implementation of directory and inode operations.
* A directory is treated as a file, and essentially contains a list
of <file name, inode #> pairs for files that are found in that
directory. The object IDs correspond to the files' inode numbers
and are allocated using a 64bit incrementing global counter.
* Each file's control block (AKA on-disk inode) is stored in its
object's attributes. This applies to both regular files and other
types (directories, device files, symlinks, etc.).
Signed-off-by: Boaz Harrosh <bharrosh@xxxxxxxxxxx>
---
fs/osdfs/Kbuild | 2 +-
fs/osdfs/dir.c | 629 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
fs/osdfs/inode.c | 267 +++++++++++++++++++++++
fs/osdfs/namei.c | 348 ++++++++++++++++++++++++++++++
fs/osdfs/osdfs.h | 26 +++
5 files changed, 1271 insertions(+), 1 deletions(-)
create mode 100644 fs/osdfs/dir.c
create mode 100644 fs/osdfs/namei.c
diff --git a/fs/osdfs/Kbuild b/fs/osdfs/Kbuild
index eddba6a..d6ac8d6 100644
--- a/fs/osdfs/Kbuild
+++ b/fs/osdfs/Kbuild
@@ -20,5 +20,5 @@ EXTRA_CFLAGS += -I$(OSD_INC)
# EXTRA_CFLAGS += -DCONFIG_OSDFS_DEBUG
endif
-osdfs-objs := osd.o inode.o file.o symlink.o
+osdfs-objs := osd.o inode.o file.o symlink.o namei.o dir.o
obj-$(CONFIG_OSDFS_FS) += osdfs.o
diff --git a/fs/osdfs/dir.c b/fs/osdfs/dir.c
new file mode 100644
index 0000000..ba28fd6
--- /dev/null
+++ b/fs/osdfs/dir.c
@@ -0,0 +1,629 @@
+/*
+ * Copyright (C) 2005, 2006
+ * Avishay Traeger (avishay@xxxxxxxxx) (avishay@xxxxxxxxxx)
+ * Copyright (C) 2005, 2006
+ * International Business Machines
+ *
+ * Copyrights for code taken from ext2:
+ * Copyright (C) 1992, 1993, 1994, 1995
+ * Remy Card (card@xxxxxxxxxxx)
+ * Laboratoire MASI - Institut Blaise Pascal
+ * Universite Pierre et Marie Curie (Paris VI)
+ * from
+ * linux/fs/minix/inode.c
+ * Copyright (C) 1991, 1992 Linus Torvalds
+ *
+ * This file is part of osdfs.
+ *
+ * osdfs 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. Since it is based on ext2, and the only
+ * valid version of GPL for the Linux kernel is version 2, the only valid
+ * version of GPL for osdfs is version 2.
+ *
+ * osdfs 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 osdfs; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include <linux/pagemap.h>
+#include <linux/smp_lock.h>
+#include "osdfs.h"
+
+static inline unsigned osdfs_chunk_size(struct inode *inode)
+{
+ return inode->i_sb->s_blocksize;
+}
+
+static inline void osdfs_put_page(struct page *page)
+{
+ kunmap(page);
+ page_cache_release(page);
+}
+
+static inline unsigned long dir_pages(struct inode *inode)
+{
+ return (inode->i_size+PAGE_CACHE_SIZE-1)>>PAGE_CACHE_SHIFT;
+}
+
+static unsigned osdfs_last_byte(struct inode *inode, unsigned long page_nr)
+{
+ unsigned last_byte = inode->i_size;
+
+ last_byte -= page_nr << PAGE_CACHE_SHIFT;
+ if (last_byte > PAGE_CACHE_SIZE)
+ last_byte = PAGE_CACHE_SIZE;
+ return last_byte;
+}
+
+static int osdfs_commit_chunk(struct page *page, unsigned from, unsigned to)
+{
+ struct inode *dir = page->mapping->host;
+ int err = 0;
+ dir->i_version++;
+ page->mapping->a_ops->commit_write(NULL, page, from, to);
+ if (IS_DIRSYNC(dir))
