */ - struct sun_info { /* Partition hdrs, sec 2 */ - __be16 id; - __be16 flags; - } infos[8]; - __be16 padding; /* Alignment padding */ - __be32 bootinfo[3]; /* Info needed by mboot */ - __be32 sanity; /* To verify vtoc sanity */ - __be32 reserved[10]; /* Free space */ - __be32 timestamp[8]; /* Partition timestamp */ - } vtoc; - __be32 write_reinstruct; /* sectors to skip, writes */ - __be32 read_reinstruct; /* sectors to skip, reads */ - unsigned char spare[148]; /* Padding */ - __be16 rspeed; /* Disk rotational speed */ - __be16 pcylcount; /* Physical cylinder count */ - __be16 sparecyl; /* extra sects per cylinder */ - __be16 obs1; /* gap1 */ - __be16 obs2; /* gap2 */ - __be16 ilfact; /* Interleave factor */ - __be16 ncyl; /* Data cylinder count */ - __be16 nacyl; /* Alt. cylinder count */ - __be16 ntrks; /* Tracks per cylinder */ - __be16 nsect; /* Sectors per track */ - __be16 obs3; /* bhead - Label head offset */ - __be16 obs4; /* ppart - Physical Partition */ - struct sun_partition { - __be32 start_cylinder; - __be32 num_sectors; - } partitions[8]; - __be16 magic; /* Magic number */ - __be16 csum; /* Label xor'd checksum */ - } * label; - struct sun_partition *p; - unsigned long spc; - char b[BDEVNAME_SIZE]; - int use_vtoc; - int nparts; - - label = read_part_sector(state, 0, §); - if (!label) - return -1; - - p = label->partitions; - if (be16_to_cpu(label->magic) != SUN_LABEL_MAGIC) { -/* printk(KERN_INFO "Dev %s Sun disklabel: bad magic %04x\n", - bdevname(bdev, b), be16_to_cpu(label->magic)); */ - put_dev_sector(sect); - return 0; - } - /* Look at the checksum */ - ush = ((__be16 *) (label+1)) - 1; - for (csum = 0; ush >= ((__be16 *) label);) - csum ^= *ush--; - if (csum) { - printk("Dev %s Sun disklabel: Csum bad, label corrupted\n", - bdevname(state->bdev, b)); - put_dev_sector(sect); - return 0; - } - - /* Check to see if we can use the VTOC table */ - use_vtoc = ((be32_to_cpu(label->vtoc.sanity) == SUN_VTOC_SANITY) && - (be32_to_cpu(label->vtoc.version) == 1) && - (be16_to_cpu(label->vtoc.nparts) <= 8)); - - /* Use 8 partition entries if not specified in validated VTOC */ - nparts = (use_vtoc) ? be16_to_cpu(label->vtoc.nparts) : 8; - - /* - * So that old Linux-Sun partitions continue to work, - * alow the VTOC to be used under the additional condition ... - */ - use_vtoc = use_vtoc || !(label->vtoc.sanity || - label->vtoc.version || label->vtoc.nparts); - spc = be16_to_cpu(label->ntrks) * be16_to_cpu(label->nsect); - for (i = 0; i < nparts; i++, p++) { - unsigned long st_sector; - unsigned int num_sectors; - - st_sector = be32_to_cpu(p->start_cylinder) * spc; - num_sectors = be32_to_cpu(p->num_sectors); - if (num_sectors) { - put_partition(state, slot, st_sector, num_sectors); - state->parts[slot].flags = 0; - if (use_vtoc) { - if (be16_to_cpu(label->vtoc.infos[i].id) == LINUX_RAID_PARTITION) - state->parts[slot].flags |= ADDPART_FLAG_RAID; - else if (be16_to_cpu(label->vtoc.infos[i].id) == SUN_WHOLE_DISK) - state->parts[slot].flags |= ADDPART_FLAG_WHOLEDISK; - } - } - slot++; - } - strlcat(state->pp_buf, "\n", PAGE_SIZE); - put_dev_sector(sect); - return 1; -} diff --git a/fs/partitions/sun.h b/fs/partitions/sun.h deleted file mode 100644 index 2424baa8319f..000000000000 --- a/fs/partitions/sun.h +++ /dev/null @@ -1,8 +0,0 @@ -/* - * fs/partitions/sun.h - */ - -#define SUN_LABEL_MAGIC 0xDABE -#define SUN_VTOC_SANITY 0x600DDEEE - -int sun_partition(struct parsed_partitions *state); diff --git a/fs/partitions/sysv68.c b/fs/partitions/sysv68.c deleted file mode 100644 index 9627ccffc1c4..000000000000 --- a/fs/partitions/sysv68.c +++ /dev/null @@ -1,95 +0,0 @@ -/* - * fs/partitions/sysv68.c - * - * Copyright (C) 2007 Philippe De Muyter <phdm@macqel.be> - */ - -#include "check.h" -#include "sysv68.h" - -/* - * Volume ID structure: on first 256-bytes sector of disk - */ - -struct volumeid { - u8 vid_unused[248]; - u8 vid_mac[8]; /* ASCII string "MOTOROLA" */ -}; - -/* - * config block: second 256-bytes sector on disk - */ - -struct dkconfig { - u8 ios_unused0[128]; - __be32 ios_slcblk; /* Slice table block number */ - __be16 ios_slccnt; /* Number of entries in slice table */ - u8 ios_unused1[122]; -}; - -/* - * combined volumeid and dkconfig block - */ - -struct dkblk0 { - struct volumeid dk_vid; - struct dkconfig dk_ios; -}; - -/* - * Slice Table Structure - */ - -struct slice { - __be32 nblocks; /* slice size (in blocks) */ - __be32 blkoff; /* block offset of slice */ -}; - - -int sysv68_partition(struct parsed_partitions *state) -{ - int i, slices; - int slot = 1; - Sector sect; - unsigned char *data; - struct dkblk0 *b; - struct slice *slice; - char tmp[64]; - - data = read_part_sector(state, 0, §); - if (!data) - return -1; - - b = (struct dkblk0 *)data; - if (memcmp(b->dk_vid.vid_mac, "MOTOROLA", sizeof(b->dk_vid.vid_mac))) { - put_dev_sector(sect); - return 0; - } - slices = be16_to_cpu(b->dk_ios.ios_slccnt); - i = be32_to_cpu(b->dk_ios.ios_slcblk); - put_dev_sector(sect); - - data = read_part_sector(state, i, §); - if (!data) - return -1; - - slices -= 1; /* last slice is the whole disk */ - snprintf(tmp, sizeof(tmp), "sysV68: %s(s%u)", state->name, slices); - strlcat(state->pp_buf, tmp, PAGE_SIZE); - slice = (struct slice *)data; - for (i = 0; i < slices; i++, slice++) { - if (slot == state->limit) - break; - if (be32_to_cpu(slice->nblocks)) { - put_partition(state, slot, - be32_to_cpu(slice->blkoff), - be32_to_cpu(slice->nblocks)); - snprintf(tmp, sizeof(tmp), "(s%u)", i); - strlcat(state->pp_buf, tmp, PAGE_SIZE); - } - slot++; - } - strlcat(state->pp_buf, "\n", PAGE_SIZE); - put_dev_sector(sect); - return 1; -} diff --git a/fs/partitions/sysv68.h b/fs/partitions/sysv68.h deleted file mode 100644 index bf2f5ffa97ac..000000000000 --- a/fs/partitions/sysv68.h +++ /dev/null @@ -1 +0,0 @@ -extern int sysv68_partition(struct parsed_partitions *state); diff --git a/fs/partitions/ultrix.c b/fs/partitions/ultrix.c deleted file mode 100644 index 8dbaf9f77a99..000000000000 --- a/fs/partitions/ultrix.c +++ /dev/null @@ -1,48 +0,0 @@ -/* - * fs/partitions/ultrix.c - * - * Code extracted from drivers/block/genhd.c - * - * Re-organised Jul 1999 Russell King - */ - -#include "check.h" -#include "ultrix.h" - -int ultrix_partition(struct parsed_partitions *state) -{ - int i; - Sector sect; - unsigned char *data; - struct ultrix_disklabel { - s32 pt_magic; /* magic no. indicating part. info exits */ - s32 pt_valid; /* set by driver if pt is current */ - struct pt_info { - s32 pi_nblocks; /* no. of sectors */ - u32 pi_blkoff; /* block offset for start */ - } pt_part[8]; - } *label; - -#define PT_MAGIC 0x032957 /* Partition magic number */ -#define PT_VALID 1 /* Indicates if struct is valid */ - - data = read_part_sector(state, (16384 - sizeof(*label))/512, §); - if (!data) - return -1; - - label = (struct ultrix_disklabel *)(data + 512 - sizeof(*label)); - - if (label->pt_magic == PT_MAGIC && label->pt_valid == PT_VALID) { - for (i=0; i<8; i++) - if (label->pt_part[i].pi_nblocks) - put_partition(state, i+1, - label->pt_part[i].pi_blkoff, - label->pt_part[i].pi_nblocks); - put_dev_sector(sect); - strlcat(state->pp_buf, "\n", PAGE_SIZE); - return 1; - } else { - put_dev_sector(sect); - return 0; - } -} diff --git a/fs/partitions/ultrix.h b/fs/partitions/ultrix.h deleted file mode 100644 index a3cc00b2bded..000000000000 --- a/fs/partitions/ultrix.h +++ /dev/null @@ -1,5 +0,0 @@ -/* - * fs/partitions/ultrix.h - */ - -int ultrix_partition(struct parsed_partitions *state); diff --git a/fs/pipe.c b/fs/pipe.c index 4065f07366b3..a932ced92a16 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -1137,7 +1137,7 @@ static long pipe_set_size(struct pipe_inode_info *pipe, unsigned long nr_pages) if (nr_pages < pipe->nrbufs) return -EBUSY; - bufs = kcalloc(nr_pages, sizeof(struct pipe_buffer), GFP_KERNEL); + bufs = kcalloc(nr_pages, sizeof(*bufs), GFP_KERNEL | __GFP_NOWARN); if (unlikely(!bufs)) return -ENOMEM; @@ -1290,11 +1290,4 @@ static int __init init_pipe_fs(void) return err; } -static void __exit exit_pipe_fs(void) -{ - kern_unmount(pipe_mnt); - unregister_filesystem(&pipe_fs_type); -} - fs_initcall(init_pipe_fs); -module_exit(exit_pipe_fs); diff --git a/fs/pnode.c b/fs/pnode.c index d42514e32380..ab5fa9e1a79a 100644 --- a/fs/pnode.c +++ b/fs/pnode.c @@ -13,45 +13,30 @@ #include "pnode.h" /* return the next shared peer mount of @p */ -static inline struct vfsmount *next_peer(struct vfsmount *p) +static inline struct mount *next_peer(struct mount *p) { - return list_entry(p->mnt_share.next, struct vfsmount, mnt_share); + return list_entry(p->mnt_share.next, struct mount, mnt_share); } -static inline struct vfsmount *first_slave(struct vfsmount *p) +static inline struct mount *first_slave(struct mount *p) { - return list_entry(p->mnt_slave_list.next, struct vfsmount, mnt_slave); + return list_entry(p->mnt_slave_list.next, struct mount, mnt_slave); } -static inline struct vfsmount *next_slave(struct vfsmount *p) +static inline struct mount *next_slave(struct mount *p) { - return list_entry(p->mnt_slave.next, struct vfsmount, mnt_slave); + return list_entry(p->mnt_slave.next, struct mount, mnt_slave); } -/* - * Return true if path is reachable from root - * - * namespace_sem is held, and mnt is attached - */ -static bool is_path_reachable(struct vfsmount *mnt, struct dentry *dentry, - const struct path *root) -{ - while (mnt != root->mnt && mnt->mnt_parent != mnt) { - dentry = mnt->mnt_mountpoint; - mnt = mnt->mnt_parent; - } - return mnt == root->mnt && is_subdir(dentry, root->dentry); -} - -static struct vfsmount *get_peer_under_root(struct vfsmount *mnt, - struct mnt_namespace *ns, - const struct path *root) +static struct mount *get_peer_under_root(struct mount *mnt, + struct mnt_namespace *ns, + const struct path *root) { - struct vfsmount *m = mnt; + struct mount *m = mnt; do { /* Check the namespace first for optimization */ - if (m->mnt_ns == ns && is_path_reachable(m, m->mnt_root, root)) + if (m->mnt_ns == ns && is_path_reachable(m, m->mnt.mnt_root, root)) return m; m = next_peer(m); @@ -66,12 +51,12 @@ static struct vfsmount *get_peer_under_root(struct vfsmount *mnt, * * Caller must hold namespace_sem */ -int get_dominating_id(struct vfsmount *mnt, const struct path *root) +int get_dominating_id(struct mount *mnt, const struct path *root) { - struct vfsmount *m; + struct mount *m; for (m = mnt->mnt_master; m != NULL; m = m->mnt_master) { - struct vfsmount *d = get_peer_under_root(m, mnt->mnt_ns, root); + struct mount *d = get_peer_under_root(m, mnt->mnt_ns, root); if (d) return d->mnt_group_id; } @@ -79,10 +64,10 @@ int get_dominating_id(struct vfsmount *mnt, const struct path *root) return 0; } -static int do_make_slave(struct vfsmount *mnt) +static int do_make_slave(struct mount *mnt) { - struct vfsmount *peer_mnt = mnt, *master = mnt->mnt_master; - struct vfsmount *slave_mnt; + struct mount *peer_mnt = mnt, *master = mnt->mnt_master; + struct mount *slave_mnt; /* * slave 'mnt' to a peer mount that has the @@ -90,7 +75,7 @@ static int do_make_slave(struct vfsmount *mnt) * slave it to anything that is available. */ while ((peer_mnt = next_peer(peer_mnt)) != mnt && - peer_mnt->mnt_root != mnt->mnt_root) ; + peer_mnt->mnt.mnt_root != mnt->mnt.mnt_root) ; if (peer_mnt == mnt) { peer_mnt = next_peer(mnt); @@ -116,7 +101,7 @@ static int do_make_slave(struct vfsmount *mnt) struct list_head *p = &mnt->mnt_slave_list; while (!list_empty(p)) { slave_mnt = list_first_entry(p, - struct vfsmount, mnt_slave); + struct mount, mnt_slave); list_del_init(&slave_mnt->mnt_slave); slave_mnt->mnt_master = NULL; } @@ -129,7 +114,7 @@ static int do_make_slave(struct vfsmount *mnt) /* * vfsmount lock must be held for write */ -void change_mnt_propagation(struct vfsmount *mnt, int type) +void change_mnt_propagation(struct mount *mnt, int type) { if (type == MS_SHARED) { set_mnt_shared(mnt); @@ -140,9 +125,9 @@ void change_mnt_propagation(struct vfsmount *mnt, int type) list_del_init(&mnt->mnt_slave); mnt->mnt_master = NULL; if (type == MS_UNBINDABLE) - mnt->mnt_flags |= MNT_UNBINDABLE; + mnt->mnt.mnt_flags |= MNT_UNBINDABLE; else - mnt->mnt_flags &= ~MNT_UNBINDABLE; + mnt->mnt.mnt_flags &= ~MNT_UNBINDABLE; } } @@ -156,20 +141,19 @@ void change_mnt_propagation(struct vfsmount *mnt, int type) * vfsmount found while iterating with propagation_next() is * a peer of one we'd found earlier. */ -static struct vfsmount *propagation_next(struct vfsmount *m, - struct vfsmount *origin) +static struct mount *propagation_next(struct mount *m, + struct mount *origin) { /* are there any slaves of this mount? */ if (!IS_MNT_NEW(m) && !list_empty(&m->mnt_slave_list)) return first_slave(m); while (1) { - struct vfsmount *next; - struct vfsmount *master = m->mnt_master; + struct mount *master = m->mnt_master; if (master == origin->mnt_master) { - next = next_peer(m); - return ((next == origin) ? NULL : next); + struct mount *next = next_peer(m); + return (next == origin) ? NULL : next; } else if (m->mnt_slave.next != &master->mnt_slave_list) return next_slave(m); @@ -187,13 +171,13 @@ static struct vfsmount *propagation_next(struct vfsmount *m, * @type return CL_SLAVE if the new mount has to be * cloned as a slave. */ -static struct vfsmount *get_source(struct vfsmount *dest, - struct vfsmount *last_dest, - struct vfsmount *last_src, - int *type) +static struct mount *get_source(struct mount *dest, + struct mount *last_dest, + struct mount *last_src, + int *type) { - struct vfsmount *p_last_src = NULL; - struct vfsmount *p_last_dest = NULL; + struct mount *p_last_src = NULL; + struct mount *p_last_dest = NULL; while (last_dest != dest->mnt_master) { p_last_dest = last_dest; @@ -233,33 +217,33 @@ static struct vfsmount *get_source(struct vfsmount *dest, * @source_mnt: source mount. * @tree_list : list of heads of trees to be attached. */ -int propagate_mnt(struct vfsmount *dest_mnt, struct dentry *dest_dentry, - struct vfsmount *source_mnt, struct list_head *tree_list) +int propagate_mnt(struct mount *dest_mnt, struct dentry *dest_dentry, + struct mount *source_mnt, struct list_head *tree_list) { - struct vfsmount *m, *child; + struct mount *m, *child; int ret = 0; - struct vfsmount *prev_dest_mnt = dest_mnt; - struct vfsmount *prev_src_mnt = source_mnt; + struct mount *prev_dest_mnt = dest_mnt; + struct mount *prev_src_mnt = source_mnt; LIST_HEAD(tmp_list); LIST_HEAD(umount_list); for (m = propagation_next(dest_mnt, dest_mnt); m; m = propagation_next(m, dest_mnt)) { int type; - struct vfsmount *source; + struct mount *source; if (IS_MNT_NEW(m)) continue; source = get_source(m, prev_dest_mnt, prev_src_mnt, &type); - if (!(child = copy_tree(source, source->mnt_root, type))) { + if (!(child = copy_tree(source, source->mnt.mnt_root, type))) { ret = -ENOMEM; list_splice(tree_list, tmp_list.prev); goto out; } - if (is_subdir(dest_dentry, m->mnt_root)) { + if (is_subdir(dest_dentry, m->mnt.mnt_root)) { mnt_set_mountpoint(m, dest_dentry, child); list_add_tail(&child->mnt_hash, tree_list); } else { @@ -275,7 +259,7 @@ int propagate_mnt(struct vfsmount *dest_mnt, struct dentry *dest_dentry, out: br_write_lock(vfsmount_lock); while (!list_empty(&tmp_list)) { - child = list_first_entry(&tmp_list, struct vfsmount, mnt_hash); + child = list_first_entry(&tmp_list, struct mount, mnt_hash); umount_tree(child, 0, &umount_list); } br_write_unlock(vfsmount_lock); @@ -286,7 +270,7 @@ out: /* * return true if the refcount is greater than count */ -static inline int do_refcount_check(struct vfsmount *mnt, int count) +static inline int do_refcount_check(struct mount *mnt, int count) { int mycount = mnt_get_count(mnt) - mnt->mnt_ghosts; return (mycount > count); @@ -302,10 +286,10 @@ static inline int do_refcount_check(struct vfsmount *mnt, int count) * * vfsmount lock must be held for write */ -int propagate_mount_busy(struct vfsmount *mnt, int refcnt) +int propagate_mount_busy(struct mount *mnt, int refcnt) { - struct vfsmount *m, *child; - struct vfsmount *parent = mnt->mnt_parent; + struct mount *m, *child; + struct mount *parent = mnt->mnt_parent; int ret = 0; if (mnt == parent) @@ -321,7 +305,7 @@ int propagate_mount_busy(struct vfsmount *mnt, int refcnt) for (m = propagation_next(parent, parent); m; m = propagation_next(m, parent)) { - child = __lookup_mnt(m, mnt->mnt_mountpoint, 0); + child = __lookup_mnt(&m->mnt, mnt->mnt_mountpoint, 0); if (child && list_empty(&child->mnt_mounts) && (ret = do_refcount_check(child, 1))) break; @@ -333,17 +317,17 @@ int propagate_mount_busy(struct vfsmount *mnt, int refcnt) * NOTE: unmounting 'mnt' naturally propagates to all other mounts its * parent propagates to. */ -static void __propagate_umount(struct vfsmount *mnt) +static void __propagate_umount(struct mount *mnt) { - struct vfsmount *parent = mnt->mnt_parent; - struct vfsmount *m; + struct mount *parent = mnt->mnt_parent; + struct mount *m; BUG_ON(parent == mnt); for (m = propagation_next(parent, parent); m; m = propagation_next(m, parent)) { - struct vfsmount *child = __lookup_mnt(m, + struct mount *child = __lookup_mnt(&m->mnt, mnt->mnt_mountpoint, 0); /* * umount the child only if the child has no @@ -363,7 +347,7 @@ static void __propagate_umount(struct vfsmount *mnt) */ int propagate_umount(struct list_head *list) { - struct vfsmount *mnt; + struct mount *mnt; list_for_each_entry(mnt, list, mnt_hash) __propagate_umount(mnt); diff --git a/fs/pnode.h b/fs/pnode.h index 1ea4ae1efcd3..65c60979d541 100644 --- a/fs/pnode.h +++ b/fs/pnode.h @@ -9,13 +9,13 @@ #define _LINUX_PNODE_H #include <linux/list.h> -#include <linux/mount.h> +#include "mount.h" -#define IS_MNT_SHARED(mnt) (mnt->mnt_flags & MNT_SHARED) -#define IS_MNT_SLAVE(mnt) (mnt->mnt_master) -#define IS_MNT_NEW(mnt) (!mnt->mnt_ns) -#define CLEAR_MNT_SHARED(mnt) (mnt->mnt_flags &= ~MNT_SHARED) -#define IS_MNT_UNBINDABLE(mnt) (mnt->mnt_flags & MNT_UNBINDABLE) +#define IS_MNT_SHARED(m) ((m)->mnt.mnt_flags & MNT_SHARED) +#define IS_MNT_SLAVE(m) ((m)->mnt_master) +#define IS_MNT_NEW(m) (!(m)->mnt_ns) +#define CLEAR_MNT_SHARED(m) ((m)->mnt.mnt_flags &= ~MNT_SHARED) +#define IS_MNT_UNBINDABLE(m) ((m)->mnt.mnt_flags & MNT_UNBINDABLE) #define CL_EXPIRE 0x01 #define CL_SLAVE 0x02 @@ -23,17 +23,25 @@ #define CL_MAKE_SHARED 0x08 #define CL_PRIVATE 0x10 -static inline void set_mnt_shared(struct vfsmount *mnt) +static inline void set_mnt_shared(struct mount *mnt) { - mnt->mnt_flags &= ~MNT_SHARED_MASK; - mnt->mnt_flags |= MNT_SHARED; + mnt->mnt.mnt_flags &= ~MNT_SHARED_MASK; + mnt->mnt.mnt_flags |= MNT_SHARED; } -void change_mnt_propagation(struct vfsmount *, int); -int propagate_mnt(struct vfsmount *, struct dentry *, struct vfsmount *, +void change_mnt_propagation(struct mount *, int); +int propagate_mnt(struct mount *, struct dentry *, struct mount *, struct list_head *); int propagate_umount(struct list_head *); -int propagate_mount_busy(struct vfsmount *, int); -void mnt_release_group_id(struct vfsmount *); -int get_dominating_id(struct vfsmount *mnt, const struct path *root); +int propagate_mount_busy(struct mount *, int); +void mnt_release_group_id(struct mount *); +int get_dominating_id(struct mount *mnt, const struct path *root); +unsigned int mnt_get_count(struct mount *mnt); +void mnt_set_mountpoint(struct mount *, struct dentry *, + struct mount *); +void release_mounts(struct list_head *); +void umount_tree(struct mount *, int, struct list_head *); +struct mount *copy_tree(struct mount *, struct dentry *, int); +bool is_path_reachable(struct mount *, struct dentry *, + const struct path *root); #endif /* _LINUX_PNODE_H */ diff --git a/fs/proc/array.c b/fs/proc/array.c index 3a1dafd228d1..c602b8d20f06 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -380,7 +380,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, state = *get_task_state(task); vsize = eip = esp = 0; - permitted = ptrace_may_access(task, PTRACE_MODE_READ); + permitted = ptrace_may_access(task, PTRACE_MODE_READ | PTRACE_MODE_NOAUDIT); mm = get_task_mm(task); if (mm) { vsize = task_vsize(mm); @@ -394,8 +394,8 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, sigemptyset(&sigign); sigemptyset(&sigcatch); - cutime = cstime = utime = stime = cputime_zero; - cgtime = gtime = cputime_zero; + cutime = cstime = utime = stime = 0; + cgtime = gtime = 0; if (lock_task_sighand(task, &flags)) { struct signal_struct *sig = task->signal; @@ -423,14 +423,14 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, do { min_flt += t->min_flt; maj_flt += t->maj_flt; - gtime = cputime_add(gtime, t->gtime); + gtime += t->gtime; t = next_thread(t); } while (t != task); min_flt += sig->min_flt; maj_flt += sig->maj_flt; thread_group_times(task, &utime, &stime); - gtime = cputime_add(gtime, sig->gtime); + gtime += sig->gtime; } sid = task_session_nr_ns(task, ns); @@ -464,7 +464,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, seq_printf(m, "%d (%s) %c %d %d %d %d %d %u %lu \ %lu %lu %lu %lu %lu %ld %ld %ld %ld %d 0 %llu %lu %ld %lu %lu %lu %lu %lu \ -%lu %lu %lu %lu %lu %lu %lu %lu %d %d %u %u %llu %lu %ld\n", +%lu %lu %lu %lu %lu %lu %lu %lu %d %d %u %u %llu %lu %ld %lu %lu %lu\n", pid_nr_ns(pid, ns), tcomm, state, @@ -511,7 +511,10 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, task->policy, (unsigned long long)delayacct_blkio_ticks(task), cputime_to_clock_t(gtime), - cputime_to_clock_t(cgtime)); + cputime_to_clock_t(cgtime), + (mm && permitted) ? mm->start_data : 0, + (mm && permitted) ? mm->end_data : 0, + (mm && permitted) ? mm->start_brk : 0); if (mm) mmput(mm); return 0; diff --git a/fs/proc/base.c b/fs/proc/base.c index 851ba3dcdc29..5485a5388ecb 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -83,9 +83,11 @@ #include <linux/pid_namespace.h> #include <linux/fs_struct.h> #include <linux/slab.h> +#include <linux/flex_array.h> #ifdef CONFIG_HARDWALL #include <asm/hardwall.h> #endif +#include <trace/events/oom.h> #include "internal.h" /* NOTE: @@ -101,7 +103,7 @@ struct pid_entry { char *name; int len; - mode_t mode; + umode_t mode; const struct inode_operations *iop; const struct file_operations *fop; union proc_op op; @@ -133,6 +135,8 @@ struct pid_entry { NULL, &proc_single_file_operations, \ { .proc_show = show } ) +static int proc_fd_permission(struct inode *inode, int mask); + /* * Count the number of hardlinks for the pid_entry table, excluding the . * and .. links. @@ -165,9 +169,9 @@ static int get_task_root(struct task_struct *task, struct path *root) return result; } -static int proc_cwd_link(struct inode *inode, struct path *path) +static int proc_cwd_link(struct dentry *dentry, struct path *path) { - struct task_struct *task = get_proc_task(inode); + struct task_struct *task = get_proc_task(dentry->d_inode); int result = -ENOENT; if (task) { @@ -182,9 +186,9 @@ static int proc_cwd_link(struct inode *inode, struct path *path) return result; } -static int proc_root_link(struct inode *inode, struct path *path) +static int proc_root_link(struct dentry *dentry, struct path *path) { - struct task_struct *task = get_proc_task(inode); + struct task_struct *task = get_proc_task(dentry->d_inode); int result = -ENOENT; if (task) { @@ -627,122 +631,54 @@ int proc_setattr(struct dentry *dentry, struct iattr *attr) return 0; } -static const struct inode_operations proc_def_inode_operations = { - .setattr = proc_setattr, -}; - -static int mounts_open_common(struct inode *inode, struct file *file, - const struct seq_operations *op) +/* + * May current process learn task's sched/cmdline info (for hide_pid_min=1) + * or euid/egid (for hide_pid_min=2)? + */ +static bool has_pid_permissions(struct pid_namespace *pid, + struct task_struct *task, + int hide_pid_min) { - struct task_struct *task = get_proc_task(inode); - struct nsproxy *nsp; - struct mnt_namespace *ns = NULL; - struct path root; - struct proc_mounts *p; - int ret = -EINVAL; - - if (task) { - rcu_read_lock(); - nsp = task_nsproxy(task); - if (nsp) { - ns = nsp->mnt_ns; - if (ns) - get_mnt_ns(ns); - } - rcu_read_unlock(); - if (ns && get_task_root(task, &root) == 0) - ret = 0; - put_task_struct(task); - } - - if (!ns) - goto err; - if (ret) - goto err_put_ns; - - ret = -ENOMEM; - p = kmalloc(sizeof(struct proc_mounts), GFP_KERNEL); - if (!p) - goto err_put_path; - - file->private_data = &p->m; - ret = seq_open(file, op); - if (ret) - goto err_free; - - p->m.private = p; - p->ns = ns; - p->root = root; - p->m.poll_event = ns->event; - - return 0; - - err_free: - kfree(p); - err_put_path: - path_put(&root); - err_put_ns: - put_mnt_ns(ns); - err: - return ret; + if (pid->hide_pid < hide_pid_min) + return true; + if (in_group_p(pid->pid_gid)) + return true; + return ptrace_may_access(task, PTRACE_MODE_READ); } -static int mounts_release(struct inode *inode, struct file *file) -{ - struct proc_mounts *p = file->private_data; - path_put(&p->root); - put_mnt_ns(p->ns); - return seq_release(inode, file); -} -static unsigned mounts_poll(struct file *file, poll_table *wait) +static int proc_pid_permission(struct inode *inode, int mask) { - struct proc_mounts *p = file->private_data; - unsigned res = POLLIN | POLLRDNORM; - - poll_wait(file, &p->ns->poll, wait); - if (mnt_had_events(p)) - res |= POLLERR | POLLPRI; - - return res; -} + struct pid_namespace *pid = inode->i_sb->s_fs_info; + struct task_struct *task; + bool has_perms; -static int mounts_open(struct inode *inode, struct file *file) -{ - return mounts_open_common(inode, file, &mounts_op); -} + task = get_proc_task(inode); + if (!task) + return -ESRCH; + has_perms = has_pid_permissions(pid, task, 1); + put_task_struct(task); -static const struct file_operations proc_mounts_operations = { - .open = mounts_open, - .read = seq_read, - .llseek = seq_lseek, - .release = mounts_release, - .poll = mounts_poll, -}; + if (!has_perms) { + if (pid->hide_pid == 2) { + /* + * Let's make getdents(), stat(), and open() + * consistent with each other. If a process + * may not stat() a file, it shouldn't be seen + * in procfs at all. + */ + return -ENOENT; + } -static int mountinfo_open(struct inode *inode, struct file *file) -{ - return mounts_open_common(inode, file, &mountinfo_op); + return -EPERM; + } + return generic_permission(inode, mask); } -static const struct file_operations proc_mountinfo_operations = { - .open = mountinfo_open, - .read = seq_read, - .llseek = seq_lseek, - .release = mounts_release, - .poll = mounts_poll, -}; -static int mountstats_open(struct inode *inode, struct file *file) -{ - return mounts_open_common(inode, file, &mountstats_op); -} -static const struct file_operations proc_mountstats_operations = { - .open = mountstats_open, - .read = seq_read, - .llseek = seq_lseek, - .release = mounts_release, +static const struct inode_operations proc_def_inode_operations = { + .setattr = proc_setattr, }; #define PROC_BLOCK_SIZE (3*1024) /* 4K page size but our output routines use some slack for overruns */ @@ -1124,6 +1060,7 @@ static ssize_t oom_adjust_write(struct file *file, const char __user *buf, else task->signal->oom_score_adj = (oom_adjust * OOM_SCORE_ADJ_MAX) / -OOM_DISABLE; + trace_oom_score_adj_update(task); err_sighand: unlock_task_sighand(task, &flags); err_task_lock: @@ -1211,6 +1148,7 @@ static ssize_t oom_score_adj_write(struct file *file, const char __user *buf, task->signal->oom_score_adj = oom_score_adj; if (has_capability_noaudit(current, CAP_SYS_RESOURCE)) task->signal->oom_score_adj_min = oom_score_adj; + trace_oom_score_adj_update(task); /* * Scale /proc/pid/oom_adj appropriately ensuring that OOM_DISABLE is * always attainable. @@ -1567,13 +1505,13 @@ static const struct file_operations proc_pid_set_comm_operations = { .release = single_release, }; -static int proc_exe_link(struct inode *inode, struct path *exe_path) +static int proc_exe_link(struct dentry *dentry, struct path *exe_path) { struct task_struct *task; struct mm_struct *mm; struct file *exe_file; - task = get_proc_task(inode); + task = get_proc_task(dentry->d_inode); if (!task) return -ENOENT; mm = get_task_mm(task); @@ -1603,7 +1541,7 @@ static void *proc_pid_follow_link(struct dentry *dentry, struct nameidata *nd) if (!proc_fd_access_allowed(inode)) goto out; - error = PROC_I(inode)->op.proc_get_link(inode, &nd->path); + error = PROC_I(inode)->op.proc_get_link(dentry, &nd->path); out: return ERR_PTR(error); } @@ -1642,7 +1580,7 @@ static int proc_pid_readlink(struct dentry * dentry, char __user * buffer, int b if (!proc_fd_access_allowed(inode)) goto out; - error = PROC_I(inode)->op.proc_get_link(inode, &path); + error = PROC_I(inode)->op.proc_get_link(dentry, &path); if (error) goto out; @@ -1723,6 +1661,7 @@ int pid_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) struct inode *inode = dentry->d_inode; struct task_struct *task; const struct cred *cred; + struct pid_namespace *pid = dentry->d_sb->s_fs_info; generic_fillattr(inode, stat); @@ -1731,6 +1670,14 @@ int pid_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) stat->gid = 0; task = pid_task(proc_pid(inode), PIDTYPE_PID); if (task) { + if (!has_pid_permissions(pid, task, 2)) { + rcu_read_unlock(); + /* + * This doesn't prevent learning whether PID exists, + * it only makes getattr() consistent with readdir(). + */ + return -ENOENT; + } if ((inode->i_mode == (S_IFDIR|S_IRUGO|S_IXUGO)) || task_dumpable(task)) { cred = __task_cred(task); @@ -1934,9 +1881,9 @@ static int proc_fd_info(struct inode *inode, struct path *path, char *info) return -ENOENT; } -static int proc_fd_link(struct inode *inode, struct path *path) +static int proc_fd_link(struct dentry *dentry, struct path *path) { - return proc_fd_info(inode, path, NULL); + return proc_fd_info(dentry->d_inode, path, NULL); } static int tid_fd_revalidate(struct dentry *dentry, struct nameidata *nd) @@ -2157,6 +2104,355 @@ static const struct file_operations proc_fd_operations = { .llseek = default_llseek, }; +#ifdef CONFIG_CHECKPOINT_RESTORE + +/* + * dname_to_vma_addr - maps a dentry name into two unsigned longs + * which represent vma start and end addresses. + */ +static int dname_to_vma_addr(struct dentry *dentry, + unsigned long *start, unsigned long *end) +{ + if (sscanf(dentry->d_name.name, "%lx-%lx", start, end) != 2) + return -EINVAL; + + return 0; +} + +static int map_files_d_revalidate(struct dentry *dentry, struct nameidata *nd) +{ + unsigned long vm_start, vm_end; + bool exact_vma_exists = false; + struct mm_struct *mm = NULL; + struct task_struct *task; + const struct cred *cred; + struct inode *inode; + int status = 0; + + if (nd && nd->flags & LOOKUP_RCU) + return -ECHILD; + + if (!capable(CAP_SYS_ADMIN)) { + status = -EACCES; + goto out_notask; + } + + inode = dentry->d_inode; + task = get_proc_task(inode); + if (!task) + goto out_notask; + + if (!ptrace_may_access(task, PTRACE_MODE_READ)) + goto out; + + mm = get_task_mm(task); + if (!mm) + goto out; + + if (!dname_to_vma_addr(dentry, &vm_start, &vm_end)) { + down_read(&mm->mmap_sem); + exact_vma_exists = !!find_exact_vma(mm, vm_start, vm_end); + up_read(&mm->mmap_sem); + } + + mmput(mm); + + if (exact_vma_exists) { + if (task_dumpable(task)) { + rcu_read_lock(); + cred = __task_cred(task); + inode->i_uid = cred->euid; + inode->i_gid = cred->egid; + rcu_read_unlock(); + } else { + inode->i_uid = 0; + inode->i_gid = 0; + } + security_task_to_inode(task, inode); + status = 1; + } + +out: + put_task_struct(task); + +out_notask: + if (status <= 0) + d_drop(dentry); + + return status; +} + +static const struct dentry_operations tid_map_files_dentry_operations = { + .d_revalidate = map_files_d_revalidate, + .d_delete = pid_delete_dentry, +}; + +static int proc_map_files_get_link(struct dentry *dentry, struct path *path) +{ + unsigned long vm_start, vm_end; + struct vm_area_struct *vma; + struct task_struct *task; + struct mm_struct *mm; + int rc; + + rc = -ENOENT; + task = get_proc_task(dentry->d_inode); + if (!task) + goto out; + + mm = get_task_mm(task); + put_task_struct(task); + if (!mm) + goto out; + + rc = dname_to_vma_addr(dentry, &vm_start, &vm_end); + if (rc) + goto out_mmput; + + down_read(&mm->mmap_sem); + vma = find_exact_vma(mm, vm_start, vm_end); + if (vma && vma->vm_file) { + *path = vma->vm_file->f_path; + path_get(path); + rc = 0; + } + up_read(&mm->mmap_sem); + +out_mmput: + mmput(mm); +out: + return rc; +} + +struct map_files_info { + struct file *file; + unsigned long len; + unsigned char name[4*sizeof(long)+2]; /* max: %lx-%lx\0 */ +}; + +static struct dentry * +proc_map_files_instantiate(struct inode *dir, struct dentry *dentry, + struct task_struct *task, const void *ptr) +{ + const struct file *file = ptr; + struct proc_inode *ei; + struct inode *inode; + + if (!file) + return ERR_PTR(-ENOENT); + + inode = proc_pid_make_inode(dir->i_sb, task); + if (!inode) + return ERR_PTR(-ENOENT); + + ei = PROC_I(inode); + ei->op.proc_get_link = proc_map_files_get_link; + + inode->i_op = &proc_pid_link_inode_operations; + inode->i_size = 64; + inode->i_mode = S_IFLNK; + + if (file->f_mode & FMODE_READ) + inode->i_mode |= S_IRUSR; + if (file->f_mode & FMODE_WRITE) + inode->i_mode |= S_IWUSR; + + d_set_d_op(dentry, &tid_map_files_dentry_operations); + d_add(dentry, inode); + + return NULL; +} + +static struct dentry *proc_map_files_lookup(struct inode *dir, + struct dentry *dentry, struct nameidata *nd) +{ + unsigned long vm_start, vm_end; + struct vm_area_struct *vma; + struct task_struct *task; + struct dentry *result; + struct mm_struct *mm; + + result = ERR_PTR(-EACCES); + if (!capable(CAP_SYS_ADMIN)) + goto out; + + result = ERR_PTR(-ENOENT); + task = get_proc_task(dir); + if (!task) + goto out; + + result = ERR_PTR(-EACCES); + if (lock_trace(task)) + goto out_put_task; + + result = ERR_PTR(-ENOENT); + if (dname_to_vma_addr(dentry, &vm_start, &vm_end)) + goto out_unlock; + + mm = get_task_mm(task); + if (!mm) + goto out_unlock; + + down_read(&mm->mmap_sem); + vma = find_exact_vma(mm, vm_start, vm_end); + if (!vma) + goto out_no_vma; + + result = proc_map_files_instantiate(dir, dentry, task, vma->vm_file); + +out_no_vma: + up_read(&mm->mmap_sem); + mmput(mm); +out_unlock: + unlock_trace(task); +out_put_task: + put_task_struct(task); +out: + return result; +} + +static const struct inode_operations proc_map_files_inode_operations = { + .lookup = proc_map_files_lookup, + .permission = proc_fd_permission, + .setattr = proc_setattr, +}; + +static int +proc_map_files_readdir(struct file *filp, void *dirent, filldir_t filldir) +{ + struct dentry *dentry = filp->f_path.dentry; + struct inode *inode = dentry->d_inode; + struct vm_area_struct *vma; + struct task_struct *task; + struct mm_struct *mm; + ino_t ino; + int ret; + + ret = -EACCES; + if (!capable(CAP_SYS_ADMIN)) + goto out; + + ret = -ENOENT; + task = get_proc_task(inode); + if (!task) + goto out; + + ret = -EACCES; + if (lock_trace(task)) + goto out_put_task; + + ret = 0; + switch (filp->f_pos) { + case 0: + ino = inode->i_ino; + if (filldir(dirent, ".", 1, 0, ino, DT_DIR) < 0) + goto out_unlock; + filp->f_pos++; + case 1: + ino = parent_ino(dentry); + if (filldir(dirent, "..", 2, 1, ino, DT_DIR) < 0) + goto out_unlock; + filp->f_pos++; + default: + { + unsigned long nr_files, pos, i; + struct flex_array *fa = NULL; + struct map_files_info info; + struct map_files_info *p; + + mm = get_task_mm(task); + if (!mm) + goto out_unlock; + down_read(&mm->mmap_sem); + + nr_files = 0; + + /* + * We need two passes here: + * + * 1) Collect vmas of mapped files with mmap_sem taken + * 2) Release mmap_sem and instantiate entries + * + * otherwise we get lockdep complained, since filldir() + * routine might require mmap_sem taken in might_fault(). + */ + + for (vma = mm->mmap, pos = 2; vma; vma = vma->vm_next) { + if (vma->vm_file && ++pos > filp->f_pos) + nr_files++; + } + + if (nr_files) { + fa = flex_array_alloc(sizeof(info), nr_files, + GFP_KERNEL); + if (!fa || flex_array_prealloc(fa, 0, nr_files, + GFP_KERNEL)) { + ret = -ENOMEM; + if (fa) + flex_array_free(fa); + up_read(&mm->mmap_sem); + mmput(mm); + goto out_unlock; + } + for (i = 0, vma = mm->mmap, pos = 2; vma; + vma = vma->vm_next) { + if (!vma->vm_file) + continue; + if (++pos <= filp->f_pos) + continue; + + get_file(vma->vm_file); + info.file = vma->vm_file; + info.len = snprintf(info.name, + sizeof(info.name), "%lx-%lx", + vma->vm_start, vma->vm_end); + if (flex_array_put(fa, i++, &info, GFP_KERNEL)) + BUG(); + } + } + up_read(&mm->mmap_sem); + + for (i = 0; i < nr_files; i++) { + p = flex_array_get(fa, i); + ret = proc_fill_cache(filp, dirent, filldir, + p->name, p->len, + proc_map_files_instantiate, + task, p->file); + if (ret) + break; + filp->f_pos++; + fput(p->file); + } + for (; i < nr_files; i++) { + /* + * In case of error don't forget + * to put rest of file refs. + */ + p = flex_array_get(fa, i); + fput(p->file); + } + if (fa) + flex_array_free(fa); + mmput(mm); + } + } + +out_unlock: + unlock_trace(task); +out_put_task: + put_task_struct(task); +out: + return ret; +} + +static const struct file_operations proc_map_files_operations = { + .read = generic_read_dir, + .readdir = proc_map_files_readdir, + .llseek = default_llseek, +}; + +#endif /* CONFIG_CHECKPOINT_RESTORE */ + /* * /proc/pid/fd needs a special permission handler so that a process can still * access /proc/self/fd after it has executed a setuid(). @@ -2772,6 +3068,9 @@ static const struct inode_operations proc_task_inode_operations; static const struct pid_entry tgid_base_stuff[] = { DIR("task", S_IRUGO|S_IXUGO, proc_task_inode_operations, proc_task_operations), DIR("fd", S_IRUSR|S_IXUSR, proc_fd_inode_operations, proc_fd_operations), +#ifdef CONFIG_CHECKPOINT_RESTORE + DIR("map_files", S_IRUSR|S_IXUSR, proc_map_files_inode_operations, proc_map_files_operations), +#endif DIR("fdinfo", S_IRUSR|S_IXUSR, proc_fdinfo_inode_operations, proc_fdinfo_operations), DIR("ns", S_IRUSR|S_IXUGO, proc_ns_dir_inode_operations, proc_ns_dir_operations), #ifdef CONFIG_NET @@ -2875,6 +3174,7 @@ static const struct inode_operations proc_tgid_base_inode_operations = { .lookup = proc_tgid_base_lookup, .getattr = pid_getattr, .setattr = proc_setattr, + .permission = proc_pid_permission, }; static void proc_flush_task_mnt(struct vfsmount *mnt, pid_t pid, pid_t tgid) @@ -3078,6 +3378,12 @@ static int proc_pid_fill_cache(struct file *filp, void *dirent, filldir_t filldi proc_pid_instantiate, iter.task, NULL); } +static int fake_filldir(void *buf, const char *name, int namelen, + loff_t offset, u64 ino, unsigned d_type) +{ + return 0; +} + /* for the /proc/ directory itself, after non-process stuff has been done */ int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir) { @@ -3085,6 +3391,7 @@ int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir) struct task_struct *reaper; struct tgid_iter iter; struct pid_namespace *ns; + filldir_t __filldir; if (filp->f_pos >= PID_MAX_LIMIT + TGID_OFFSET) goto out_no_task; @@ -3106,8 +3413,13 @@ int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir) for (iter = next_tgid(ns, iter); iter.task; iter.tgid += 1, iter = next_tgid(ns, iter)) { + if (has_pid_permissions(ns, iter.task, 2)) + __filldir = filldir; + else + __filldir = fake_filldir; + filp->f_pos = iter.tgid + TGID_OFFSET; - if (proc_pid_fill_cache(filp, dirent, filldir, iter) < 0) { + if (proc_pid_fill_cache(filp, dirent, __filldir, iter) < 0) { put_task_struct(iter.task); goto out; } @@ -3442,6 +3754,7 @@ static const struct inode_operations proc_task_inode_operations = { .lookup = proc_task_lookup, .getattr = proc_task_getattr, .setattr = proc_setattr, + .permission = proc_pid_permission, }; static const struct file_operations proc_task_operations = { diff --git a/fs/proc/generic.c b/fs/proc/generic.c index 10090d9c7ad5..2edf34f2eb61 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -597,7 +597,7 @@ static int proc_register(struct proc_dir_entry * dir, struct proc_dir_entry * dp static struct proc_dir_entry *__proc_create(struct proc_dir_entry **parent, const char *name, - mode_t mode, + umode_t mode, nlink_t nlink) { struct proc_dir_entry *ent = NULL; @@ -659,7 +659,7 @@ struct proc_dir_entry *proc_symlink(const char *name, } EXPORT_SYMBOL(proc_symlink); -struct proc_dir_entry *proc_mkdir_mode(const char *name, mode_t mode, +struct proc_dir_entry *proc_mkdir_mode(const char *name, umode_t mode, struct proc_dir_entry *parent) { struct proc_dir_entry *ent; @@ -699,7 +699,7 @@ struct proc_dir_entry *proc_mkdir(const char *name, } EXPORT_SYMBOL(proc_mkdir); -struct proc_dir_entry *create_proc_entry(const char *name, mode_t mode, +struct proc_dir_entry *create_proc_entry(const char *name, umode_t mode, struct proc_dir_entry *parent) { struct proc_dir_entry *ent; @@ -728,7 +728,7 @@ struct proc_dir_entry *create_proc_entry(const char *name, mode_t mode, } EXPORT_SYMBOL(create_proc_entry); -struct proc_dir_entry *proc_create_data(const char *name, mode_t mode, +struct proc_dir_entry *proc_create_data(const char *name, umode_t mode, struct proc_dir_entry *parent, const struct file_operations *proc_fops, void *data) diff --git a/fs/proc/inode.c b/fs/proc/inode.c index 7737c5468a40..84fd3235a590 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -7,6 +7,7 @@ #include <linux/time.h> #include <linux/proc_fs.h> #include <linux/kernel.h> +#include <linux/pid_namespace.h> #include <linux/mm.h> #include <linux/string.h> #include <linux/stat.h> @@ -17,7 +18,9 @@ #include <linux/init.h> #include <linux/module.h> #include <linux/sysctl.h> +#include <linux/seq_file.h> #include <linux/slab.h> +#include <linux/mount.h> #include <asm/system.h> #include <asm/uaccess.h> @@ -77,7 +80,6 @@ static struct inode *proc_alloc_inode(struct super_block *sb) static void proc_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(proc_inode_cachep, PROC_I(inode)); } @@ -102,12 +104,27 @@ void __init proc_init_inodecache(void) init_once); } +static int proc_show_options(struct seq_file *seq, struct dentry *root) +{ + struct super_block *sb = root->d_sb; + struct pid_namespace *pid = sb->s_fs_info; + + if (pid->pid_gid) + seq_printf(seq, ",gid=%lu", (unsigned long)pid->pid_gid); + if (pid->hide_pid != 0) + seq_printf(seq, ",hidepid=%u", pid->hide_pid); + + return 0; +} + static const struct super_operations proc_sops = { .alloc_inode = proc_alloc_inode, .destroy_inode = proc_destroy_inode, .drop_inode = generic_delete_inode, .evict_inode = proc_evict_inode, .statfs = simple_statfs, + .remount_fs = proc_remount, + .show_options = proc_show_options, }; static void __pde_users_dec(struct proc_dir_entry *pde) diff --git a/fs/proc/internal.h b/fs/proc/internal.h index 7838e5cfec14..292577531ad1 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -117,6 +117,7 @@ void pde_put(struct proc_dir_entry *pde); int proc_fill_super(struct super_block *); struct inode *proc_get_inode(struct super_block *, struct proc_dir_entry *); +int proc_remount(struct super_block *sb, int *flags, char *data); /* * These are generic /proc routines that use the internal diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c index 586174168e2a..80e4645f7990 100644 --- a/fs/proc/meminfo.c +++ b/fs/proc/meminfo.c @@ -131,12 +131,13 @@ static int meminfo_proc_show(struct seq_file *m, void *v) K(i.freeswap), K(global_page_state(NR_FILE_DIRTY)), K(global_page_state(NR_WRITEBACK)), - K(global_page_state(NR_ANON_PAGES) #ifdef CONFIG_TRANSPARENT_HUGEPAGE + K(global_page_state(NR_ANON_PAGES) + global_page_state(NR_ANON_TRANSPARENT_HUGEPAGES) * - HPAGE_PMD_NR + HPAGE_PMD_NR), +#else + K(global_page_state(NR_ANON_PAGES)), #endif - ), K(global_page_state(NR_FILE_MAPPED)), K(global_page_state(NR_SHMEM)), K(global_page_state(NR_SLAB_RECLAIMABLE) + diff --git a/fs/proc/namespaces.c b/fs/proc/namespaces.c index be177f702acb..27da860115c6 100644 --- a/fs/proc/namespaces.c +++ b/fs/proc/namespaces.c @@ -9,7 +9,6 @@ #include <linux/file.h> #include <linux/utsname.h> #include <net/net_namespace.h> -#include <linux/mnt_namespace.h> #include <linux/ipc_namespace.h> #include <linux/pid_namespace.h> #include "internal.h" diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c index f738024ccc8e..06e1cc17caf6 100644 --- a/fs/proc/proc_net.c +++ b/fs/proc/proc_net.c @@ -179,7 +179,7 @@ const struct file_operations proc_net_operations = { struct proc_dir_entry *proc_net_fops_create(struct net *net, - const char *name, mode_t mode, const struct file_operations *fops) + const char *name, umode_t mode, const struct file_operations *fops) { return proc_create(name, mode, net->proc_net, fops); } diff --git a/fs/proc/root.c b/fs/proc/root.c index 9a8a2b77b874..46a15d8a29ca 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -18,6 +18,7 @@ #include <linux/bitops.h> #include <linux/mount.h> #include <linux/pid_namespace.h> +#include <linux/parser.h> #include "internal.h" @@ -36,6 +37,63 @@ static int proc_set_super(struct super_block *sb, void *data) return err; } +enum { + Opt_gid, Opt_hidepid, Opt_err, +}; + +static const match_table_t tokens = { + {Opt_hidepid, "hidepid=%u"}, + {Opt_gid, "gid=%u"}, + {Opt_err, NULL}, +}; + +static int proc_parse_options(char *options, struct pid_namespace *pid) +{ + char *p; + substring_t args[MAX_OPT_ARGS]; + int option; + + if (!options) + return 1; + + while ((p = strsep(&options, ",")) != NULL) { + int token; + if (!*p) + continue; + + args[0].to = args[0].from = 0; + token = match_token(p, tokens, args); + switch (token) { + case Opt_gid: + if (match_int(&args[0], &option)) + return 0; + pid->pid_gid = option; + break; + case Opt_hidepid: + if (match_int(&args[0], &option)) + return 0; + if (option < 0 || option > 2) { + pr_err("proc: hidepid value must be between 0 and 2.\n"); + return 0; + } + pid->hide_pid = option; + break; + default: + pr_err("proc: unrecognized mount option \"%s\" " + "or missing value\n", p); + return 0; + } + } + + return 1; +} + +int proc_remount(struct super_block *sb, int *flags, char *data) +{ + struct pid_namespace *pid = sb->s_fs_info; + return !proc_parse_options(data, pid); +} + static struct dentry *proc_mount(struct file_system_type *fs_type, int flags, const char *dev_name, void *data) { @@ -43,11 +101,15 @@ static struct dentry *proc_mount(struct file_system_type *fs_type, struct super_block *sb; struct pid_namespace *ns; struct proc_inode *ei; + char *options; - if (flags & MS_KERNMOUNT) + if (flags & MS_KERNMOUNT) { ns = (struct pid_namespace *)data; - else + options = NULL; + } else { ns = current->nsproxy->pid_ns; + options = data; + } sb = sget(fs_type, proc_test_super, proc_set_super, ns); if (IS_ERR(sb)) @@ -55,6 +117,10 @@ static struct dentry *proc_mount(struct file_system_type *fs_type, if (!sb->s_root) { sb->s_flags = flags; + if (!proc_parse_options(options, ns)) { + deactivate_locked_super(sb); + return ERR_PTR(-EINVAL); + } err = proc_fill_super(sb); if (err) { deactivate_locked_super(sb); @@ -91,20 +157,18 @@ static struct file_system_type proc_fs_type = { void __init proc_root_init(void) { - struct vfsmount *mnt; int err; proc_init_inodecache(); err = register_filesystem(&proc_fs_type); if (err) return; - mnt = kern_mount_data(&proc_fs_type, &init_pid_ns); - if (IS_ERR(mnt)) { + err = pid_ns_prepare_proc(&init_pid_ns); + if (err) { unregister_filesystem(&proc_fs_type); return; } - init_pid_ns.proc_mnt = mnt; proc_symlink("mounts", NULL, "self/mounts"); proc_net_init(); @@ -209,5 +273,5 @@ int pid_ns_prepare_proc(struct pid_namespace *ns) void pid_ns_release_proc(struct pid_namespace *ns) { - mntput(ns->proc_mnt); + kern_unmount(ns->proc_mnt); } diff --git a/fs/proc/stat.c b/fs/proc/stat.c index 42b274da92c3..d76ca6ae2b1b 100644 --- a/fs/proc/stat.c +++ b/fs/proc/stat.c @@ -22,31 +22,29 @@ #define arch_idle_time(cpu) 0 #endif -static cputime64_t get_idle_time(int cpu) +static u64 get_idle_time(int cpu) { - u64 idle_time = get_cpu_idle_time_us(cpu, NULL); - cputime64_t idle; + u64 idle, idle_time = get_cpu_idle_time_us(cpu, NULL); if (idle_time == -1ULL) { /* !NO_HZ so we can rely on cpustat.idle */ - idle = kstat_cpu(cpu).cpustat.idle; - idle = cputime64_add(idle, arch_idle_time(cpu)); + idle = kcpustat_cpu(cpu).cpustat[CPUTIME_IDLE]; + idle += arch_idle_time(cpu); } else - idle = usecs_to_cputime(idle_time); + idle = usecs_to_cputime64(idle_time); return idle; } -static cputime64_t get_iowait_time(int cpu) +static u64 get_iowait_time(int cpu) { - u64 iowait_time = get_cpu_iowait_time_us(cpu, NULL); - cputime64_t iowait; + u64 iowait, iowait_time = get_cpu_iowait_time_us(cpu, NULL); if (iowait_time == -1ULL) /* !NO_HZ so we can rely on cpustat.iowait */ - iowait = kstat_cpu(cpu).cpustat.iowait; + iowait = kcpustat_cpu(cpu).cpustat[CPUTIME_IOWAIT]; else - iowait = usecs_to_cputime(iowait_time); + iowait = usecs_to_cputime64(iowait_time); return iowait; } @@ -55,33 +53,30 @@ static int show_stat(struct seq_file *p, void *v) { int i, j; unsigned long jif; - cputime64_t user, nice, system, idle, iowait, irq, softirq, steal; - cputime64_t guest, guest_nice; + u64 user, nice, system, idle, iowait, irq, softirq, steal; + u64 guest, guest_nice; u64 sum = 0; u64 sum_softirq = 0; unsigned int per_softirq_sums[NR_SOFTIRQS] = {0}; struct timespec boottime; user = nice = system = idle = iowait = - irq = softirq = steal = cputime64_zero; - guest = guest_nice = cputime64_zero; + irq = softirq = steal = 0; + guest = guest_nice = 0; getboottime(&boottime); jif = boottime.tv_sec; for_each_possible_cpu(i) { - user = cputime64_add(user, kstat_cpu(i).cpustat.user); - nice = cputime64_add(nice, kstat_cpu(i).cpustat.nice); - system = cputime64_add(system, kstat_cpu(i).cpustat.system); - idle = cputime64_add(idle, get_idle_time(i)); - iowait = cputime64_add(iowait, get_iowait_time(i)); - irq = cputime64_add(irq, kstat_cpu(i).cpustat.irq); - softirq = cputime64_add(softirq, kstat_cpu(i).cpustat.softirq); - steal = cputime64_add(steal, kstat_cpu(i).cpustat.steal); - guest = cputime64_add(guest, kstat_cpu(i).cpustat.guest); - guest_nice = cputime64_add(guest_nice, - kstat_cpu(i).cpustat.guest_nice); - sum += kstat_cpu_irqs_sum(i); - sum += arch_irq_stat_cpu(i); + user += kcpustat_cpu(i).cpustat[CPUTIME_USER]; + nice += kcpustat_cpu(i).cpustat[CPUTIME_NICE]; + system += kcpustat_cpu(i).cpustat[CPUTIME_SYSTEM]; + idle += get_idle_time(i); + iowait += get_iowait_time(i); + irq += kcpustat_cpu(i).cpustat[CPUTIME_IRQ]; + softirq += kcpustat_cpu(i).cpustat[CPUTIME_SOFTIRQ]; + steal += kcpustat_cpu(i).cpustat[CPUTIME_STEAL]; + guest += kcpustat_cpu(i).cpustat[CPUTIME_GUEST]; + guest_nice += kcpustat_cpu(i).cpustat[CPUTIME_GUEST_NICE]; for (j = 0; j < NR_SOFTIRQS; j++) { unsigned int softirq_stat = kstat_softirqs_cpu(j, i); @@ -106,16 +101,16 @@ static int show_stat(struct seq_file *p, void *v) (unsigned long long)cputime64_to_clock_t(guest_nice)); for_each_online_cpu(i) { /* Copy values here to work around gcc-2.95.3, gcc-2.96 */ - user = kstat_cpu(i).cpustat.user; - nice = kstat_cpu(i).cpustat.nice; - system = kstat_cpu(i).cpustat.system; + user = kcpustat_cpu(i).cpustat[CPUTIME_USER]; + nice = kcpustat_cpu(i).cpustat[CPUTIME_NICE]; + system = kcpustat_cpu(i).cpustat[CPUTIME_SYSTEM]; idle = get_idle_time(i); iowait = get_iowait_time(i); - irq = kstat_cpu(i).cpustat.irq; - softirq = kstat_cpu(i).cpustat.softirq; - steal = kstat_cpu(i).cpustat.steal; - guest = kstat_cpu(i).cpustat.guest; - guest_nice = kstat_cpu(i).cpustat.guest_nice; + irq = kcpustat_cpu(i).cpustat[CPUTIME_IRQ]; + softirq = kcpustat_cpu(i).cpustat[CPUTIME_SOFTIRQ]; + steal = kcpustat_cpu(i).cpustat[CPUTIME_STEAL]; + guest = kcpustat_cpu(i).cpustat[CPUTIME_GUEST]; + guest_nice = kcpustat_cpu(i).cpustat[CPUTIME_GUEST_NICE]; seq_printf(p, "cpu%d %llu %llu %llu %llu %llu %llu %llu %llu %llu " "%llu\n", diff --git a/fs/proc/uptime.c b/fs/proc/uptime.c index 766b1d456050..9610ac772d7e 100644 --- a/fs/proc/uptime.c +++ b/fs/proc/uptime.c @@ -11,15 +11,20 @@ static int uptime_proc_show(struct seq_file *m, void *v) { struct timespec uptime; struct timespec idle; + u64 idletime; + u64 nsec; + u32 rem; int i; - cputime_t idletime = cputime_zero; + idletime = 0; for_each_possible_cpu(i) - idletime = cputime64_add(idletime, kstat_cpu(i).cpustat.idle); + idletime += (__force u64) kcpustat_cpu(i).cpustat[CPUTIME_IDLE]; do_posix_clock_monotonic_gettime(&uptime); monotonic_to_bootbased(&uptime); - cputime_to_timespec(idletime, &idle); + nsec = cputime64_to_jiffies64(idletime) * TICK_NSEC; + idle.tv_sec = div_u64_rem(nsec, NSEC_PER_SEC, &rem); + idle.tv_nsec = rem; seq_printf(m, "%lu.%02lu %lu.%02lu\n", (unsigned long) uptime.tv_sec, (uptime.tv_nsec / (NSEC_PER_SEC / 100)), diff --git a/fs/proc_namespace.c b/fs/proc_namespace.c new file mode 100644 index 000000000000..12412852d88a --- /dev/null +++ b/fs/proc_namespace.c @@ -0,0 +1,333 @@ +/* + * fs/proc_namespace.c - handling of /proc/<pid>/{mounts,mountinfo,mountstats} + * + * In fact, that's a piece of procfs; it's *almost* isolated from + * the rest of fs/proc, but has rather close relationships with + * fs/namespace.c, thus here instead of fs/proc + * + */ +#include <linux/mnt_namespace.h> +#include <linux/nsproxy.h> +#include <linux/security.h> +#include <linux/fs_struct.h> +#include "proc/internal.h" /* only for get_proc_task() in ->open() */ + +#include "pnode.h" +#include "internal.h" + +static unsigned mounts_poll(struct file *file, poll_table *wait) +{ + struct proc_mounts *p = file->private_data; + struct mnt_namespace *ns = p->ns; + unsigned res = POLLIN | POLLRDNORM; + + poll_wait(file, &p->ns->poll, wait); + + br_read_lock(vfsmount_lock); + if (p->m.poll_event != ns->event) { + p->m.poll_event = ns->event; + res |= POLLERR | POLLPRI; + } + br_read_unlock(vfsmount_lock); + + return res; +} + +struct proc_fs_info { + int flag; + const char *str; +}; + +static int show_sb_opts(struct seq_file *m, struct super_block *sb) +{ + static const struct proc_fs_info fs_info[] = { + { MS_SYNCHRONOUS, ",sync" }, + { MS_DIRSYNC, ",dirsync" }, + { MS_MANDLOCK, ",mand" }, + { 0, NULL } + }; + const struct proc_fs_info *fs_infop; + + for (fs_infop = fs_info; fs_infop->flag; fs_infop++) { + if (sb->s_flags & fs_infop->flag) + seq_puts(m, fs_infop->str); + } + + return security_sb_show_options(m, sb); +} + +static void show_mnt_opts(struct seq_file *m, struct vfsmount *mnt) +{ + static const struct proc_fs_info mnt_info[] = { + { MNT_NOSUID, ",nosuid" }, + { MNT_NODEV, ",nodev" }, + { MNT_NOEXEC, ",noexec" }, + { MNT_NOATIME, ",noatime" }, + { MNT_NODIRATIME, ",nodiratime" }, + { MNT_RELATIME, ",relatime" }, + { 0, NULL } + }; + const struct proc_fs_info *fs_infop; + + for (fs_infop = mnt_info; fs_infop->flag; fs_infop++) { + if (mnt->mnt_flags & fs_infop->flag) + seq_puts(m, fs_infop->str); + } +} + +static inline void mangle(struct seq_file *m, const char *s) +{ + seq_escape(m, s, " \t\n\\"); +} + +static void show_type(struct seq_file *m, struct super_block *sb) +{ + mangle(m, sb->s_type->name); + if (sb->s_subtype && sb->s_subtype[0]) { + seq_putc(m, '.'); + mangle(m, sb->s_subtype); + } +} + +static int show_vfsmnt(struct seq_file *m, struct vfsmount *mnt) +{ + struct mount *r = real_mount(mnt); + int err = 0; + struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt }; + struct super_block *sb = mnt_path.dentry->d_sb; + + if (sb->s_op->show_devname) { + err = sb->s_op->show_devname(m, mnt_path.dentry); + if (err) + goto out; + } else { + mangle(m, r->mnt_devname ? r->mnt_devname : "none"); + } + seq_putc(m, ' '); + seq_path(m, &mnt_path, " \t\n\\"); + seq_putc(m, ' '); + show_type(m, sb); + seq_puts(m, __mnt_is_readonly(mnt) ? " ro" : " rw"); + err = show_sb_opts(m, sb); + if (err) + goto out; + show_mnt_opts(m, mnt); + if (sb->s_op->show_options) + err = sb->s_op->show_options(m, mnt_path.dentry); + seq_puts(m, " 0 0\n"); +out: + return err; +} + +static int show_mountinfo(struct seq_file *m, struct vfsmount *mnt) +{ + struct proc_mounts *p = m->private; + struct mount *r = real_mount(mnt); + struct super_block *sb = mnt->mnt_sb; + struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt }; + struct path root = p->root; + int err = 0; + + seq_printf(m, "%i %i %u:%u ", r->mnt_id, r->mnt_parent->mnt_id, + MAJOR(sb->s_dev), MINOR(sb->s_dev)); + if (sb->s_op->show_path) + err = sb->s_op->show_path(m, mnt->mnt_root); + else + seq_dentry(m, mnt->mnt_root, " \t\n\\"); + if (err) + goto out; + seq_putc(m, ' '); + + /* mountpoints outside of chroot jail will give SEQ_SKIP on this */ + err = seq_path_root(m, &mnt_path, &root, " \t\n\\"); + if (err) + goto out; + + seq_puts(m, mnt->mnt_flags & MNT_READONLY ? " ro" : " rw"); + show_mnt_opts(m, mnt); + + /* Tagged fields ("foo:X" or "bar") */ + if (IS_MNT_SHARED(r)) + seq_printf(m, " shared:%i", r->mnt_group_id); + if (IS_MNT_SLAVE(r)) { + int master = r->mnt_master->mnt_group_id; + int dom = get_dominating_id(r, &p->root); + seq_printf(m, " master:%i", master); + if (dom && dom != master) + seq_printf(m, " propagate_from:%i", dom); + } + if (IS_MNT_UNBINDABLE(r)) + seq_puts(m, " unbindable"); + + /* Filesystem specific data */ + seq_puts(m, " - "); + show_type(m, sb); + seq_putc(m, ' '); + if (sb->s_op->show_devname) + err = sb->s_op->show_devname(m, mnt->mnt_root); + else + mangle(m, r->mnt_devname ? r->mnt_devname : "none"); + if (err) + goto out; + seq_puts(m, sb->s_flags & MS_RDONLY ? " ro" : " rw"); + err = show_sb_opts(m, sb); + if (err) + goto out; + if (sb->s_op->show_options) + err = sb->s_op->show_options(m, mnt->mnt_root); + seq_putc(m, '\n'); +out: + return err; +} + +static int show_vfsstat(struct seq_file *m, struct vfsmount *mnt) +{ + struct mount *r = real_mount(mnt); + struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt }; + struct super_block *sb = mnt_path.dentry->d_sb; + int err = 0; + + /* device */ + if (sb->s_op->show_devname) { + seq_puts(m, "device "); + err = sb->s_op->show_devname(m, mnt_path.dentry); + } else { + if (r->mnt_devname) { + seq_puts(m, "device "); + mangle(m, r->mnt_devname); + } else + seq_puts(m, "no device"); + } + + /* mount point */ + seq_puts(m, " mounted on "); + seq_path(m, &mnt_path, " \t\n\\"); + seq_putc(m, ' '); + + /* file system type */ + seq_puts(m, "with fstype "); + show_type(m, sb); + + /* optional statistics */ + if (sb->s_op->show_stats) { + seq_putc(m, ' '); + if (!err) + err = sb->s_op->show_stats(m, mnt_path.dentry); + } + + seq_putc(m, '\n'); + return err; +} + +static int mounts_open_common(struct inode *inode, struct file *file, + int (*show)(struct seq_file *, struct vfsmount *)) +{ + struct task_struct *task = get_proc_task(inode); + struct nsproxy *nsp; + struct mnt_namespace *ns = NULL; + struct path root; + struct proc_mounts *p; + int ret = -EINVAL; + + if (!task) + goto err; + + rcu_read_lock(); + nsp = task_nsproxy(task); + if (!nsp) { + rcu_read_unlock(); + put_task_struct(task); + goto err; + } + ns = nsp->mnt_ns; + if (!ns) { + rcu_read_unlock(); + put_task_struct(task); + goto err; + } + get_mnt_ns(ns); + rcu_read_unlock(); + task_lock(task); + if (!task->fs) { + task_unlock(task); + put_task_struct(task); + ret = -ENOENT; + goto err_put_ns; + } + get_fs_root(task->fs, &root); + task_unlock(task); + put_task_struct(task); + + ret = -ENOMEM; + p = kmalloc(sizeof(struct proc_mounts), GFP_KERNEL); + if (!p) + goto err_put_path; + + file->private_data = &p->m; + ret = seq_open(file, &mounts_op); + if (ret) + goto err_free; + + p->m.private = p; + p->ns = ns; + p->root = root; + p->m.poll_event = ns->event; + p->show = show; + + return 0; + + err_free: + kfree(p); + err_put_path: + path_put(&root); + err_put_ns: + put_mnt_ns(ns); + err: + return ret; +} + +static int mounts_release(struct inode *inode, struct file *file) +{ + struct proc_mounts *p = file->private_data; + path_put(&p->root); + put_mnt_ns(p->ns); + return seq_release(inode, file); +} + +static int mounts_open(struct inode *inode, struct file *file) +{ + return mounts_open_common(inode, file, show_vfsmnt); +} + +static int mountinfo_open(struct inode *inode, struct file *file) +{ + return mounts_open_common(inode, file, show_mountinfo); +} + +static int mountstats_open(struct inode *inode, struct file *file) +{ + return mounts_open_common(inode, file, show_vfsstat); +} + +const struct file_operations proc_mounts_operations = { + .open = mounts_open, + .read = seq_read, + .llseek = seq_lseek, + .release = mounts_release, + .poll = mounts_poll, +}; + +const struct file_operations proc_mountinfo_operations = { + .open = mountinfo_open, + .read = seq_read, + .llseek = seq_lseek, + .release = mounts_release, + .poll = mounts_poll, +}; + +const struct file_operations proc_mountstats_operations = { + .open = mountstats_open, + .read = seq_read, + .llseek = seq_lseek, + .release = mounts_release, +}; diff --git a/fs/pstore/inode.c b/fs/pstore/inode.c index 379a02dc1217..b3b426edb2fd 100644 --- a/fs/pstore/inode.c +++ b/fs/pstore/inode.c @@ -80,7 +80,8 @@ static int pstore_unlink(struct inode *dir, struct dentry *dentry) { struct pstore_private *p = dentry->d_inode->i_private; - p->psi->erase(p->type, p->id, p->psi); + if (p->psi->erase) + p->psi->erase(p->type, p->id, p->psi); return simple_unlink(dir, dentry); } diff --git a/fs/pstore/platform.c b/fs/pstore/platform.c index 2bd620f0d796..9ec22d3b4293 100644 --- a/fs/pstore/platform.c +++ b/fs/pstore/platform.c @@ -122,7 +122,7 @@ static void pstore_dump(struct kmsg_dumper *dumper, memcpy(dst, s1 + s1_start, l1_cpy); memcpy(dst + l1_cpy, s2 + s2_start, l2_cpy); - ret = psinfo->write(PSTORE_TYPE_DMESG, &id, part, + ret = psinfo->write(PSTORE_TYPE_DMESG, reason, &id, part, hsize + l1_cpy + l2_cpy, psinfo); if (ret == 0 && reason == KMSG_DUMP_OOPS && pstore_is_mounted()) pstore_new_entry = 1; @@ -167,6 +167,7 @@ int pstore_register(struct pstore_info *psi) } psinfo = psi; + mutex_init(&psinfo->read_mutex); spin_unlock(&pstore_lock); if (owner && !try_module_get(owner)) { @@ -195,30 +196,32 @@ EXPORT_SYMBOL_GPL(pstore_register); void pstore_get_records(int quiet) { struct pstore_info *psi = psinfo; + char *buf = NULL; ssize_t size; u64 id; enum pstore_type_id type; struct timespec time; int failed = 0, rc; - unsigned long flags; if (!psi) return; - spin_lock_irqsave(&psinfo->buf_lock, flags); - rc = psi->open(psi); - if (rc) + mutex_lock(&psi->read_mutex); + if (psi->open && psi->open(psi)) goto out; - while ((size = psi->read(&id, &type, &time, psi)) > 0) { - rc = pstore_mkfile(type, psi->name, id, psi->buf, (size_t)size, + while ((size = psi->read(&id, &type, &time, &buf, psi)) > 0) { + rc = pstore_mkfile(type, psi->name, id, buf, (size_t)size, time, psi); + kfree(buf); + buf = NULL; if (rc && (rc != -EEXIST || !quiet)) failed++; } - psi->close(psi); + if (psi->close) + psi->close(psi); out: - spin_unlock_irqrestore(&psinfo->buf_lock, flags); + mutex_unlock(&psi->read_mutex); if (failed) printk(KERN_WARNING "pstore: failed to load %d record(s) from '%s'\n", @@ -240,33 +243,5 @@ static void pstore_timefunc(unsigned long dummy) mod_timer(&pstore_timer, jiffies + PSTORE_INTERVAL); } -/* - * Call platform driver to write a record to the - * persistent store. - */ -int pstore_write(enum pstore_type_id type, char *buf, size_t size) -{ - u64 id; - int ret; - unsigned long flags; - - if (!psinfo) - return -ENODEV; - - if (size > psinfo->bufsize) - return -EFBIG; - - spin_lock_irqsave(&psinfo->buf_lock, flags); - memcpy(psinfo->buf, buf, size); - ret = psinfo->write(type, &id, 0, size, psinfo); - if (ret == 0 && pstore_is_mounted()) - pstore_mkfile(PSTORE_TYPE_DMESG, psinfo->name, id, psinfo->buf, - size, CURRENT_TIME, psinfo); - spin_unlock_irqrestore(&psinfo->buf_lock, flags); - - return 0; -} -EXPORT_SYMBOL_GPL(pstore_write); - module_param(backend, charp, 0444); MODULE_PARM_DESC(backend, "Pstore backend to use"); diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c index 3bdd21418432..2bfd987f4853 100644 --- a/fs/qnx4/inode.c +++ b/fs/qnx4/inode.c @@ -199,12 +199,13 @@ static const char *qnx4_checkroot(struct super_block *sb) if (!strcmp(rootdir->di_fname, QNX4_BMNAME)) { found = 1; - qnx4_sb(sb)->BitMap = kmalloc( sizeof( struct qnx4_inode_entry ), GFP_KERNEL ); + qnx4_sb(sb)->BitMap = kmemdup(rootdir, + sizeof(struct qnx4_inode_entry), + GFP_KERNEL); if (!qnx4_sb(sb)->BitMap) { brelse (bh); return "not enough memory for bitmap inode"; - } - memcpy( qnx4_sb(sb)->BitMap, rootdir, sizeof( struct qnx4_inode_entry ) ); /* keep bitmap inode known */ + }/* keep bitmap inode known */ break; } } @@ -427,7 +428,6 @@ static struct inode *qnx4_alloc_inode(struct super_block *sb) static void qnx4_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(qnx4_inode_cachep, qnx4_i(inode)); } diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c index 5b572c89e6c4..5ec59b20cf76 100644 --- a/fs/quota/dquot.c +++ b/fs/quota/dquot.c @@ -73,7 +73,6 @@ #include <linux/security.h> #include <linux/kmod.h> #include <linux/namei.h> -#include <linux/buffer_head.h> #include <linux/capability.h> #include <linux/quotaops.h> #include "../internal.h" /* ugh */ @@ -2199,7 +2198,7 @@ int dquot_quota_on(struct super_block *sb, int type, int format_id, if (error) return error; /* Quota file not on the same filesystem? */ - if (path->mnt->mnt_sb != sb) + if (path->dentry->d_sb != sb) error = -EXDEV; else error = vfs_load_quota_inode(path->dentry->d_inode, type, diff --git a/fs/quota/quota.c b/fs/quota/quota.c index 35f4b0ecdeb3..7898cd688a00 100644 --- a/fs/quota/quota.c +++ b/fs/quota/quota.c @@ -13,7 +13,6 @@ #include <linux/kernel.h> #include <linux/security.h> #include <linux/syscalls.h> -#include <linux/buffer_head.h> #include <linux/capability.h> #include <linux/quotaops.h> #include <linux/types.h> diff --git a/fs/ramfs/inode.c b/fs/ramfs/inode.c index 462ceb38fec6..aec766abe3af 100644 --- a/fs/ramfs/inode.c +++ b/fs/ramfs/inode.c @@ -52,7 +52,7 @@ static struct backing_dev_info ramfs_backing_dev_info = { }; struct inode *ramfs_get_inode(struct super_block *sb, - const struct inode *dir, int mode, dev_t dev) + const struct inode *dir, umode_t mode, dev_t dev) { struct inode * inode = new_inode(sb); @@ -92,7 +92,7 @@ struct inode *ramfs_get_inode(struct super_block *sb, */ /* SMP-safe */ static int -ramfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) +ramfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev) { struct inode * inode = ramfs_get_inode(dir->i_sb, dir, mode, dev); int error = -ENOSPC; @@ -106,7 +106,7 @@ ramfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) return error; } -static int ramfs_mkdir(struct inode * dir, struct dentry * dentry, int mode) +static int ramfs_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode) { int retval = ramfs_mknod(dir, dentry, mode | S_IFDIR, 0); if (!retval) @@ -114,7 +114,7 @@ static int ramfs_mkdir(struct inode * dir, struct dentry * dentry, int mode) return retval; } -static int ramfs_create(struct inode *dir, struct dentry *dentry, int mode, struct nameidata *nd) +static int ramfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { return ramfs_mknod(dir, dentry, mode | S_IFREG, 0); } diff --git a/fs/reiserfs/bitmap.c b/fs/reiserfs/bitmap.c index d1aca1df4f92..70de42f09f1d 100644 --- a/fs/reiserfs/bitmap.c +++ b/fs/reiserfs/bitmap.c @@ -13,6 +13,7 @@ #include <linux/reiserfs_fs_sb.h> #include <linux/reiserfs_fs_i.h> #include <linux/quotaops.h> +#include <linux/seq_file.h> #define PREALLOCATION_SIZE 9 @@ -634,6 +635,96 @@ int reiserfs_parse_alloc_options(struct super_block *s, char *options) return 0; } +static void print_sep(struct seq_file *seq, int *first) +{ + if (!