+ err = write_one_page(page, 1);
+ else
+ unlock_page(page);
+ return err;
+}
+
+static void osdfs_check_page(struct page *page)
+{
+ struct inode *dir = page->mapping->host;
+ unsigned chunk_size = osdfs_chunk_size(dir);
+ char *kaddr = page_address(page);
+ unsigned offs, rec_len;
+ unsigned limit = PAGE_CACHE_SIZE;
+ struct osdfs_dir_entry *p;
+ char *error;
+
+ /* if the page is the last one in the directory */
+ if ((dir->i_size >> PAGE_CACHE_SHIFT) == page->index) {
+ limit = dir->i_size & ~PAGE_CACHE_MASK;
+ if (limit & (chunk_size - 1))
+ goto Ebadsize;
+ if (!limit)
+ goto out;
+ }
+ for (offs = 0; offs <= limit - OSDFS_DIR_REC_LEN(1); offs += rec_len) {
+ p = (struct osdfs_dir_entry *)(kaddr + offs);
+ rec_len = p->rec_len;
+
+ if (rec_len < OSDFS_DIR_REC_LEN(1))
+ goto Eshort;
+ if (rec_len & 3)
+ goto Ealign;
+ if (rec_len < OSDFS_DIR_REC_LEN(p->name_len))
+ goto Enamelen;
+ if (((offs + rec_len - 1) ^ offs) & ~(chunk_size-1))
+ goto Espan;
+ }
+ if (offs != limit)
+ goto Eend;
+out:
+ SetPageChecked(page);
+ return;
+
+Ebadsize:
+ printk(KERN_ERR "ERROR [osdfs_check_page]: "
+ "size of directory #%lu is not a multiple of chunk size",
+ dir->i_ino
+ );
+ goto fail;
+Eshort:
+ error = "rec_len is smaller than minimal";
+ goto bad_entry;
+Ealign:
+ error = "unaligned directory entry";
+ goto bad_entry;
+Enamelen:
+ error = "rec_len is too small for name_len";
+ goto bad_entry;
+Espan:
+ error = "directory entry across blocks";
+ goto bad_entry;
+bad_entry:
+ printk(KERN_ERR
+ "ERROR [osdfs_check_page]: bad entry in directory #%lu: %s - "
+ "offset=%lu, inode=%lu, rec_len=%d, name_len=%d",
+ dir->i_ino, error, (page->index<<PAGE_CACHE_SHIFT)+offs,
+ (unsigned long) le32_to_cpu(p->inode),
+ rec_len, p->name_len);
+ goto fail;
+Eend:
+ p = (struct osdfs_dir_entry *)(kaddr + offs);
+ printk(KERN_ERR "ERROR [osdfs_check_page]: "
+ "entry in directory #%lu spans the page boundary"
+ "offset=%lu, inode=%lu",
+ dir->i_ino, (page->index<<PAGE_CACHE_SHIFT)+offs,
+ (unsigned long) le32_to_cpu(p->inode));
+fail:
+ SetPageChecked(page);
+ SetPageError(page);
+}
+
+static struct page *osdfs_get_page(struct inode *dir, unsigned long n)
+{
+ struct address_space *mapping = dir->i_mapping;
+ struct page *page = read_cache_page(mapping, n,
+ (filler_t *)mapping->a_ops->readpage, NULL);
+ if (!IS_ERR(page)) {
+ wait_on_page_locked(page);
+ kmap(page);
+ if (!PageUptodate(page))
+ goto fail;
+ if (!PageChecked(page))
+ osdfs_check_page(page);
+ if (PageError(page))
+ goto fail;
+ }
+ return page;
+
+fail:
+ osdfs_put_page(page);
+ return ERR_PTR(-EIO);
+}
+
+static inline int osdfs_match(int len, const unsigned char *name,
+ struct osdfs_dir_entry *de)
+{
+ if (len != de->name_len)
+ return 0;
+ if (!de->inode)
+ return 0;
+ return !memcmp(name, de->name, len);
+}
+
+static inline
+struct osdfs_dir_entry *osdfs_next_entry(struct osdfs_dir_entry *p)
+{
+ return (struct osdfs_dir_entry *)((char *)p + p->rec_len);
+}
+
+static inline unsigned
+osdfs_validate_entry(char *base, unsigned offset, unsigned mask)
+{
+ struct osdfs_dir_entry *de = (struct osdfs_dir_entry *)(base + offset);
+ struct osdfs_dir_entry *p =
+ (struct osdfs_dir_entry *)(base + (offset&mask));
+ while ((char *)p < (char *)de) {
+ if (p->rec_len == 0)
+ break;
+ p = osdfs_next_entry(p);
+ }
+ return (char *)p - base;
+}
+
+static unsigned char osdfs_filetype_table[OSDFS_FT_MAX] = {
+ [OSDFS_FT_UNKNOWN] = DT_UNKNOWN,
+ [OSDFS_FT_REG_FILE] = DT_REG,
+ [OSDFS_FT_DIR] = DT_DIR,
+ [OSDFS_FT_CHRDEV] = DT_CHR,
+ [OSDFS_FT_BLKDEV] = DT_BLK,
+ [OSDFS_FT_FIFO] = DT_FIFO,
+ [OSDFS_FT_SOCK] = DT_SOCK,
+ [OSDFS_FT_SYMLINK] = DT_LNK,
+};
+
+#define S_SHIFT 12
+static unsigned char osdfs_type_by_mode[S_IFMT >> S_SHIFT] = {
+ [S_IFREG >> S_SHIFT] = OSDFS_FT_REG_FILE,
+ [S_IFDIR >> S_SHIFT] = OSDFS_FT_DIR,
+ [S_IFCHR >> S_SHIFT] = OSDFS_FT_CHRDEV,
+ [S_IFBLK >> S_SHIFT] = OSDFS_FT_BLKDEV,
+ [S_IFIFO >> S_SHIFT] = OSDFS_FT_FIFO,
+ [S_IFSOCK >> S_SHIFT] = OSDFS_FT_SOCK,
+ [S_IFLNK >> S_SHIFT] = OSDFS_FT_SYMLINK,
+};
+
+static inline
+void osdfs_set_de_type(struct osdfs_dir_entry *de, struct inode *inode)
+{
+ mode_t mode = inode->i_mode;
+ de->file_type = osdfs_type_by_mode[(mode & S_IFMT)>>S_SHIFT];
+}
+
+static int
+osdfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
+{
+ loff_t pos = filp->f_pos;
+ struct inode *inode = filp->f_dentry->d_inode;
+ unsigned int offset = pos & ~PAGE_CACHE_MASK;
+ unsigned long n = pos >> PAGE_CACHE_SHIFT;
+ unsigned long npages = dir_pages(inode);
+ unsigned chunk_mask = ~(osdfs_chunk_size(inode)-1);
+ unsigned char *types = NULL;
+ int need_revalidate = (filp->f_version != inode->i_version);
+ int ret;
+
+ if (pos > inode->i_size - OSDFS_DIR_REC_LEN(1))
+ goto success;
+
+ types = osdfs_filetype_table;
+
+ for ( ; n < npages; n++, offset = 0) {
+ char *kaddr, *limit;
+ struct osdfs_dir_entry *de;
+ struct page *page = osdfs_get_page(inode, n);
+
+ if (IS_ERR(page)) {
+ printk(KERN_ERR "ERROR: "
+ "bad page in #%lu",
+ inode->i_ino);
+ filp->f_pos += PAGE_CACHE_SIZE - offset;
+ ret = -EIO;
+ goto done;
+ }
+ kaddr = page_address(page);
+ if (need_revalidate) {
+ offset = osdfs_validate_entry(kaddr, offset, chunk_mask);
+ need_revalidate = 0;
+ }
+ de = (struct osdfs_dir_entry *)(kaddr+offset);
+ limit = kaddr + osdfs_last_byte(inode, n) - OSDFS_DIR_REC_LEN(1);
+ for (; (char *)de <= limit; de = osdfs_next_entry(de)) {
+ if (de->rec_len == 0) {
+ printk(KERN_ERR "ERROR: "
+ "zero-length directory entry");
+ ret = -EIO;
+ osdfs_put_page(page);
+ goto done;
+ }
+ if (de->inode) {
+ int over;
+ unsigned char d_type = DT_UNKNOWN;
+
+ if (types && de->file_type < OSDFS_FT_MAX)
+ d_type = types[de->file_type];
+
+ offset = (char *)de - kaddr;
+ over = filldir(dirent, de->name, de->name_len,
+ (n<<PAGE_CACHE_SHIFT) | offset,
+ de->inode, d_type);
+ if (over) {
+ osdfs_put_page(page);
+ goto success;
+ }
+ }
+ filp->f_pos += de->rec_len;
+ }
+ osdfs_put_page(page);
+ }
+
+success:
+ ret = 0;
+done:
+ filp->f_version = inode->i_version;
+ return ret;
+}
+
+struct osdfs_dir_entry *osdfs_find_entry(struct inode *dir,
+ struct dentry *dentry, struct page **res_page)
+{
+ const unsigned char *name = dentry->d_name.name;
+ int namelen = dentry->d_name.len;
+ unsigned reclen = OSDFS_DIR_REC_LEN(namelen);