*first) + seq_puts(seq, ":"); + else + *first = 0; +} + +void show_alloc_options(struct seq_file *seq, struct super_block *s) +{ + int first = 1; + + if (SB_ALLOC_OPTS(s) == ((1 << _ALLOC_skip_busy) | + (1 << _ALLOC_dirid_groups) | (1 << _ALLOC_packing_groups))) + return; + + seq_puts(seq, ",alloc="); + + if (TEST_OPTION(concentrating_formatted_nodes, s)) { + print_sep(seq, &first); + if (REISERFS_SB(s)->s_alloc_options.border != 10) { + seq_printf(seq, "concentrating_formatted_nodes=%d", + 100 / REISERFS_SB(s)->s_alloc_options.border); + } else + seq_puts(seq, "concentrating_formatted_nodes"); + } + if (TEST_OPTION(displacing_large_files, s)) { + print_sep(seq, &first); + if (REISERFS_SB(s)->s_alloc_options.large_file_size != 16) { + seq_printf(seq, "displacing_large_files=%lu", + REISERFS_SB(s)->s_alloc_options.large_file_size); + } else + seq_puts(seq, "displacing_large_files"); + } + if (TEST_OPTION(displacing_new_packing_localities, s)) { + print_sep(seq, &first); + seq_puts(seq, "displacing_new_packing_localities"); + } + if (TEST_OPTION(old_hashed_relocation, s)) { + print_sep(seq, &first); + seq_puts(seq, "old_hashed_relocation"); + } + if (TEST_OPTION(new_hashed_relocation, s)) { + print_sep(seq, &first); + seq_puts(seq, "new_hashed_relocation"); + } + if (TEST_OPTION(dirid_groups, s)) { + print_sep(seq, &first); + seq_puts(seq, "dirid_groups"); + } + if (TEST_OPTION(oid_groups, s)) { + print_sep(seq, &first); + seq_puts(seq, "oid_groups"); + } + if (TEST_OPTION(packing_groups, s)) { + print_sep(seq, &first); + seq_puts(seq, "packing_groups"); + } + if (TEST_OPTION(hashed_formatted_nodes, s)) { + print_sep(seq, &first); + seq_puts(seq, "hashed_formatted_nodes"); + } + if (TEST_OPTION(skip_busy, s)) { + print_sep(seq, &first); + seq_puts(seq, "skip_busy"); + } + if (TEST_OPTION(hundredth_slices, s)) { + print_sep(seq, &first); + seq_puts(seq, "hundredth_slices"); + } + if (TEST_OPTION(old_way, s)) { + print_sep(seq, &first); + seq_puts(seq, "old_way"); + } + if (TEST_OPTION(displace_based_on_dirid, s)) { + print_sep(seq, &first); + seq_puts(seq, "displace_based_on_dirid"); + } + if (REISERFS_SB(s)->s_alloc_options.preallocmin != 0) { + print_sep(seq, &first); + seq_printf(seq, "preallocmin=%d", + REISERFS_SB(s)->s_alloc_options.preallocmin); + } + if (REISERFS_SB(s)->s_alloc_options.preallocsize != 17) { + print_sep(seq, &first); + seq_printf(seq, "preallocsize=%d", + REISERFS_SB(s)->s_alloc_options.preallocsize); + } +} + static inline void new_hashed_relocation(reiserfs_blocknr_hint_t * hint) { char *hash_in; @@ -1273,10 +1364,7 @@ int reiserfs_init_bitmap_cache(struct super_block *sb) struct reiserfs_bitmap_info *bitmap; unsigned int bmap_nr = reiserfs_bmap_count(sb); - /* Avoid lock recursion in fault case */ - reiserfs_write_unlock(sb); bitmap = vmalloc(sizeof(*bitmap) * bmap_nr); - reiserfs_write_lock(sb); if (bitmap == NULL) return -ENOMEM; diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 950f13af0951..9e8cd5acd79c 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -1766,7 +1766,7 @@ static int reiserfs_new_symlink(struct reiserfs_transaction_handle *th, struct i for the fresh inode. This can only be done outside a transaction, so if we return non-zero, we also end the transaction. */ int reiserfs_new_inode(struct reiserfs_transaction_handle *th, - struct inode *dir, int mode, const char *symname, + struct inode *dir, umode_t mode, const char *symname, /* 0 for regular, EMTRY_DIR_SIZE for dirs, strlen (symname) for symlinks) */ loff_t i_size, struct dentry *dentry, diff --git a/fs/reiserfs/ioctl.c b/fs/reiserfs/ioctl.c index 4e153051bc75..950e3d1b5c9e 100644 --- a/fs/reiserfs/ioctl.c +++ b/fs/reiserfs/ioctl.c @@ -55,7 +55,7 @@ long reiserfs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) break; } - err = mnt_want_write(filp->f_path.mnt); + err = mnt_want_write_file(filp); if (err) break; @@ -96,7 +96,7 @@ long reiserfs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) inode->i_ctime = CURRENT_TIME_SEC; mark_inode_dirty(inode); setflags_out: - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); break; } case REISERFS_IOC_GETVERSION: @@ -107,7 +107,7 @@ setflags_out: err = -EPERM; break; } - err = mnt_want_write(filp->f_path.mnt); + err = mnt_want_write_file(filp); if (err) break; if (get_user(inode->i_generation, (int __user *)arg)) { @@ -117,7 +117,7 @@ setflags_out: inode->i_ctime = CURRENT_TIME_SEC; mark_inode_dirty(inode); setversion_out: - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); break; default: err = -ENOTTY; diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index eb711060a6f2..c3cf54fd4de3 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -2678,16 +2678,10 @@ int journal_init(struct super_block *sb, const char *j_dev_name, char b[BDEVNAME_SIZE]; int ret; - /* - * Unlock here to avoid various RECLAIM-FS-ON <-> IN-RECLAIM-FS - * dependency inversion warnings. - */ - reiserfs_write_unlock(sb); journal = SB_JOURNAL(sb) = vzalloc(sizeof(struct reiserfs_journal)); if (!journal) { reiserfs_warning(sb, "journal-1256", "unable to get memory for journal structure"); - reiserfs_write_lock(sb); return 1; } INIT_LIST_HEAD(&journal->j_bitmap_nodes); @@ -2695,10 +2689,8 @@ int journal_init(struct super_block *sb, const char *j_dev_name, INIT_LIST_HEAD(&journal->j_working_list); INIT_LIST_HEAD(&journal->j_journal_list); journal->j_persistent_trans = 0; - ret = reiserfs_allocate_list_bitmaps(sb, journal->j_list_bitmap, - reiserfs_bmap_count(sb)); - reiserfs_write_lock(sb); - if (ret) + if (reiserfs_allocate_list_bitmaps(sb, journal->j_list_bitmap, + reiserfs_bmap_count(sb))) goto free_and_return; allocate_bitmap_nodes(sb); @@ -2727,27 +2719,11 @@ int journal_init(struct super_block *sb, const char *j_dev_name, goto free_and_return; } - /* - * We need to unlock here to avoid creating the following - * dependency: - * reiserfs_lock -> sysfs_mutex - * Because the reiserfs mmap path creates the following dependency: - * mm->mmap -> reiserfs_lock, hence we have - * mm->mmap -> reiserfs_lock ->sysfs_mutex - * This would ends up in a circular dependency with sysfs readdir path - * which does sysfs_mutex -> mm->mmap_sem - * This is fine because the reiserfs lock is useless in mount path, - * at least until we call journal_begin. We keep it for paranoid - * reasons. - */ - reiserfs_write_unlock(sb); if (journal_init_dev(sb, journal, j_dev_name) != 0) { - reiserfs_write_lock(sb); reiserfs_warning(sb, "sh-462", "unable to initialize jornal device"); goto free_and_return; } - reiserfs_write_lock(sb); rs = SB_DISK_SUPER_BLOCK(sb); @@ -2829,9 +2805,7 @@ int journal_init(struct super_block *sb, const char *j_dev_name, journal->j_mount_id = 10; journal->j_state = 0; atomic_set(&(journal->j_jlock), 0); - reiserfs_write_unlock(sb); journal->j_cnode_free_list = allocate_cnodes(num_cnodes); - reiserfs_write_lock(sb); journal->j_cnode_free_orig = journal->j_cnode_free_list; journal->j_cnode_free = journal->j_cnode_free_list ? num_cnodes : 0; journal->j_cnode_used = 0; @@ -2848,24 +2822,37 @@ int journal_init(struct super_block *sb, const char *j_dev_name, init_journal_hash(sb); jl = journal->j_current_jl; + + /* + * get_list_bitmap() may call flush_commit_list() which + * requires the lock. Calling flush_commit_list() shouldn't happen + * this early but I like to be paranoid. + */ + reiserfs_write_lock(sb); jl->j_list_bitmap = get_list_bitmap(sb, jl); + reiserfs_write_unlock(sb); if (!jl->j_list_bitmap) { reiserfs_warning(sb, "journal-2005", "get_list_bitmap failed for journal list 0"); goto free_and_return; } - if (journal_read(sb) < 0) { + + /* + * Journal_read needs to be inspected in order to push down + * the lock further inside (or even remove it). + */ + reiserfs_write_lock(sb); + ret = journal_read(sb); + reiserfs_write_unlock(sb); + if (ret < 0) { reiserfs_warning(sb, "reiserfs-2006", "Replay Failure, unable to mount"); goto free_and_return; } reiserfs_mounted_fs_count++; - if (reiserfs_mounted_fs_count <= 1) { - reiserfs_write_unlock(sb); + if (reiserfs_mounted_fs_count <= 1) commit_wq = alloc_workqueue("reiserfs", WQ_MEM_RECLAIM, 0); - reiserfs_write_lock(sb); - } INIT_DELAYED_WORK(&journal->j_work, flush_async_commits); journal->j_work_sb = sb; @@ -2896,14 +2883,13 @@ int journal_transaction_should_end(struct reiserfs_transaction_handle *th, journal->j_cnode_free < (journal->j_trans_max * 3)) { return 1; } - /* protected by the BKL here */ + journal->j_len_alloc += new_alloc; th->t_blocks_allocated += new_alloc ; return 0; } -/* this must be called inside a transaction, and requires the -** kernel_lock to be held +/* this must be called inside a transaction */ void reiserfs_block_writes(struct reiserfs_transaction_handle *th) { @@ -2914,8 +2900,7 @@ void reiserfs_block_writes(struct reiserfs_transaction_handle *th) return; } -/* this must be called without a transaction started, and does not -** require BKL +/* this must be called without a transaction started */ void reiserfs_allow_writes(struct super_block *s) { @@ -2924,8 +2909,7 @@ void reiserfs_allow_writes(struct super_block *s) wake_up(&journal->j_join_wait); } -/* this must be called without a transaction started, and does not -** require BKL +/* this must be called without a transaction started */ void reiserfs_wait_on_write_block(struct super_block *s) { diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c index 80058e8ce361..146378865239 100644 --- a/fs/reiserfs/namei.c +++ b/fs/reiserfs/namei.c @@ -559,7 +559,7 @@ static int drop_new_inode(struct inode *inode) ** outside of a transaction, so we had to pull some bits of ** reiserfs_new_inode out into this func. */ -static int new_inode_init(struct inode *inode, struct inode *dir, int mode) +static int new_inode_init(struct inode *inode, struct inode *dir, umode_t mode) { /* Make inode invalid - just in case we are going to drop it before * the initialization happens */ @@ -572,7 +572,7 @@ static int new_inode_init(struct inode *inode, struct inode *dir, int mode) return 0; } -static int reiserfs_create(struct inode *dir, struct dentry *dentry, int mode, +static int reiserfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { int retval; @@ -643,7 +643,7 @@ static int reiserfs_create(struct inode *dir, struct dentry *dentry, int mode, return retval; } -static int reiserfs_mknod(struct inode *dir, struct dentry *dentry, int mode, +static int reiserfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) { int retval; @@ -721,7 +721,7 @@ static int reiserfs_mknod(struct inode *dir, struct dentry *dentry, int mode, return retval; } -static int reiserfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int reiserfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { int retval; struct inode *inode; diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 14363b96b6af..e12d8b97cd4d 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -28,6 +28,7 @@ #include <linux/mount.h> #include <linux/namei.h> #include <linux/crc32.h> +#include <linux/seq_file.h> struct file_system_type reiserfs_fs_type; @@ -61,6 +62,7 @@ static int is_any_reiserfs_magic_string(struct reiserfs_super_block *rs) static int reiserfs_remount(struct super_block *s, int *flags, char *data); static int reiserfs_statfs(struct dentry *dentry, struct kstatfs *buf); +void show_alloc_options(struct seq_file *seq, struct super_block *s); static int reiserfs_sync_fs(struct super_block *s, int wait) { @@ -453,16 +455,20 @@ int remove_save_link(struct inode *inode, int truncate) static void reiserfs_kill_sb(struct super_block *s) { if (REISERFS_SB(s)) { - if (REISERFS_SB(s)->xattr_root) { - d_invalidate(REISERFS_SB(s)->xattr_root); - dput(REISERFS_SB(s)->xattr_root); - REISERFS_SB(s)->xattr_root = NULL; - } - if (REISERFS_SB(s)->priv_root) { - d_invalidate(REISERFS_SB(s)->priv_root); - dput(REISERFS_SB(s)->priv_root); - REISERFS_SB(s)->priv_root = NULL; - } + /* + * Force any pending inode evictions to occur now. Any + * inodes to be removed that have extended attributes + * associated with them need to clean them up before + * we can release the extended attribute root dentries. + * shrink_dcache_for_umount will BUG if we don't release + * those before it's called so ->put_super is too late. + */ + shrink_dcache_sb(s); + + dput(REISERFS_SB(s)->xattr_root); + REISERFS_SB(s)->xattr_root = NULL; + dput(REISERFS_SB(s)->priv_root); + REISERFS_SB(s)->priv_root = NULL; } kill_block_super(s); @@ -532,7 +538,6 @@ static struct inode *reiserfs_alloc_inode(struct super_block *sb) static void reiserfs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(reiserfs_inode_cachep, REISERFS_I(inode)); } @@ -597,6 +602,82 @@ out: reiserfs_write_unlock_once(inode->i_sb, lock_depth); } +static int reiserfs_show_options(struct seq_file *seq, struct dentry *root) +{ + struct super_block *s = root->d_sb; + struct reiserfs_journal *journal = SB_JOURNAL(s); + long opts = REISERFS_SB(s)->s_mount_opt; + + if (opts & (1 << REISERFS_LARGETAIL)) + seq_puts(seq, ",tails=on"); + else if (!(opts & (1 << REISERFS_SMALLTAIL))) + seq_puts(seq, ",notail"); + /* tails=small is default so we don't show it */ + + if (!(opts & (1 << REISERFS_BARRIER_FLUSH))) + seq_puts(seq, ",barrier=none"); + /* barrier=flush is default so we don't show it */ + + if (opts & (1 << REISERFS_ERROR_CONTINUE)) + seq_puts(seq, ",errors=continue"); + else if (opts & (1 << REISERFS_ERROR_PANIC)) + seq_puts(seq, ",errors=panic"); + /* errors=ro is default so we don't show it */ + + if (opts & (1 << REISERFS_DATA_LOG)) + seq_puts(seq, ",data=journal"); + else if (opts & (1 << REISERFS_DATA_WRITEBACK)) + seq_puts(seq, ",data=writeback"); + /* data=ordered is default so we don't show it */ + + if (opts & (1 << REISERFS_ATTRS)) + seq_puts(seq, ",attrs"); + + if (opts & (1 << REISERFS_XATTRS_USER)) + seq_puts(seq, ",user_xattr"); + + if (opts & (1 << REISERFS_EXPOSE_PRIVROOT)) + seq_puts(seq, ",expose_privroot"); + + if (opts & (1 << REISERFS_POSIXACL)) + seq_puts(seq, ",acl"); + + if (REISERFS_SB(s)->s_jdev) + seq_printf(seq, ",jdev=%s", REISERFS_SB(s)->s_jdev); + + if (journal->j_max_commit_age != journal->j_default_max_commit_age) + seq_printf(seq, ",commit=%d", journal->j_max_commit_age); + +#ifdef CONFIG_QUOTA + if (REISERFS_SB(s)->s_qf_names[USRQUOTA]) + seq_printf(seq, ",usrjquota=%s", REISERFS_SB(s)->s_qf_names[USRQUOTA]); + else if (opts & (1 << REISERFS_USRQUOTA)) + seq_puts(seq, ",usrquota"); + if (REISERFS_SB(s)->s_qf_names[GRPQUOTA]) + seq_printf(seq, ",grpjquota=%s", REISERFS_SB(s)->s_qf_names[GRPQUOTA]); + else if (opts & (1 << REISERFS_GRPQUOTA)) + seq_puts(seq, ",grpquota"); + if (REISERFS_SB(s)->s_jquota_fmt) { + if (REISERFS_SB(s)->s_jquota_fmt == QFMT_VFS_OLD) + seq_puts(seq, ",jqfmt=vfsold"); + else if (REISERFS_SB(s)->s_jquota_fmt == QFMT_VFS_V0) + seq_puts(seq, ",jqfmt=vfsv0"); + } +#endif + + /* Block allocator options */ + if (opts & (1 << REISERFS_NO_BORDER)) + seq_puts(seq, ",block-allocator=noborder"); + if (opts & (1 << REISERFS_NO_UNHASHED_RELOCATION)) + seq_puts(seq, ",block-allocator=no_unhashed_relocation"); + if (opts & (1 << REISERFS_HASHED_RELOCATION)) + seq_puts(seq, ",block-allocator=hashed_relocation"); + if (opts & (1 << REISERFS_TEST4)) + seq_puts(seq, ",block-allocator=test4"); + show_alloc_options(seq, s); + return 0; +} + #ifdef CONFIG_QUOTA static ssize_t reiserfs_quota_write(struct super_block *, int, const char *, size_t, loff_t); @@ -617,7 +698,7 @@ static const struct super_operations reiserfs_sops = { .unfreeze_fs = reiserfs_unfreeze, .statfs = reiserfs_statfs, .remount_fs = reiserfs_remount, - .show_options = generic_show_options, + .show_options = reiserfs_show_options, #ifdef CONFIG_QUOTA .quota_read = reiserfs_quota_read, .quota_write = reiserfs_quota_write, @@ -915,9 +996,9 @@ static int reiserfs_parse_options(struct super_block *s, char *options, /* strin {"jdev",.arg_required = 'j',.values = NULL}, {"nolargeio",.arg_required = 'w',.values = NULL}, {"commit",.arg_required = 'c',.values = NULL}, - {"usrquota",.setmask = 1 << REISERFS_QUOTA}, - {"grpquota",.setmask = 1 << REISERFS_QUOTA}, - {"noquota",.clrmask = 1 << REISERFS_QUOTA}, + {"usrquota",.setmask = 1 << REISERFS_USRQUOTA}, + {"grpquota",.setmask = 1 << REISERFS_GRPQUOTA}, + {"noquota",.clrmask = 1 << REISERFS_USRQUOTA | 1 << REISERFS_GRPQUOTA}, {"errors",.arg_required = 'e',.values = error_actions}, {"usrjquota",.arg_required = 'u' | (1 << REISERFS_OPT_ALLOWEMPTY),.values = NULL}, @@ -1031,12 +1112,19 @@ static int reiserfs_parse_options(struct super_block *s, char *options, /* strin return 0; } strcpy(qf_names[qtype], arg); - *mount_options |= 1 << REISERFS_QUOTA; + if (qtype == USRQUOTA) + *mount_options |= 1 << REISERFS_USRQUOTA; + else + *mount_options |= 1 << REISERFS_GRPQUOTA; } else { if (qf_names[qtype] != REISERFS_SB(s)->s_qf_names[qtype]) kfree(qf_names[qtype]); qf_names[qtype] = NULL; + if (qtype == USRQUOTA) + *mount_options &= ~(1 << REISERFS_USRQUOTA); + else + *mount_options &= ~(1 << REISERFS_GRPQUOTA); } } if (c == 'f') { @@ -1075,9 +1163,10 @@ static int reiserfs_parse_options(struct super_block *s, char *options, /* strin "journaled quota format not specified."); return 0; } - /* This checking is not precise wrt the quota type but for our purposes it is sufficient */ - if (!(*mount_options & (1 << REISERFS_QUOTA)) - && sb_any_quota_loaded(s)) { + if ((!(*mount_options & (1 << REISERFS_USRQUOTA)) && + sb_has_quota_loaded(s, USRQUOTA)) || + (!(*mount_options & (1 << REISERFS_GRPQUOTA)) && + sb_has_quota_loaded(s, GRPQUOTA))) { reiserfs_warning(s, "super-6516", "quota options must " "be present when quota is turned on."); return 0; @@ -1164,7 +1253,8 @@ static void handle_quota_files(struct super_block *s, char **qf_names, kfree(REISERFS_SB(s)->s_qf_names[i]); REISERFS_SB(s)->s_qf_names[i] = qf_names[i]; } - REISERFS_SB(s)->s_jquota_fmt = *qfmt; + if (*qfmt) + REISERFS_SB(s)->s_jquota_fmt = *qfmt; } #endif @@ -1225,7 +1315,8 @@ static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg) safe_mask |= 1 << REISERFS_ERROR_RO; safe_mask |= 1 << REISERFS_ERROR_CONTINUE; safe_mask |= 1 << REISERFS_ERROR_PANIC; - safe_mask |= 1 << REISERFS_QUOTA; + safe_mask |= 1 << REISERFS_USRQUOTA; + safe_mask |= 1 << REISERFS_GRPQUOTA; /* Update the bitmask, taking care to keep * the bits we're not allowed to change here */ @@ -1428,9 +1519,7 @@ static int read_super_block(struct super_block *s, int offset) static int reread_meta_blocks(struct super_block *s) { ll_rw_block(READ, 1, &(SB_BUFFER_WITH_SB(s))); - reiserfs_write_unlock(s); wait_on_buffer(SB_BUFFER_WITH_SB(s)); - reiserfs_write_lock(s); if (!buffer_uptodate(SB_BUFFER_WITH_SB(s))) { reiserfs_warning(s, "reiserfs-2504", "error reading the super"); return 1; @@ -1655,22 +1744,19 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent) mutex_init(&REISERFS_SB(s)->lock); REISERFS_SB(s)->lock_depth = -1; - /* - * This function is called with the bkl, which also was the old - * locking used here. - * do_journal_begin() will soon check if we hold the lock (ie: was the - * bkl). This is likely because do_journal_begin() has several another - * callers because at this time, it doesn't seem to be necessary to - * protect against anything. - * Anyway, let's be conservative and lock for now. - */ - reiserfs_write_lock(s); - jdev_name = NULL; if (reiserfs_parse_options (s, (char *)data, &(sbi->s_mount_opt), &blocks, &jdev_name, &commit_max_age, qf_names, &qfmt) == 0) { - goto error; + goto error_unlocked; + } + if (jdev_name && jdev_name[0]) { + REISERFS_SB(s)->s_jdev = kstrdup(jdev_name, GFP_KERNEL); + if (!REISERFS_SB(s)->s_jdev) { + SWARN(silent, s, "", "Cannot allocate memory for " + "journal device name"); + goto error; + } } #ifdef CONFIG_QUOTA handle_quota_files(s, qf_names, &qfmt); @@ -1678,7 +1764,7 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent) if (blocks) { SWARN(silent, s, "jmacd-7", "resize option for remount only"); - goto error; + goto error_unlocked; } /* try old format (undistributed bitmap, super block in 8-th 1k block of a device) */ @@ -1688,7 +1774,7 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent) else if (read_super_block(s, REISERFS_DISK_OFFSET_IN_BYTES)) { SWARN(silent, s, "sh-2021", "can not find reiserfs on %s", reiserfs_bdevname(s)); - goto error; + goto error_unlocked; } rs = SB_DISK_SUPER_BLOCK(s); @@ -1704,7 +1790,7 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent) "or increase size of your LVM partition"); SWARN(silent, s, "", "Or may be you forgot to " "reboot after fdisk when it told you to"); - goto error; + goto error_unlocked; } sbi->s_mount_state = SB_REISERFS_STATE(s); @@ -1712,8 +1798,9 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent) if ((errval = reiserfs_init_bitmap_cache(s))) { SWARN(silent, s, "jmacd-8", "unable to read bitmap"); - goto error; + goto error_unlocked; } + errval = -EINVAL; #ifdef CONFIG_REISERFS_CHECK SWARN(silent, s, "", "CONFIG_REISERFS_CHECK is set ON"); @@ -1736,24 +1823,26 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent) if (reiserfs_barrier_flush(s)) { printk("reiserfs: using flush barriers\n"); } + // set_device_ro(s->s_dev, 1) ; if (journal_init(s, jdev_name, old_format, commit_max_age)) { SWARN(silent, s, "sh-2022", "unable to initialize journal space"); - goto error; + goto error_unlocked; } else { jinit_done = 1; /* once this is set, journal_release must be called ** if we error out of the mount */ } + if (reread_meta_blocks(s)) { SWARN(silent, s, "jmacd-9", "unable to reread meta blocks after journal init"); - goto error; + goto error_unlocked; } if (replay_only(s)) - goto error; + goto error_unlocked; if (bdev_read_only(s->s_bdev) && !(s->s_flags & MS_RDONLY)) { SWARN(silent, s, "clm-7000", @@ -1767,9 +1856,19 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent) reiserfs_init_locked_inode, (void *)(&args)); if (!root_inode) { SWARN(silent, s, "jmacd-10", "get root inode failed"); - goto error; + goto error_unlocked; } + /* + * This path assumed to be called with the BKL in the old times. + * Now we have inherited the big reiserfs lock from it and many + * reiserfs helpers called in the mount path and elsewhere require + * this lock to be held even if it's not always necessary. Let's be + * conservative and hold it early. The window can be reduced after + * careful review of the code. + */ + reiserfs_write_lock(s); + if (root_inode->i_state & I_NEW) { reiserfs_read_locked_inode(root_inode, &args); unlock_new_inode(root_inode); @@ -1896,12 +1995,16 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent) return (0); error: - if (jinit_done) { /* kill the commit thread, free journal ram */ + reiserfs_write_unlock(s); + +error_unlocked: + /* kill the commit thread, free journal ram */ + if (jinit_done) { + reiserfs_write_lock(s); journal_release_error(NULL, s); + reiserfs_write_unlock(s); } - reiserfs_write_unlock(s); - reiserfs_free_bitmap_cache(s); if (SB_BUFFER_WITH_SB(s)) brelse(SB_BUFFER_WITH_SB(s)); @@ -2054,12 +2157,13 @@ static int reiserfs_quota_on(struct super_block *sb, int type, int format_id, int err; struct inode *inode; struct reiserfs_transaction_handle th; + int opt = type == USRQUOTA ? REISERFS_USRQUOTA : REISERFS_GRPQUOTA; - if (!(REISERFS_SB(sb)->s_mount_opt & (1 << REISERFS_QUOTA))) + if (!(REISERFS_SB(sb)->s_mount_opt & (1 << opt))) return -EINVAL; /* Quotafile not on the same filesystem? */ - if (path->mnt->mnt_sb != sb) { + if (path->dentry->d_sb != sb) { err = -EXDEV; goto out; } diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index 6bc346c160e7..c24deda8a8bc 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -66,7 +66,7 @@ static int xattr_create(struct inode *dir, struct dentry *dentry, int mode) } #endif -static int xattr_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int xattr_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { BUG_ON(!mutex_is_locked(&dir->i_mutex)); return dir->i_op->mkdir(dir, dentry, mode); diff --git a/fs/romfs/mmap-nommu.c b/fs/romfs/mmap-nommu.c index eed99428f104..e1a7779dd3cb 100644 --- a/fs/romfs/mmap-nommu.c +++ b/fs/romfs/mmap-nommu.c @@ -28,9 +28,10 @@ static unsigned long romfs_get_unmapped_area(struct file *file, struct inode *inode = file->f_mapping->host; struct mtd_info *mtd = inode->i_sb->s_mtd; unsigned long isize, offset, maxpages, lpages; + int ret; if (!mtd) - goto cant_map_directly; + return (unsigned long) -ENOSYS; /* the mapping mustn't extend beyond the EOF */ lpages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT; @@ -41,23 +42,20 @@ static unsigned long romfs_get_unmapped_area(struct file *file, if ((pgoff >= maxpages) || (maxpages - pgoff < lpages)) return (unsigned long) -EINVAL; - /* we need to call down to the MTD layer to do the actual mapping */ - if (mtd->get_unmapped_area) { - if (addr != 0) - return (unsigned long) -EINVAL; - - if (len > mtd->size || pgoff >= (mtd->size >> PAGE_SHIFT)) - return (unsigned long) -EINVAL; + if (addr != 0) + return (unsigned long) -EINVAL; - offset += ROMFS_I(inode)->i_dataoffset; - if (offset > mtd->size - len) - return (unsigned long) -EINVAL; + if (len > mtd->size || pgoff >= (mtd->size >> PAGE_SHIFT)) + return (unsigned long) -EINVAL; - return mtd->get_unmapped_area(mtd, len, offset, flags); - } + offset += ROMFS_I(inode)->i_dataoffset; + if (offset > mtd->size - len) + return (unsigned long) -EINVAL; -cant_map_directly: - return (unsigned long) -ENOSYS; + ret = mtd_get_unmapped_area(mtd, len, offset, flags); + if (ret == -EOPNOTSUPP) + ret = -ENOSYS; + return (unsigned long) ret; } /* diff --git a/fs/romfs/super.c b/fs/romfs/super.c index 8b4089f30408..bb36ab74eb45 100644 --- a/fs/romfs/super.c +++ b/fs/romfs/super.c @@ -403,7 +403,6 @@ static struct inode *romfs_alloc_inode(struct super_block *sb) static void romfs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(romfs_inode_cachep, ROMFS_I(inode)); } diff --git a/fs/seq_file.c b/fs/seq_file.c index 05d6b0e78c95..4023d6be939b 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -397,7 +397,7 @@ EXPORT_SYMBOL(seq_printf); * Returns pointer past last written character in @s, or NULL in case of * failure. */ -char *mangle_path(char *s, char *p, char *esc) +char *mangle_path(char *s, const char *p, const char *esc) { while (s <= p) { char c = *p++; @@ -427,7 +427,7 @@ EXPORT_SYMBOL(mangle_path); * return the absolute path of 'path', as represented by the * dentry / mnt pair in the path parameter. */ -int seq_path(struct seq_file *m, struct path *path, char *esc) +int seq_path(struct seq_file *m, const struct path *path, const char *esc) { char *buf; size_t size = seq_get_buf(m, &buf); @@ -449,11 +449,9 @@ EXPORT_SYMBOL(seq_path); /* * Same as seq_path, but relative to supplied root. - * - * root may be changed, see __d_path(). */ -int seq_path_root(struct seq_file *m, struct path *path, struct path *root, - char *esc) +int seq_path_root(struct seq_file *m, const struct path *path, + const struct path *root, const char *esc) { char *buf; size_t size = seq_get_buf(m, &buf); @@ -463,6 +461,8 @@ int seq_path_root(struct seq_file *m, struct path *path, struct path *root, char *p; p = __d_path(path, root, buf, size); + if (!p) + return SEQ_SKIP; res = PTR_ERR(p); if (!IS_ERR(p)) { char *end = mangle_path(buf, p, esc); @@ -474,13 +474,13 @@ int seq_path_root(struct seq_file *m, struct path *path, struct path *root, } seq_commit(m, res); - return res < 0 ? res : 0; + return res < 0 && res != -ENAMETOOLONG ? res : 0; } /* * returns the path of the 'dentry' from the root of its filesystem. */ -int seq_dentry(struct seq_file *m, struct dentry *dentry, char *esc) +int seq_dentry(struct seq_file *m, struct dentry *dentry, const char *esc) { char *buf; size_t size = seq_get_buf(m, &buf); diff --git a/fs/splice.c b/fs/splice.c index fa2defa8afcf..1ec0493266b3 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -25,7 +25,6 @@ #include <linux/mm_inline.h> #include <linux/swap.h> #include <linux/writeback.h> -#include <linux/buffer_head.h> #include <linux/module.h> #include <linux/syscalls.h> #include <linux/uio.h> diff --git a/fs/squashfs/cache.c b/fs/squashfs/cache.c index f744be98cd5a..af0b73802592 100644 --- a/fs/squashfs/cache.c +++ b/fs/squashfs/cache.c @@ -70,11 +70,15 @@ struct squashfs_cache_entry *squashfs_cache_get(struct super_block *sb, spin_lock(&cache->lock); while (1) { - for (i = 0; i < cache->entries; i++) - if (cache->entry[i].block == block) + for (i = cache->curr_blk, n = 0; n < cache->entries; n++) { + if (cache->entry[i].block == block) { + cache->curr_blk = i; break; + } + i = (i + 1) % cache->entries; + } - if (i == cache->entries) { + if (n == cache->entries) { /* * Block not in cache, if all cache entries are used * go to sleep waiting for one to become available. @@ -245,6 +249,7 @@ struct squashfs_cache *squashfs_cache_init(char *name, int entries, goto cleanup; } + cache->curr_blk = 0; cache->next_blk = 0; cache->unused = entries; cache->entries = entries; @@ -332,17 +337,20 @@ int squashfs_read_metadata(struct super_block *sb, void *buffer, u64 *block, int *offset, int length) { struct squashfs_sb_info *msblk = sb->s_fs_info; - int bytes, copied = length; + int bytes, res = length; struct squashfs_cache_entry *entry; TRACE("Entered squashfs_read_metadata [%llx:%x]\n", *block, *offset); while (length) { entry = squashfs_cache_get(sb, msblk->block_cache, *block, 0); - if (entry->error) - return entry->error; - else if (*offset >= entry->length) - return -EIO; + if (entry->error) { + res = entry->error; + goto error; + } else if (*offset >= entry->length) { + res = -EIO; + goto error; + } bytes = squashfs_copy_data(buffer, entry, *offset, length); if (buffer) @@ -358,7 +366,11 @@ int squashfs_read_metadata(struct super_block *sb, void *buffer, squashfs_cache_put(entry); } - return copied; + return res; + +error: + squashfs_cache_put(entry); + return res; } diff --git a/fs/squashfs/inode.c b/fs/squashfs/inode.c index fd7b3b3bda13..81afbccfa843 100644 --- a/fs/squashfs/inode.c +++ b/fs/squashfs/inode.c @@ -208,8 +208,8 @@ int squashfs_read_inode(struct inode *inode, long long ino) inode->i_op = &squashfs_inode_ops; inode->i_fop = &generic_ro_fops; inode->i_mode |= S_IFREG; - inode->i_blocks = ((inode->i_size - - le64_to_cpu(sqsh_ino->sparse) - 1) >> 9) + 1; + inode->i_blocks = (inode->i_size - + le64_to_cpu(sqsh_ino->sparse) + 511) >> 9; squashfs_i(inode)->fragment_block = frag_blk; squashfs_i(inode)->fragment_size = frag_size; diff --git a/fs/squashfs/squashfs_fs_sb.h b/fs/squashfs/squashfs_fs_sb.h index 651f0b31d296..52934a22f296 100644 --- a/fs/squashfs/squashfs_fs_sb.h +++ b/fs/squashfs/squashfs_fs_sb.h @@ -28,6 +28,7 @@ struct squashfs_cache { char *name; int entries; + int curr_blk; int next_blk; int num_waiters; int unused; diff --git a/fs/squashfs/super.c b/fs/squashfs/super.c index 2da1715452ac..ecaa2f7bdb8f 100644 --- a/fs/squashfs/super.c +++ b/fs/squashfs/super.c @@ -290,7 +290,7 @@ handle_fragments: check_directory_table: /* Sanity check directory_table */ - if (msblk->directory_table >= next_table) { + if (msblk->directory_table > next_table) { err = -EINVAL; goto failed_mount; } @@ -464,7 +464,6 @@ static struct inode *squashfs_alloc_inode(struct super_block *sb) static void squashfs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(squashfs_inode_cachep, squashfs_i(inode)); } diff --git a/fs/statfs.c b/fs/statfs.c index 9cf04a118965..2aa6a22e0be2 100644 --- a/fs/statfs.c +++ b/fs/statfs.c @@ -7,6 +7,7 @@ #include <linux/statfs.h> #include <linux/security.h> #include <linux/uaccess.h> +#include "internal.h" static int flags_by_mnt(int mnt_flags) { @@ -45,7 +46,7 @@ static int calculate_f_flags(struct vfsmount *mnt) flags_by_sb(mnt->mnt_sb->s_flags); } -int statfs_by_dentry(struct dentry *dentry, struct kstatfs *buf) +static int statfs_by_dentry(struct dentry *dentry, struct kstatfs *buf) { int retval; @@ -205,19 +206,23 @@ SYSCALL_DEFINE3(fstatfs64, unsigned int, fd, size_t, sz, struct statfs64 __user return error; } -SYSCALL_DEFINE2(ustat, unsigned, dev, struct ustat __user *, ubuf) +int vfs_ustat(dev_t dev, struct kstatfs *sbuf) { - struct super_block *s; - struct ustat tmp; - struct kstatfs sbuf; + struct super_block *s = user_get_super(dev); int err; - - s = user_get_super(new_decode_dev(dev)); if (!s) return -EINVAL; - err = statfs_by_dentry(s->s_root, &sbuf); + err = statfs_by_dentry(s->s_root, sbuf); drop_super(s); + return err; +} + +SYSCALL_DEFINE2(ustat, unsigned, dev, struct ustat __user *, ubuf) +{ + struct ustat tmp; + struct kstatfs sbuf; + int err = vfs_ustat(new_decode_dev(dev), &sbuf); if (err) return err; diff --git a/fs/super.c b/fs/super.c index afd0f1ad45e0..de41e1e46f09 100644 --- a/fs/super.c +++ b/fs/super.c @@ -136,12 +136,13 @@ static struct super_block *alloc_super(struct file_system_type *type) INIT_LIST_HEAD(&s->s_files); #endif s->s_bdi = &default_backing_dev_info; - INIT_LIST_HEAD(&s->s_instances); + INIT_HLIST_NODE(&s->s_instances); INIT_HLIST_BL_HEAD(&s->s_anon); INIT_LIST_HEAD(&s->s_inodes); INIT_LIST_HEAD(&s->s_dentry_lru); INIT_LIST_HEAD(&s->s_inode_lru); spin_lock_init(&s->s_inode_lru_lock); + INIT_LIST_HEAD(&s->s_mounts); init_rwsem(&s->s_umount); mutex_init(&s->s_lock); lockdep_set_class(&s->s_umount, &type->s_umount_key); @@ -200,6 +201,7 @@ static inline void destroy_super(struct super_block *s) free_percpu(s->s_files); #endif security_sb_free(s); + WARN_ON(!list_empty(&s->s_mounts)); kfree(s->s_subtype); kfree(s->s_options); kfree(s); @@ -210,7 +212,7 @@ static inline void destroy_super(struct super_block *s) /* * Drop a superblock's refcount. The caller must hold sb_lock. */ -void __put_super(struct super_block *sb) +static void __put_super(struct super_block *sb) { if (!