+ unsigned long start, n;
+ unsigned long npages = dir_pages(dir);
+ struct page *page = NULL;
+ struct osdfs_i_info *oi = OSDFS_I(dir);
+ struct osdfs_dir_entry *de;
+
+ if (npages == 0)
+ goto out;
+
+ *res_page = NULL;
+
+ start = oi->i_dir_start_lookup;
+ if (start >= npages)
+ start = 0;
+ n = start;
+ do {
+ char *kaddr;
+ page = osdfs_get_page(dir, n);
+ if (!IS_ERR(page)) {
+ kaddr = page_address(page);
+ de = (struct osdfs_dir_entry *) kaddr;
+ kaddr += osdfs_last_byte(dir, n) - reclen;
+ while ((char *) de <= kaddr) {
+ if (de->rec_len == 0) {
+ printk(KERN_ERR
+ "ERROR: osdfs_find_entry: "
+ "zero-length directory entry");
+ osdfs_put_page(page);
+ goto out;
+ }
+ if (osdfs_match(namelen, name, de))
+ goto found;
+ de = osdfs_next_entry(de);
+ }
+ osdfs_put_page(page);
+ }
+ if (++n >= npages)
+ n = 0;
+ } while (n != start);
+out:
+ return NULL;
+
+found:
+ *res_page = page;
+ oi->i_dir_start_lookup = n;
+ return de;
+}
+
+struct osdfs_dir_entry *osdfs_dotdot(struct inode *dir, struct page **p)
+{
+ struct page *page = osdfs_get_page(dir, 0);
+ struct osdfs_dir_entry *de = NULL;
+
+ if (!IS_ERR(page)) {
+ de = osdfs_next_entry(
+ (struct osdfs_dir_entry *)page_address(page));
+ *p = page;
+ }
+ return de;
+}
+
+ino_t osdfs_inode_by_name(struct inode *dir, struct dentry *dentry)
+{
+ ino_t res = 0;
+ struct osdfs_dir_entry *de;
+ struct page *page;
+
+ de = osdfs_find_entry(dir, dentry, &page);
+ if (de) {
+ res = de->inode;
+ kunmap(page);
+ page_cache_release(page);
+ }
+ return res;
+}
+
+void osdfs_set_link(struct inode *dir, struct osdfs_dir_entry *de,
+ struct page *page, struct inode *inode)
+{
+ unsigned from = (char *) de - (char *) page_address(page);
+ unsigned to = from + de->rec_len;
+ int err;
+
+ lock_page(page);
+ err = page->mapping->a_ops->prepare_write(NULL, page, from, to);
+ if (err)
+ BUG();
+ de->inode = inode->i_ino;
+ osdfs_set_de_type(de, inode);
+ err = osdfs_commit_chunk(page, from, to);
+ osdfs_put_page(page);
+ dir->i_mtime = dir->i_ctime = CURRENT_TIME;
+ mark_inode_dirty(dir);
+}
+
+int osdfs_add_link(struct dentry *dentry, struct inode *inode)
+{
+ struct inode *dir = dentry->d_parent->d_inode;
+ const unsigned char *name = dentry->d_name.name;
+ int namelen = dentry->d_name.len;
+ unsigned chunk_size = osdfs_chunk_size(dir);
+ unsigned reclen = OSDFS_DIR_REC_LEN(namelen);
+ unsigned short rec_len, name_len;
+ struct page *page = NULL;
+ struct osdfs_sb_info *sbi = inode->i_sb->s_fs_info;
+ struct osdfs_dir_entry *de;
+ unsigned long npages = dir_pages(dir);
+ unsigned long n;
+ char *kaddr;
+ unsigned from, to;
+ int err;
+
+ for (n = 0; n <= npages; n++) {
+ char *dir_end;
+
+ page = osdfs_get_page(dir, n);
+ err = PTR_ERR(page);
+ if (IS_ERR(page))
+ goto out;
+ lock_page(page);
+ kaddr = page_address(page);
+ dir_end = kaddr + osdfs_last_byte(dir, n);
+ de = (struct osdfs_dir_entry *)kaddr;
+ kaddr += PAGE_CACHE_SIZE - reclen;
+ while ((char *)de <= kaddr) {
+ if ((char *)de == dir_end) {
+ name_len = 0;
+ rec_len = chunk_size;
+ de->rec_len = chunk_size;
+ de->inode = 0;
+ goto got_it;
+ }
+ if (de->rec_len == 0) {