--sb->s_count) { list_del_init(&sb->s_list); @@ -225,7 +227,7 @@ void __put_super(struct super_block *sb) * Drops a temporary reference, frees superblock if there's no * references left. */ -void put_super(struct super_block *sb) +static void put_super(struct super_block *sb) { spin_lock(&sb_lock); __put_super(sb); @@ -328,7 +330,7 @@ static int grab_super(struct super_block *s) __releases(sb_lock) bool grab_super_passive(struct super_block *sb) { spin_lock(&sb_lock); - if (list_empty(&sb->s_instances)) { + if (hlist_unhashed(&sb->s_instances)) { spin_unlock(&sb_lock); return false; } @@ -337,7 +339,7 @@ bool grab_super_passive(struct super_block *sb) spin_unlock(&sb_lock); if (down_read_trylock(&sb->s_umount)) { - if (sb->s_root) + if (sb->s_root && (sb->s_flags & MS_BORN)) return true; up_read(&sb->s_umount); } @@ -400,7 +402,7 @@ void generic_shutdown_super(struct super_block *sb) } spin_lock(&sb_lock); /* should be initialized for __put_super_and_need_restart() */ - list_del_init(&sb->s_instances); + hlist_del_init(&sb->s_instances); spin_unlock(&sb_lock); up_write(&sb->s_umount); } @@ -420,13 +422,14 @@ struct super_block *sget(struct file_system_type *type, void *data) { struct super_block *s = NULL; + struct hlist_node *node; struct super_block *old; int err; retry: spin_lock(&sb_lock); if (test) { - list_for_each_entry(old, &type->fs_supers, s_instances) { + hlist_for_each_entry(old, node, &type->fs_supers, s_instances) { if (!test(old, data)) continue; if (!grab_super(old)) @@ -462,7 +465,7 @@ retry: s->s_type = type; strlcpy(s->s_id, type->name, sizeof(s->s_id)); list_add_tail(&s->s_list, &super_blocks); - list_add(&s->s_instances, &type->fs_supers); + hlist_add_head(&s->s_instances, &type->fs_supers); spin_unlock(&sb_lock); get_filesystem(type); register_shrinker(&s->s_shrink); @@ -497,14 +500,14 @@ void sync_supers(void) spin_lock(&sb_lock); list_for_each_entry(sb, &super_blocks, s_list) { - if (list_empty(&sb->s_instances)) + if (hlist_unhashed(&sb->s_instances)) continue; if (sb->s_op->write_super && sb->s_dirt) { sb->s_count++; spin_unlock(&sb_lock); down_read(&sb->s_umount); - if (sb->s_root && sb->s_dirt) + if (sb->s_root && sb->s_dirt && (sb->s_flags & MS_BORN)) sb->s_op->write_super(sb); up_read(&sb->s_umount); @@ -533,13 +536,13 @@ void iterate_supers(void (*f)(struct super_block *, void *), void *arg) spin_lock(&sb_lock); list_for_each_entry(sb, &super_blocks, s_list) { - if (list_empty(&sb->s_instances)) + if (hlist_unhashed(&sb->s_instances)) continue; sb->s_count++; spin_unlock(&sb_lock); down_read(&sb->s_umount); - if (sb->s_root) + if (sb->s_root && (sb->s_flags & MS_BORN)) f(sb, arg); up_read(&sb->s_umount); @@ -566,14 +569,15 @@ void iterate_supers_type(struct file_system_type *type, void (*f)(struct super_block *, void *), void *arg) { struct super_block *sb, *p = NULL; + struct hlist_node *node; spin_lock(&sb_lock); - list_for_each_entry(sb, &type->fs_supers, s_instances) { + hlist_for_each_entry(sb, node, &type->fs_supers, s_instances) { sb->s_count++; spin_unlock(&sb_lock); down_read(&sb->s_umount); - if (sb->s_root) + if (sb->s_root && (sb->s_flags & MS_BORN)) f(sb, arg); up_read(&sb->s_umount); @@ -607,14 +611,14 @@ struct super_block *get_super(struct block_device *bdev) spin_lock(&sb_lock); rescan: list_for_each_entry(sb, &super_blocks, s_list) { - if (list_empty(&sb->s_instances)) + if (hlist_unhashed(&sb->s_instances)) continue; if (sb->s_bdev == bdev) { sb->s_count++; spin_unlock(&sb_lock); down_read(&sb->s_umount); /* still alive? */ - if (sb->s_root) + if (sb->s_root && (sb->s_flags & MS_BORN)) return sb; up_read(&sb->s_umount); /* nope, got unmounted */ @@ -647,7 +651,7 @@ struct super_block *get_active_super(struct block_device *bdev) restart: spin_lock(&sb_lock); list_for_each_entry(sb, &super_blocks, s_list) { - if (list_empty(&sb->s_instances)) + if (hlist_unhashed(&sb->s_instances)) continue; if (sb->s_bdev == bdev) { if (grab_super(sb)) /* drops sb_lock */ @@ -667,14 +671,14 @@ struct super_block *user_get_super(dev_t dev) spin_lock(&sb_lock); rescan: list_for_each_entry(sb, &super_blocks, s_list) { - if (list_empty(&sb->s_instances)) + if (hlist_unhashed(&sb->s_instances)) continue; if (sb->s_dev == dev) { sb->s_count++; spin_unlock(&sb_lock); down_read(&sb->s_umount); /* still alive? */ - if (sb->s_root) + if (sb->s_root && (sb->s_flags & MS_BORN)) return sb; up_read(&sb->s_umount); /* nope, got unmounted */ @@ -719,23 +723,29 @@ int do_remount_sb(struct super_block *sb, int flags, void *data, int force) /* If we are remounting RDONLY and current sb is read/write, make sure there are no rw files opened */ if (remount_ro) { - if (force) + if (force) { mark_files_ro(sb); - else if (!fs_may_remount_ro(sb)) - return -EBUSY; + } else { + retval = sb_prepare_remount_readonly(sb); + if (retval) + return retval; + } } if (sb->s_op->remount_fs) { retval = sb->s_op->remount_fs(sb, &flags, data); if (retval) { if (!force) - return retval; + goto cancel_readonly; /* If forced remount, go ahead despite any errors */ WARN(1, "forced remount of a %s fs returned %i\n", sb->s_type->name, retval); } } sb->s_flags = (sb->s_flags & ~MS_RMT_MASK) | (flags & MS_RMT_MASK); + /* Needs to be ordered wrt mnt_is_readonly() */ + smp_wmb(); + sb->s_readonly_remount = 0; /* * Some filesystems modify their metadata via some other path than the @@ -748,6 +758,10 @@ int do_remount_sb(struct super_block *sb, int flags, void *data, int force) if (remount_ro && sb->s_bdev) invalidate_bdev(sb->s_bdev); return 0; + +cancel_readonly: + sb->s_readonly_remount = 0; + return retval; } static void do_emergency_remount(struct work_struct *work) @@ -756,12 +770,13 @@ static void do_emergency_remount(struct work_struct *work) spin_lock(&sb_lock); list_for_each_entry(sb, &super_blocks, s_list) { - if (list_empty(&sb->s_instances)) + if (hlist_unhashed(&sb->s_instances)) continue; sb->s_count++; spin_unlock(&sb_lock); down_write(&sb->s_umount); - if (sb->s_root && sb->s_bdev && !(sb->s_flags & MS_RDONLY)) { + if (sb->s_root && sb->s_bdev && (sb->s_flags & MS_BORN) && + !(sb->s_flags & MS_RDONLY)) { /* * What lock protects sb->s_flags?? */ @@ -1144,6 +1159,11 @@ int freeze_super(struct super_block *sb) return -EBUSY; } + if (!(sb->s_flags & MS_BORN)) { + up_write(&sb->s_umount); + return 0; /* sic - it's "nothing to do" */ + } + if (sb->s_flags & MS_RDONLY) { sb->s_frozen = SB_FREEZE_TRANS; smp_wmb(); diff --git a/fs/sync.c b/fs/sync.c index 101b8ef901d7..f3501ef39235 100644 --- a/fs/sync.c +++ b/fs/sync.c @@ -14,7 +14,6 @@ #include <linux/linkage.h> #include <linux/pagemap.h> #include <linux/quotaops.h> -#include <linux/buffer_head.h> #include <linux/backing-dev.h> #include "internal.h" diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index d4e6080b4b20..62f4fb37789e 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -518,7 +518,7 @@ out: } int sysfs_add_file_mode(struct sysfs_dirent *dir_sd, - const struct attribute *attr, int type, mode_t amode) + const struct attribute *attr, int type, umode_t amode) { umode_t mode = (amode & S_IALLUGO) | S_IFREG; struct sysfs_addrm_cxt acxt; @@ -618,7 +618,7 @@ EXPORT_SYMBOL_GPL(sysfs_add_file_to_group); * */ int sysfs_chmod_file(struct kobject *kobj, const struct attribute *attr, - mode_t mode) + umode_t mode) { struct sysfs_dirent *sd; struct iattr newattrs; diff --git a/fs/sysfs/group.c b/fs/sysfs/group.c index 194414f8298c..dd1701caecc9 100644 --- a/fs/sysfs/group.c +++ b/fs/sysfs/group.c @@ -33,7 +33,7 @@ static int create_files(struct sysfs_dirent *dir_sd, struct kobject *kobj, int error = 0, i; for (i = 0, attr = grp->attrs; *attr && !error; i++, attr++) { - mode_t mode = 0; + umode_t mode = 0; /* in update mode, we're changing the permissions or * visibility. Do this by first removing then diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c index c81b22f3ace1..4a802b4a9056 100644 --- a/fs/sysfs/inode.c +++ b/fs/sysfs/inode.c @@ -187,7 +187,7 @@ out: return error; } -static inline void set_default_inode_attr(struct inode * inode, mode_t mode) +static inline void set_default_inode_attr(struct inode * inode, umode_t mode) { inode->i_mode = mode; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h index ce29e28b766d..7484a36ee678 100644 --- a/fs/sysfs/sysfs.h +++ b/fs/sysfs/sysfs.h @@ -79,7 +79,7 @@ struct sysfs_dirent { }; unsigned int s_flags; - unsigned short s_mode; + umode_t s_mode; ino_t s_ino; struct sysfs_inode_attrs *s_iattr; }; @@ -229,7 +229,7 @@ int sysfs_add_file(struct sysfs_dirent *dir_sd, const struct attribute *attr, int type); int sysfs_add_file_mode(struct sysfs_dirent *dir_sd, - const struct attribute *attr, int type, mode_t amode); + const struct attribute *attr, int type, umode_t amode); /* * bin.c */ diff --git a/fs/sysv/ialloc.c b/fs/sysv/ialloc.c index 0c96c98bd1db..8233b02eccae 100644 --- a/fs/sysv/ialloc.c +++ b/fs/sysv/ialloc.c @@ -132,7 +132,7 @@ void sysv_free_inode(struct inode * inode) brelse(bh); } -struct inode * sysv_new_inode(const struct inode * dir, mode_t mode) +struct inode * sysv_new_inode(const struct inode * dir, umode_t mode) { struct super_block *sb = dir->i_sb; struct sysv_sb_info *sbi = SYSV_SB(sb); diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c index 25ffb3e9a3f8..3da5ce25faf0 100644 --- a/fs/sysv/inode.c +++ b/fs/sysv/inode.c @@ -336,7 +336,6 @@ static struct inode *sysv_alloc_inode(struct super_block *sb) static void sysv_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(sysv_inode_cachep, SYSV_I(inode)); } diff --git a/fs/sysv/itree.c b/fs/sysv/itree.c index fa8d43c92bb8..90b54b438789 100644 --- a/fs/sysv/itree.c +++ b/fs/sysv/itree.c @@ -442,7 +442,7 @@ static unsigned sysv_nblocks(struct super_block *s, loff_t size) int sysv_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) { - struct super_block *s = mnt->mnt_sb; + struct super_block *s = dentry->d_sb; generic_fillattr(dentry->d_inode, stat); stat->blocks = (s->s_blocksize / 512) * sysv_nblocks(s, stat->size); stat->blksize = s->s_blocksize; diff --git a/fs/sysv/namei.c b/fs/sysv/namei.c index e474fbcf8bde..b217797e621b 100644 --- a/fs/sysv/namei.c +++ b/fs/sysv/namei.c @@ -61,7 +61,7 @@ static struct dentry *sysv_lookup(struct inode * dir, struct dentry * dentry, st return NULL; } -static int sysv_mknod(struct inode * dir, struct dentry * dentry, int mode, dev_t rdev) +static int sysv_mknod(struct inode * dir, struct dentry * dentry, umode_t mode, dev_t rdev) { struct inode * inode; int err; @@ -80,7 +80,7 @@ static int sysv_mknod(struct inode * dir, struct dentry * dentry, int mode, dev_ return err; } -static int sysv_create(struct inode * dir, struct dentry * dentry, int mode, struct nameidata *nd) +static int sysv_create(struct inode * dir, struct dentry * dentry, umode_t mode, struct nameidata *nd) { return sysv_mknod(dir, dentry, mode, 0); } @@ -131,7 +131,7 @@ static int sysv_link(struct dentry * old_dentry, struct inode * dir, return add_nondir(dentry, inode); } -static int sysv_mkdir(struct inode * dir, struct dentry *dentry, int mode) +static int sysv_mkdir(struct inode * dir, struct dentry *dentry, umode_t mode) { struct inode * inode; int err = -EMLINK; diff --git a/fs/sysv/sysv.h b/fs/sysv/sysv.h index bb55cdb394bf..0e4b821c5691 100644 --- a/fs/sysv/sysv.h +++ b/fs/sysv/sysv.h @@ -125,7 +125,7 @@ static inline void dirty_sb(struct super_block *sb) /* ialloc.c */ extern struct sysv_inode *sysv_raw_inode(struct super_block *, unsigned, struct buffer_head **); -extern struct inode * sysv_new_inode(const struct inode *, mode_t); +extern struct inode * sysv_new_inode(const struct inode *, umode_t); extern void sysv_free_inode(struct inode *); extern unsigned long sysv_count_free_inodes(struct super_block *); diff --git a/fs/ubifs/debug.c b/fs/ubifs/debug.c index b09ba2dd8b62..f922cbacdb96 100644 --- a/fs/ubifs/debug.c +++ b/fs/ubifs/debug.c @@ -38,9 +38,6 @@ DEFINE_SPINLOCK(dbg_lock); -static char dbg_key_buf0[128]; -static char dbg_key_buf1[128]; - static const char *get_key_fmt(int fmt) { switch (fmt) { @@ -103,8 +100,8 @@ static const char *get_dent_type(int type) } } -static void sprintf_key(const struct ubifs_info *c, const union ubifs_key *key, - char *buffer) +const char *dbg_snprintf_key(const struct ubifs_info *c, + const union ubifs_key *key, char *buffer, int len) { char *p = buffer; int type = key_type(c, key); @@ -112,45 +109,34 @@ static void sprintf_key(const struct ubifs_info *c, const union ubifs_key *key, if (c->key_fmt == UBIFS_SIMPLE_KEY_FMT) { switch (type) { case UBIFS_INO_KEY: - sprintf(p, "(%lu, %s)", (unsigned long)key_inum(c, key), - get_key_type(type)); + len -= snprintf(p, len, "(%lu, %s)", + (unsigned long)key_inum(c, key), + get_key_type(type)); break; case UBIFS_DENT_KEY: case UBIFS_XENT_KEY: - sprintf(p, "(%lu, %s, %#08x)", - (unsigned long)key_inum(c, key), - get_key_type(type), key_hash(c, key)); + len -= snprintf(p, len, "(%lu, %s, %#08x)", + (unsigned long)key_inum(c, key), + get_key_type(type), key_hash(c, key)); break; case UBIFS_DATA_KEY: - sprintf(p, "(%lu, %s, %u)", - (unsigned long)key_inum(c, key), - get_key_type(type), key_block(c, key)); + len -= snprintf(p, len, "(%lu, %s, %u)", + (unsigned long)key_inum(c, key), + get_key_type(type), key_block(c, key)); break; case UBIFS_TRUN_KEY: - sprintf(p, "(%lu, %s)", - (unsigned long)key_inum(c, key), - get_key_type(type)); + len -= snprintf(p, len, "(%lu, %s)", + (unsigned long)key_inum(c, key), + get_key_type(type)); break; default: - sprintf(p, "(bad key type: %#08x, %#08x)", - key->u32[0], key->u32[1]); + len -= snprintf(p, len, "(bad key type: %#08x, %#08x)", + key->u32[0], key->u32[1]); } } else - sprintf(p, "bad key format %d", c->key_fmt); -} - -const char *dbg_key_str0(const struct ubifs_info *c, const union ubifs_key *key) -{ - /* dbg_lock must be held */ - sprintf_key(c, key, dbg_key_buf0); - return dbg_key_buf0; -} - -const char *dbg_key_str1(const struct ubifs_info *c, const union ubifs_key *key) -{ - /* dbg_lock must be held */ - sprintf_key(c, key, dbg_key_buf1); - return dbg_key_buf1; + len -= snprintf(p, len, "bad key format %d", c->key_fmt); + ubifs_assert(len > 0); + return p; } const char *dbg_ntype(int type) @@ -319,6 +305,7 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node) int i, n; union ubifs_key key; const struct ubifs_ch *ch = node; + char key_buf[DBG_KEY_BUF_LEN]; if (dbg_is_tst_rcvry(c)) return; @@ -474,7 +461,8 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node) const struct ubifs_ino_node *ino = node; key_read(c, &ino->key, &key); - printk(KERN_DEBUG "\tkey %s\n", DBGKEY(&key)); + printk(KERN_DEBUG "\tkey %s\n", + dbg_snprintf_key(c, &key, key_buf, DBG_KEY_BUF_LEN)); printk(KERN_DEBUG "\tcreat_sqnum %llu\n", (unsigned long long)le64_to_cpu(ino->creat_sqnum)); printk(KERN_DEBUG "\tsize %llu\n", @@ -517,7 +505,8 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node) int nlen = le16_to_cpu(dent->nlen); key_read(c, &dent->key, &key); - printk(KERN_DEBUG "\tkey %s\n", DBGKEY(&key)); + printk(KERN_DEBUG "\tkey %s\n", + dbg_snprintf_key(c, &key, key_buf, DBG_KEY_BUF_LEN)); printk(KERN_DEBUG "\tinum %llu\n", (unsigned long long)le64_to_cpu(dent->inum)); printk(KERN_DEBUG "\ttype %d\n", (int)dent->type); @@ -541,7 +530,8 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node) int dlen = le32_to_cpu(ch->len) - UBIFS_DATA_NODE_SZ; key_read(c, &dn->key, &key); - printk(KERN_DEBUG "\tkey %s\n", DBGKEY(&key)); + printk(KERN_DEBUG "\tkey %s\n", + dbg_snprintf_key(c, &key, key_buf, DBG_KEY_BUF_LEN)); printk(KERN_DEBUG "\tsize %u\n", le32_to_cpu(dn->size)); printk(KERN_DEBUG "\tcompr_typ %d\n", @@ -582,7 +572,9 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node) key_read(c, &br->key, &key); printk(KERN_DEBUG "\t%d: LEB %d:%d len %d key %s\n", i, le32_to_cpu(br->lnum), le32_to_cpu(br->offs), - le32_to_cpu(br->len), DBGKEY(&key)); + le32_to_cpu(br->len), + dbg_snprintf_key(c, &key, key_buf, + DBG_KEY_BUF_LEN)); } break; } @@ -934,6 +926,7 @@ void dbg_dump_znode(const struct ubifs_info *c, { int n; const struct ubifs_zbranch *zbr; + char key_buf[DBG_KEY_BUF_LEN]; spin_lock(&dbg_lock); if (znode->parent) @@ -958,12 +951,16 @@ void dbg_dump_znode(const struct ubifs_info *c, printk(KERN_DEBUG "\t%d: znode %p LEB %d:%d len %d key " "%s\n", n, zbr->znode, zbr->lnum, zbr->offs, zbr->len, - DBGKEY(&zbr->key)); + dbg_snprintf_key(c, &zbr->key, + key_buf, + DBG_KEY_BUF_LEN)); else printk(KERN_DEBUG "\t%d: LNC %p LEB %d:%d len %d key " "%s\n", n, zbr->znode, zbr->lnum, zbr->offs, zbr->len, - DBGKEY(&zbr->key)); + dbg_snprintf_key(c, &zbr->key, + key_buf, + DBG_KEY_BUF_LEN)); } spin_unlock(&dbg_lock); } @@ -1260,6 +1257,7 @@ static int dbg_check_key_order(struct ubifs_info *c, struct ubifs_zbranch *zbr1, int err, nlen1, nlen2, cmp; struct ubifs_dent_node *dent1, *dent2; union ubifs_key key; + char key_buf[DBG_KEY_BUF_LEN]; ubifs_assert(!keys_cmp(c, &zbr1->key, &zbr2->key)); dent1 = kmalloc(UBIFS_MAX_DENT_NODE_SZ, GFP_NOFS); @@ -1290,9 +1288,11 @@ static int dbg_check_key_order(struct ubifs_info *c, struct ubifs_zbranch *zbr1, key_read(c, &dent1->key, &key); if (keys_cmp(c, &zbr1->key, &key)) { dbg_err("1st entry at %d:%d has key %s", zbr1->lnum, - zbr1->offs, DBGKEY(&key)); + zbr1->offs, dbg_snprintf_key(c, &key, key_buf, + DBG_KEY_BUF_LEN)); dbg_err("but it should have key %s according to tnc", - DBGKEY(&zbr1->key)); + dbg_snprintf_key(c, &zbr1->key, key_buf, + DBG_KEY_BUF_LEN)); dbg_dump_node(c, dent1); goto out_free; } @@ -1300,9 +1300,11 @@ static int dbg_check_key_order(struct ubifs_info *c, struct ubifs_zbranch *zbr1, key_read(c, &dent2->key, &key); if (keys_cmp(c, &zbr2->key, &key)) { dbg_err("2nd entry at %d:%d has key %s", zbr1->lnum, - zbr1->offs, DBGKEY(&key)); + zbr1->offs, dbg_snprintf_key(c, &key, key_buf, + DBG_KEY_BUF_LEN)); dbg_err("but it should have key %s according to tnc", - DBGKEY(&zbr2->key)); + dbg_snprintf_key(c, &zbr2->key, key_buf, + DBG_KEY_BUF_LEN)); dbg_dump_node(c, dent2); goto out_free; } @@ -1319,7 +1321,7 @@ static int dbg_check_key_order(struct ubifs_info *c, struct ubifs_zbranch *zbr1, dbg_err("2 xent/dent nodes with the same name"); else dbg_err("bad order of colliding key %s", - DBGKEY(&key)); + dbg_snprintf_key(c, &key, key_buf, DBG_KEY_BUF_LEN)); ubifs_msg("first node at %d:%d\n", zbr1->lnum, zbr1->offs); dbg_dump_node(c, dent1); diff --git a/fs/ubifs/debug.h b/fs/ubifs/debug.h index 8d9c46810189..ad1a6fee6010 100644 --- a/fs/ubifs/debug.h +++ b/fs/ubifs/debug.h @@ -169,40 +169,39 @@ struct ubifs_global_debug_info { spin_unlock(&dbg_lock); \ } while (0) -const char *dbg_key_str0(const struct ubifs_info *c, - const union ubifs_key *key); -const char *dbg_key_str1(const struct ubifs_info *c, - const union ubifs_key *key); - -/* - * DBGKEY macros require @dbg_lock to be held, which it is in the dbg message - * macros. - */ -#define DBGKEY(key) dbg_key_str0(c, (key)) -#define DBGKEY1(key) dbg_key_str1(c, (key)) - -extern spinlock_t dbg_lock; - -#define ubifs_dbg_msg(type, fmt, ...) do { \ - spin_lock(&dbg_lock); \ - pr_debug("UBIFS DBG " type ": " fmt "\n", ##__VA_ARGS__); \ - spin_unlock(&dbg_lock); \ +#define ubifs_dbg_msg(type, fmt, ...) \ + pr_debug("UBIFS DBG " type ": " fmt "\n", ##__VA_ARGS__) + +#define DBG_KEY_BUF_LEN 32 +#define ubifs_dbg_msg_key(type, key, fmt, ...) do { \ + char __tmp_key_buf[DBG_KEY_BUF_LEN]; \ + pr_debug("UBIFS DBG " type ": " fmt "%s\n", ##__VA_ARGS__, \ + dbg_snprintf_key(c, key, __tmp_key_buf, DBG_KEY_BUF_LEN)); \ } while (0) /* Just a debugging messages not related to any specific UBIFS subsystem */ -#define dbg_msg(fmt, ...) ubifs_dbg_msg("msg", fmt, ##__VA_ARGS__) +#define dbg_msg(fmt, ...) \ + printk(KERN_DEBUG "UBIFS DBG (pid %d): %s: " fmt "\n", current->pid, \ + __func__, ##__VA_ARGS__) + /* General messages */ #define dbg_gen(fmt, ...) ubifs_dbg_msg("gen", fmt, ##__VA_ARGS__) /* Additional journal messages */ #define dbg_jnl(fmt, ...) ubifs_dbg_msg("jnl", fmt, ##__VA_ARGS__) +#define dbg_jnlk(key, fmt, ...) \ + ubifs_dbg_msg_key("jnl", key, fmt, ##__VA_ARGS__) /* Additional TNC messages */ #define dbg_tnc(fmt, ...) ubifs_dbg_msg("tnc", fmt, ##__VA_ARGS__) +#define dbg_tnck(key, fmt, ...) \ + ubifs_dbg_msg_key("tnc", key, fmt, ##__VA_ARGS__) /* Additional lprops messages */ #define dbg_lp(fmt, ...) ubifs_dbg_msg("lp", fmt, ##__VA_ARGS__) /* Additional LEB find messages */ #define dbg_find(fmt, ...) ubifs_dbg_msg("find", fmt, ##__VA_ARGS__) /* Additional mount messages */ #define dbg_mnt(fmt, ...) ubifs_dbg_msg("mnt", fmt, ##__VA_ARGS__) +#define dbg_mntk(key, fmt, ...) \ + ubifs_dbg_msg_key("mnt", key, fmt, ##__VA_ARGS__) /* Additional I/O messages */ #define dbg_io(fmt, ...) ubifs_dbg_msg("io", fmt, ##__VA_ARGS__) /* Additional commit messages */ @@ -218,6 +217,7 @@ extern spinlock_t dbg_lock; /* Additional recovery messages */ #define dbg_rcvry(fmt, ...) ubifs_dbg_msg("rcvry", fmt, ##__VA_ARGS__) +extern spinlock_t dbg_lock; extern struct ubifs_global_debug_info ubifs_dbg; static inline int dbg_is_chk_gen(const struct ubifs_info *c) @@ -258,6 +258,8 @@ const char *dbg_cstate(int cmt_state); const char *dbg_jhead(int jhead); const char *dbg_get_key_dump(const struct ubifs_info *c, const union ubifs_key *key); +const char *dbg_snprintf_key(const struct ubifs_info *c, + const union ubifs_key *key, char *buffer, int len); void dbg_dump_inode(struct ubifs_info *c, const struct inode *inode); void dbg_dump_node(const struct ubifs_info *c, const void *node); void dbg_dump_lpt_node(const struct ubifs_info *c, void *node, int lnum, @@ -345,20 +347,23 @@ void dbg_debugfs_exit_fs(struct ubifs_info *c); #define dbg_dump_stack() #define ubifs_assert_cmt_locked(c) -#define dbg_msg(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_gen(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_jnl(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_tnc(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_lp(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_find(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_mnt(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_io(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_cmt(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_budg(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_log(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_gc(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_scan(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_rcvry(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_msg(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_gen(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_jnl(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_jnlk(key, fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_tnc(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_tnck(key, fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_lp(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_find(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_mnt(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_mntk(key, fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_io(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_cmt(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_budg(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_log(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_gc(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_scan(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_rcvry(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) static inline int ubifs_debugging_init(struct ubifs_info *c) { return 0; } static inline void ubifs_debugging_exit(struct ubifs_info *c) { return; } @@ -368,6 +373,10 @@ static inline const char *dbg_jhead(int jhead) { return ""; } static inline const char * dbg_get_key_dump(const struct ubifs_info *c, const union ubifs_key *key) { return ""; } +static inline const char * +dbg_snprintf_key(const struct ubifs_info *c, + const union ubifs_key *key, char *buffer, + int len) { return ""; } static inline void dbg_dump_inode(struct ubifs_info *c, const struct inode *inode) { return; } static inline void dbg_dump_node(const struct ubifs_info *c, diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c index 683492043317..d6fe1c79f18b 100644 --- a/fs/ubifs/dir.c +++ b/fs/ubifs/dir.c @@ -56,7 +56,7 @@ * * This function returns the inherited flags. */ -static int inherit_flags(const struct inode *dir, int mode) +static int inherit_flags(const struct inode *dir, umode_t mode) { int flags; const struct ubifs_inode *ui = ubifs_inode(dir); @@ -86,7 +86,7 @@ static int inherit_flags(const struct inode *dir, int mode) * case of failure. */ struct inode *ubifs_new_inode(struct ubifs_info *c, const struct inode *dir, - int mode) + umode_t mode) { struct inode *inode; struct ubifs_inode *ui; @@ -253,7 +253,7 @@ out: return ERR_PTR(err); } -static int ubifs_create(struct inode *dir, struct dentry *dentry, int mode, +static int ubifs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { struct inode *inode; @@ -268,7 +268,7 @@ static int ubifs_create(struct inode *dir, struct dentry *dentry, int mode, * parent directory inode. */ - dbg_gen("dent '%.*s', mode %#x in dir ino %lu", + dbg_gen("dent '%.*s', mode %#hx in dir ino %lu", dentry->d_name.len, dentry->d_name.name, mode, dir->i_ino); err = ubifs_budget_space(c, &req); @@ -712,7 +712,7 @@ out_cancel: return err; } -static int ubifs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int ubifs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct inode *inode; struct ubifs_inode *dir_ui = ubifs_inode(dir); @@ -725,7 +725,7 @@ static int ubifs_mkdir(struct inode *dir, struct dentry *dentry, int mode) * directory inode. */ - dbg_gen("dent '%.*s', mode %#x in dir ino %lu", + dbg_gen("dent '%.*s', mode %#hx in dir ino %lu", dentry->d_name.len, dentry->d_name.name, mode, dir->i_ino); err = ubifs_budget_space(c, &req); @@ -769,7 +769,7 @@ out_budg: } static int ubifs_mknod(struct inode *dir, struct dentry *dentry, - int mode, dev_t rdev) + umode_t mode, dev_t rdev) { struct inode *inode; struct ubifs_inode *ui; diff --git a/fs/ubifs/ioctl.c b/fs/ubifs/ioctl.c index 548acf494afd..1a7e2d8bdbe9 100644 --- a/fs/ubifs/ioctl.c +++ b/fs/ubifs/ioctl.c @@ -173,12 +173,12 @@ long ubifs_ioctl(struct file *file, unsigned int cmd, unsigned long arg) * Make sure the file-system is read-write and make sure it * will not become read-only while we are changing the flags. */ - err = mnt_want_write(file->f_path.mnt); + err = mnt_want_write_file(file); if (err) return err; dbg_gen("set flags: %#x, i_flags %#x", flags, inode->i_flags); err = setflags(inode, flags); - mnt_drop_write(file->f_path.mnt); + mnt_drop_write_file(file); return err; } diff --git a/fs/ubifs/journal.c b/fs/ubifs/journal.c index cef0460f4c54..2f438ab2e7a2 100644 --- a/fs/ubifs/journal.c +++ b/fs/ubifs/journal.c @@ -697,9 +697,8 @@ int ubifs_jnl_write_data(struct ubifs_info *c, const struct inode *inode, int dlen = COMPRESSED_DATA_NODE_BUF_SZ, allocated = 1; struct ubifs_inode *ui = ubifs_inode(inode); - dbg_jnl("ino %lu, blk %u, len %d, key %s", - (unsigned long)key_inum(c, key), key_block(c, key), len, - DBGKEY(key)); + dbg_jnlk(key, "ino %lu, blk %u, len %d, key ", + (unsigned long)key_inum(c, key), key_block(c, key), len); ubifs_assert(len <= UBIFS_BLOCK_SIZE); data = kmalloc(dlen, GFP_NOFS | __GFP_NOWARN); @@ -1177,7 +1176,7 @@ int ubifs_jnl_truncate(struct ubifs_info *c, const struct inode *inode, dn = (void *)trun + UBIFS_TRUN_NODE_SZ; blk = new_size >> UBIFS_BLOCK_SHIFT; data_key_init(c, &key, inum, blk); - dbg_jnl("last block key %s", DBGKEY(&key)); + dbg_jnlk(&key, "last block key "); err = ubifs_tnc_lookup(c, &key, dn); if (err == -ENOENT) dlen = 0; /* Not found (so it is a hole) */ diff --git a/fs/ubifs/lpt.c b/fs/ubifs/lpt.c index 6189c74d97f0..66d59d0a1402 100644 --- a/fs/ubifs/lpt.c +++ b/fs/ubifs/lpt.c @@ -1986,12 +1986,11 @@ again: if (path[h].in_tree) continue; - nnode = kmalloc(sz, GFP_NOFS); + nnode = kmemdup(&path[h].nnode, sz, GFP_NOFS); if (!nnode) { err = -ENOMEM; goto out; } - memcpy(nnode, &path[h].nnode, sz); parent = nnode->parent; parent->nbranch[nnode->iip].nnode = nnode; path[h].ptr.nnode = nnode; @@ -2004,12 +2003,11 @@ again: const size_t sz = sizeof(struct ubifs_pnode); struct ubifs_nnode *parent; - pnode = kmalloc(sz, GFP_NOFS); + pnode = kmemdup(&path[h].pnode, sz, GFP_NOFS); if (!pnode) { err = -ENOMEM; goto out; } - memcpy(pnode, &path[h].pnode, sz); parent = pnode->parent; parent->nbranch[pnode->iip].pnode = pnode; path[h].ptr.pnode = pnode; diff --git a/fs/ubifs/replay.c b/fs/ubifs/replay.c index ccabaf1164b3..b007637f0406 100644 --- a/fs/ubifs/replay.c +++ b/fs/ubifs/replay.c @@ -221,8 +221,8 @@ static int apply_replay_entry(struct ubifs_info *c, struct replay_entry *r) { int err; - dbg_mnt("LEB %d:%d len %d deletion %d sqnum %llu %s", r->lnum, - r->offs, r->len, r->deletion, r->sqnum, DBGKEY(&r->key)); + dbg_mntk(&r->key, "LEB %d:%d len %d deletion %d sqnum %llu key ", + r->lnum, r->offs, r->len, r->deletion, r->sqnum); /* Set c->replay_sqnum to help deal with dangling branches. */ c->replay_sqnum = r->sqnum; @@ -361,7 +361,7 @@ static int insert_node(struct ubifs_info *c, int lnum, int offs, int len, { struct replay_entry *r; - dbg_mnt("add LEB %d:%d, key %s", lnum, offs, DBGKEY(key)); + dbg_mntk(key, "add LEB %d:%d, key ", lnum, offs); if (key_inum(c, key) >= c->highest_inum) c->highest_inum = key_inum(c, key); @@ -409,7 +409,7 @@ static int insert_dent(struct ubifs_info *c, int lnum, int offs, int len, struct replay_entry *r; char *nbuf; - dbg_mnt("add LEB %d:%d, key %s", lnum, offs, DBGKEY(key)); + dbg_mntk(key, "add LEB %d:%d, key ", lnum, offs); if (key_inum(c, key) >= c->highest_inum) c->highest_inum = key_inum(c, key); diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index 20403dc5d437..63765d58445b 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -276,7 +276,6 @@ static void ubifs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); struct ubifs_inode *ui = ubifs_inode(inode); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(ubifs_inode_slab, ui); } @@ -420,9 +419,9 @@ static int ubifs_statfs(struct dentry *dentry, struct kstatfs *buf) return 0; } -static int ubifs_show_options(struct seq_file *s, struct vfsmount *mnt) +static int ubifs_show_options(struct seq_file *s, struct dentry *root) { - struct ubifs_info *c = mnt->mnt_sb->s_fs_info; + struct ubifs_info *c = root->d_sb->s_fs_info; if (c->mount_opts.unmount_mode == 2) seq_printf(s, ",fast_unmount"); @@ -2264,19 +2263,12 @@ static int __init ubifs_init(void) return -EINVAL; } - err = register_filesystem(&ubifs_fs_type); - if (err) { - ubifs_err("cannot register file system, error %d", err); - return err; - } - - err = -ENOMEM; ubifs_inode_slab = kmem_cache_create("ubifs_inode_slab", sizeof(struct ubifs_inode), 0, SLAB_MEM_SPREAD | SLAB_RECLAIM_ACCOUNT, &inode_slab_ctor); if (!ubifs_inode_slab) - goto out_reg; + return -ENOMEM; register_shrinker(&ubifs_shrinker_info); @@ -2288,15 +2280,20 @@ static int __init ubifs_init(void) if (err) goto out_compr; + err = register_filesystem(&ubifs_fs_type); + if (err) { + ubifs_err("cannot register file system, error %d", err); + goto out_dbg; + } return 0; +out_dbg: + dbg_debugfs_exit(); out_compr: ubifs_compressors_exit(); out_shrinker: unregister_shrinker(&ubifs_shrinker_info); kmem_cache_destroy(ubifs_inode_slab); -out_reg: - unregister_filesystem(&ubifs_fs_type); return err; } /* late_initcall to let compressors initialize first */ diff --git a/fs/ubifs/tnc.c b/fs/ubifs/tnc.c index 066738647685..16ad84d8402f 100644 --- a/fs/ubifs/tnc.c +++ b/fs/ubifs/tnc.c @@ -344,12 +344,11 @@ static int lnc_add(struct ubifs_info *c, struct ubifs_zbranch *zbr, return err; } - lnc_node = kmalloc(zbr->len, GFP_NOFS); + lnc_node = kmemdup(node, zbr->len, GFP_NOFS); if (!lnc_node) /* We don't have to have the cache, so no error */ return 0; - memcpy(lnc_node, node, zbr->len); zbr->leaf = lnc_node; return 0; } @@ -506,7 +505,7 @@ static int fallible_read_node(struct ubifs_info *c, const union ubifs_key *key, { int ret; - dbg_tnc("LEB %d:%d, key %s", zbr->lnum, zbr->offs, DBGKEY(key)); + dbg_tnck(key, "LEB %d:%d, key ", zbr->lnum, zbr->offs); ret = try_read_node(c, node, key_type(c, key), zbr->len, zbr->lnum, zbr->offs); @@ -520,8 +519,8 @@ static int fallible_read_node(struct ubifs_info *c, const union ubifs_key *key, ret = 0; } if (ret == 0 && c->replaying) - dbg_mnt("dangling branch LEB %d:%d len %d, key %s", - zbr->lnum, zbr->offs, zbr->len, DBGKEY(key)); + dbg_mntk(key, "dangling branch LEB %d:%d len %d, key ", + zbr->lnum, zbr->offs, zbr->len); return ret; } @@ -996,9 +995,9 @@ static int fallible_resolve_collision(struct ubifs_info *c, if (adding || !o_znode) return 0; - dbg_mnt("dangling match LEB %d:%d len %d %s", + dbg_mntk(key, "dangling match LEB %d:%d len %d key ", o_znode->zbranch[o_n].lnum, o_znode->zbranch[o_n].offs, - o_znode->zbranch[o_n].len, DBGKEY(key)); + o_znode->zbranch[o_n].len); *zn = o_znode; *n = o_n; return 1; @@ -1180,7 +1179,7 @@ int ubifs_lookup_level0(struct ubifs_info *c, const union ubifs_key *key, struct ubifs_znode *znode; unsigned long time = get_seconds(); - dbg_tnc("search key %s", DBGKEY(key)); + dbg_tnck(key, "search key "); ubifs_assert(key_type(c, key) < UBIFS_INVALID_KEY); znode = c->zroot.znode; @@ -1316,7 +1315,7 @@ static int lookup_level0_dirty(struct ubifs_info *c, const union ubifs_key *key, struct ubifs_znode *znode; unsigned long time = get_seconds(); - dbg_tnc("search and dirty key %s", DBGKEY(key)); + dbg_tnck(key, "search and dirty key "); znode = c->zroot.znode; if (unlikely(!znode)) { @@ -1723,8 +1722,8 @@ static int validate_data_node(struct ubifs_info *c, void *buf, if (!keys_eq(c, &zbr->key, &key1)) { ubifs_err("bad key in node at LEB %d:%d", zbr->lnum, zbr->offs); - dbg_tnc("looked for key %s found node's key %s", - DBGKEY(&zbr->key), DBGKEY1(&key1)); + dbg_tnck(&zbr->key, "looked for key "); + dbg_tnck(&key1, "found node's key "); goto out_err; } @@ -1777,7 +1776,7 @@ int ubifs_tnc_bulk_read(struct ubifs_info *c, struct bu_info *bu) ubifs_err("failed to read from LEB %d:%d, error %d", lnum, offs, err); dbg_dump_stack(); - dbg_tnc("key %s", DBGKEY(&bu->key)); + dbg_tnck(&bu->key, "key "); return err; } @@ -1812,7 +1811,7 @@ static int do_lookup_nm(struct ubifs_info *c, const union ubifs_key *key, int found, n, err; struct ubifs_znode *znode; - dbg_tnc("name '%.