+ printk(KERN_ERR "ERROR: osdfs_add_link: "
+ "zero-length directory entry");
+ err = -EIO;
+ goto out_unlock;
+ }
+ err = -EEXIST;
+ if (osdfs_match(namelen, name, de))
+ goto out_unlock;
+ name_len = OSDFS_DIR_REC_LEN(de->name_len);
+ rec_len = de->rec_len;
+ if (!de->inode && rec_len >= reclen)
+ goto got_it;
+ if (rec_len >= name_len + reclen)
+ goto got_it;
+ de = (struct osdfs_dir_entry *) ((char *) de + rec_len);
+ }
+ unlock_page(page);
+ osdfs_put_page(page);
+ }
+ BUG();
+ return -EINVAL;
+
+got_it:
+ from = (char *)de - (char *)page_address(page);
+ to = from + rec_len;
+ err = page->mapping->a_ops->prepare_write(NULL, page, from, to);
+ if (err)
+ goto out_unlock;
+ if (de->inode) {
+ struct osdfs_dir_entry *de1 =
+ (struct osdfs_dir_entry *)((char *)de + name_len);
+ de1->rec_len = rec_len - name_len;
+ de->rec_len = name_len;
+ de = de1;
+ }
+ de->name_len = namelen;
+ memcpy(de->name, name, namelen);
+ de->inode = inode->i_ino;
+ osdfs_set_de_type(de, inode);
+ err = osdfs_commit_chunk(page, from, to);
+ dir->i_mtime = dir->i_ctime = CURRENT_TIME;
+ mark_inode_dirty(dir);
+ sbi->s_numfiles++;
+
+out_put:
+ osdfs_put_page(page);
+out:
+ return err;
+out_unlock:
+ unlock_page(page);
+ goto out_put;
+}
+
+int osdfs_delete_entry(struct osdfs_dir_entry *dir, struct page *page)
+{
+ struct address_space *mapping = page->mapping;
+ struct inode *inode = mapping->host;
+ struct osdfs_sb_info *sbi = inode->i_sb->s_fs_info;
+ char *kaddr = page_address(page);
+ unsigned from = ((char *)dir - kaddr) & ~(osdfs_chunk_size(inode)-1);
+ unsigned to = ((char *)dir - kaddr) + dir->rec_len;
+ struct osdfs_dir_entry *pde = NULL;
+ struct osdfs_dir_entry *de = (struct osdfs_dir_entry *) (kaddr + from);
+ int err;
+
+ while ((char *)de < (char *)dir) {
+ if (de->rec_len == 0) {
+ printk(KERN_ERR "ERROR: osdfs_delete_entry:"
+ "zero-length directory entry");
+ err = -EIO;
+ goto out;
+ }
+ pde = de;
+ de = osdfs_next_entry(de);
+ }
+ if (pde)
+ from = (char *)pde - (char *)page_address(page);
+ lock_page(page);
+ err = mapping->a_ops->prepare_write(NULL, page, from, to);
+ if (err)
+ BUG();
+ if (pde)
+ pde->rec_len = cpu_to_le16(to-from);
+ dir->inode = 0;
+ err = osdfs_commit_chunk(page, from, to);
+ inode->i_ctime = inode->i_mtime = CURRENT_TIME;
+ mark_inode_dirty(inode);
+ sbi->s_numfiles--;
+out:
+ osdfs_put_page(page);
+ return err;
+}
+
+int osdfs_make_empty(struct inode *inode, struct inode *parent)
+{
+ struct address_space *mapping = inode->i_mapping;
+ struct page *page = grab_cache_page(mapping, 0);
+ unsigned chunk_size = osdfs_chunk_size(inode);
+ struct osdfs_dir_entry *de;
+ int err;
+ void *kaddr;
+
+ if (!page)
+ return -ENOMEM;
+ err = mapping->a_ops->prepare_write(NULL, page, 0, chunk_size);
+ if (err) {
+ unlock_page(page);
+ goto fail;
+ }
+
+ kaddr = kmap_atomic(page, KM_USER0);
+ de = (struct osdfs_dir_entry *)kaddr;
+ de->name_len = 1;
+ de->rec_len = OSDFS_DIR_REC_LEN(1);
+ memcpy(de->name, ".\0\0", 4);
+ de->inode = inode->i_ino;
+ osdfs_set_de_type(de, inode);
+
+ de = (struct osdfs_dir_entry *)(kaddr + OSDFS_DIR_REC_LEN(1));
+ de->name_len = 2;
+ de->rec_len = chunk_size - OSDFS_DIR_REC_LEN(1);