*s' key %s", nm->len, nm->name, DBGKEY(key)); + dbg_tnck(key, "name '%.*s' key ", nm->len, nm->name); mutex_lock(&c->tnc_mutex); found = ubifs_lookup_level0(c, key, &znode, &n); if (!found) { @@ -1986,8 +1985,7 @@ again: zp = znode->parent; if (znode->child_cnt < c->fanout) { ubifs_assert(n != c->fanout); - dbg_tnc("inserted at %d level %d, key %s", n, znode->level, - DBGKEY(key)); + dbg_tnck(key, "inserted at %d level %d, key ", n, znode->level); insert_zbranch(znode, zbr, n); @@ -2002,7 +2000,7 @@ again: * Unfortunately, @znode does not have more empty slots and we have to * split it. */ - dbg_tnc("splitting level %d, key %s", znode->level, DBGKEY(key)); + dbg_tnck(key, "splitting level %d, key ", znode->level); if (znode->alt) /* @@ -2096,7 +2094,7 @@ do_split: } /* Insert new key and branch */ - dbg_tnc("inserting at %d level %d, key %s", n, zn->level, DBGKEY(key)); + dbg_tnck(key, "inserting at %d level %d, key ", n, zn->level); insert_zbranch(zi, zbr, n); @@ -2172,7 +2170,7 @@ int ubifs_tnc_add(struct ubifs_info *c, const union ubifs_key *key, int lnum, struct ubifs_znode *znode; mutex_lock(&c->tnc_mutex); - dbg_tnc("%d:%d, len %d, key %s", lnum, offs, len, DBGKEY(key)); + dbg_tnck(key, "%d:%d, len %d, key ", lnum, offs, len); found = lookup_level0_dirty(c, key, &znode, &n); if (!found) { struct ubifs_zbranch zbr; @@ -2221,8 +2219,8 @@ int ubifs_tnc_replace(struct ubifs_info *c, const union ubifs_key *key, struct ubifs_znode *znode; mutex_lock(&c->tnc_mutex); - dbg_tnc("old LEB %d:%d, new LEB %d:%d, len %d, key %s", old_lnum, - old_offs, lnum, offs, len, DBGKEY(key)); + dbg_tnck(key, "old LEB %d:%d, new LEB %d:%d, len %d, key ", old_lnum, + old_offs, lnum, offs, len); found = lookup_level0_dirty(c, key, &znode, &n); if (found < 0) { err = found; @@ -2304,8 +2302,8 @@ int ubifs_tnc_add_nm(struct ubifs_info *c, const union ubifs_key *key, struct ubifs_znode *znode; mutex_lock(&c->tnc_mutex); - dbg_tnc("LEB %d:%d, name '%.*s', key %s", lnum, offs, nm->len, nm->name, - DBGKEY(key)); + dbg_tnck(key, "LEB %d:%d, name '%.*s', key ", + lnum, offs, nm->len, nm->name); found = lookup_level0_dirty(c, key, &znode, &n); if (found < 0) { err = found; @@ -2398,7 +2396,7 @@ static int tnc_delete(struct ubifs_info *c, struct ubifs_znode *znode, int n) /* Delete without merge for now */ ubifs_assert(znode->level == 0); ubifs_assert(n >= 0 && n < c->fanout); - dbg_tnc("deleting %s", DBGKEY(&znode->zbranch[n].key)); + dbg_tnck(&znode->zbranch[n].key, "deleting key "); zbr = &znode->zbranch[n]; lnc_free(zbr); @@ -2508,7 +2506,7 @@ int ubifs_tnc_remove(struct ubifs_info *c, const union ubifs_key *key) struct ubifs_znode *znode; mutex_lock(&c->tnc_mutex); - dbg_tnc("key %s", DBGKEY(key)); + dbg_tnck(key, "key "); found = lookup_level0_dirty(c, key, &znode, &n); if (found < 0) { err = found; @@ -2539,7 +2537,7 @@ int ubifs_tnc_remove_nm(struct ubifs_info *c, const union ubifs_key *key, struct ubifs_znode *znode; mutex_lock(&c->tnc_mutex); - dbg_tnc("%.*s, key %s", nm->len, nm->name, DBGKEY(key)); + dbg_tnck(key, "%.*s, key ", nm->len, nm->name); err = lookup_level0_dirty(c, key, &znode, &n); if (err < 0) goto out_unlock; @@ -2654,7 +2652,7 @@ int ubifs_tnc_remove_range(struct ubifs_info *c, union ubifs_key *from_key, dbg_dump_znode(c, znode); goto out_unlock; } - dbg_tnc("removing %s", DBGKEY(key)); + dbg_tnck(key, "removing key "); } if (k) { for (i = n + 1 + k; i < znode->child_cnt; i++) @@ -2774,7 +2772,7 @@ struct ubifs_dent_node *ubifs_tnc_next_ent(struct ubifs_info *c, struct ubifs_zbranch *zbr; union ubifs_key *dkey; - dbg_tnc("%s %s", nm->name ? (char *)nm->name : "(lowest)", DBGKEY(key)); + dbg_tnck(key, "%s ", nm->name ? (char *)nm->name : "(lowest)"); ubifs_assert(is_hash_key(c, key)); mutex_lock(&c->tnc_mutex); @@ -3333,9 +3331,9 @@ int dbg_check_inode_size(struct ubifs_info *c, const struct inode *inode, out_dump: block = key_block(c, key); - ubifs_err("inode %lu has size %lld, but there are data at offset %lld " - "(data key %s)", (unsigned long)inode->i_ino, size, - ((loff_t)block) << UBIFS_BLOCK_SHIFT, DBGKEY(key)); + ubifs_err("inode %lu has size %lld, but there are data at offset %lld", + (unsigned long)inode->i_ino, size, + ((loff_t)block) << UBIFS_BLOCK_SHIFT); mutex_unlock(&c->tnc_mutex); dbg_dump_inode(c, inode); dbg_dump_stack(); diff --git a/fs/ubifs/tnc_misc.c b/fs/ubifs/tnc_misc.c index b48db999903e..dc28fe6ec07a 100644 --- a/fs/ubifs/tnc_misc.c +++ b/fs/ubifs/tnc_misc.c @@ -328,8 +328,8 @@ static int read_znode(struct ubifs_info *c, int lnum, int offs, int len, case UBIFS_XENT_KEY: break; default: - dbg_msg("bad key type at slot %d: %s", i, - DBGKEY(&zbr->key)); + dbg_msg("bad key type at slot %d: %d", + i, key_type(c, &zbr->key)); err = 3; goto out_dump; } @@ -475,7 +475,7 @@ int ubifs_tnc_read_node(struct ubifs_info *c, struct ubifs_zbranch *zbr, zbr->offs); if (err) { - dbg_tnc("key %s", DBGKEY(key)); + dbg_tnck(key, "key "); return err; } @@ -484,8 +484,8 @@ int ubifs_tnc_read_node(struct ubifs_info *c, struct ubifs_zbranch *zbr, if (!keys_eq(c, key, &key1)) { ubifs_err("bad key in node at LEB %d:%d", zbr->lnum, zbr->offs); - dbg_tnc("looked for key %s found node's key %s", - DBGKEY(key), DBGKEY1(&key1)); + dbg_tnck(key, "looked for key "); + dbg_tnck(&key1, "but found node's key "); dbg_dump_node(c, node); return -EINVAL; } diff --git a/fs/ubifs/ubifs.h b/fs/ubifs/ubifs.h index 27f22551f805..12e94774aa88 100644 --- a/fs/ubifs/ubifs.h +++ b/fs/ubifs/ubifs.h @@ -1734,7 +1734,7 @@ int ubifs_setattr(struct dentry *dentry, struct iattr *attr); /* dir.c */ struct inode *ubifs_new_inode(struct ubifs_info *c, const struct inode *dir, - int mode); + umode_t mode); int ubifs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat); diff --git a/fs/ubifs/xattr.c b/fs/ubifs/xattr.c index bf18f7a04544..85b272268754 100644 --- a/fs/ubifs/xattr.c +++ b/fs/ubifs/xattr.c @@ -138,12 +138,11 @@ static int create_xattr(struct ubifs_info *c, struct inode *host, ui = ubifs_inode(inode); ui->xattr = 1; ui->flags |= UBIFS_XATTR_FL; - ui->data = kmalloc(size, GFP_NOFS); + ui->data = kmemdup(value, size, GFP_NOFS); if (!ui->data) { err = -ENOMEM; goto out_free; } - memcpy(ui->data, value, size); inode->i_size = ui->ui_size = size; ui->data_len = size; @@ -204,12 +203,11 @@ static int change_xattr(struct ubifs_info *c, struct inode *host, return err; kfree(ui->data); - ui->data = kmalloc(size, GFP_NOFS); + ui->data = kmemdup(value, size, GFP_NOFS); if (!ui->data) { err = -ENOMEM; goto out_free; } - memcpy(ui->data, value, size); inode->i_size = ui->ui_size = size; ui->data_len = size; diff --git a/fs/udf/file.c b/fs/udf/file.c index d8ffa7cc661d..dca0c3881e82 100644 --- a/fs/udf/file.c +++ b/fs/udf/file.c @@ -125,7 +125,6 @@ static ssize_t udf_file_aio_write(struct kiocb *iocb, const struct iovec *iov, err = udf_expand_file_adinicb(inode); if (err) { udf_debug("udf_expand_adinicb: err=%d\n", err); - up_write(&iinfo->i_data_sem); return err; } } else { @@ -133,9 +132,10 @@ static ssize_t udf_file_aio_write(struct kiocb *iocb, const struct iovec *iov, iinfo->i_lenAlloc = pos + count; else iinfo->i_lenAlloc = inode->i_size; + up_write(&iinfo->i_data_sem); } - } - up_write(&iinfo->i_data_sem); + } else + up_write(&iinfo->i_data_sem); retval = generic_file_aio_write(iocb, iov, nr_segs, ppos); if (retval > 0) diff --git a/fs/udf/ialloc.c b/fs/udf/ialloc.c index 6fb7e0adcda0..05ab48195be9 100644 --- a/fs/udf/ialloc.c +++ b/fs/udf/ialloc.c @@ -46,7 +46,7 @@ void udf_free_inode(struct inode *inode) udf_free_blocks(sb, NULL, &UDF_I(inode)->i_location, 0, 1); } -struct inode *udf_new_inode(struct inode *dir, int mode, int *err) +struct inode *udf_new_inode(struct inode *dir, umode_t mode, int *err) { struct super_block *sb = dir->i_sb; struct udf_sb_info *sbi = UDF_SB(sb); diff --git a/fs/udf/inode.c b/fs/udf/inode.c index 4fd1d809738c..7699df7b3198 100644 --- a/fs/udf/inode.c +++ b/fs/udf/inode.c @@ -48,13 +48,12 @@ MODULE_LICENSE("GPL"); #define EXTENT_MERGE_SIZE 5 -static mode_t udf_convert_permissions(struct fileEntry *); +static umode_t udf_convert_permissions(struct fileEntry *); static int udf_update_inode(struct inode *, int); static void udf_fill_inode(struct inode *, struct buffer_head *); static int udf_sync_inode(struct inode *inode); static int udf_alloc_i_data(struct inode *inode, size_t size); -static struct buffer_head *inode_getblk(struct inode *, sector_t, int *, - sector_t *, int *); +static sector_t inode_getblk(struct inode *, sector_t, int *, int *); static int8_t udf_insert_aext(struct inode *, struct extent_position, struct kernel_lb_addr, uint32_t); static void udf_split_extents(struct inode *, int *, int, int, @@ -151,6 +150,12 @@ const struct address_space_operations udf_aops = { .bmap = udf_bmap, }; +/* + * Expand file stored in ICB to a normal one-block-file + * + * This function requires i_data_sem for writing and releases it. + * This function requires i_mutex held + */ int udf_expand_file_adinicb(struct inode *inode) { struct page *page; @@ -169,9 +174,15 @@ int udf_expand_file_adinicb(struct inode *inode) iinfo->i_alloc_type = ICBTAG_FLAG_AD_LONG; /* from now on we have normal address_space methods */ inode->i_data.a_ops = &udf_aops; + up_write(&iinfo->i_data_sem); mark_inode_dirty(inode); return 0; } + /* + * Release i_data_sem so that we can lock a page - page lock ranks + * above i_data_sem. i_mutex still protects us against file changes. + */ + up_write(&iinfo->i_data_sem); page = find_or_create_page(inode->i_mapping, 0, GFP_NOFS); if (!page) @@ -187,6 +198,7 @@ int udf_expand_file_adinicb(struct inode *inode) SetPageUptodate(page); kunmap(page); } + down_write(&iinfo->i_data_sem); memset(iinfo->i_ext.i_data + iinfo->i_lenEAttr, 0x00, iinfo->i_lenAlloc); iinfo->i_lenAlloc = 0; @@ -196,17 +208,20 @@ int udf_expand_file_adinicb(struct inode *inode) iinfo->i_alloc_type = ICBTAG_FLAG_AD_LONG; /* from now on we have normal address_space methods */ inode->i_data.a_ops = &udf_aops; + up_write(&iinfo->i_data_sem); err = inode->i_data.a_ops->writepage(page, &udf_wbc); if (err) { /* Restore everything back so that we don't lose data... */ lock_page(page); kaddr = kmap(page); + down_write(&iinfo->i_data_sem); memcpy(iinfo->i_ext.i_data + iinfo->i_lenEAttr, kaddr, inode->i_size); kunmap(page); unlock_page(page); iinfo->i_alloc_type = ICBTAG_FLAG_AD_IN_ICB; inode->i_data.a_ops = &udf_adinicb_aops; + up_write(&iinfo->i_data_sem); } page_cache_release(page); mark_inode_dirty(inode); @@ -310,7 +325,6 @@ static int udf_get_block(struct inode *inode, sector_t block, struct buffer_head *bh_result, int create) { int err, new; - struct buffer_head *bh; sector_t phys = 0; struct udf_inode_info *iinfo; @@ -323,7 +337,6 @@ static int udf_get_block(struct inode *inode, sector_t block, err = -EIO; new = 0; - bh = NULL; iinfo = UDF_I(inode); down_write(&iinfo->i_data_sem); @@ -332,13 +345,10 @@ static int udf_get_block(struct inode *inode, sector_t block, iinfo->i_next_alloc_goal++; } - err = 0; - bh = inode_getblk(inode, block, &err, &phys, &new); - BUG_ON(bh); - if (err) + phys = inode_getblk(inode, block, &err, &new); + if (!phys) goto abort; - BUG_ON(!phys); if (new) set_buffer_new(bh_result); @@ -547,11 +557,10 @@ out: return err; } -static struct buffer_head *inode_getblk(struct inode *inode, sector_t block, - int *err, sector_t *phys, int *new) +static sector_t inode_getblk(struct inode *inode, sector_t block, + int *err, int *new) { static sector_t last_block; - struct buffer_head *result = NULL; struct kernel_long_ad laarr[EXTENT_MERGE_SIZE]; struct extent_position prev_epos, cur_epos, next_epos; int count = 0, startnum = 0, endnum = 0; @@ -566,6 +575,8 @@ static struct buffer_head *inode_getblk(struct inode *inode, sector_t block, int goal = 0, pgoal = iinfo->i_location.logicalBlockNum; int lastblock = 0; + *err = 0; + *new = 0; prev_epos.offset = udf_file_entry_alloc_offset(inode); prev_epos.block = iinfo->i_location; prev_epos.bh = NULL; @@ -635,8 +646,7 @@ static struct buffer_head *inode_getblk(struct inode *inode, sector_t block, brelse(cur_epos.bh); brelse(next_epos.bh); newblock = udf_get_lb_pblock(inode->i_sb, &eloc, offset); - *phys = newblock; - return NULL; + return newblock; } last_block = block; @@ -664,7 +674,7 @@ static struct buffer_head *inode_getblk(struct inode *inode, sector_t block, brelse(cur_epos.bh); brelse(next_epos.bh); *err = ret; - return NULL; + return 0; } c = 0; offset = 0; @@ -729,7 +739,7 @@ static struct buffer_head *inode_getblk(struct inode *inode, sector_t block, if (!newblocknum) { brelse(prev_epos.bh); *err = -ENOSPC; - return NULL; + return 0; } iinfo->i_lenExtents += inode->i_sb->s_blocksize; } @@ -761,10 +771,10 @@ static struct buffer_head *inode_getblk(struct inode *inode, sector_t block, newblock = udf_get_pblock(inode->i_sb, newblocknum, iinfo->i_location.partitionReferenceNum, 0); - if (!newblock) - return NULL; - *phys = newblock; - *err = 0; + if (!newblock) { + *err = -EIO; + return 0; + } *new = 1; iinfo->i_next_alloc_block = block; iinfo->i_next_alloc_goal = newblocknum; @@ -775,7 +785,7 @@ static struct buffer_head *inode_getblk(struct inode *inode, sector_t block, else mark_inode_dirty(inode); - return result; + return newblock; } static void udf_split_extents(struct inode *inode, int *c, int offset, @@ -1111,10 +1121,9 @@ int udf_setsize(struct inode *inode, loff_t newsize) if (bsize < (udf_file_entry_alloc_offset(inode) + newsize)) { err = udf_expand_file_adinicb(inode); - if (err) { - up_write(&iinfo->i_data_sem); + if (err) return err; - } + down_write(&iinfo->i_data_sem); } else iinfo->i_lenAlloc = newsize; } @@ -1452,9 +1461,9 @@ static int udf_alloc_i_data(struct inode *inode, size_t size) return 0; } -static mode_t udf_convert_permissions(struct fileEntry *fe) +static umode_t udf_convert_permissions(struct fileEntry *fe) { - mode_t mode; + umode_t mode; uint32_t permissions; uint32_t flags; diff --git a/fs/udf/namei.c b/fs/udf/namei.c index 4639e137222f..08bf46edf9c4 100644 --- a/fs/udf/namei.c +++ b/fs/udf/namei.c @@ -552,7 +552,7 @@ static int udf_delete_entry(struct inode *inode, struct fileIdentDesc *fi, return udf_write_fi(inode, cfi, fi, fibh, NULL, NULL); } -static int udf_create(struct inode *dir, struct dentry *dentry, int mode, +static int udf_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *nd) { struct udf_fileident_bh fibh; @@ -596,7 +596,7 @@ static int udf_create(struct inode *dir, struct dentry *dentry, int mode, return 0; } -static int udf_mknod(struct inode *dir, struct dentry *dentry, int mode, +static int udf_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) { struct inode *inode; @@ -640,7 +640,7 @@ out: return err; } -static int udf_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int udf_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct inode *inode; struct udf_fileident_bh fibh; diff --git a/fs/udf/super.c b/fs/udf/super.c index e185253470df..c09a84daaf50 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -89,7 +89,7 @@ static void udf_open_lvid(struct super_block *); static void udf_close_lvid(struct super_block *); static unsigned int udf_count_free(struct super_block *); static int udf_statfs(struct dentry *, struct kstatfs *); -static int udf_show_options(struct seq_file *, struct vfsmount *); +static int udf_show_options(struct seq_file *, struct dentry *); struct logicalVolIntegrityDescImpUse *udf_sb_lvidiu(struct udf_sb_info *sbi) { @@ -138,7 +138,6 @@ static struct inode *udf_alloc_inode(struct super_block *sb) static void udf_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(udf_inode_cachep, UDF_I(inode)); } @@ -196,11 +195,11 @@ struct udf_options { unsigned int fileset; unsigned int rootdir; unsigned int flags; - mode_t umask; + umode_t umask; gid_t gid; uid_t uid; - mode_t fmode; - mode_t dmode; + umode_t fmode; + umode_t dmode; struct nls_table *nls_map; }; @@ -250,9 +249,9 @@ static int udf_sb_alloc_partition_maps(struct super_block *sb, u32 count) return 0; } -static int udf_show_options(struct seq_file *seq, struct vfsmount *mnt) +static int udf_show_options(struct seq_file *seq, struct dentry *root) { - struct super_block *sb = mnt->mnt_sb; + struct super_block *sb = root->d_sb; struct udf_sb_info *sbi = UDF_SB(sb); if (!UDF_QUERY_FLAG(sb, UDF_FLAG_STRICT)) @@ -280,11 +279,11 @@ static int udf_show_options(struct seq_file *seq, struct vfsmount *mnt) if (UDF_QUERY_FLAG(sb, UDF_FLAG_GID_SET)) seq_printf(seq, ",gid=%u", sbi->s_gid); if (sbi->s_umask != 0) - seq_printf(seq, ",umask=%o", sbi->s_umask); + seq_printf(seq, ",umask=%ho", sbi->s_umask); if (sbi->s_fmode != UDF_INVALID_MODE) - seq_printf(seq, ",mode=%o", sbi->s_fmode); + seq_printf(seq, ",mode=%ho", sbi->s_fmode); if (sbi->s_dmode != UDF_INVALID_MODE) - seq_printf(seq, ",dmode=%o", sbi->s_dmode); + seq_printf(seq, ",dmode=%ho", sbi->s_dmode); if (UDF_QUERY_FLAG(sb, UDF_FLAG_SESSION_SET)) seq_printf(seq, ",session=%u", sbi->s_session); if (UDF_QUERY_FLAG(sb, UDF_FLAG_LASTBLOCK_SET)) @@ -1799,6 +1798,12 @@ static void udf_close_lvid(struct super_block *sb) le16_to_cpu(lvid->descTag.descCRCLength))); lvid->descTag.tagChecksum = udf_tag_checksum(&lvid->descTag); + /* + * We set buffer uptodate unconditionally here to avoid spurious + * warnings from mark_buffer_dirty() when previous EIO has marked + * the buffer as !uptodate + */ + set_buffer_uptodate(bh); mark_buffer_dirty(bh); sbi->s_lvid_dirty = 0; mutex_unlock(&sbi->s_alloc_mutex); diff --git a/fs/udf/symlink.c b/fs/udf/symlink.c index b1d4488b0f14..d7c6dbe4194b 100644 --- a/fs/udf/symlink.c +++ b/fs/udf/symlink.c @@ -41,10 +41,16 @@ static void udf_pc_to_char(struct super_block *sb, unsigned char *from, pc = (struct pathComponent *)(from + elen); switch (pc->componentType) { case 1: - if (pc->lengthComponentIdent == 0) { - p = to; - *p++ = '/'; - } + /* + * Symlink points to some place which should be agreed + * upon between originator and receiver of the media. Ignore. + */ + if (pc->lengthComponentIdent > 0) + break; + /* Fall through */ + case 2: + p = to; + *p++ = '/'; break; case 3: memcpy(p, "../", 3); diff --git a/fs/udf/udf_sb.h b/fs/udf/udf_sb.h index 5142a82e3276..42ad69ac9576 100644 --- a/fs/udf/udf_sb.h +++ b/fs/udf/udf_sb.h @@ -50,7 +50,7 @@ #define UDF_SPARABLE_MAP15 0x1522U #define UDF_METADATA_MAP25 0x2511U -#define UDF_INVALID_MODE ((mode_t)-1) +#define UDF_INVALID_MODE ((umode_t)-1) #pragma pack(1) /* XXX(hch): Why? This file just defines in-core structures */ @@ -127,11 +127,11 @@ struct udf_sb_info { struct buffer_head *s_lvid_bh; /* Default permissions */ - mode_t s_umask; + umode_t s_umask; gid_t s_gid; uid_t s_uid; - mode_t s_fmode; - mode_t s_dmode; + umode_t s_fmode; + umode_t s_dmode; /* Lock protecting consistency of above permission settings */ rwlock_t s_cred_lock; diff --git a/fs/udf/udfdecl.h b/fs/udf/udfdecl.h index f34e6fc0cdaa..ebe10314e512 100644 --- a/fs/udf/udfdecl.h +++ b/fs/udf/udfdecl.h @@ -215,7 +215,7 @@ extern int udf_CS0toUTF8(struct ustr *, const struct ustr *); /* ialloc.c */ extern void udf_free_inode(struct inode *); -extern struct inode *udf_new_inode(struct inode *, int, int *); +extern struct inode *udf_new_inode(struct inode *, umode_t, int *); /* truncate.c */ extern void udf_truncate_tail_extent(struct inode *); diff --git a/fs/ufs/ialloc.c b/fs/ufs/ialloc.c index 78a4c70d46b5..4ec5c1085a87 100644 --- a/fs/ufs/ialloc.c +++ b/fs/ufs/ialloc.c @@ -170,7 +170,7 @@ static void ufs2_init_inodes_chunk(struct super_block *sb, * For other inodes, search forward from the parent directory's block * group to find a free inode. */ -struct inode * ufs_new_inode(struct inode * dir, int mode) +struct inode *ufs_new_inode(struct inode *dir, umode_t mode) { struct super_block * sb; struct ufs_sb_info * sbi; diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c index 879b13436fa4..9094e1d917be 100644 --- a/fs/ufs/inode.c +++ b/fs/ufs/inode.c @@ -583,7 +583,7 @@ static int ufs1_read_inode(struct inode *inode, struct ufs_inode *ufs_inode) { struct ufs_inode_info *ufsi = UFS_I(inode); struct super_block *sb = inode->i_sb; - mode_t mode; + umode_t mode; /* * Copy data to the in-core inode. @@ -630,7 +630,7 @@ static int ufs2_read_inode(struct inode *inode, struct ufs2_inode *ufs2_inode) { struct ufs_inode_info *ufsi = UFS_I(inode); struct super_block *sb = inode->i_sb; - mode_t mode; + umode_t mode; UFSD("Reading ufs2 inode, ino %lu\n", inode->i_ino); /* diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c index 639d49162241..38cac199edff 100644 --- a/fs/ufs/namei.c +++ b/fs/ufs/namei.c @@ -70,7 +70,7 @@ static struct dentry *ufs_lookup(struct inode * dir, struct dentry *dentry, stru * If the create succeeds, we fill in the inode information * with d_instantiate(). */ -static int ufs_create (struct inode * dir, struct dentry * dentry, int mode, +static int ufs_create (struct inode * dir, struct dentry * dentry, umode_t mode, struct nameidata *nd) { struct inode *inode; @@ -94,7 +94,7 @@ static int ufs_create (struct inode * dir, struct dentry * dentry, int mode, return err; } -static int ufs_mknod (struct inode * dir, struct dentry *dentry, int mode, dev_t rdev) +static int ufs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) { struct inode *inode; int err; @@ -180,7 +180,7 @@ static int ufs_link (struct dentry * old_dentry, struct inode * dir, return error; } -static int ufs_mkdir(struct inode * dir, struct dentry * dentry, int mode) +static int ufs_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode) { struct inode * inode; int err = -EMLINK; diff --git a/fs/ufs/super.c b/fs/ufs/super.c index 3915ade6f9a8..5246ee3e5607 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -1351,9 +1351,9 @@ static int ufs_remount (struct super_block *sb, int *mount_flags, char *data) return 0; } -static int ufs_show_options(struct seq_file *seq, struct vfsmount *vfs) +static int ufs_show_options(struct seq_file *seq, struct dentry *root) { - struct ufs_sb_info *sbi = UFS_SB(vfs->mnt_sb); + struct ufs_sb_info *sbi = UFS_SB(root->d_sb); unsigned mval = sbi->s_mount_opt & UFS_MOUNT_UFSTYPE; const struct match_token *tp = tokens; @@ -1425,7 +1425,6 @@ static struct inode *ufs_alloc_inode(struct super_block *sb) static void ufs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(ufs_inode_cachep, UFS_I(inode)); } diff --git a/fs/ufs/ufs.h b/fs/ufs/ufs.h index c26f2bcec264..528750b7e701 100644 --- a/fs/ufs/ufs.h +++ b/fs/ufs/ufs.h @@ -104,7 +104,7 @@ extern const struct address_space_operations ufs_aops; /* ialloc.c */ extern void ufs_free_inode (struct inode *inode); -extern struct inode * ufs_new_inode (struct inode *, int); +extern struct inode * ufs_new_inode (struct inode *, umode_t); /* inode.c */ extern struct inode *ufs_iget(struct super_block *, unsigned long); diff --git a/fs/xattr.c b/fs/xattr.c index 67583de8218c..82f43376c7cd 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -397,7 +397,7 @@ SYSCALL_DEFINE5(fsetxattr, int, fd, const char __user *, name, error = mnt_want_write_file(f); if (!error) { error = setxattr(dentry, name, value, size, flags); - mnt_drop_write(f->f_path.mnt); + mnt_drop_write_file(f); } fput(f); return error; @@ -624,7 +624,7 @@ SYSCALL_DEFINE2(fremovexattr, int, fd, const char __user *, name) error = mnt_want_write_file(f); if (!error) { error = removexattr(dentry, name); - mnt_drop_write(f->f_path.mnt); + mnt_drop_write_file(f); } fput(f); return error; diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c index b6c4b3795c4a..ac702a6eab9b 100644 --- a/fs/xfs/xfs_acl.c +++ b/fs/xfs/xfs_acl.c @@ -39,9 +39,11 @@ xfs_acl_from_disk(struct xfs_acl *aclp) struct posix_acl_entry *acl_e; struct posix_acl *acl; struct xfs_acl_entry *ace; - int count, i; + unsigned int count, i; count = be32_to_cpu(aclp->acl_cnt); + if (count > XFS_ACL_MAX_ENTRIES) + return ERR_PTR(-EFSCORRUPTED); acl = posix_acl_alloc(count, GFP_KERNEL); if (!acl) diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c index d4906e7c9787..c1b55e596551 100644 --- a/fs/xfs/xfs_attr_leaf.c +++ b/fs/xfs/xfs_attr_leaf.c @@ -110,6 +110,7 @@ xfs_attr_namesp_match(int arg_flags, int ondisk_flags) /* * Query whether the requested number of additional bytes of extended * attribute space will be able to fit inline. + * * Returns zero if not, else the di_forkoff fork offset to be used in the * literal area for attribute data once the new bytes have been added. * @@ -122,7 +123,7 @@ xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes) int offset; int minforkoff; /* lower limit on valid forkoff locations */ int maxforkoff; /* upper limit on valid forkoff locations */ - int dsize; + int dsize; xfs_mount_t *mp = dp->i_mount; offset = (XFS_LITINO(mp) - bytes) >> 3; /* rounded down */ @@ -136,47 +137,60 @@ xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes) return (offset >= minforkoff) ? minforkoff : 0; } - if (!(mp->m_flags & XFS_MOUNT_ATTR2)) { - if (bytes <= XFS_IFORK_ASIZE(dp)) - return dp->i_d.di_forkoff; + /* + * If the requested numbers of bytes is smaller or equal to the + * current attribute fork size we can always proceed. + * + * Note that if_bytes in the data fork might actually be larger than + * the current data fork size is due to delalloc extents. In that + * case either the extent count will go down when they are converted + * to real extents, or the delalloc conversion will take care of the + * literal area rebalancing. + */ + if (bytes <= XFS_IFORK_ASIZE(dp)) + return dp->i_d.di_forkoff; + + /* + * For attr2 we can try to move the forkoff if there is space in the + * literal area, but for the old format we are done if there is no + * space in the fixed attribute fork. + */ + if (!(mp->m_flags & XFS_MOUNT_ATTR2)) return 0; - } dsize = dp->i_df.if_bytes; - + switch (dp->i_d.di_format) { case XFS_DINODE_FMT_EXTENTS: - /* + /* * If there is no attr fork and the data fork is extents, - * determine if creating the default attr fork will result - * in the extents form migrating to btree. If so, the - * minimum offset only needs to be the space required for + * determine if creating the default attr fork will result + * in the extents form migrating to btree. If so, the + * minimum offset only needs to be the space required for * the btree root. - */ + */ if (!dp->i_d.di_forkoff && dp->i_df.if_bytes > xfs_default_attroffset(dp)) dsize = XFS_BMDR_SPACE_CALC(MINDBTPTRS); break; - case XFS_DINODE_FMT_BTREE: /* - * If have data btree then keep forkoff if we have one, - * otherwise we are adding a new attr, so then we set - * minforkoff to where the btree root can finish so we have + * If we have a data btree then keep forkoff if we have one, + * otherwise we are adding a new attr, so then we set + * minforkoff to where the btree root can finish so we have * plenty of room for attrs */ if (dp->i_d.di_forkoff) { - if (offset < dp->i_d.di_forkoff) + if (offset < dp->i_d.di_forkoff) return 0; - else - return dp->i_d.di_forkoff; - } else - dsize = XFS_BMAP_BROOT_SPACE(dp->i_df.if_broot); + return dp->i_d.di_forkoff; + } + dsize = XFS_BMAP_BROOT_SPACE(dp->i_df.if_broot); break; } - - /* - * A data fork btree root must have space for at least + + /* + * A data fork btree root must have space for at least * MINDBTPTRS key/ptr pairs if the data fork is small or empty. */ minforkoff = MAX(dsize, XFS_BMDR_SPACE_CALC(MINDBTPTRS)); @@ -186,10 +200,10 @@ xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes) maxforkoff = XFS_LITINO(mp) - XFS_BMDR_SPACE_CALC(MINABTPTRS); maxforkoff = maxforkoff >> 3; /* rounded down */ - if (offset >= minforkoff && offset < maxforkoff) - return offset; if (offset >= maxforkoff) return maxforkoff; + if (offset >= minforkoff) + return offset; return 0; } diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index c68baeb0974a..d0ab78837057 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -2383,6 +2383,8 @@ xfs_bmap_btalloc( int tryagain; int error; + ASSERT(ap->length); + mp = ap->ip->i_mount; align = ap->userdata ? xfs_get_extsz_hint(ap->ip) : 0; if (unlikely(align)) { @@ -4629,6 +4631,8 @@ xfs_bmapi_allocate( int error; int rt; + ASSERT(bma->length > 0); + rt = (whichfork == XFS_DATA_FORK) && XFS_IS_REALTIME_INODE(bma->ip); /* @@ -4849,6 +4853,7 @@ xfs_bmapi_write( ASSERT(*nmap <= XFS_BMAP_MAX_NMAP); ASSERT(!(flags & XFS_BMAPI_IGSTATE)); ASSERT(tp != NULL); + ASSERT(len > 0); whichfork = (flags & XFS_BMAPI_ATTRFORK) ? XFS_ATTR_FORK : XFS_DATA_FORK; @@ -4918,9 +4923,22 @@ xfs_bmapi_write( bma.eof = eof; bma.conv = !!