+ de->inode = parent->i_ino;
+ memcpy(de->name, "..\0", 4);
+ osdfs_set_de_type(de, inode);
+ kunmap_atomic(page, KM_USER0);
+ err = osdfs_commit_chunk(page, 0, chunk_size);
+fail:
+ page_cache_release(page);
+ return err;
+}
+
+int osdfs_empty_dir(struct inode *inode)
+{
+ struct page *page = NULL;
+ unsigned long i, npages = dir_pages(inode);
+
+ for (i = 0; i < npages; i++) {
+ char *kaddr;
+ struct osdfs_dir_entry *de;
+ page = osdfs_get_page(inode, i);
+
+ if (IS_ERR(page))
+ continue;
+
+ kaddr = page_address(page);
+ de = (struct osdfs_dir_entry *)kaddr;
+ kaddr += osdfs_last_byte(inode, i) - OSDFS_DIR_REC_LEN(1);
+
+ while ((char *)de <= kaddr) {
+ if (de->rec_len == 0) {
+ printk(KERN_ERR "ERROR: osdfs_empty_dir: "
+ "zero-length directory entry");
+ printk("kaddr=%p, de=%p\n", kaddr, de);
+ goto not_empty;
+ }
+ if (de->inode != 0) {
+ /* check for . and .. */
+ if (de->name[0] != '.')
+ goto not_empty;
+ if (de->name_len > 2)
+ goto not_empty;
+ if (de->name_len < 2) {
+ if (de->inode !=
+ inode->i_ino)
+ goto not_empty;
+ } else if (de->name[1] != '.')
+ goto not_empty;
+ }
+ de = osdfs_next_entry(de);
+ }
+ osdfs_put_page(page);
+ }
+ return 1;
+
+not_empty:
+ osdfs_put_page(page);
+ return 0;
+}
+
+struct file_operations osdfs_dir_operations = {
+ .llseek = generic_file_llseek,
+ .read = generic_read_dir,
+ .readdir = osdfs_readdir,
+};
diff --git a/fs/osdfs/inode.c b/fs/osdfs/inode.c
index bfd82b1..478805e 100644
--- a/fs/osdfs/inode.c
+++ b/fs/osdfs/inode.c
@@ -408,6 +408,178 @@ fail:
}
/*
+ * Read an inode from the OSD, and return it as is. We also return the size
+ * attribute in the 'sanity' argument if we got compiled with debugging turned
+ * on.
+ */
+int osdfs_get_inode(struct super_block *sb, struct osdfs_i_info *oi,
+ struct osdfs_fcb *inode, uint64_t *sanity)
+{
+ struct osdfs_sb_info *sbi = sb->s_fs_info;
+ struct osd_request *req = NULL;
+ uint32_t page;
+ uint32_t attr;
+ uint16_t expected;
+ uint8_t *buf;
+ uint64_t o_id;
+ int ret;
+
+ o_id = oi->vfs_inode.i_ino + OSDFS_OBJ_OFF;
+
+ make_credential(oi->i_cred, sbi->s_pid, o_id);
+
+ req = prepare_osd_get_attr(sbi->s_dev, sbi->s_pid, o_id);
+ if (!req) {
+ printk(KERN_ERR "ERROR: prepare get_attr failed.\n");
+ return -ENOMEM;
+ }
+
+ /* we need the inode attribute */
+ prepare_get_attr_list_add_entry(req,
+ OSD_PAGE_NUM_IBM_UOBJ_FS_DATA,
+ OSD_ATTR_NUM_IBM_UOBJ_FS_DATA_INODE,
+ OSDFS_INO_ATTR_SIZE);
+
+#ifdef OSDFS_DEBUG
+ /* we get the size attributes to do a sanity check */
+ prepare_get_attr_list_add_entry(req,
+ OSD_APAGE_OBJECT_INFORMATION,
+ OSD_ATTR_OI_LOGICAL_LENGTH, 8);
+#endif
+
+ ret = osdfs_sync_op(req, sbi->s_timeout, oi->i_cred);
+ if (ret)
+ goto out;
+
+ page = OSD_PAGE_NUM_IBM_UOBJ_FS_DATA;
+ attr = OSD_ATTR_NUM_IBM_UOBJ_FS_DATA_INODE;
+ expected = OSDFS_INO_ATTR_SIZE;
+ ret = extract_next_attr_from_req(req, &page, &attr, &expected, &buf);
+ if (ret) {
+ printk(KERN_ERR "ERROR: extract attr from req failed\n");
+ goto out;
+ }
+ memcpy(inode, buf, sizeof(struct osdfs_fcb));
+
+#ifdef OSDFS_DEBUG
+ page = OSD_APAGE_OBJECT_INFORMATION;
+ attr = OSD_ATTR_OI_LOGICAL_LENGTH;
+ expected = 8;
+ ret = extract_next_attr_from_req(req, &page, &attr, &expected, &buf);
+ if (ret) {
+ printk(KERN_ERR "ERROR: extract attr from req failed\n");
+ goto out;
+ }
+ *sanity = be64_to_cpu(*((uint64_t *) buf));
+#endif
+
+out:
+ free_osd_req(req);
+ return ret;
+}
+
+/*
+ * Fill in an inode read from the OSD and set it up for use
+ */
+struct inode *osdfs_iget(struct super_block *sb, unsigned long ino)
+{
+ struct osdfs_i_info *oi;
+ struct osdfs_fcb fcb;
+ struct inode *inode;
+ uint64_t sanity;
+ int ret;
+ int n;
+
+ inode = iget_locked(sb, ino);
+ if (!inode)
+ return ERR_PTR(-ENOMEM);
+ if (!(inode->i_state & I_NEW))
+ return inode;
+ oi = OSDFS_I(inode);
+
+ /* read the inode from the osd */
+ ret = osdfs_get_inode(sb, oi, &fcb, &sanity);
+ if (ret)
+ goto bad_inode;
+
+ init_waitqueue_head(&oi->i_wq);
+ SetObjCreated(oi);
+
+ /* copy stuff from on-disk struct to in-memory struct */
+ inode->i_mode = be16_to_cpu(fcb.i_mode);
+ inode->i_uid = be32_to_cpu(fcb.i_uid);
+ inode->i_gid = be32_to_cpu(fcb.i_gid);
+ inode->i_nlink = be16_to_cpu(fcb.i_links_count);
+ inode->i_ctime.tv_sec = be32_to_cpu(fcb.i_ctime);
+ inode->i_atime.tv_sec = be32_to_cpu(fcb.i_atime);
+ inode->i_mtime.tv_sec = be32_to_cpu(fcb.i_mtime);
+ inode->i_atime.tv_nsec = inode->i_mtime.tv_nsec =
+ inode->i_ctime.tv_nsec = 0;
+ i_size_write(inode, be64_to_cpu(fcb.i_size));
+ inode->i_blkbits = OSDFS_BLKSHIFT;
+ inode->i_generation = be32_to_cpu(fcb.i_generation);
+
+#ifdef OSDFS_DEBUG
+ if ((inode->i_size != sanity) &&
+ (!osdfs_inode_is_fast_symlink(inode))) {
+ printk(KERN_WARNING
+ "WARNING: Size of object from inode and "
+ "attributes differ (%lld != %llu)\n",
+ inode->i_size, sanity);
+ }
+#endif
+
+ oi->i_objs = fcb.i_objs;
+ oi->i_dir_start_lookup = 0;
+
+ if ((inode->i_nlink == 0) && (inode->i_mode == 0)) {
+ ret = -ESTALE;
+ goto bad_inode;
+ }
+
+ if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode)) {
+ if (fcb.i_data[0])
+ inode->i_rdev = old_decode_dev(fcb.i_data[0]);
+ else
+ inode->i_rdev = new_decode_dev(fcb.i_data[1]);
+ } else
+ for (n = 0; n < OSDFS_IDATA; n++)
+ oi->i_data[n] = fcb.i_data[n];
+
+ if (S_ISREG(inode->i_mode)) {
+ inode->i_op = &osdfs_file_inode_operations;
+ inode->i_fop = &osdfs_file_operations;
+ inode->i_mapping->a_ops = &osdfs_aops;
+ } else if (S_ISDIR(inode->i_mode)) {
+ inode->i_op = &osdfs_dir_inode_operations;
+ inode->i_fop = &osdfs_dir_operations;
+ inode->i_mapping->a_ops = &osdfs_aops;
+ } else if (S_ISLNK(inode->i_mode)) {
+ if (osdfs_inode_is_fast_symlink(inode))
+ inode->i_op = &osdfs_fast_symlink_inode_operations;
+ else {
+ inode->i_op = &osdfs_symlink_inode_operations;
+ inode->i_mapping->a_ops = &osdfs_aops;
+ }
+ } else {
+ inode->i_op = &osdfs_special_inode_operations;
+ if (fcb.i_data[0])
+ init_special_inode(inode, inode->i_mode,
+ old_decode_dev(le32_to_cpu(fcb.i_data[0])));
+ else
+ init_special_inode(inode, inode->i_mode,
+ new_decode_dev(le32_to_cpu(fcb.i_data[1])));
+ }
+
+ unlock_new_inode(inode);
+ return inode;
+
+bad_inode:
+ iget_failed(inode);
+ return ERR_PTR(ret);
+}
+
+/*
* Set inode attributes - just call generic functions.