(flags & XFS_BMAPI_CONVERT); bma.wasdel = wasdelay; - bma.length = len; bma.offset = bno; + /* + * There's a 32/64 bit type mismatch between the + * allocation length request (which can be 64 bits in + * length) and the bma length request, which is + * xfs_extlen_t and therefore 32 bits. Hence we have to + * check for 32-bit overflows and handle them here. + */ + if (len > (xfs_filblks_t)MAXEXTLEN) + bma.length = MAXEXTLEN; + else + bma.length = len; + + ASSERT(len > 0); + ASSERT(bma.length > 0); error = xfs_bmapi_allocate(&bma, flags); if (error) goto error0; diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index cf0ac056815f..4dff85c7d7eb 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -1370,7 +1370,7 @@ restart: goto restart; } /* - * clear the LRU reference count so the bufer doesn't get + * clear the LRU reference count so the buffer doesn't get * ignored in xfs_buf_rele(). */ atomic_set(&bp->b_lru_ref, 0); @@ -1701,12 +1701,8 @@ xfsbufd( struct list_head tmp; struct blk_plug plug; - if (unlikely(freezing(current))) { - set_bit(XBT_FORCE_SLEEP, &target->bt_flags); - refrigerator(); - } else { - clear_bit(XBT_FORCE_SLEEP, &target->bt_flags); - } + if (unlikely(freezing(current))) + try_to_freeze(); /* sleep for a long time if there is nothing to do. */ if (list_empty(&target->bt_delwri_queue)) diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h index 5bab046e859f..df7ffb0affe7 100644 --- a/fs/xfs/xfs_buf.h +++ b/fs/xfs/xfs_buf.h @@ -90,8 +90,7 @@ typedef unsigned int xfs_buf_flags_t; { _XBF_DELWRI_Q, "DELWRI_Q" } typedef enum { - XBT_FORCE_SLEEP = 0, - XBT_FORCE_FLUSH = 1, + XBT_FORCE_FLUSH = 0, } xfs_buftarg_flags_t; typedef struct xfs_buftarg { diff --git a/fs/xfs/xfs_discard.c b/fs/xfs/xfs_discard.c index 8a24f0c6c860..286a051f12cf 100644 --- a/fs/xfs/xfs_discard.c +++ b/fs/xfs/xfs_discard.c @@ -68,7 +68,7 @@ xfs_trim_extents( * Look up the longest btree in the AGF and start with it. */ error = xfs_alloc_lookup_le(cur, 0, - XFS_BUF_TO_AGF(agbp)->agf_longest, &i); + be32_to_cpu(XFS_BUF_TO_AGF(agbp)->agf_longest), &i); if (error) goto out_del_cursor; @@ -84,7 +84,7 @@ xfs_trim_extents( if (error) goto out_del_cursor; XFS_WANT_CORRUPTED_GOTO(i == 1, out_del_cursor); - ASSERT(flen <= XFS_BUF_TO_AGF(agbp)->agf_longest); + ASSERT(flen <= be32_to_cpu(XFS_BUF_TO_AGF(agbp)->agf_longest)); /* * Too small? Give up. diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c index 25d7280e9f6b..b4ff40b5f918 100644 --- a/fs/xfs/xfs_dquot.c +++ b/fs/xfs/xfs_dquot.c @@ -39,20 +39,19 @@ #include "xfs_qm.h" #include "xfs_trace.h" - /* - LOCK ORDER - - inode lock (ilock) - dquot hash-chain lock (hashlock) - xqm dquot freelist lock (freelistlock - mount's dquot list lock (mplistlock) - user dquot lock - lock ordering among dquots is based on the uid or gid - group dquot lock - similar to udquots. Between the two dquots, the udquot - has to be locked first. - pin lock - the dquot lock must be held to take this lock. - flush lock - ditto. -*/ + * Lock order: + * + * ip->i_lock + * qh->qh_lock + * qi->qi_dqlist_lock + * dquot->q_qlock (xfs_dqlock() and friends) + * dquot->q_flush (xfs_dqflock() and friends) + * xfs_Gqm->qm_dqfrlist_lock + * + * If two dquots need to be locked the order is user before group/project, + * otherwise by the lowest id first, see xfs_dqlock2. + */ #ifdef DEBUG xfs_buftarg_t *xfs_dqerror_target; @@ -155,24 +154,6 @@ xfs_qm_dqdestroy( } /* - * This is what a 'fresh' dquot inside a dquot chunk looks like on disk. - */ -STATIC void -xfs_qm_dqinit_core( - xfs_dqid_t id, - uint type, - xfs_dqblk_t *d) -{ - /* - * Caller has zero'd the entire dquot 'chunk' already. - */ - d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC); - d->dd_diskdq.d_version = XFS_DQUOT_VERSION; - d->dd_diskdq.d_id = cpu_to_be32(id); - d->dd_diskdq.d_flags = type; -} - -/* * If default limits are in force, push them into the dquot now. * We overwrite the dquot limits only if they are zero and this * is not the root dquot. @@ -328,8 +309,13 @@ xfs_qm_init_dquot_blk( curid = id - (id % q->qi_dqperchunk); ASSERT(curid >= 0); memset(d, 0, BBTOB(q->qi_dqchunklen)); - for (i = 0; i < q->qi_dqperchunk; i++, d++, curid++) - xfs_qm_dqinit_core(curid, type, d); + for (i = 0; i < q->qi_dqperchunk; i++, d++, curid++) { + d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC); + d->dd_diskdq.d_version = XFS_DQUOT_VERSION; + d->dd_diskdq.d_id = cpu_to_be32(curid); + d->dd_diskdq.d_flags = type; + } + xfs_trans_dquot_buf(tp, bp, (type & XFS_DQ_USER ? XFS_BLF_UDQUOT_BUF : ((type & XFS_DQ_PROJ) ? XFS_BLF_PDQUOT_BUF : @@ -564,36 +550,62 @@ xfs_qm_dqtobp( * Read in the ondisk dquot using dqtobp() then copy it to an incore version, * and release the buffer immediately. * + * If XFS_QMOPT_DQALLOC is set, allocate a dquot on disk if it needed. */ -/* ARGSUSED */ -STATIC int +int xfs_qm_dqread( - xfs_trans_t **tpp, - xfs_dqid_t id, - xfs_dquot_t *dqp, /* dquot to get filled in */ - uint flags) + struct xfs_mount *mp, + xfs_dqid_t id, + uint type, + uint flags, + struct xfs_dquot **O_dqpp) { - xfs_disk_dquot_t *ddqp; - xfs_buf_t *bp; - int error; - xfs_trans_t *tp; + struct xfs_dquot *dqp; + struct xfs_disk_dquot *ddqp; + struct xfs_buf *bp; + struct xfs_trans *tp = NULL; + int error; + int cancelflags = 0; - ASSERT(tpp); + dqp = xfs_qm_dqinit(mp, id, type); trace_xfs_dqread(dqp); + if (flags & XFS_QMOPT_DQALLOC) { + tp = xfs_trans_alloc(mp, XFS_TRANS_QM_DQALLOC); + error = xfs_trans_reserve(tp, XFS_QM_DQALLOC_SPACE_RES(mp), + XFS_WRITE_LOG_RES(mp) + + /* + * Round the chunklen up to the next multiple + * of 128 (buf log item chunk size)). + */ + BBTOB(mp->m_quotainfo->qi_dqchunklen) - 1 + 128, + 0, + XFS_TRANS_PERM_LOG_RES, + XFS_WRITE_LOG_COUNT); + if (error) + goto error1; + cancelflags = XFS_TRANS_RELEASE_LOG_RES; + } + /* * get a pointer to the on-disk dquot and the buffer containing it * dqp already knows its own type (GROUP/USER). */ - if ((error = xfs_qm_dqtobp(tpp, dqp, &ddqp, &bp, flags))) { - return (error); + error = xfs_qm_dqtobp(&tp, dqp, &ddqp, &bp, flags); + if (error) { + /* + * This can happen if quotas got turned off (ESRCH), + * or if the dquot didn't exist on disk and we ask to + * allocate (ENOENT). + */ + trace_xfs_dqread_fail(dqp); + cancelflags |= XFS_TRANS_ABORT; + goto error1; } - tp = *tpp; /* copy everything from disk dquot to the incore dquot */ memcpy(&dqp->q_core, ddqp, sizeof(xfs_disk_dquot_t)); - ASSERT(be32_to_cpu(dqp->q_core.d_id) == id); xfs_qm_dquot_logitem_init(dqp); /* @@ -622,77 +634,22 @@ xfs_qm_dqread( ASSERT(xfs_buf_islocked(bp)); xfs_trans_brelse(tp, bp); - return (error); -} - - -/* - * allocate an incore dquot from the kernel heap, - * and fill its core with quota information kept on disk. - * If XFS_QMOPT_DQALLOC is set, it'll allocate a dquot on disk - * if it wasn't already allocated. - */ -STATIC int -xfs_qm_idtodq( - xfs_mount_t *mp, - xfs_dqid_t id, /* gid or uid, depending on type */ - uint type, /* UDQUOT or GDQUOT */ - uint flags, /* DQALLOC, DQREPAIR */ - xfs_dquot_t **O_dqpp)/* OUT : incore dquot, not locked */ -{ - xfs_dquot_t *dqp; - int error; - xfs_trans_t *tp; - int cancelflags=0; - - dqp = xfs_qm_dqinit(mp, id, type); - tp = NULL; - if (flags & XFS_QMOPT_DQALLOC) { - tp = xfs_trans_alloc(mp, XFS_TRANS_QM_DQALLOC); - error = xfs_trans_reserve(tp, XFS_QM_DQALLOC_SPACE_RES(mp), - XFS_WRITE_LOG_RES(mp) + - BBTOB(mp->m_quotainfo->qi_dqchunklen) - 1 + - 128, - 0, - XFS_TRANS_PERM_LOG_RES, - XFS_WRITE_LOG_COUNT); - if (error) { - cancelflags = 0; - goto error0; - } - cancelflags = XFS_TRANS_RELEASE_LOG_RES; - } - - /* - * Read it from disk; xfs_dqread() takes care of - * all the necessary initialization of dquot's fields (locks, etc) - */ - if ((error = xfs_qm_dqread(&tp, id, dqp, flags))) { - /* - * This can happen if quotas got turned off (ESRCH), - * or if the dquot didn't exist on disk and we ask to - * allocate (ENOENT). - */ - trace_xfs_dqread_fail(dqp); - cancelflags |= XFS_TRANS_ABORT; - goto error0; - } if (tp) { - if ((error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES))) - goto error1; + error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); + if (error) + goto error0; } *O_dqpp = dqp; - return (0); + return error; - error0: - ASSERT(error); +error1: if (tp) xfs_trans_cancel(tp, cancelflags); - error1: +error0: xfs_qm_dqdestroy(dqp); *O_dqpp = NULL; - return (error); + return error; } /* @@ -710,12 +667,9 @@ xfs_qm_dqlookup( xfs_dquot_t **O_dqpp) { xfs_dquot_t *dqp; - uint flist_locked; ASSERT(mutex_is_locked(&qh->qh_lock)); - flist_locked = B_FALSE; - /* * Traverse the hashchain looking for a match */ @@ -725,70 +679,31 @@ xfs_qm_dqlookup( * dqlock to look at the id field of the dquot, since the * id can't be modified without the hashlock anyway. */ - if (be32_to_cpu(dqp->q_core.d_id) == id && dqp->q_mount == mp) { - trace_xfs_dqlookup_found(dqp); + if (be32_to_cpu(dqp->q_core.d_id) != id || dqp->q_mount != mp) + continue; - /* - * All in core dquots must be on the dqlist of mp - */ - ASSERT(!list_empty(&dqp->q_mplist)); - - xfs_dqlock(dqp); - if (dqp->q_nrefs == 0) { - ASSERT(!list_empty(&dqp->q_freelist)); - if (!mutex_trylock(&xfs_Gqm->qm_dqfrlist_lock)) { - trace_xfs_dqlookup_want(dqp); - - /* - * We may have raced with dqreclaim_one() - * (and lost). So, flag that we don't - * want the dquot to be reclaimed. - */ - dqp->dq_flags |= XFS_DQ_WANT; - xfs_dqunlock(dqp); - mutex_lock(&xfs_Gqm->qm_dqfrlist_lock); - xfs_dqlock(dqp); - dqp->dq_flags &= ~(XFS_DQ_WANT); - } - flist_locked = B_TRUE; - } + trace_xfs_dqlookup_found(dqp); - /* - * id couldn't have changed; we had the hashlock all - * along - */ - ASSERT(be32_to_cpu(dqp->q_core.d_id) == id); - - if (flist_locked) { - if (dqp->q_nrefs != 0) { - mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); - flist_locked = B_FALSE; - } else { - /* take it off the freelist */ - trace_xfs_dqlookup_freelist(dqp); - list_del_init(&dqp->q_freelist); - xfs_Gqm->qm_dqfrlist_cnt--; - } - } + xfs_dqlock(dqp); + if (dqp->dq_flags & XFS_DQ_FREEING) { + *O_dqpp = NULL; + xfs_dqunlock(dqp); + return -1; + } - XFS_DQHOLD(dqp); + dqp->q_nrefs++; - if (flist_locked) - mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); - /* - * move the dquot to the front of the hashchain - */ - ASSERT(mutex_is_locked(&qh->qh_lock)); - list_move(&dqp->q_hashlist, &qh->qh_list); - trace_xfs_dqlookup_done(dqp); - *O_dqpp = dqp; - return 0; - } + /* + * move the dquot to the front of the hashchain + */ + list_move(&dqp->q_hashlist, &qh->qh_list); + trace_xfs_dqlookup_done(dqp); + *O_dqpp = dqp; + return 0; } *O_dqpp = NULL; - ASSERT(mutex_is_locked(&qh->qh_lock)); - return (1); + return 1; } /* @@ -829,11 +744,7 @@ xfs_qm_dqget( return (EIO); } } -#endif - - again: -#ifdef DEBUG ASSERT(type == XFS_DQ_USER || type == XFS_DQ_PROJ || type == XFS_DQ_GROUP); @@ -845,13 +756,21 @@ xfs_qm_dqget( ASSERT(ip->i_gdquot == NULL); } #endif + +restart: mutex_lock(&h->qh_lock); /* * Look in the cache (hashtable). * The chain is kept locked during lookup. */ - if (xfs_qm_dqlookup(mp, id, h, O_dqpp) == 0) { + switch (xfs_qm_dqlookup(mp, id, h, O_dqpp)) { + case -1: + XQM_STATS_INC(xqmstats.xs_qm_dquot_dups); + mutex_unlock(&h->qh_lock); + delay(1); + goto restart; + case 0: XQM_STATS_INC(xqmstats.xs_qm_dqcachehits); /* * The dquot was found, moved to the front of the chain, @@ -862,9 +781,11 @@ xfs_qm_dqget( ASSERT(XFS_DQ_IS_LOCKED(*O_dqpp)); mutex_unlock(&h->qh_lock); trace_xfs_dqget_hit(*O_dqpp); - return (0); /* success */ + return 0; /* success */ + default: + XQM_STATS_INC(xqmstats.xs_qm_dqcachemisses); + break; } - XQM_STATS_INC(xqmstats.xs_qm_dqcachemisses); /* * Dquot cache miss. We don't want to keep the inode lock across @@ -882,41 +803,18 @@ xfs_qm_dqget( version = h->qh_version; mutex_unlock(&h->qh_lock); - /* - * Allocate the dquot on the kernel heap, and read the ondisk - * portion off the disk. Also, do all the necessary initialization - * This can return ENOENT if dquot didn't exist on disk and we didn't - * ask it to allocate; ESRCH if quotas got turned off suddenly. - */ - if ((error = xfs_qm_idtodq(mp, id, type, - flags & (XFS_QMOPT_DQALLOC|XFS_QMOPT_DQREPAIR| - XFS_QMOPT_DOWARN), - &dqp))) { - if (ip) - xfs_ilock(ip, XFS_ILOCK_EXCL); - return (error); - } + error = xfs_qm_dqread(mp, id, type, flags, &dqp); - /* - * See if this is mount code calling to look at the overall quota limits - * which are stored in the id == 0 user or group's dquot. - * Since we may not have done a quotacheck by this point, just return - * the dquot without attaching it to any hashtables, lists, etc, or even - * taking a reference. - * The caller must dqdestroy this once done. - */ - if (flags & XFS_QMOPT_DQSUSER) { - ASSERT(id == 0); - ASSERT(! ip); - goto dqret; - } + if (ip) + xfs_ilock(ip, XFS_ILOCK_EXCL); + + if (error) + return error; /* * Dquot lock comes after hashlock in the lock ordering */ if (ip) { - xfs_ilock(ip, XFS_ILOCK_EXCL); - /* * A dquot could be attached to this inode by now, since * we had dropped the ilock. @@ -961,16 +859,21 @@ xfs_qm_dqget( * lock order between the two dquots here since dqp isn't * on any findable lists yet. */ - if (xfs_qm_dqlookup(mp, id, h, &tmpdqp) == 0) { + switch (xfs_qm_dqlookup(mp, id, h, &tmpdqp)) { + case 0: + case -1: /* - * Duplicate found. Just throw away the new dquot - * and start over. + * Duplicate found, either in cache or on its way out. + * Just throw away the new dquot and start over. */ - xfs_qm_dqput(tmpdqp); + if (tmpdqp) + xfs_qm_dqput(tmpdqp); mutex_unlock(&h->qh_lock); xfs_qm_dqdestroy(dqp); XQM_STATS_INC(xqmstats.xs_qm_dquot_dups); - goto again; + goto restart; + default: + break; } } @@ -1015,67 +918,49 @@ xfs_qm_dqget( */ void xfs_qm_dqput( - xfs_dquot_t *dqp) + struct xfs_dquot *dqp) { - xfs_dquot_t *gdqp; + struct xfs_dquot *gdqp; ASSERT(dqp->q_nrefs > 0); ASSERT(XFS_DQ_IS_LOCKED(dqp)); trace_xfs_dqput(dqp); - if (dqp->q_nrefs != 1) { - dqp->q_nrefs--; +recurse: + if (--dqp->q_nrefs > 0) { xfs_dqunlock(dqp); return; } + trace_xfs_dqput_free(dqp); + + mutex_lock(&xfs_Gqm->qm_dqfrlist_lock); + if (list_empty(&dqp->q_freelist)) { + list_add_tail(&dqp->q_freelist, &xfs_Gqm->qm_dqfrlist); + xfs_Gqm->qm_dqfrlist_cnt++; + } + mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); + /* - * drop the dqlock and acquire the freelist and dqlock - * in the right order; but try to get it out-of-order first + * If we just added a udquot to the freelist, then we want to release + * the gdquot reference that it (probably) has. Otherwise it'll keep + * the gdquot from getting reclaimed. */ - if (!mutex_trylock(&xfs_Gqm->qm_dqfrlist_lock)) { - trace_xfs_dqput_wait(dqp); - xfs_dqunlock(dqp); - mutex_lock(&xfs_Gqm->qm_dqfrlist_lock); - xfs_dqlock(dqp); + gdqp = dqp->q_gdquot; + if (gdqp) { + xfs_dqlock(gdqp); + dqp->q_gdquot = NULL; } + xfs_dqunlock(dqp); - while (1) { - gdqp = NULL; - - /* We can't depend on nrefs being == 1 here */ - if (--dqp->q_nrefs == 0) { - trace_xfs_dqput_free(dqp); - - list_add_tail(&dqp->q_freelist, &xfs_Gqm->qm_dqfrlist); - xfs_Gqm->qm_dqfrlist_cnt++; - - /* - * If we just added a udquot to the freelist, then - * we want to release the gdquot reference that - * it (probably) has. Otherwise it'll keep the - * gdquot from getting reclaimed. - */ - if ((gdqp = dqp->q_gdquot)) { - /* - * Avoid a recursive dqput call - */ - xfs_dqlock(gdqp); - dqp->q_gdquot = NULL; - } - } - xfs_dqunlock(dqp); - - /* - * If we had a group quota inside the user quota as a hint, - * release it now. - */ - if (! gdqp) - break; + /* + * If we had a group quota hint, release it now. + */ + if (gdqp) { dqp = gdqp; + goto recurse; } - mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); } /* @@ -1169,7 +1054,7 @@ xfs_qm_dqflush( * If not dirty, or it's pinned and we are not supposed to block, nada. */ if (!XFS_DQ_IS_DIRTY(dqp) || - (!(flags & SYNC_WAIT) && atomic_read(&dqp->q_pincount) > 0)) { + ((flags & SYNC_TRYLOCK) && atomic_read(&dqp->q_pincount) > 0)) { xfs_dqfunlock(dqp); return 0; } @@ -1257,40 +1142,17 @@ xfs_qm_dqflush( } -int -xfs_qm_dqlock_nowait( - xfs_dquot_t *dqp) -{ - return mutex_trylock(&dqp->q_qlock); -} - -void -xfs_dqlock( - xfs_dquot_t *dqp) -{ - mutex_lock(&dqp->q_qlock); -} - void xfs_dqunlock( xfs_dquot_t *dqp) { - mutex_unlock(&(dqp->q_qlock)); + xfs_dqunlock_nonotify(dqp); if (dqp->q_logitem.qli_dquot == dqp) { - /* Once was dqp->q_mount, but might just have been cleared */ xfs_trans_unlocked_item(dqp->q_logitem.qli_item.li_ailp, - (xfs_log_item_t*)&(dqp->q_logitem)); + &dqp->q_logitem.qli_item); } } - -void -xfs_dqunlock_nonotify( - xfs_dquot_t *dqp) -{ - mutex_unlock(&(dqp->q_qlock)); -} - /* * Lock two xfs_dquot structures. * @@ -1319,43 +1181,18 @@ xfs_dqlock2( } } - /* - * Take a dquot out of the mount's dqlist as well as the hashlist. - * This is called via unmount as well as quotaoff, and the purge - * will always succeed unless there are soft (temp) references - * outstanding. - * - * This returns 0 if it was purged, 1 if it wasn't. It's not an error code - * that we're returning! XXXsup - not cool. + * Take a dquot out of the mount's dqlist as well as the hashlist. This is + * called via unmount as well as quotaoff, and the purge will always succeed. */ -/* ARGSUSED */ -int +void xfs_qm_dqpurge( - xfs_dquot_t *dqp) + struct xfs_dquot *dqp) { - xfs_dqhash_t *qh = dqp->q_hash; - xfs_mount_t *mp = dqp->q_mount; - - ASSERT(mutex_is_locked(&mp->m_quotainfo->qi_dqlist_lock)); - ASSERT(mutex_is_locked(&dqp->q_hash->qh_lock)); + struct xfs_mount *mp = dqp->q_mount; + struct xfs_dqhash *qh = dqp->q_hash; xfs_dqlock(dqp); - /* - * We really can't afford to purge a dquot that is - * referenced, because these are hard refs. - * It shouldn't happen in general because we went thru _all_ inodes in - * dqrele_all_inodes before calling this and didn't let the mountlock go. - * However it is possible that we have dquots with temporary - * references that are not attached to an inode. e.g. see xfs_setattr(). - */ - if (dqp->q_nrefs != 0) { - xfs_dqunlock(dqp); - mutex_unlock(&dqp->q_hash->qh_lock); - return (1); - } - - ASSERT(!list_empty(&dqp->q_freelist)); /* * If we're turning off quotas, we have to make sure that, for @@ -1370,23 +1207,18 @@ xfs_qm_dqpurge( * Block on the flush lock after nudging dquot buffer, * if it is incore. */ - xfs_qm_dqflock_pushbuf_wait(dqp); + xfs_dqflock_pushbuf_wait(dqp); } /* - * XXXIf we're turning this type of quotas off, we don't care + * If we are turning this type of quotas off, we don't care * about the dirty metadata sitting in this dquot. OTOH, if * we're unmounting, we do care, so we flush it and wait. */ if (XFS_DQ_IS_DIRTY(dqp)) { int error; - /* dqflush unlocks dqflock */ /* - * Given that dqpurge is a very rare occurrence, it is OK - * that we're holding the hashlist and mplist locks - * across the disk write. But, ... XXXsup - * * We don't care about getting disk errors here. We need * to purge this dquot anyway, so we go ahead regardless. */ @@ -1396,38 +1228,44 @@ xfs_qm_dqpurge( __func__, dqp); xfs_dqflock(dqp); } + ASSERT(atomic_read(&dqp->q_pincount) == 0); ASSERT(XFS_FORCED_SHUTDOWN(mp) || !(dqp->q_logitem.qli_item.li_flags & XFS_LI_IN_AIL)); + xfs_dqfunlock(dqp); + xfs_dqunlock(dqp); + + mutex_lock(&qh->qh_lock); list_del_init(&dqp->q_hashlist); qh->qh_version++; + mutex_unlock(&qh->qh_lock); + + mutex_lock(&mp->m_quotainfo->qi_dqlist_lock); list_del_init(&dqp->q_mplist); mp->m_quotainfo->qi_dqreclaims++; mp->m_quotainfo->qi_dquots--; + mutex_unlock(&mp->m_quotainfo->qi_dqlist_lock); + /* - * XXX Move this to the front of the freelist, if we can get the - * freelist lock. + * We move dquots to the freelist as soon as their reference count + * hits zero, so it really should be on the freelist here. */ + mutex_lock(&xfs_Gqm->qm_dqfrlist_lock); ASSERT(!list_empty(&dqp->q_freelist)); + list_del_init(&dqp->q_freelist); + xfs_Gqm->qm_dqfrlist_cnt--; + mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); - dqp->q_mount = NULL; - dqp->q_hash = NULL; - dqp->dq_flags = XFS_DQ_INACTIVE; - memset(&dqp->q_core, 0, sizeof(dqp->q_core)); - xfs_dqfunlock(dqp); - xfs_dqunlock(dqp); - mutex_unlock(&qh->qh_lock); - return (0); + xfs_qm_dqdestroy(dqp); } - /* * Give the buffer a little push if it is incore and * wait on the flush lock. */ void -xfs_qm_dqflock_pushbuf_wait( +xfs_dqflock_pushbuf_wait( xfs_dquot_t *dqp) { xfs_mount_t *mp = dqp->q_mount; diff --git a/fs/xfs/xfs_dquot.h b/fs/xfs/xfs_dquot.h index 34b7e945dbfa..a1d91d8f1802 100644 --- a/fs/xfs/xfs_dquot.h +++ b/fs/xfs/xfs_dquot.h @@ -80,8 +80,6 @@ enum { XFS_QLOCK_NESTED, }; -#define XFS_DQHOLD(dqp) ((dqp)->q_nrefs++) - /* * Manage the q_flush completion queue embedded in the dquot. This completion * queue synchronizes processes attempting to flush the in-core dquot back to @@ -102,6 +100,21 @@ static inline void xfs_dqfunlock(xfs_dquot_t *dqp) complete(&dqp->q_flush); } +static inline int xfs_dqlock_nowait(struct xfs_dquot *dqp) +{ + return mutex_trylock(&dqp->q_qlock); +} + +static inline void xfs_dqlock(struct xfs_dquot *dqp) +{ + mutex_lock(&dqp->q_qlock); +} + +static inline void xfs_dqunlock_nonotify(struct xfs_dquot *dqp) +{ + mutex_unlock(&dqp->q_qlock); +} + #define XFS_DQ_IS_LOCKED(dqp) (mutex_is_locked(&((dqp)->q_qlock))) #define XFS_DQ_IS_DIRTY(dqp) ((dqp)->dq_flags & XFS_DQ_DIRTY) #define XFS_QM_ISUDQ(dqp) ((dqp)->dq_flags & XFS_DQ_USER) @@ -116,12 +129,12 @@ static inline void xfs_dqfunlock(xfs_dquot_t *dqp) (XFS_IS_UQUOTA_ON((d)->q_mount)) : \ (XFS_IS_OQUOTA_ON((d)->q_mount)))) +extern int xfs_qm_dqread(struct xfs_mount *, xfs_dqid_t, uint, + uint, struct xfs_dquot **); extern void xfs_qm_dqdestroy(xfs_dquot_t *); extern int xfs_qm_dqflush(xfs_dquot_t *, uint); -extern int xfs_qm_dqpurge(xfs_dquot_t *); +extern void xfs_qm_dqpurge(xfs_dquot_t *); extern void xfs_qm_dqunpin_wait(xfs_dquot_t *); -extern int xfs_qm_dqlock_nowait(xfs_dquot_t *); -extern void xfs_qm_dqflock_pushbuf_wait(xfs_dquot_t *dqp); extern void xfs_qm_adjust_dqtimers(xfs_mount_t *, xfs_disk_dquot_t *); extern void xfs_qm_adjust_dqlimits(xfs_mount_t *, @@ -129,9 +142,17 @@ extern void xfs_qm_adjust_dqlimits(xfs_mount_t *, extern int xfs_qm_dqget(xfs_mount_t *, xfs_inode_t *, xfs_dqid_t, uint, uint, xfs_dquot_t **); extern void xfs_qm_dqput(xfs_dquot_t *); -extern void xfs_dqlock(xfs_dquot_t *); -extern void xfs_dqlock2(xfs_dquot_t *, xfs_dquot_t *); -extern void xfs_dqunlock(xfs_dquot_t *); -extern void xfs_dqunlock_nonotify(xfs_dquot_t *); + +extern void xfs_dqlock2(struct xfs_dquot *, struct xfs_dquot *); +extern void xfs_dqunlock(struct xfs_dquot *); +extern void xfs_dqflock_pushbuf_wait(struct xfs_dquot *dqp); + +static inline struct xfs_dquot *xfs_qm_dqhold(struct xfs_dquot *dqp) +{ + xfs_dqlock(dqp); + dqp->q_nrefs++; + xfs_dqunlock(dqp); + return dqp; +} #endif /* __XFS_DQUOT_H__ */ diff --git a/fs/xfs/xfs_dquot_item.c b/fs/xfs/xfs_dquot_item.c index 0dee0b71029d..34baeae45265 100644 --- a/fs/xfs/xfs_dquot_item.c +++ b/fs/xfs/xfs_dquot_item.c @@ -73,7 +73,6 @@ xfs_qm_dquot_logitem_format( logvec->i_len = sizeof(xfs_disk_dquot_t); logvec->i_type = XLOG_REG_TYPE_DQUOT; - ASSERT(2 == lip->li_desc->lid_size); qlip->qli_format.qlf_size = 2; } @@ -134,7 +133,7 @@ xfs_qm_dquot_logitem_push( * lock without sleeping, then there must not have been * anyone in the process of flushing the dquot. */ - error = xfs_qm_dqflush(dqp, 0); + error = xfs_qm_dqflush(dqp, SYNC_TRYLOCK); if (error) xfs_warn(dqp->q_mount, "%s: push error %d on dqp %p", __func__, error, dqp); @@ -237,7 +236,7 @@ xfs_qm_dquot_logitem_trylock( if (atomic_read(&dqp->q_pincount) > 0) return XFS_ITEM_PINNED; - if (!xfs_qm_dqlock_nowait(dqp)) + if (!xfs_dqlock_nowait(dqp)) return XFS_ITEM_LOCKED; if (!xfs_dqflock_nowait(dqp)) { diff --git a/fs/xfs/xfs_export.c b/fs/xfs/xfs_export.c index da108977b21f..558910f5e3c0 100644 --- a/fs/xfs/xfs_export.c +++ b/fs/xfs/xfs_export.c @@ -98,22 +98,22 @@ xfs_fs_encode_fh( switch (fileid_type) { case FILEID_INO32_GEN_PARENT: spin_lock(&dentry->d_lock); - fid->i32.parent_ino = dentry->d_parent->d_inode->i_ino; + fid->i32.parent_ino = XFS_I(dentry->d_parent->d_inode)->i_ino; fid->i32.parent_gen = dentry->d_parent->d_inode->i_generation; spin_unlock(&dentry->d_lock); /*FALLTHRU*/ case FILEID_INO32_GEN: - fid->i32.ino = inode->i_ino; + fid->i32.ino = XFS_I(inode)->i_ino; fid->i32.gen = inode->i_generation; break; case FILEID_INO32_GEN_PARENT | XFS_FILEID_TYPE_64FLAG: spin_lock(&dentry->d_lock); - fid64->parent_ino = dentry->d_parent->d_inode->i_ino; + fid64->parent_ino = XFS_I(dentry->d_parent->d_inode)->i_ino; fid64->parent_gen = dentry->d_parent->d_inode->i_generation; spin_unlock(&dentry->d_lock); /*FALLTHRU*/ case FILEID_INO32_GEN | XFS_FILEID_TYPE_64FLAG: - fid64->ino = inode->i_ino; + fid64->ino = XFS_I(inode)->i_ino; fid64->gen = inode->i_generation; break; } diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 753ed9b5c70b..f675f3d9d7b3 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -209,10 +209,10 @@ xfs_file_fsync( /* * First check if the VFS inode is marked dirty. All the dirtying - * of non-transactional updates no goes through mark_inode_dirty*, - * which allows us to distinguish beteeen pure timestamp updates + * of non-transactional updates do not go through mark_inode_dirty*, + * which allows us to distinguish between pure timestamp updates * and i_size updates which need to be caught for fdatasync. - * After that also theck for the dirty state in the XFS inode, which + * After that also check for the dirty state in the XFS inode, which * might gets cleared when the inode gets written out via the AIL * or xfs_iflush_cluster. */ diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c index 169380e66057..dad1a31aa4fc 100644 --- a/fs/xfs/xfs_ialloc.c +++ b/fs/xfs/xfs_ialloc.c @@ -447,7 +447,7 @@ STATIC xfs_buf_t * /* allocation group buffer */ xfs_ialloc_ag_select( xfs_trans_t *tp, /* transaction pointer */ xfs_ino_t parent, /* parent directory inode number */ - mode_t mode, /* bits set to indicate file type */ + umode_t mode, /* bits set to indicate file type */ int okalloc) /* ok to allocate more space */ { xfs_buf_t *agbp; /* allocation group header buffer */ @@ -640,7 +640,7 @@ int xfs_dialloc( xfs_trans_t *tp, /* transaction pointer */ xfs_ino_t parent, /* parent inode (directory) */ - mode_t mode, /* mode bits for new inode */ + umode_t mode, /* mode bits for new inode */ int okalloc, /* ok to allocate more space */ xfs_buf_t **IO_agbp, /* in/out ag header's buffer */ boolean_t *alloc_done, /* true if we needed to replenish diff --git a/fs/xfs/xfs_ialloc.h b/fs/xfs/xfs_ialloc.h index bb5385475e1f..666a037398d6 100644 --- a/fs/xfs/xfs_ialloc.h +++ b/fs/xfs/xfs_ialloc.h @@ -81,7 +81,7 @@ int /* error */ xfs_dialloc( struct xfs_trans *tp, /* transaction pointer */ xfs_ino_t parent, /* parent inode (directory) */ - mode_t mode, /* mode bits for new inode */ + umode_t mode, /* mode bits for new inode */ int okalloc, /* ok to allocate more space */ struct xfs_buf **agbp, /* buf for a.g. inode header */ boolean_t *alloc_done, /* an allocation was done to replenish diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index 0fa98b1c70ea..3960a066d7ff 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -107,7 +107,6 @@ xfs_inode_free_callback( struct inode *inode = container_of(head, struct inode, i_rcu); struct xfs_inode *ip = XFS_I(inode); - INIT_LIST_HEAD(&inode->i_dentry); kmem_zone_free(xfs_inode_zone, ip); } diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index c0237c602f11..9dda7cc32848 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -961,7 +961,7 @@ int xfs_ialloc( xfs_trans_t *tp, xfs_inode_t *pip, - mode_t mode, + umode_t mode, xfs_nlink_t nlink, xfs_dev_t rdev, prid_t prid, @@ -1002,7 +1002,7 @@ xfs_ialloc( return error; ASSERT(ip != NULL); - ip->i_d.di_mode = (__uint16_t)mode; + ip->i_d.di_mode = mode; ip->i_d.di_onlink = 0; ip->i_d.di_nlink = nlink; ASSERT(ip->i_d.di_nlink == nlink); @@ -2835,6 +2835,27 @@ corrupt_out: return XFS_ERROR(EFSCORRUPTED); } +void +xfs_promote_inode( + struct xfs_inode *ip) +{ + struct xfs_buf *bp; + + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED)); + + bp = xfs_incore(ip->i_mount->m_ddev_targp, ip->i_imap.im_blkno, + ip->i_imap.im_len, XBF_TRYLOCK); + if (!bp) + return; + + if (XFS_BUF_ISDELAYWRITE(bp)) { + xfs_buf_delwri_promote(bp); + wake_up_process(ip->i_mount->m_ddev_targp->bt_task); + } + + xfs_buf_relse(bp); +} + /* * Return a pointer to the extent record at file index idx. */ diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 760140d1dd66..f0e6b151ba37 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -481,7 +481,7 @@ void xfs_inode_free(struct xfs_inode *ip); /* * xfs_inode.c prototypes. */ -int xfs_ialloc(struct xfs_trans *, xfs_inode_t *, mode_t, +int xfs_ialloc(struct xfs_trans *, xfs_inode_t *, umode_t, xfs_nlink_t, xfs_dev_t, prid_t, int, struct xfs_buf **, boolean_t *, xfs_inode_t **); @@ -498,6 +498,7 @@ int xfs_iunlink(struct xfs_trans *, xfs_inode_t *); void xfs_iext_realloc(xfs_inode_t *, int, int); void xfs_iunpin_wait(xfs_inode_t *); int xfs_iflush(xfs_inode_t *, uint); +void xfs_promote_inode(struct xfs_inode *); void xfs_lock_inodes(xfs_inode_t **, int, uint); void xfs_lock_two_inodes(xfs_inode_t *, xfs_inode_t *, uint); diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c index abaafdbb3e65..cfd6c7f8cc3c 100644 --- a/fs/xfs/xfs_inode_item.c +++ b/fs/xfs/xfs_inode_item.c @@ -437,7 +437,6 @@ xfs_inode_item_format( * Assert that no attribute-related log flags are set. */ if (!XFS_IFORK_Q(ip)) { - ASSERT(nvecs == lip->li_desc->lid_size); iip->ili_format.ilf_size = nvecs; ASSERT(!(iip->ili_format.ilf_fields & (XFS_ILOG_ADATA | XFS_ILOG_ABROOT | XFS_ILOG_AEXT))); @@ -521,7 +520,6 @@ xfs_inode_item_format( break; } - ASSERT(nvecs == lip->li_desc->lid_size); iip->ili_format.ilf_size = nvecs; } diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index d99a90518909..76f3ca5cfc36 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -559,23 +559,23 @@ xfs_attrmulti_by_handle( ops[i].am_flags); break; case ATTR_OP_SET: - ops[i].am_error = mnt_want_write(parfilp->f_path.mnt); + ops[i].am_error = mnt_want_write_file(parfilp); if (ops[i].am_error) break; ops[i].am_error = xfs_attrmulti_attr_set( dentry->d_inode, attr_name, ops[i].am_attrvalue, ops[i].am_length, ops[i].am_flags); - mnt_drop_write(parfilp->f_path.mnt); + mnt_drop_write_file(parfilp); break; case ATTR_OP_REMOVE: - ops[i].am_error = mnt_want_write(parfilp->f_path.mnt); + ops[i].am_error = mnt_want_write_file(parfilp); if (ops[i].am_error) break; ops[i].am_error = xfs_attrmulti_attr_remove( dentry->d_inode, attr_name, ops[i].am_flags); - mnt_drop_write(parfilp->f_path.mnt); + mnt_drop_write_file(parfilp); break; default: ops[i].am_error = EINVAL; diff --git a/fs/xfs/xfs_ioctl32.c b/fs/xfs/xfs_ioctl32.c index 54e623bfbb85..f9ccb7b7c043 100644 --- a/fs/xfs/xfs_ioctl32.c +++ b/fs/xfs/xfs_ioctl32.c @@ -454,23 +454,23 @@ xfs_compat_attrmulti_by_handle( &ops[i].am_length, ops[i].am_flags); break; case ATTR_OP_SET: - ops[i].am_error = mnt_want_write(parfilp->f_path.mnt); + ops[i].am_error = mnt_want_write_file(parfilp); if (ops[i].am_error) break; ops[i].am_error = xfs_attrmulti_attr_set( dentry->d_inode, attr_name, compat_ptr(ops[i].am_attrvalue), ops[i].am_length, ops[i].am_flags); - mnt_drop_write(parfilp->f_path.mnt); + mnt_drop_write_file(parfilp); break; case ATTR_OP_REMOVE: - ops[i].am_error = mnt_want_write(parfilp->f_path.mnt); + ops[i].am_error = mnt_want_write_file(parfilp); if (ops[i].am_error) break; ops[i].am_error = xfs_attrmulti_attr_remove( dentry->d_inode, attr_name, ops[i].am_flags); - mnt_drop_write(parfilp->f_path.mnt); + mnt_drop_write_file(parfilp); break; default: ops[i].am_error = EINVAL; diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index 23ce927973a4..f9babd179223 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -168,7 +168,7 @@ STATIC int xfs_vn_mknod( struct inode *dir, struct dentry *dentry, - int mode, + umode_t mode, dev_t rdev) { struct inode *inode; @@ -231,7 +231,7 @@ STATIC int xfs_vn_create( struct inode *dir, struct dentry *dentry, - int mode, + umode_t mode, struct nameidata *nd) { return xfs_vn_mknod(dir, dentry, mode, 0); @@ -241,7 +241,7 @@ STATIC int xfs_vn_mkdir( struct inode *dir, struct dentry *dentry, - int mode) + umode_t mode) { return xfs_vn_mknod(dir, dentry, mode|S_IFDIR, 0); } @@ -366,7 +366,7 @@ xfs_vn_symlink( struct xfs_inode *cip = NULL; struct xfs_name name; int error; - mode_t mode; + umode_t mode; mode = S_IFLNK | (irix_symlink_mode ? 0777 & ~current_umask() : S_IRWXUGO); diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index a14cd89fe465..e2cc3568c299 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -150,6 +150,117 @@ xlog_grant_add_space( } while (head_val != old); } +STATIC bool +xlog_reserveq_wake( + struct log *log, + int *free_bytes) +{ + struct xlog_ticket *tic; + int need_bytes; + + list_for_each_entry(tic, &log->l_reserveq, t_queue) { + if (tic->t_flags & XLOG_TIC_PERM_RESERV) + need_bytes = tic->t_unit_res * tic->t_cnt; + else + need_bytes = tic->t_unit_res; + + if (*free_bytes < need_bytes) + return false; + *free_bytes -= need_bytes; + + trace_xfs_log_grant_wake_up(log, tic); + wake_up(&tic->t_wait); + } + + return true; +} + +STATIC bool +xlog_writeq_wake( + struct log *log, + int *free_bytes) +{ + struct xlog_ticket *tic; + int need_bytes; + + list_for_each_entry(tic, &log->l_writeq, t_queue) { + ASSERT(tic->t_flags & XLOG_TIC_PERM_RESERV); + + need_bytes = tic->t_unit_res; + + if (*free_bytes < need_bytes) + return false; + *free_bytes -= need_bytes; + + trace_xfs_log_regrant_write_wake_up(log, tic); + wake_up(&tic->t_wait); + } + + return true; +} + +STATIC int +xlog_reserveq_wait( + struct log *log, + struct xlog_ticket *tic, + int need_bytes) +{ + list_add_tail(&tic->t_queue, &log->l_reserveq); + + do { + if (XLOG_FORCED_SHUTDOWN(log)) + goto shutdown; + xlog_grant_push_ail(log, need_bytes); + + XFS_STATS_INC(xs_sleep_logspace); + trace_xfs_log_grant_sleep(log, tic); + + xlog_wait(&tic->t_wait, &log->l_grant_reserve_lock); + trace_xfs_log_grant_wake(log, tic); + + spin_lock(&log->l_grant_reserve_lock); + if (XLOG_FORCED_SHUTDOWN(log)) + goto shutdown; + } while (xlog_space_left(log, &log->l_grant_reserve_head) < need_bytes); + + list_del_init(&tic->t_queue); + return 0; +shutdown: + list_del_init(&tic->t_queue); + return XFS_ERROR(EIO); +} + +STATIC int +xlog_writeq_wait( + struct log *log, + struct xlog_ticket *tic, + int need_bytes) +{ + list_add_tail(&tic->t_queue, &log->l_writeq); + + do { + if (XLOG_FORCED_SHUTDOWN(log)) + goto shutdown; + xlog_grant_push_ail(log, need_bytes); + + XFS_STATS_INC(xs_sleep_logspace); + trace_xfs_log_regrant_write_sleep(log, tic); + + xlog_wait(&tic->t_wait, &log->l_grant_write_lock); + trace_xfs_log_regrant_write_wake(log, tic); + + spin_lock(&log->l_grant_write_lock); + if (XLOG_FORCED_SHUTDOWN(log)) + goto shutdown; + } while (xlog_space_left(log, &log->l_grant_write_head) < need_bytes); + + list_del_init(&tic->t_queue); + return 0; +shutdown: + list_del_init(&tic->t_queue); + return XFS_ERROR(EIO); +} + static void xlog_tic_reset_res(xlog_ticket_t *tic) { @@ -350,8 +461,19 @@ xfs_log_reserve( retval = xlog_grant_log_space(log, internal_ticket); } + if (unlikely(retval)) { + /* + * If we are failing, make sure the ticket doesn't have any + * current reservations. We don't want to add this back + * when the ticket/ transaction gets cancelled. + */ + internal_ticket->t_curr_res = 0; + /* ungrant will give back unit_res * t_cnt. */ + internal_ticket->t_cnt = 0; + } + return retval; -} /* xfs_log_reserve */ +} /* @@ -638,38 +760,6 @@ xfs_log_item_init( INIT_LIST_HEAD(&item->li_cil); } -/* - * Write region vectors to log. The write happens using the space reservation - * of the ticket (tic). It is not a requirement that all writes for a given - * transaction occur with one call to xfs_log_write(). However, it is important - * to note that the transaction reservation code makes an assumption about the - * number of log headers a transaction requires that may be violated if you - * don't pass all the transaction vectors in one call.... - */ -int -xfs_log_write( - struct xfs_mount *mp, - struct xfs_log_iovec reg[], - int nentries, - struct xlog_ticket *tic, - xfs_lsn_t *start_lsn) -{ - struct log *log = mp->m_log; - int error; - struct xfs_log_vec vec = { - .lv_niovecs = nentries, - .lv_iovecp = reg, - }; - - if (XLOG_FORCED_SHUTDOWN(log)) - return XFS_ERROR(EIO); - - error = xlog_write(log, &vec, tic, start_lsn, NULL, 0); - if (error) - xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR); - return error; -} - void xfs_log_move_tail(xfs_mount_t *mp, xfs_lsn_t tail_lsn) @@ -1563,7 +1653,7 @@ xlog_print_tic_res( }; xfs_warn(mp, - "xfs_log_write: reservation summary:\n" + "xlog_write: reservation summary:\n" " trans type = %s (%u)\n" " unit res = %d bytes\n" " current res = %d bytes\n" @@ -1592,7 +1682,7 @@ xlog_print_tic_res( } xfs_alert_tag(mp, XFS_PTAG_LOGRES, - "xfs_log_write: reservation ran out. Need to up reservation"); + "xlog_write: reservation ran out. Need to up reservation"); xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); } @@ -1846,23 +1936,21 @@ xlog_write( *start_lsn = 0; len = xlog_write_calc_vec_length(ticket, log_vector); - if (log->l_cilp) { - /* - * Region headers and bytes are already accounted for. - * We only need to take into account start records and - * split regions in this function. - */ - if (ticket->t_flags & XLOG_TIC_INITED) - ticket->t_curr_res -= sizeof(xlog_op_header_t); - /* - * Commit record headers need to be accounted for. These - * come in as separate writes so are easy to detect. - */ - if (flags & (XLOG_COMMIT_TRANS | XLOG_UNMOUNT_TRANS)) - ticket->t_curr_res -= sizeof(xlog_op_header_t); - } else - ticket->t_curr_res -= len; + /* + * Region headers and bytes are already accounted for. + * We only need to take into account start records and + * split regions in this function. + */ + if (ticket->t_flags & XLOG_TIC_INITED) + ticket->t_curr_res -= sizeof(xlog_op_header_t); + + /* + * Commit record headers need to be accounted for. These + * come in as separate writes so are easy to detect. + */ + if (flags & (XLOG_COMMIT_TRANS | XLOG_UNMOUNT_TRANS)) + ticket->t_curr_res -= sizeof(xlog_op_header_t); if (ticket->t_curr_res < 0) xlog_print_tic_res(log->l_mp, ticket); @@ -2481,8 +2569,8 @@ restart: /* * Atomically get the log space required for a log ticket. * - * Once a ticket gets put onto the reserveq, it will only return after - * the needed reservation is satisfied. + * Once a ticket gets put onto the reserveq, it will only return after the + * needed reservation is satisfied. * * This function is structured so that it has a lock free fast path. This is * necessary because every new transaction reservation will come through this @@ -2490,113 +2578,53 @@ restart: * every pass. * * As tickets are only ever moved on and off the reserveq under the - * l_grant_reserve_lock, we only need to take that lock if we are going - * to add the ticket to the queue and sleep. We can avoid taking the lock if the - * ticket was never added to the reserveq because the t_queue list head will be - * empty and we hold the only reference to it so it can safely be checked - * unlocked. + * l_grant_reserve_lock, we only need to take that lock if we are going to add + * the ticket to the queue and sleep. We can avoid taking the lock if the ticket + * was never added to the reserveq because the t_queue list head will be empty + * and we hold the only reference to it so it can safely be checked unlocked. */ STATIC int -xlog_grant_log_space(xlog_t *log, - xlog_ticket_t *tic) +xlog_grant_log_space( + struct log *log, + struct xlog_ticket *tic) { - int free_bytes; - int need_bytes; + int free_bytes, need_bytes; + int error = 0; -#ifdef DEBUG - if (log->l_flags & XLOG_ACTIVE_RECOVERY) - panic("grant Recovery problem"); -#endif + ASSERT(!(log->l_flags & XLOG_ACTIVE_RECOVERY)); trace_xfs_log_grant_enter(log, tic); + /* + * If there are other waiters on the queue then give them a chance at + * logspace before us. Wake up the first waiters, if we do not wake + * up all the waiters then go to sleep waiting for more free space, + * otherwise try to get some space for this transaction. + */ need_bytes = tic->t_unit_res; if (tic->t_flags & XFS_LOG_PERM_RESERV) need_bytes *= tic->t_ocnt; - - /* something is already sleeping; insert new transaction at end */ - if (!list_empty_careful(&log->l_reserveq)) { - spin_lock(&log->l_grant_reserve_lock); - /* recheck the queue now we are locked */ - if (list_empty(&log->l_reserveq)) { - spin_unlock(&log->l_grant_reserve_lock); - goto redo; - } - list_add_tail(&tic->t_queue, &log->l_reserveq); - - trace_xfs_log_grant_sleep1(log, tic); - - /* - * Gotta check this before going to sleep, while we're - * holding the grant lock. - */ - if (XLOG_FORCED_SHUTDOWN(log)) - goto error_return; - - XFS_STATS_INC(xs_sleep_logspace); - xlog_wait(&tic->t_wait, &log->l_grant_reserve_lock); - - /* - * If we got an error, and the filesystem is shutting down, - * we'll catch it down below. So just continue... - */ - trace_xfs_log_grant_wake1(log, tic); - } - -redo: - if (XLOG_FORCED_SHUTDOWN(log)) - goto error_return_unlocked; - free_bytes = xlog_space_left(log, &log->l_grant_reserve_head); - if (free_bytes < need_bytes) { + if (!list_empty_careful(&log->l_reserveq)) { spin_lock(&log->l_grant_reserve_lock); - if (list_empty(&tic->t_queue)) - list_add_tail(&tic->t_queue, &log->l_reserveq); - - trace_xfs_log_grant_sleep2(log, tic); - - if (XLOG_FORCED_SHUTDOWN(log)) - goto error_return; - - xlog_grant_push_ail(log, need_bytes); - - XFS_STATS_INC(xs_sleep_logspace); - xlog_wait(&tic->t_wait, &log->l_grant_reserve_lock); - - trace_xfs_log_grant_wake2(log, tic); - goto redo; - } - - if (!list_empty(&tic->t_queue)) { + if (!xlog_reserveq_wake(log, &free_bytes) || + free_bytes < need_bytes) + error = xlog_reserveq_wait(log, tic, need_bytes); + spin_unlock(&log->l_grant_reserve_lock); + } else if (free_bytes < need_bytes) { spin_lock(&log->l_grant_reserve_lock); - list_del_init(&tic->t_queue); + error = xlog_reserveq_wait(log, tic, need_bytes); spin_unlock(&log->l_grant_reserve_lock); } + if (error) + return error; - /* we've got enough space */ xlog_grant_add_space(log, &log->l_grant_reserve_head, need_bytes); xlog_grant_add_space(log, &log->l_grant_write_head, need_bytes); trace_xfs_log_grant_exit(log, tic); xlog_verify_grant_tail(log); return 0; - -error_return_unlocked: - spin_lock(&log->l_grant_reserve_lock); -error_return: - list_del_init(&tic->t_queue); - spin_unlock(&log->l_grant_reserve_lock); - trace_xfs_log_grant_error(log, tic); - - /* - * If we are failing, make sure the ticket doesn't have any - * current reservations. We don't want to add this back when - * the ticket/transaction gets cancelled. - */ - tic->t_curr_res = 0; - tic->t_cnt = 0; /* ungrant will give back unit_res * t_cnt. */ - return XFS_ERROR(EIO); -} /* xlog_grant_log_space */ - +} /* * Replenish the byte reservation required by moving the grant write head. @@ -2605,10 +2633,12 @@ error_return: * free fast path. */ STATIC int -xlog_regrant_write_log_space(xlog_t *log, - xlog_ticket_t *tic) +xlog_regrant_write_log_space( + struct log *log, + struct xlog_ticket *tic) { - int free_bytes, need_bytes; + int free_bytes, need_bytes; + int error = 0; tic->t_curr_res = tic->t_unit_res; xlog_tic_reset_res(tic); @@ -2616,104 +2646,38 @@ xlog_regrant_write_log_space(xlog_t *log, if (tic->t_cnt > 0) return 0; -#ifdef DEBUG - if (log->l_flags & XLOG_ACTIVE_RECOVERY) - panic("regrant Recovery problem"); -#endif + ASSERT(!(log->l_flags & XLOG_ACTIVE_RECOVERY)); trace_xfs_log_regrant_write_enter(log, tic); - if (XLOG_FORCED_SHUTDOWN(log)) - goto error_return_unlocked; - /* If there are other waiters on the queue then give them a - * chance at logspace before us. Wake up the first waiters, - * if we do not wake up all the waiters then go to sleep waiting - * for more free space, otherwise try to get some space for - * this transaction. + /* + * If there are other waiters on the queue then give them a chance at + * logspace before us. Wake up the first waiters, if we do not wake + * up all the waiters then go to sleep waiting for more free space, + * otherwise try to get some space for this transaction. */ need_bytes = tic->t_unit_res; - if (!list_empty_careful(&log->l_writeq)) { - struct xlog_ticket *ntic; - - spin_lock(&log->l_grant_write_lock); - free_bytes = xlog_space_left(log, &log->l_grant_write_head); - list_for_each_entry(ntic, &log->l_writeq, t_queue) { - ASSERT(ntic->t_flags & XLOG_TIC_PERM_RESERV); - - if (free_bytes < ntic->t_unit_res) - break; - free_bytes -= ntic->t_unit_res; - wake_up(&ntic->t_wait); - } - - if (ntic != list_first_entry(&log->l_writeq, - struct xlog_ticket, t_queue)) { - if (list_empty(&tic->t_queue)) - list_add_tail(&tic->t_queue, &log->l_writeq); - trace_xfs_log_regrant_write_sleep1(log, tic); - - xlog_grant_push_ail(log, need_bytes); - - XFS_STATS_INC(xs_sleep_logspace); - xlog_wait(&tic->t_wait, &log->l_grant_write_lock); - trace_xfs_log_regrant_write_wake1(log, tic); - } else - spin_unlock(&log->l_grant_write_lock); - } - -redo: - if (XLOG_FORCED_SHUTDOWN(log)) - goto error_return_unlocked; - free_bytes = xlog_space_left(log, &log->l_grant_write_head); - if (free_bytes < need_bytes) { + if (!list_empty_careful(&log->l_writeq)) { spin_lock(&log->l_grant_write_lock); - if (list_empty(&tic->t_queue)) - list_add_tail(&tic->t_queue, &log->l_writeq); - - if (XLOG_FORCED_SHUTDOWN(log)) - goto error_return; - - xlog_grant_push_ail(log, need_bytes); - - XFS_STATS_INC(xs_sleep_logspace); - trace_xfs_log_regrant_write_sleep2(log, tic); - xlog_wait(&tic->t_wait, &log->l_grant_write_lock); - - trace_xfs_log_regrant_write_wake2(log, tic); - goto redo; - } - - if (!list_empty(&tic->t_queue)) { + if (!xlog_writeq_wake(log, &free_bytes) || + free_bytes < need_bytes) + error = xlog_writeq_wait(log, tic, need_bytes); + spin_unlock(&log->l_grant_write_lock); + } else if (free_bytes < need_bytes) { spin_lock(&log->l_grant_write_lock); - list_del_init(&tic->t_queue); + error = xlog_writeq_wait(log, tic, need_bytes); spin_unlock(&log->l_grant_write_lock); } - /* we've got enough space */ + if (error) + return error; + xlog_grant_add_space(log, &log->l_grant_write_head, need_bytes); trace_xfs_log_regrant_write_exit(log, tic); xlog_verify_grant_tail(log); return 0; - - - error_return_unlocked: - spin_lock(&log->l_grant_write_lock); - error_return: - list_del_init(&tic->t_queue); - spin_unlock(&log->l_grant_write_lock); - trace_xfs_log_regrant_write_error(log, tic); - - /* - * If we are failing, make sure the ticket doesn't have any - * current reservations. We don't want to add this back when - * the ticket/transaction gets cancelled. - */ - tic->t_curr_res = 0; - tic->t_cnt = 0; /* ungrant will give back unit_res * t_cnt. */ - return XFS_ERROR(EIO); -} /* xlog_regrant_write_log_space */ - +} /* The first cnt-1 times through here we don't need to * move the grant write head because the permanent @@ -2933,8 +2897,7 @@ _xfs_log_force( XFS_STATS_INC(xs_log_force); - if (log->l_cilp) - xlog_cil_force(log); + xlog_cil_force(log); spin_lock(&log->l_icloglock); @@ -3083,11 +3046,9 @@ _xfs_log_force_lsn( XFS_STATS_INC(xs_log_force); - if (log->l_cilp) { - lsn = xlog_cil_force_lsn(log, lsn); - if (lsn == NULLCOMMITLSN) - return 0; - } + lsn = xlog_cil_force_lsn(log, lsn); + if (lsn == NULLCOMMITLSN) + return 0; try_again: spin_lock(&log->l_icloglock); @@ -3655,7 +3616,7 @@ xfs_log_force_umount( * completed transactions are flushed to disk with the xfs_log_force() * call below. */ - if (!logerror && (mp->m_flags & XFS_MOUNT_DELAYLOG)) + if (!logerror) xlog_cil_force(log); /* diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h index 3f7bf451c034..2aee3b22d29c 100644 --- a/fs/xfs/xfs_log.h +++ b/fs/xfs/xfs_log.h @@ -174,11 +174,6 @@ int xfs_log_reserve(struct xfs_mount *mp, __uint8_t clientid, uint flags, uint t_type); -int xfs_log_write(struct xfs_mount *mp, - xfs_log_iovec_t region[], - int nentries, - struct xlog_ticket *ticket, - xfs_lsn_t *start_lsn); int xfs_log_unmount_write(struct xfs_mount *mp); void xfs_log_unmount(struct xfs_mount *mp); int xfs_log_force_umount(struct xfs_mount *mp, int logerror); @@ -189,8 +184,7 @@ void xlog_iodone(struct xfs_buf *); struct xlog_ticket *xfs_log_ticket_get(struct xlog_ticket *ticket); void xfs_log_ticket_put(struct xlog_ticket *ticket); -void xfs_log_commit_cil(struct xfs_mount *mp, struct xfs_trans *tp, - struct xfs_log_vec *log_vector, +int xfs_log_commit_cil(struct xfs_mount *mp, struct xfs_trans *tp, xfs_lsn_t *commit_lsn, int flags); bool xfs_log_item_in_current_chkpt(struct xfs_log_item *lip); diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c index c7755d5a5fbe..d4fadbe8ac90 100644 --- a/fs/xfs/xfs_log_cil.c +++ b/fs/xfs/xfs_log_cil.c @@ -32,10 +32,7 @@ #include "xfs_discard.h" /* - * Perform initial CIL structure initialisation. If the CIL is not - * enabled in this filesystem, ensure the log->l_cilp is null so - * we can check this conditional to determine if we are doing delayed - * logging or not. + * Perform initial CIL structure initialisation. */ int xlog_cil_init( @@ -44,10 +41,6 @@ xlog_cil_init( struct xfs_cil *cil; struct xfs_cil_ctx *ctx; - log->l_cilp = NULL; - if (!(log->l_mp->m_flags & XFS_MOUNT_DELAYLOG)) - return 0; - cil = kmem_zalloc(sizeof(*cil), KM_SLEEP|KM_MAYFAIL); if (!cil) return ENOMEM; @@ -80,9 +73,6 @@ void xlog_cil_destroy( struct log *log) { - if (!log->l_cilp) - return; - if (log->l_cilp->xc_ctx) { if (log->l_cilp->xc_ctx->ticket) xfs_log_ticket_put(log->l_cilp->xc_ctx->ticket); @@ -137,9 +127,6 @@ void xlog_cil_init_post_recovery( struct log *log) { - if (!log->l_cilp) - return; - log->l_cilp->xc_ctx->ticket = xlog_cil_ticket_alloc(log); log->l_cilp->xc_ctx->sequence = 1; log->l_cilp->xc_ctx->commit_lsn = xlog_assign_lsn(log->l_curr_cycle, @@ -172,37 +159,73 @@ xlog_cil_init_post_recovery( * format the regions into the iclog as though they are being formatted * directly out of the objects themselves. */ -static void -xlog_cil_format_items( - struct log *log, - struct xfs_log_vec *log_vector) +static struct xfs_log_vec * +xlog_cil_prepare_log_vecs( + struct xfs_trans *tp) { - struct xfs_log_vec *lv; + struct xfs_log_item_desc *lidp; + struct xfs_log_vec *lv = NULL; + struct xfs_log_vec *ret_lv = NULL; - ASSERT(log_vector); - for (lv = log_vector; lv; lv = lv->lv_next) { + + /* Bail out if we didn't find a log item. */ + if (list_empty(&tp->t_items)) { + ASSERT(0); + return NULL; + } + + list_for_each_entry(lidp, &tp->t_items, lid_trans) { + struct xfs_log_vec *new_lv; void *ptr; int index; int len = 0; + uint niovecs; + + /* Skip items which aren't dirty in this transaction. */ + if (!(lidp->lid_flags & XFS_LID_DIRTY)) + continue; + + /* Skip items that do not have any vectors for writing */ + niovecs = IOP_SIZE(lidp->lid_item); + if (!niovecs) + continue; + + new_lv = kmem_zalloc(sizeof(*new_lv) + + niovecs * sizeof(struct xfs_log_iovec), + KM_SLEEP); + + /* The allocated iovec region lies beyond the log vector. */ + new_lv->lv_iovecp = (struct xfs_log_iovec *)&new_lv[1]; + new_lv->lv_niovecs = niovecs; + new_lv->lv_item = lidp->lid_item; /* build the vector array and calculate it's length */ - IOP_FORMAT(lv->lv_item, lv->lv_iovecp); - for (index = 0; index < lv->lv_niovecs; index++) - len += lv->lv_iovecp[index].i_len; + IOP_FORMAT(new_lv->lv_item, new_lv->lv_iovecp); + for (index = 0; index < new_lv->lv_niovecs; index++) + len += new_lv->lv_iovecp[index].i_len; - lv->lv_buf_len = len; - lv->lv_buf = kmem_alloc(lv->lv_buf_len, KM_SLEEP|KM_NOFS); - ptr = lv->lv_buf; + new_lv->lv_buf_len = len; + new_lv->lv_buf = kmem_alloc(new_lv->lv_buf_len, + KM_SLEEP|KM_NOFS); + ptr = new_lv->lv_buf; - for (index = 0; index < lv->lv_niovecs; index++) { - struct xfs_log_iovec *vec = &lv->lv_iovecp[index]; + for (index = 0; index < new_lv->lv_niovecs; index++) { + struct xfs_log_iovec *vec = &new_lv->lv_iovecp[index]; memcpy(ptr, vec->i_addr, vec->i_len); vec->i_addr = ptr; ptr += vec->i_len; } - ASSERT(ptr == lv->lv_buf + lv->lv_buf_len); + ASSERT(ptr == new_lv->lv_buf + new_lv->lv_buf_len); + + if (!ret_lv) + ret_lv = new_lv; + else + lv->lv_next = new_lv; + lv = new_lv; } + + return ret_lv; } /* @@ -256,7 +279,7 @@ xfs_cil_prepare_item( * Insert the log items into the CIL and calculate the difference in space * consumed by the item. Add the space to the checkpoint ticket and calculate * if the change requires additional log metadata. If it does, take that space - * as well. Remove the amount of space we addded to the checkpoint ticket from + * as well. Remove the amount of space we added to the checkpoint ticket from * the current transaction ticket so that the accounting works out correctly. */ static void @@ -635,28 +658,30 @@ out_abort: * background commit, returns without it held once background commits are * allowed again. */ -void +int xfs_log_commit_cil( struct xfs_mount *mp, struct xfs_trans *tp, - struct xfs_log_vec *log_vector, xfs_lsn_t *commit_lsn, int flags) { struct log *log = mp->m_log; int log_flags = 0; int push = 0; + struct xfs_log_vec *log_vector; if (flags & XFS_TRANS_RELEASE_LOG_RES) log_flags = XFS_LOG_REL_PERM_RESERV; /* - * do all the hard work of formatting items (including memory + * Do all the hard work of formatting items (including memory * allocation) outside the CIL context lock. This prevents stalling CIL * pushes when we are low on memory and a transaction commit spends a * lot of time in memory reclaim. */ - xlog_cil_format_items(log, log_vector); + log_vector = xlog_cil_prepare_log_vecs(tp); + if (!log_vector) + return ENOMEM; /* lock out background commit */ down_read(&log->l_cilp->xc_ctx_lock); @@ -709,6 +734,7 @@ xfs_log_commit_cil( */ if (push) xlog_cil_push(log, 0); + return 0; } /* @@ -786,8 +812,6 @@ xfs_log_item_in_current_chkpt( { struct xfs_cil_ctx *ctx; - if (!(lip->li_mountp->m_flags & XFS_MOUNT_DELAYLOG)) - return false; if (list_empty(&lip->li_cil)) return false; diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index bb24dac42a25..19f69e232509 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -219,7 +219,6 @@ typedef struct xfs_mount { #define XFS_MOUNT_WSYNC (1ULL << 0) /* for nfs - all metadata ops must be synchronous except for space allocations */ -#define XFS_MOUNT_DELAYLOG (1ULL << 1) /* delayed logging is enabled */ #define XFS_MOUNT_WAS_CLEAN (1ULL << 3) #define XFS_MOUNT_FS_SHUTDOWN (1ULL << 4) /* atomic stop of all filesystem operations, typically for diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c index 5cff443f6cdb..671f37eae1c7 100644 --- a/fs/xfs/xfs_qm.c +++ b/fs/xfs/xfs_qm.c @@ -154,12 +154,17 @@ STATIC void xfs_qm_destroy( struct xfs_qm *xqm) { - struct xfs_dquot *dqp, *n; int hsize, i; ASSERT(xqm != NULL); ASSERT(xqm->qm_nrefs == 0); + unregister_shrinker(&xfs_qm_shaker); + + mutex_lock(&xqm->qm_dqfrlist_lock); + ASSERT(list_empty(&xqm->qm_dqfrlist)); + mutex_unlock(&xqm->qm_dqfrlist_lock); + hsize = xqm->qm_dqhashmask + 1; for (i = 0; i < hsize; i++) { xfs_qm_list_destroy(&(xqm->qm_usr_dqhtable[i])); @@ -171,17 +176,6 @@ xfs_qm_destroy( xqm->qm_grp_dqhtable = NULL; xqm->qm_dqhashmask = 0; - /* frlist cleanup */ - mutex_lock(&xqm->qm_dqfrlist_lock); - list_for_each_entry_safe(dqp, n, &xqm->qm_dqfrlist, q_freelist) { - xfs_dqlock(dqp); - list_del_init(&dqp->q_freelist); - xfs_Gqm->qm_dqfrlist_cnt--; - xfs_dqunlock(dqp); - xfs_qm_dqdestroy(dqp); - } - mutex_unlock(&xqm->qm_dqfrlist_lock); - mutex_destroy(&xqm->qm_dqfrlist_lock); kmem_free(xqm); } @@ -232,34 +226,10 @@ STATIC void xfs_qm_rele_quotafs_ref( struct xfs_mount *mp) { - xfs_dquot_t *dqp, *n; - ASSERT(xfs_Gqm); ASSERT(xfs_Gqm->qm_nrefs > 0); /* - * Go thru the freelist and destroy all inactive dquots. - */ - mutex_lock(&xfs_Gqm->qm_dqfrlist_lock); - - list_for_each_entry_safe(dqp, n, &xfs_Gqm->qm_dqfrlist, q_freelist) { - xfs_dqlock(dqp); - if (dqp->dq_flags & XFS_DQ_INACTIVE) { - ASSERT(dqp->q_mount == NULL); - ASSERT(! XFS_DQ_IS_DIRTY(dqp)); - ASSERT(list_empty(&dqp->q_hashlist)); - ASSERT(list_empty(&dqp->q_mplist)); - list_del_init(&dqp->q_freelist); - xfs_Gqm->qm_dqfrlist_cnt--; - xfs_dqunlock(dqp); - xfs_qm_dqdestroy(dqp); - } else { - xfs_dqunlock(dqp); - } - } - mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); - - /* * Destroy the entire XQM. If somebody mounts with quotaon, this'll * be restarted. */ @@ -415,8 +385,7 @@ xfs_qm_unmount_quotas( */ STATIC int xfs_qm_dqflush_all( - struct xfs_mount *mp, - int sync_mode) + struct xfs_mount *mp) { struct xfs_quotainfo *q = mp->m_quotainfo; int recl; @@ -429,7 +398,8 @@ again: mutex_lock(&q->qi_dqlist_lock); list_for_each_entry(dqp, &q->qi_dqlist, q_mplist) { xfs_dqlock(dqp); - if (! XFS_DQ_IS_DIRTY(dqp)) { + if ((dqp->dq_flags & XFS_DQ_FREEING) || + !XFS_DQ_IS_DIRTY(dqp)) { xfs_dqunlock(dqp); continue; } @@ -444,14 +414,14 @@ again: * out immediately. We'll be able to acquire * the flush lock when the I/O completes. */ - xfs_qm_dqflock_pushbuf_wait(dqp); + xfs_dqflock_pushbuf_wait(dqp); } /* * Let go of the mplist lock. We don't want to hold it * across a disk write. */ mutex_unlock(&q->qi_dqlist_lock); - error = xfs_qm_dqflush(dqp, sync_mode); + error = xfs_qm_dqflush(dqp, 0); xfs_dqunlock(dqp); if (error) return error; @@ -468,6 +438,7 @@ again: /* return ! busy */ return 0; } + /* * Release the group dquot pointers the user dquots may be * carrying around as a hint. mplist is locked on entry and exit. @@ -478,31 +449,26 @@ xfs_qm_detach_gdquots( { struct xfs_quotainfo *q = mp->m_quotainfo; struct xfs_dquot *dqp, *gdqp; - int nrecl; again: ASSERT(mutex_is_locked(&q->qi_dqlist_lock)); list_for_each_entry(dqp, &q->qi_dqlist, q_mplist) { xfs_dqlock(dqp); - if ((gdqp = dqp->q_gdquot)) { - xfs_dqlock(gdqp); - dqp->q_gdquot = NULL; - } - xfs_dqunlock(dqp); - - if (gdqp) { - /* - * Can't hold the mplist lock across a dqput. - * XXXmust convert to marker based iterations here. - */ - nrecl = q->qi_dqreclaims; + if (dqp->dq_flags & XFS_DQ_FREEING) { + xfs_dqunlock(dqp); mutex_unlock(&q->qi_dqlist_lock); - xfs_qm_dqput(gdqp); - + delay(1); mutex_lock(&q->qi_dqlist_lock); - if (nrecl != q->qi_dqreclaims) - goto again; + goto again; } + + gdqp = dqp->q_gdquot; + if (gdqp) + dqp->q_gdquot = NULL; + xfs_dqunlock(dqp); + + if (gdqp) + xfs_qm_dqrele(gdqp); } } @@ -520,8 +486,8 @@ xfs_qm_dqpurge_int( struct xfs_quotainfo *q = mp->m_quotainfo; struct xfs_dquot *dqp, *n; uint dqtype; - int nrecl; - int nmisses; + int nmisses = 0; + LIST_HEAD (dispose_list); if (!q) return 0; @@ -540,47 +506,26 @@ xfs_qm_dqpurge_int( */ xfs_qm_detach_gdquots(mp); - again: - nmisses = 0; - ASSERT(mutex_is_locked(&q->qi_dqlist_lock)); /* - * Try to get rid of all of the unwanted dquots. The idea is to - * get them off mplist and hashlist, but leave them on freelist. + * Try to get rid of all of the unwanted dquots. */ list_for_each_entry_safe(dqp, n, &q->qi_dqlist, q_mplist) { - /* - * It's OK to look at the type without taking dqlock here. - * We're holding the mplist lock here, and that's needed for - * a dqreclaim. - */ - if ((dqp->dq_flags & dqtype) == 0) - continue; - - if (!mutex_trylock(&dqp->q_hash->qh_lock)) { - nrecl = q->qi_dqreclaims; - mutex_unlock(&q->qi_dqlist_lock); - mutex_lock(&dqp->q_hash->qh_lock); - mutex_lock(&q->qi_dqlist_lock); - - /* - * XXXTheoretically, we can get into a very long - * ping pong game here. - * No one can be adding dquots to the mplist at - * this point, but somebody might be taking things off. - */ - if (nrecl != q->qi_dqreclaims) { - mutex_unlock(&dqp->q_hash->qh_lock); - goto again; - } + xfs_dqlock(dqp); + if ((dqp->dq_flags & dqtype) != 0 && + !(dqp->dq_flags & XFS_DQ_FREEING)) { + if (dqp->q_nrefs == 0) { + dqp->dq_flags |= XFS_DQ_FREEING; + list_move_tail(&dqp->q_mplist, &dispose_list); + } else + nmisses++; } - - /* - * Take the dquot off the mplist and hashlist. It may remain on - * freelist in INACTIVE state. - */ - nmisses += xfs_qm_dqpurge(dqp); + xfs_dqunlock(dqp); } mutex_unlock(&q->qi_dqlist_lock); + + list_for_each_entry_safe(dqp, n, &dispose_list, q_mplist) + xfs_qm_dqpurge(dqp); + return nmisses; } @@ -648,12 +593,9 @@ xfs_qm_dqattach_one( */ dqp = udqhint->q_gdquot; if (dqp && be32_to_cpu(dqp->q_core.d_id) == id) { - xfs_dqlock(dqp); - XFS_DQHOLD(dqp); ASSERT(*IO_idqpp == NULL); - *IO_idqpp = dqp; - xfs_dqunlock(dqp); + *IO_idqpp = xfs_qm_dqhold(dqp); xfs_dqunlock(udqhint); return 0; } @@ -674,7 +616,8 @@ xfs_qm_dqattach_one( * disk and we didn't ask it to allocate; * ESRCH if quotas got turned off suddenly. */ - error = xfs_qm_dqget(ip->i_mount, ip, id, type, XFS_QMOPT_DOWARN, &dqp); + error = xfs_qm_dqget(ip->i_mount, ip, id, type, + doalloc | XFS_QMOPT_DOWARN, &dqp); if (error) return error; @@ -692,11 +635,7 @@ xfs_qm_dqattach_one( /* * Given a udquot and gdquot, attach a ptr to the group dquot in the - * udquot as a hint for future lookups. The idea sounds simple, but the - * execution isn't, because the udquot might have a group dquot attached - * already and getting rid of that gets us into lock ordering constraints. - * The process is complicated more by the fact that the dquots may or may not - * be locked on entry. + * udquot as a hint for future lookups. */ STATIC void xfs_qm_dqattach_grouphint( @@ -707,45 +646,17 @@ xfs_qm_dqattach_grouphint( xfs_dqlock(udq); - if ((tmp = udq->q_gdquot)) { - if (tmp == gdq) { - xfs_dqunlock(udq); - return; - } + tmp = udq->q_gdquot; + if (tmp) { + if (tmp == gdq) + goto done; udq->q_gdquot = NULL; - /* - * We can't keep any dqlocks when calling dqrele, - * because the freelist lock comes before dqlocks. - */ - xfs_dqunlock(udq); - /* - * we took a hard reference once upon a time in dqget, - * so give it back when the udquot no longer points at it - * dqput() does the unlocking of the dquot. - */ xfs_qm_dqrele(tmp); - - xfs_dqlock(udq); - xfs_dqlock(gdq); - - } else { - ASSERT(XFS_DQ_IS_LOCKED(udq)); - xfs_dqlock(gdq); - } - - ASSERT(XFS_DQ_IS_LOCKED(udq)); - ASSERT(XFS_DQ_IS_LOCKED(gdq)); - /* - * Somebody could have attached a gdquot here, - * when we dropped the uqlock. If so, just do nothing. - */ - if (udq->q_gdquot == NULL) { - XFS_DQHOLD(gdq); - udq->q_gdquot = gdq; } - xfs_dqunlock(gdq); + udq->q_gdquot = xfs_qm_dqhold(gdq); +done: xfs_dqunlock(udq); } @@ -812,17 +723,13 @@ xfs_qm_dqattach_locked( ASSERT(ip->i_gdquot); /* - * We may or may not have the i_udquot locked at this point, - * but this check is OK since we don't depend on the i_gdquot to - * be accurate 100% all the time. It is just a hint, and this - * will succeed in general. - */ - if (ip->i_udquot->q_gdquot == ip->i_gdquot) - goto done; - /* - * Attach i_gdquot to the gdquot hint inside the i_udquot. + * We do not have i_udquot locked at this point, but this check + * is OK since we don't depend on the i_gdquot to be accurate + * 100% all the time. It is just a hint, and this will + * succeed in general. */ - xfs_qm_dqattach_grouphint(ip->i_udquot, ip->i_gdquot); + if (ip->i_udquot->q_gdquot != ip->i_gdquot) + xfs_qm_dqattach_grouphint(ip->i_udquot, ip->i_gdquot); } done: @@ -878,100 +785,6 @@ xfs_qm_dqdetach( } } -int -xfs_qm_sync( - struct xfs_mount *mp, - int flags) -{ - struct xfs_quotainfo *q = mp->m_quotainfo; - int recl, restarts; - struct xfs_dquot *dqp; - int error; - - if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp)) - return 0; - - restarts = 0; - - again: - mutex_lock(&q->qi_dqlist_lock); - /* - * dqpurge_all() also takes the mplist lock and iterate thru all dquots - * in quotaoff. However, if the QUOTA_ACTIVE bits are not cleared - * when we have the mplist lock, we know that dquots will be consistent - * as long as we have it locked. - */ - if (!XFS_IS_QUOTA_ON(mp)) { - mutex_unlock(&q->qi_dqlist_lock); - return 0; - } - ASSERT(mutex_is_locked(&q->qi_dqlist_lock)); - list_for_each_entry(dqp, &q->qi_dqlist, q_mplist) { - /* - * If this is vfs_sync calling, then skip the dquots that - * don't 'seem' to be dirty. ie. don't acquire dqlock. - * This is very similar to what xfs_sync does with inodes. - */ - if (flags & SYNC_TRYLOCK) { - if (!XFS_DQ_IS_DIRTY(dqp)) - continue; - if (!xfs_qm_dqlock_nowait(dqp)) - continue; - } else { - xfs_dqlock(dqp); - } - - /* - * Now, find out for sure if this dquot is dirty or not. - */ - if (! XFS_DQ_IS_DIRTY(dqp)) { - xfs_dqunlock(dqp); - continue; - } - - /* XXX a sentinel would be better */ - recl = q->qi_dqreclaims; - if (!xfs_dqflock_nowait(dqp)) { - if (flags & SYNC_TRYLOCK) { - xfs_dqunlock(dqp); - continue; - } - /* - * If we can't grab the flush lock then if the caller - * really wanted us to give this our best shot, so - * see if we can give a push to the buffer before we wait - * on the flush lock. At this point, we know that - * even though the dquot is being flushed, - * it has (new) dirty data. - */ - xfs_qm_dqflock_pushbuf_wait(dqp); - } - /* - * Let go of the mplist lock. We don't want to hold it - * across a disk write - */ - mutex_unlock(&q->qi_dqlist_lock); - error = xfs_qm_dqflush(dqp, flags); - xfs_dqunlock(dqp); - if (error && XFS_FORCED_SHUTDOWN(mp)) - return 0; /* Need to prevent umount failure */ - else if (error) - return error; - - mutex_lock(&q->qi_dqlist_lock); - if (recl != q->qi_dqreclaims) { - if (++restarts >= XFS_QM_SYNC_MAX_RESTARTS) - break; - - mutex_unlock(&q->qi_dqlist_lock); - goto again; - } - } - - mutex_unlock(&q->qi_dqlist_lock); - return 0; -} - /* * The hash chains and the mplist use the same xfs_dqhash structure as * their list head, but we can take the mplist qh_lock and one of the @@ -1033,18 +846,21 @@ xfs_qm_init_quotainfo( /* * We try to get the limits from the superuser's limits fields. * This is quite hacky, but it is standard quota practice. + * * We look at the USR dquot with id == 0 first, but if user quotas * are not enabled we goto the GRP dquot with id == 0. * We don't really care to keep separate default limits for user * and group quotas, at least not at this point. + * + * Since we may not have done a quotacheck by this point, just read + * the dquot without attaching it to any hashtables or lists. */ - error = xfs_qm_dqget(mp, NULL, (xfs_dqid_t)0, - XFS_IS_UQUOTA_RUNNING(mp) ? XFS_DQ_USER : - (XFS_IS_GQUOTA_RUNNING(mp) ? XFS_DQ_GROUP : - XFS_DQ_PROJ), - XFS_QMOPT_DQSUSER|XFS_QMOPT_DOWARN, - &dqp); - if (! error) { + error = xfs_qm_dqread(mp, 0, + XFS_IS_UQUOTA_RUNNING(mp) ? XFS_DQ_USER : + (XFS_IS_GQUOTA_RUNNING(mp) ? XFS_DQ_GROUP : + XFS_DQ_PROJ), + XFS_QMOPT_DOWARN, &dqp); + if (!error) { xfs_disk_dquot_t *ddqp = &dqp->q_core; /* @@ -1071,11 +887,6 @@ xfs_qm_init_quotainfo( qinf->qi_rtbhardlimit = be64_to_cpu(ddqp->d_rtb_hardlimit); qinf->qi_rtbsoftlimit = be64_to_cpu(ddqp->d_rtb_softlimit); - /* - * We sent the XFS_QMOPT_DQSUSER flag to dqget because - * we don't want this dquot cached. We haven't done a - * quotacheck yet, and quotacheck doesn't like incore dquots. - */ xfs_qm_dqdestroy(dqp); } else { qinf->qi_btimelimit = XFS_QM_BTIMELIMIT; @@ -1660,7 +1471,7 @@ xfs_qm_quotacheck( * successfully. */ if (!error) - error = xfs_qm_dqflush_all(mp, 0); + error = xfs_qm_dqflush_all(mp); /* * We can get this error if we couldn't do a dquot allocation inside @@ -1792,59 +1603,33 @@ xfs_qm_init_quotainos( /* - * Just pop the least recently used dquot off the freelist and - * recycle it. The returned dquot is locked. + * Pop the least recently used dquot off the freelist and recycle it. */ -STATIC xfs_dquot_t * +STATIC struct xfs_dquot * xfs_qm_dqreclaim_one(void) { - xfs_dquot_t *dqpout; - xfs_dquot_t *dqp; - int restarts; - int startagain; - - restarts = 0; - dqpout = NULL; + struct xfs_dquot *dqp; + int restarts = 0; - /* lockorder: hashchainlock, freelistlock, mplistlock, dqlock, dqflock */ -again: - startagain = 0; mutex_lock(&xfs_Gqm->qm_dqfrlist_lock); - +restart: list_for_each_entry(dqp, &xfs_Gqm->qm_dqfrlist, q_freelist) { struct xfs_mount *mp = dqp->q_mount; - xfs_dqlock(dqp); + + if (!xfs_dqlock_nowait(dqp)) + continue; /* - * We are racing with dqlookup here. Naturally we don't - * want to reclaim a dquot that lookup wants. We release the - * freelist lock and start over, so that lookup will grab - * both the dquot and the freelistlock. + * This dquot has already been grabbed by dqlookup. + * Remove it from the freelist and try again. */ - if (dqp->dq_flags & XFS_DQ_WANT) { - ASSERT(! (dqp->dq_flags & XFS_DQ_INACTIVE)); - + if (dqp->q_nrefs) { trace_xfs_dqreclaim_want(dqp); XQM_STATS_INC(xqmstats.xs_qm_dqwants); - restarts++; - startagain = 1; - goto dqunlock; - } - /* - * If the dquot is inactive, we are assured that it is - * not on the mplist or the hashlist, and that makes our - * life easier. - */ - if (dqp->dq_flags & XFS_DQ_INACTIVE) { - ASSERT(mp == NULL); - ASSERT(! XFS_DQ_IS_DIRTY(dqp)); - ASSERT(list_empty(&dqp->q_hashlist)); - ASSERT(list_empty(&dqp->q_mplist)); list_del_init(&dqp->q_freelist); xfs_Gqm->qm_dqfrlist_cnt--; - dqpout = dqp; - XQM_STATS_INC(xqmstats.xs_qm_dqinact_reclaims); + restarts++; goto dqunlock; } @@ -1873,64 +1658,49 @@ again: * We flush it delayed write, so don't bother * releasing the freelist lock. */ - error = xfs_qm_dqflush(dqp, 0); + error = xfs_qm_dqflush(dqp, SYNC_TRYLOCK); if (error) { xfs_warn(mp, "%s: dquot %p flush failed", __func__, dqp); } goto dqunlock; } + xfs_dqfunlock(dqp); /* - * We're trying to get the hashlock out of order. This races - * with dqlookup; so, we giveup and goto the next dquot if - * we couldn't get the hashlock. This way, we won't starve - * a dqlookup process that holds the hashlock that is - * waiting for the freelist lock. + * Prevent lookup now that we are going to reclaim the dquot. + * Once XFS_DQ_FREEING is set lookup won't touch the dquot, + * thus we can drop the lock now. */ - if (!mutex_trylock(&dqp->q_hash->qh_lock)) { - restarts++; - goto dqfunlock; - } + dqp->dq_flags |= XFS_DQ_FREEING; + xfs_dqunlock(dqp); - /* - * This races with dquot allocation code as well as dqflush_all - * and reclaim code. So, if we failed to grab the mplist lock, - * giveup everything and start over. - */ - if (!mutex_trylock(&mp->m_quotainfo->qi_dqlist_lock)) { - restarts++; - startagain = 1; - goto qhunlock; - } + mutex_lock(&dqp->q_hash->qh_lock); + list_del_init(&dqp->q_hashlist); + dqp->q_hash->qh_version++; + mutex_unlock(&dqp->q_hash->qh_lock); - ASSERT(dqp->q_nrefs == 0); + mutex_lock(&mp->m_quotainfo->qi_dqlist_lock); list_del_init(&dqp->q_mplist); mp->m_quotainfo->qi_dquots--; mp->m_quotainfo->qi_dqreclaims++; - list_del_init(&dqp->q_hashlist); - dqp->q_hash->qh_version++; + mutex_unlock(&mp->m_quotainfo->qi_dqlist_lock); + + ASSERT(dqp->q_nrefs == 0); list_del_init(&dqp->q_freelist); xfs_Gqm->qm_dqfrlist_cnt--; - dqpout = dqp; - mutex_unlock(&mp->m_quotainfo->qi_dqlist_lock); -qhunlock: - mutex_unlock(&dqp->q_hash->qh_lock); -dqfunlock: - xfs_dqfunlock(dqp); + + mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); + return dqp; dqunlock: xfs_dqunlock(dqp); - if (dqpout) - break; if (restarts >= XFS_QM_RECLAIM_MAX_RESTARTS) break; - if (startagain) { - mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); - goto again; - } + goto restart; } + mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); - return dqpout; + return NULL; } /* @@ -2150,10 +1920,7 @@ xfs_qm_vop_dqalloc( * this to caller */ ASSERT(ip->i_udquot); - uq = ip->i_udquot; - xfs_dqlock(uq); - XFS_DQHOLD(uq); - xfs_dqunlock(uq); + uq = xfs_qm_dqhold(ip->i_udquot); } } if ((flags & XFS_QMOPT_GQUOTA) && XFS_IS_GQUOTA_ON(mp)) { @@ -2174,10 +1941,7 @@ xfs_qm_vop_dqalloc( xfs_ilock(ip, lockflags); } else { ASSERT(ip->i_gdquot); - gq = ip->i_gdquot; - xfs_dqlock(gq); - XFS_DQHOLD(gq); - xfs_dqunlock(gq); + gq = xfs_qm_dqhold(ip->i_gdquot); } } else if ((flags & XFS_QMOPT_PQUOTA) && XFS_IS_PQUOTA_ON(mp)) { if (xfs_get_projid(ip) != prid) { @@ -2197,10 +1961,7 @@ xfs_qm_vop_dqalloc( xfs_ilock(ip, lockflags); } else { ASSERT(ip->i_gdquot); - gq = ip->i_gdquot; - xfs_dqlock(gq); - XFS_DQHOLD(gq); - xfs_dqunlock(gq); + gq = xfs_qm_dqhold(ip->i_gdquot); } } if (uq) @@ -2250,14 +2011,10 @@ xfs_qm_vop_chown( xfs_trans_mod_dquot(tp, newdq, XFS_TRANS_DQ_ICOUNT, 1); /* - * Take an extra reference, because the inode - * is going to keep this dquot pointer even - * after the trans_commit. + * Take an extra reference, because the inode is going to keep + * this dquot pointer even after the trans_commit. */ - xfs_dqlock(newdq); - XFS_DQHOLD(newdq); - xfs_dqunlock(newdq); - *IO_olddq = newdq; + *IO_olddq = xfs_qm_dqhold(newdq); return prevdq; } @@ -2389,25 +2146,21 @@ xfs_qm_vop_create_dqattach( ASSERT(XFS_IS_QUOTA_RUNNING(mp)); if (udqp) { - xfs_dqlock(udqp); - XFS_DQHOLD(udqp); - xfs_dqunlock(udqp); ASSERT(ip->i_udquot == NULL); - ip->i_udquot = udqp; ASSERT(XFS_IS_UQUOTA_ON(mp)); ASSERT(ip->i_d.di_uid == be32_to_cpu(udqp->q_core.d_id)); + + ip->i_udquot = xfs_qm_dqhold(udqp); xfs_trans_mod_dquot(tp, udqp, XFS_TRANS_DQ_ICOUNT, 1); } if (gdqp) { - xfs_dqlock(gdqp); - XFS_DQHOLD(gdqp); - xfs_dqunlock(gdqp); ASSERT(ip->i_gdquot == NULL); - ip->i_gdquot = gdqp; ASSERT(XFS_IS_OQUOTA_ON(mp)); ASSERT((XFS_IS_GQUOTA_ON(mp) ? ip->i_d.di_gid : xfs_get_projid(ip)) == be32_to_cpu(gdqp->q_core.d_id)); + + ip->i_gdquot = xfs_qm_dqhold(gdqp); xfs_trans_mod_dquot(tp, gdqp, XFS_TRANS_DQ_ICOUNT, 1); } } diff --git a/fs/xfs/xfs_qm.h b/fs/xfs/xfs_qm.h index 43b9abe1052c..9b4f3adefbc5 100644 --- a/fs/xfs/xfs_qm.h +++ b/fs/xfs/xfs_qm.h @@ -33,12 +33,6 @@ extern kmem_zone_t *qm_dqzone; extern kmem_zone_t *qm_dqtrxzone; /* - * Used in xfs_qm_sync called by xfs_sync to count the max times that it can - * iterate over the mountpt's dquot list in one call. - */ -#define XFS_QM_SYNC_MAX_RESTARTS 7 - -/* * Ditto, for xfs_qm_dqreclaim_one. */ #define XFS_QM_RECLAIM_MAX_RESTARTS 4 diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h index a595f29567fe..8a0807e0f979 100644 --- a/fs/xfs/xfs_quota.h +++ b/fs/xfs/xfs_quota.h @@ -87,8 +87,7 @@ typedef struct xfs_dqblk { #define XFS_DQ_PROJ 0x0002 /* project quota */ #define XFS_DQ_GROUP 0x0004 /* a group quota */ #define XFS_DQ_DIRTY 0x0008 /* dquot is dirty */ -#define XFS_DQ_WANT 0x0010 /* for lookup/reclaim race */ -#define XFS_DQ_INACTIVE 0x0020 /* dq off mplist & hashlist */ +#define XFS_DQ_FREEING 0x0010 /* dquot is beeing torn down */ #define XFS_DQ_ALLTYPES (XFS_DQ_USER|XFS_DQ_PROJ|XFS_DQ_GROUP) @@ -97,8 +96,7 @@ typedef struct xfs_dqblk { { XFS_DQ_PROJ, "PROJ" }, \ { XFS_DQ_GROUP, "GROUP" }, \ { XFS_DQ_DIRTY, "DIRTY" }, \ - { XFS_DQ_WANT, "WANT" }, \ - { XFS_DQ_INACTIVE, "INACTIVE" } + { XFS_DQ_FREEING, "FREEING" } /* * In the worst case, when both user and group quotas are on, @@ -199,7 +197,6 @@ typedef struct xfs_qoff_logformat { #define XFS_QMOPT_UQUOTA 0x0000004 /* user dquot requested */ #define XFS_QMOPT_PQUOTA 0x0000008 /* project dquot requested */ #define XFS_QMOPT_FORCE_RES 0x0000010 /* ignore quota limits */ -#define XFS_QMOPT_DQSUSER 0x0000020 /* don't cache super users dquot */ #define XFS_QMOPT_SBVERSION 0x0000040 /* change superblock version num */ #define XFS_QMOPT_DOWARN 0x0000400 /* increase warning cnt if needed */ #define XFS_QMOPT_DQREPAIR 0x0001000 /* repair dquot if damaged */ @@ -326,7 +323,6 @@ extern int xfs_qm_dqattach_locked(struct xfs_inode *, uint); extern void xfs_qm_dqdetach(struct xfs_inode *); extern void xfs_qm_dqrele(struct xfs_dquot *); extern void xfs_qm_statvfs(struct xfs_inode *, struct kstatfs *); -extern int xfs_qm_sync(struct xfs_mount *, int); extern int xfs_qm_newmount(struct xfs_mount *, uint *, uint *); extern void xfs_qm_mount_quotas(struct xfs_mount *); extern void xfs_qm_unmount(struct xfs_mount *); @@ -366,10 +362,6 @@ static inline int xfs_trans_reserve_quota_bydquots(struct xfs_trans *tp, #define xfs_qm_dqdetach(ip) #define xfs_qm_dqrele(d) #define xfs_qm_statvfs(ip, s) -static inline int xfs_qm_sync(struct xfs_mount *mp, int flags) -{ - return 0; -} #define xfs_qm_newmount(mp, a, b) (0) #define xfs_qm_mount_quotas(mp) #define xfs_qm_unmount(mp) diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 3eca58f51ae9..281961c1d81a 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -199,7 +199,6 @@ xfs_parseargs( mp->m_flags |= XFS_MOUNT_BARRIER; mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE; mp->m_flags |= XFS_MOUNT_SMALL_INUMS; - mp->m_flags |= XFS_MOUNT_DELAYLOG; /* * These can be overridden by the mount option parsing. @@ -353,11 +352,11 @@ xfs_parseargs( mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE); mp->m_qflags &= ~XFS_OQUOTA_ENFD; } else if (!strcmp(this_char, MNTOPT_DELAYLOG)) { - mp->m_flags |= XFS_MOUNT_DELAYLOG; + xfs_warn(mp, + "delaylog is the default now, option is deprecated."); } else if (!strcmp(this_char, MNTOPT_NODELAYLOG)) { - mp->m_flags &= ~XFS_MOUNT_DELAYLOG; xfs_warn(mp, - "nodelaylog is deprecated and will be removed in Linux 3.3"); + "nodelaylog support has been removed, option is deprecated."); } else if (!strcmp(this_char, MNTOPT_DISCARD)) { mp->m_flags |= XFS_MOUNT_DISCARD; } else if (!strcmp(this_char, MNTOPT_NODISCARD)) { @@ -395,13 +394,6 @@ xfs_parseargs( return EINVAL; } - if ((mp->m_flags & XFS_MOUNT_DISCARD) && - !(mp->m_flags & XFS_MOUNT_DELAYLOG)) { - xfs_warn(mp, - "the discard option is incompatible with the nodelaylog option"); - return EINVAL; - } - #ifndef CONFIG_XFS_QUOTA if (XFS_IS_QUOTA_RUNNING(mp)) { xfs_warn(mp, "quota support not available in this kernel."); @@ -501,7 +493,6 @@ xfs_showargs( { XFS_MOUNT_ATTR2, "," MNTOPT_ATTR2 }, { XFS_MOUNT_FILESTREAMS, "," MNTOPT_FILESTREAM }, { XFS_MOUNT_GRPID, "," MNTOPT_GRPID }, - { XFS_MOUNT_DELAYLOG, "," MNTOPT_DELAYLOG }, { XFS_MOUNT_DISCARD, "," MNTOPT_DISCARD }, { 0, NULL } }; @@ -869,27 +860,6 @@ xfs_fs_dirty_inode( } STATIC int -xfs_log_inode( - struct xfs_inode *ip) -{ - struct xfs_mount *mp = ip->i_mount; - struct xfs_trans *tp; - int error; - - tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS); - error = xfs_trans_reserve(tp, 0, XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0); - if (error) { - xfs_trans_cancel(tp, 0); - return error; - } - - xfs_ilock(ip, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - return xfs_trans_commit(tp, 0); -} - -STATIC int xfs_fs_write_inode( struct inode *inode, struct writeback_control *wbc) @@ -902,10 +872,8 @@ xfs_fs_write_inode( if (XFS_FORCED_SHUTDOWN(mp)) return -XFS_ERROR(EIO); - if (!ip->i_update_core) - return 0; - if (wbc->sync_mode == WB_SYNC_ALL) { + if (wbc->sync_mode == WB_SYNC_ALL || wbc->for_kupdate) { /* * Make sure the inode has made it it into the log. Instead * of forcing it all the way to stable storage using a @@ -913,11 +881,14 @@ xfs_fs_write_inode( * ->sync_fs call do that for thus, which reduces the number * of synchronous log forces dramatically. */ - error = xfs_log_inode(ip); + error = xfs_log_dirty_inode(ip, NULL, 0); if (error) goto out; return 0; } else { + if (!ip->i_update_core) + return 0; + /* * We make this non-blocking if the inode is contended, return * EAGAIN to indicate to the caller that they did not succeed. @@ -1034,17 +1005,10 @@ xfs_fs_sync_fs( int error; /* - * Not much we can do for the first async pass. Writing out the - * superblock would be counter-productive as we are going to redirty - * when writing out other data and metadata (and writing out a single - * block is quite fast anyway). - * - * Try to asynchronously kick off quota syncing at least. + * Doing anything during the async pass would be counterproductive. */ - if (!wait) { - xfs_qm_sync(mp, SYNC_TRYLOCK); + if (!wait) return 0; - } error = xfs_quiesce_data(mp); if (error) @@ -1258,9 +1222,9 @@ xfs_fs_unfreeze( STATIC int xfs_fs_show_options( struct seq_file *m, - struct vfsmount *mnt) + struct dentry *root) { - return -xfs_showargs(XFS_M(mnt->mnt_sb), m); + return -xfs_showargs(XFS_M(root->d_sb), m); } /* @@ -1641,12 +1605,12 @@ STATIC int __init xfs_init_workqueues(void) { /* - * max_active is set to 8 to give enough concurency to allow - * multiple work operations on each CPU to run. This allows multiple - * filesystems to be running sync work concurrently, and scales with - * the number of CPUs in the system. + * We never want to the same work item to run twice, reclaiming inodes + * or idling the log is not going to get any faster by multiple CPUs + * competing for ressources. Use the default large max_active value + * so that even lots of filesystems can perform these task in parallel. */ - xfs_syncd_wq = alloc_workqueue("xfssyncd", WQ_CPU_INTENSIVE, 8); + xfs_syncd_wq = alloc_workqueue("xfssyncd", WQ_NON_REENTRANT, 0); if (!xfs_syncd_wq) return -ENOMEM; return 0; diff --git a/fs/xfs/xfs_sync.c b/fs/xfs/xfs_sync.c index aa3dc1a4d53d..72c01a1c16e7 100644 --- a/fs/xfs/xfs_sync.c +++ b/fs/xfs/xfs_sync.c @@ -336,6 +336,32 @@ xfs_sync_fsdata( return error; } +int +xfs_log_dirty_inode( + struct xfs_inode *ip, + struct xfs_perag *pag, + int flags) +{ + struct xfs_mount *mp = ip->i_mount; + struct xfs_trans *tp; + int error; + + if (!ip->i_update_core) + return 0; + + tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS); + error = xfs_trans_reserve(tp, 0, XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0); + if (error) { + xfs_trans_cancel(tp, 0); + return error; + } + + xfs_ilock(ip, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + return xfs_trans_commit(tp, 0); +} + /* * When remounting a filesystem read-only or freezing the filesystem, we have * two phases to execute. This first phase is syncing the data before we @@ -359,10 +385,17 @@ xfs_quiesce_data( { int error, error2 = 0; - xfs_qm_sync(mp, SYNC_TRYLOCK); - xfs_qm_sync(mp, SYNC_WAIT); + /* + * Log all pending size and timestamp updates. The vfs writeback + * code is supposed to do this, but due to its overagressive + * livelock detection it will skip inodes where appending writes + * were written out in the first non-blocking sync phase if their + * completion took long enough that it happened after taking the + * timestamp for the cut-off in the blocking phase. + */ + xfs_inode_ag_iterator(mp, xfs_log_dirty_inode, 0); - /* force out the newly dirtied log buffers */ + /* force out the log */ xfs_log_force(mp, XFS_LOG_SYNC); /* write superblock and hoover up shutdown errors */ @@ -470,7 +503,6 @@ xfs_sync_worker( error = xfs_fs_log_dummy(mp); else xfs_log_force(mp, 0); - error = xfs_qm_sync(mp, SYNC_TRYLOCK); /* start pushing all the metadata that is currently dirty */ xfs_ail_push_all(mp->m_ail); @@ -770,6 +802,17 @@ restart: if (!xfs_iflock_nowait(ip)) { if (!(sync_mode & SYNC_WAIT)) goto out; + + /* + * If we only have a single dirty inode in a cluster there is + * a fair chance that the AIL push may have pushed it into + * the buffer, but xfsbufd won't touch it until 30 seconds + * from now, and thus we will lock up here. + * + * Promote the inode buffer to the front of the delwri list + * and wake up xfsbufd now. + */ + xfs_promote_inode(ip); xfs_iflock(ip); } diff --git a/fs/xfs/xfs_sync.h b/fs/xfs/xfs_sync.h index 941202e7ac6e..fa965479d788 100644 --- a/fs/xfs/xfs_sync.h +++ b/fs/xfs/xfs_sync.h @@ -34,6 +34,8 @@ void xfs_quiesce_attr(struct xfs_mount *mp); void xfs_flush_inodes(struct xfs_inode *ip); +int xfs_log_dirty_inode(struct xfs_inode *ip, struct xfs_perag *pag, int flags); + int xfs_reclaim_inodes(struct xfs_mount *mp, int mode); int xfs_reclaim_inodes_count(struct xfs_mount *mp); void xfs_reclaim_inodes_nr(struct xfs_mount *mp, int nr_to_scan); diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index f1d2802b2f07..a9d5b1e06efe 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -743,8 +743,6 @@ DEFINE_DQUOT_EVENT(xfs_dqtobp_read); DEFINE_DQUOT_EVENT(xfs_dqread); DEFINE_DQUOT_EVENT(xfs_dqread_fail); DEFINE_DQUOT_EVENT(xfs_dqlookup_found); -DEFINE_DQUOT_EVENT(xfs_dqlookup_want); -DEFINE_DQUOT_EVENT(xfs_dqlookup_freelist); DEFINE_DQUOT_EVENT(xfs_dqlookup_done); DEFINE_DQUOT_EVENT(xfs_dqget_hit); DEFINE_DQUOT_EVENT(xfs_dqget_miss); @@ -834,18 +832,14 @@ DEFINE_LOGGRANT_EVENT(xfs_log_umount_write); DEFINE_LOGGRANT_EVENT(xfs_log_grant_enter); DEFINE_LOGGRANT_EVENT(xfs_log_grant_exit); DEFINE_LOGGRANT_EVENT(xfs_log_grant_error); -DEFINE_LOGGRANT_EVENT(xfs_log_grant_sleep1); -DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake1); -DEFINE_LOGGRANT_EVENT(xfs_log_grant_sleep2); -DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake2); +DEFINE_LOGGRANT_EVENT(xfs_log_grant_sleep); +DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake); DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake_up); DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_enter); DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_exit); DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_error); -DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_sleep1); -DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake1); -DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_sleep2); -DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake2); +DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_sleep); +DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake); DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake_up); DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_enter); DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_exit); diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index 1f35b2feca97..329b06aba1c2 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -1158,7 +1158,6 @@ xfs_trans_add_item( lidp->lid_item = lip; lidp->lid_flags = 0; - lidp->lid_size = 0; list_add_tail(&lidp->lid_trans, &tp->t_items); lip->li_desc = lidp; @@ -1210,219 +1209,6 @@ xfs_trans_free_items( } } -/* - * Unlock the items associated with a transaction. - * - * Items which were not logged should be freed. Those which were logged must - * still be tracked so they can be unpinned when the transaction commits. - */ -STATIC void -xfs_trans_unlock_items( - struct xfs_trans *tp, - xfs_lsn_t commit_lsn) -{ - struct xfs_log_item_desc *lidp, *next; - - list_for_each_entry_safe(lidp, next, &tp->t_items, lid_trans) { - struct xfs_log_item *lip = lidp->lid_item; - - lip->li_desc = NULL; - - if (commit_lsn != NULLCOMMITLSN) - IOP_COMMITTING(lip, commit_lsn); - IOP_UNLOCK(lip); - - /* - * Free the descriptor if the item is not dirty - * within this transaction. - */ - if (!(lidp->lid_flags & XFS_LID_DIRTY)) - xfs_trans_free_item_desc(lidp); - } -} - -/* - * Total up the number of log iovecs needed to commit this - * transaction. The transaction itself needs one for the - * transaction header. Ask each dirty item in turn how many - * it needs to get the total. - */ -static uint -xfs_trans_count_vecs( - struct xfs_trans *tp) -{ - int nvecs; - struct xfs_log_item_desc *lidp; - - nvecs = 1; - - /* In the non-debug case we need to start bailing out if we - * didn't find a log_item here, return zero and let trans_commit - * deal with it. - */ - if (list_empty(&tp->t_items)) { - ASSERT(0); - return 0; - } - - list_for_each_entry(lidp, &tp->t_items, lid_trans) { - /* - * Skip items which aren't dirty in this transaction. - */ - if (!(lidp->lid_flags & XFS_LID_DIRTY)) - continue; - lidp->lid_size = IOP_SIZE(lidp->lid_item); - nvecs += lidp->lid_size; - } - - return nvecs; -} - -/* - * Fill in the vector with pointers to data to be logged - * by this transaction. The transaction header takes - * the first vector, and then each dirty item takes the - * number of vectors it indicated it needed in xfs_trans_count_vecs(). - * - * As each item fills in the entries it needs, also pin the item - * so that it cannot be flushed out until the log write completes. - */ -static void -xfs_trans_fill_vecs( - struct xfs_trans *tp, - struct xfs_log_iovec *log_vector) -{ - struct xfs_log_item_desc *lidp; - struct xfs_log_iovec *vecp; - uint nitems; - - /* - * Skip over the entry for the transaction header, we'll - * fill that in at the end. - */ - vecp = log_vector + 1; - - nitems = 0; - ASSERT(!list_empty(&tp->t_items)); - list_for_each_entry(lidp, &tp->t_items, lid_trans) { - /* Skip items which aren't dirty in this transaction. */ - if (!(lidp->lid_flags & XFS_LID_DIRTY)) - continue; - - /* - * The item may be marked dirty but not log anything. This can - * be used to get called when a transaction is committed. - */ - if (lidp->lid_size) - nitems++; - IOP_FORMAT(lidp->lid_item, vecp); - vecp += lidp->lid_size; - IOP_PIN(lidp->lid_item); - } - - /* - * Now that we've counted the number of items in this transaction, fill - * in the transaction header. Note that the transaction header does not - * have a log item. - */ - tp->t_header.th_magic = XFS_TRANS_HEADER_MAGIC; - tp->t_header.th_type = tp->t_type; - tp->t_header.th_num_items = nitems; - log_vector->i_addr = (xfs_caddr_t)&tp->t_header; - log_vector->i_len = sizeof(xfs_trans_header_t); - log_vector->i_type = XLOG_REG_TYPE_TRANSHDR; -} - -/* - * The committed item processing consists of calling the committed routine of - * each logged item, updating the item's position in the AIL if necessary, and - * unpinning each item. If the committed routine returns -1, then do nothing - * further with the item because it may have been freed. - * - * Since items are unlocked when they are copied to the incore log, it is - * possible for two transactions to be completing and manipulating the same - * item simultaneously. The AIL lock will protect the lsn field of each item. - * The value of this field can never go backwards. - * - * We unpin the items after repositioning them in the AIL, because otherwise - * they could be immediately flushed and we'd have to race with the flusher - * trying to pull the item from the AIL as we add it. - */ -static void -xfs_trans_item_committed( - struct xfs_log_item *lip, - xfs_lsn_t commit_lsn, - int aborted) -{ - xfs_lsn_t item_lsn; - struct xfs_ail *ailp; - - if (aborted) - lip->li_flags |= XFS_LI_ABORTED; - item_lsn = IOP_COMMITTED(lip, commit_lsn); - - /* item_lsn of -1 means the item needs no further processing */ - if (XFS_LSN_CMP(item_lsn, (xfs_lsn_t)-1) == 0) - return; - - /* - * If the returned lsn is greater than what it contained before, update - * the location of the item in the AIL. If it is not, then do nothing. - * Items can never move backwards in the AIL. - * - * While the new lsn should usually be greater, it is possible that a - * later transaction completing simultaneously with an earlier one - * using the same item could complete first with a higher lsn. This - * would cause the earlier transaction to fail the test below. - */ - ailp = lip->li_ailp; - spin_lock(&ailp->xa_lock); - if (XFS_LSN_CMP(item_lsn, lip->li_lsn) > 0) { - /* - * This will set the item's lsn to item_lsn and update the - * position of the item in the AIL. - * - * xfs_trans_ail_update() drops the AIL lock. - */ - xfs_trans_ail_update(ailp, lip, item_lsn); - } else { - spin_unlock(&ailp->xa_lock); - } - - /* - * Now that we've repositioned the item in the AIL, unpin it so it can - * be flushed. Pass information about buffer stale state down from the - * log item flags, if anyone else stales the buffer we do not want to - * pay any attention to it. - */ - IOP_UNPIN(lip, 0); -} - -/* - * This is typically called by the LM when a transaction has been fully - * committed to disk. It needs to unpin the items which have - * been logged by the transaction and update their positions - * in the AIL if necessary. - * - * This also gets called when the transactions didn't get written out - * because of an I/O error. Abortflag & XFS_LI_ABORTED is set then. - */ -STATIC void -xfs_trans_committed( - void *arg, - int abortflag) -{ - struct xfs_trans *tp = arg; - struct xfs_log_item_desc *lidp, *next; - - list_for_each_entry_safe(lidp, next, &tp->t_items, lid_trans) { - xfs_trans_item_committed(lidp->lid_item, tp->t_lsn, abortflag); - xfs_trans_free_item_desc(lidp); - } - - xfs_trans_free(tp); -} - static inline void xfs_log_item_batch_insert( struct xfs_ail *ailp, @@ -1538,258 +1324,6 @@ xfs_trans_committed_bulk( } /* - * Called from the trans_commit code when we notice that the filesystem is in - * the middle of a forced shutdown. - * - * When we are called here, we have already pinned all the items in the - * transaction. However, neither IOP_COMMITTING or IOP_UNLOCK has been called - * so we can simply walk the items in the transaction, unpin them with an abort - * flag and then free the items. Note that unpinning the items can result in - * them being freed immediately, so we need to use a safe list traversal method - * here. - */ -STATIC void -xfs_trans_uncommit( - struct xfs_trans *tp, - uint flags) -{ - struct xfs_log_item_desc *lidp, *n; - - list_for_each_entry_safe(lidp, n, &tp->t_items, lid_trans) { - if (lidp->lid_flags & XFS_LID_DIRTY) - IOP_UNPIN(lidp->lid_item, 1); - } - - xfs_trans_unreserve_and_mod_sb(tp); - xfs_trans_unreserve_and_mod_dquots(tp); - - xfs_trans_free_items(tp, NULLCOMMITLSN, flags); - xfs_trans_free(tp); -} - -/* - * Format the transaction direct to the iclog. This isolates the physical - * transaction commit operation from the logical operation and hence allows - * other methods to be introduced without affecting the existing commit path. - */ -static int -xfs_trans_commit_iclog( - struct xfs_mount *mp, - struct xfs_trans *tp, - xfs_lsn_t *commit_lsn, - int flags) -{ - int shutdown; - int error; - int log_flags = 0; - struct xlog_in_core *commit_iclog; -#define XFS_TRANS_LOGVEC_COUNT 16 - struct xfs_log_iovec log_vector_fast[XFS_TRANS_LOGVEC_COUNT]; - struct xfs_log_iovec *log_vector; - uint nvec; - - - /* - * Ask each log item how many log_vector entries it will - * need so we can figure out how many to allocate. - * Try to avoid the kmem_alloc() call in the common case - * by using a vector from the stack when it fits. - */ - nvec = xfs_trans_count_vecs(tp); - if (nvec == 0) { - return ENOMEM; /* triggers a shutdown! */ - } else if (nvec <= XFS_TRANS_LOGVEC_COUNT) { - log_vector = log_vector_fast; - } else { - log_vector = (xfs_log_iovec_t *)kmem_alloc(nvec * - sizeof(xfs_log_iovec_t), - KM_SLEEP); - } - - /* - * Fill in the log_vector and pin the logged items, and - * then write the transaction to the log. - */ - xfs_trans_fill_vecs(tp, log_vector); - - if (flags & XFS_TRANS_RELEASE_LOG_RES) - log_flags = XFS_LOG_REL_PERM_RESERV; - - error = xfs_log_write(mp, log_vector, nvec, tp->t_ticket, &(tp->t_lsn)); - - /* - * The transaction is committed incore here, and can go out to disk - * at any time after this call. However, all the items associated - * with the transaction are still locked and pinned in memory. - */ - *commit_lsn = xfs_log_done(mp, tp->t_ticket, &commit_iclog, log_flags); - - tp->t_commit_lsn = *commit_lsn; - trace_xfs_trans_commit_lsn(tp); - - if (nvec > XFS_TRANS_LOGVEC_COUNT) - kmem_free(log_vector); - - /* - * If we got a log write error. Unpin the logitems that we - * had pinned, clean up, free trans structure, and return error. - */ - if (error || *commit_lsn == -1) { - current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS); - xfs_trans_uncommit(tp, flags|XFS_TRANS_ABORT); - return XFS_ERROR(EIO); - } - - /* - * Once the transaction has committed, unused - * reservations need to be released and changes to - * the superblock need to be reflected in the in-core - * version. Do that now. - */ - xfs_trans_unreserve_and_mod_sb(tp); - - /* - * Tell the LM to call the transaction completion routine - * when the log write with LSN commit_lsn completes (e.g. - * when the transaction commit really hits the on-disk log). - * After this call we cannot reference tp, because the call - * can happen at any time and the call will free the transaction - * structure pointed to by tp. The only case where we call - * the completion routine (xfs_trans_committed) directly is - * if the log is turned off on a debug kernel or we're - * running in simulation mode (the log is explicitly turned - * off). - */ - tp->t_logcb.cb_func = xfs_trans_committed; - tp->t_logcb.cb_arg = tp; - - /* - * We need to pass the iclog buffer which was used for the - * transaction commit record into this function, and attach - * the callback to it. The callback must be attached before - * the items are unlocked to avoid racing with other threads - * waiting for an item to unlock. - */ - shutdown = xfs_log_notify(mp, commit_iclog, &(tp->t_logcb)); - - /* - * Mark this thread as no longer being in a transaction - */ - current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS); - - /* - * Once all the items of the transaction have been copied - * to the in core log and the callback is attached, the - * items can be unlocked. - * - * This will free descriptors pointing to items which were - * not logged since there is nothing more to do with them. - * For items which were logged, we will keep pointers to them - * so they can be unpinned after the transaction commits to disk. - * This will also stamp each modified meta-data item with - * the commit lsn of this transaction for dependency tracking - * purposes. - */ - xfs_trans_unlock_items(tp, *commit_lsn); - - /* - * If we detected a log error earlier, finish committing - * the transaction now (unpin log items, etc). - * - * Order is critical here, to avoid using the transaction - * pointer after its been freed (by xfs_trans_committed - * either here now, or as a callback). We cannot do this - * step inside xfs_log_notify as was done earlier because - * of this issue. - */ - if (shutdown) - xfs_trans_committed(tp, XFS_LI_ABORTED); - - /* - * Now that the xfs_trans_committed callback has been attached, - * and the items are released we can finally allow the iclog to - * go to disk. - */ - return xfs_log_release_iclog(mp, commit_iclog); -} - -/* - * Walk the log items and allocate log vector structures for - * each item large enough to fit all the vectors they require. - * Note that this format differs from the old log vector format in - * that there is no transaction header in these log vectors. - */ -STATIC struct xfs_log_vec * -xfs_trans_alloc_log_vecs( - xfs_trans_t *tp) -{ - struct xfs_log_item_desc *lidp; - struct xfs_log_vec *lv = NULL; - struct xfs_log_vec *ret_lv = NULL; - - - /* Bail out if we didn't find a log item. */ - if (list_empty(&tp->t_items)) { - ASSERT(0); - return NULL; - } - - list_for_each_entry(lidp, &tp->t_items, lid_trans) { - struct xfs_log_vec *new_lv; - - /* Skip items which aren't dirty in this transaction. */ - if (!(lidp->lid_flags & XFS_LID_DIRTY)) - continue; - - /* Skip items that do not have any vectors for writing */ - lidp->lid_size = IOP_SIZE(lidp->lid_item); - if (!lidp->lid_size) - continue; - - new_lv = kmem_zalloc(sizeof(*new_lv) + - lidp->lid_size * sizeof(struct xfs_log_iovec), - KM_SLEEP); - - /* The allocated iovec region lies beyond the log vector. */ - new_lv->lv_iovecp = (struct xfs_log_iovec *)&new_lv[1]; - new_lv->lv_niovecs = lidp->lid_size; - new_lv->lv_item = lidp->lid_item; - if (!ret_lv) - ret_lv = new_lv; - else - lv->lv_next = new_lv; - lv = new_lv; - } - - return ret_lv; -} - -static int -xfs_trans_commit_cil( - struct xfs_mount *mp, - struct xfs_trans *tp, - xfs_lsn_t *commit_lsn, - int flags) -{ - struct xfs_log_vec *log_vector; - - /* - * Get each log item to allocate a vector structure for - * the log item to to pass to the log write code. The - * CIL commit code will format the vector and save it away. - */ - log_vector = xfs_trans_alloc_log_vecs(tp); - if (!log_vector) - return ENOMEM; - - xfs_log_commit_cil(mp, tp, log_vector, commit_lsn, flags); - - current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS); - xfs_trans_free(tp); - return 0; -} - -/* * Commit the given transaction to the log. * * XFS disk error handling mechanism is not based on a typical @@ -1845,17 +1379,16 @@ xfs_trans_commit( xfs_trans_apply_sb_deltas(tp); xfs_trans_apply_dquot_deltas(tp); - if (mp->m_flags & XFS_MOUNT_DELAYLOG) - error = xfs_trans_commit_cil(mp, tp, &commit_lsn, flags); - else - error = xfs_trans_commit_iclog(mp, tp, &commit_lsn, flags); - + error = xfs_log_commit_cil(mp, tp, &commit_lsn, flags); if (error == ENOMEM) { xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR); error = XFS_ERROR(EIO); goto out_unreserve; } + current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS); + xfs_trans_free(tp); + /* * If the transaction needs to be synchronous, then force the * log out now and wait for it. diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h index 3ae713c0abd9..f6118703f20d 100644 --- a/fs/xfs/xfs_trans.h +++ b/fs/xfs/xfs_trans.h @@ -163,9 +163,8 @@ typedef struct xfs_trans_header { */ struct xfs_log_item_desc { struct xfs_log_item *lid_item; - ushort lid_size; - unsigned char lid_flags; struct list_head lid_trans; + unsigned char lid_flags; }; #define XFS_LID_DIRTY 0x1 diff --git a/fs/xfs/xfs_utils.c b/fs/xfs/xfs_utils.c index 8b32d1a4c5a1..89dbb4a50872 100644 --- a/fs/xfs/xfs_utils.c +++ b/fs/xfs/xfs_utils.c @@ -53,7 +53,7 @@ xfs_dir_ialloc( output: may be a new transaction. */ xfs_inode_t *dp, /* directory within whose allocate the inode. */ - mode_t mode, + umode_t mode, xfs_nlink_t nlink, xfs_dev_t rdev, prid_t prid, /* project id */ diff --git a/fs/xfs/xfs_utils.h b/fs/xfs/xfs_utils.h index 456fca314933..5eeab4690cfe 100644 --- a/fs/xfs/xfs_utils.h +++ b/fs/xfs/xfs_utils.h @@ -18,7 +18,7 @@ #ifndef __XFS_UTILS_H__ #define __XFS_UTILS_H__ -extern int xfs_dir_ialloc(xfs_trans_t **, xfs_inode_t *, mode_t, xfs_nlink_t, +extern int xfs_dir_ialloc(xfs_trans_t **, xfs_inode_t *, umode_t, xfs_nlink_t, xfs_dev_t, prid_t, int, xfs_inode_t **, int *); extern int xfs_droplink(xfs_trans_t *, xfs_inode_t *); extern int xfs_bumplink(xfs_trans_t *, xfs_inode_t *); diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index ce9268a2f56b..f2fea868d4db 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -822,7 +822,7 @@ int xfs_create( xfs_inode_t *dp, struct xfs_name *name, - mode_t mode, + umode_t mode, xfs_dev_t rdev, xfs_inode_t **ipp) { @@ -1481,7 +1481,7 @@ xfs_symlink( xfs_inode_t *dp, struct xfs_name *link_name, const char *target_path, - mode_t mode, + umode_t mode, xfs_inode_t **ipp) { xfs_mount_t *mp = dp->i_mount; diff --git a/fs/xfs/xfs_vnodeops.h b/fs/xfs/xfs_vnodeops.h index 35d3d513e1e9..0c877cbde142 100644 --- a/fs/xfs/xfs_vnodeops.h +++ b/fs/xfs/xfs_vnodeops.h @@ -26,7 +26,7 @@ int xfs_release(struct xfs_inode *ip); int xfs_inactive(struct xfs_inode *ip); int xfs_lookup(struct xfs_inode *dp, struct xfs_name *name, struct xfs_inode **ipp, struct xfs_name *ci_name); -int xfs_create(struct xfs_inode *dp, struct xfs_name *name, mode_t mode, +int xfs_create(struct xfs_inode *dp, struct xfs_name *name, umode_t mode, xfs_dev_t rdev, struct xfs_inode **ipp); int xfs_remove(struct xfs_inode *dp, struct xfs_name *name, struct xfs_inode *ip); @@ -35,7 +35,7 @@ int xfs_link(struct xfs_inode *tdp, struct xfs_inode *sip, int xfs_readdir(struct xfs_inode *dp, void *dirent, size_t bufsize, xfs_off_t *offset, filldir_t filldir); int xfs_symlink(struct xfs_inode *dp, struct xfs_name *link_name, - const char *target_path, mode_t mode, struct xfs_inode **ipp); + const char *target_path, umode_t mode, struct xfs_inode **ipp); int xfs_set_dmattrs(struct xfs_inode *ip, u_int evmask, u_int16_t state); int xfs_change_file_space(struct xfs_inode *ip, int cmd, xfs_flock64_t *bf, xfs_off_t offset, int attr_flags); |