*/
int osdfs_setattr(struct dentry *dentry, struct iattr *iattr)
@@ -422,3 +594,98 @@ int osdfs_setattr(struct dentry *dentry, struct iattr *iattr)
error = inode_setattr(inode, iattr);
return error;
}
+
+/*
+ * Callback function from osdfs_new_inode(). The important thing is that we
+ * set the ObjCreated flag so that other methods know that the object exists on
+ * the OSD.
+ */
+void create_done(struct osd_request *req, void *p)
+{
+ struct inode *inode = (struct inode *)p;
+ struct osdfs_i_info *oi = OSDFS_I(inode);
+ struct osdfs_sb_info *sbi = inode->i_sb->s_fs_info;
+ int ret;
+
+ ret = check_ok(req);
+ free_osd_req(req);
+ atomic_dec(&sbi->s_curr_pending);
+
+ if (ret)
+ make_bad_inode(inode);
+ else
+ SetObjCreated(oi);
+
+ atomic_dec(&inode->i_count);
+}
+
+/*
+ * Set up a new inode and create an object for it on the OSD
+ */
+struct inode *osdfs_new_inode(struct inode *dir, int mode)
+{
+ struct super_block *sb;
+ struct inode *inode;
+ struct osdfs_i_info *oi;
+ struct osdfs_sb_info *sbi;
+ struct osd_request *req = NULL;
+ int ret;
+
+ sb = dir->i_sb;
+ inode = new_inode(sb);
+ if (!inode)
+ return ERR_PTR(-ENOMEM);
+
+ oi = OSDFS_I(inode);
+
+ init_waitqueue_head(&oi->i_wq);
+ SetObj2BCreated(oi);
+
+ sbi = sb->s_fs_info;
+
+ sb->s_dirt = 1;
+ inode->i_uid = current->fsuid;
+ if (dir->i_mode & S_ISGID) {
+ inode->i_gid = dir->i_gid;
+ if (S_ISDIR(mode))
+ mode |= S_ISGID;
+ } else
+ inode->i_gid = current->fsgid;
+ inode->i_mode = mode;
+
+ inode->i_ino = sbi->s_nextid++;
+ inode->i_blkbits = OSDFS_BLKSHIFT;
+ inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
+ inode->i_size = 0;
+ spin_lock(&sbi->s_next_gen_lock);
+ inode->i_generation = sbi->s_next_generation++;
+ spin_unlock(&sbi->s_next_gen_lock);
+ insert_inode_hash(inode);
+
+ mark_inode_dirty(inode);
+
+ req = prepare_osd_create(sbi->s_dev, sbi->s_pid,
+ inode->i_ino + OSDFS_OBJ_OFF);
+ if (!req) {
+ printk(KERN_ERR "ERROR: prepare_osd_create failed\n");
+ return ERR_PTR(-EIO);
+ }
+
+ make_credential(oi->i_cred, sbi->s_pid, inode->i_ino + OSDFS_OBJ_OFF);
+
+ /* increment the refcount so that the inode will still be around when we
+ * reach the callback
+ */
+ atomic_inc(&inode->i_count);
+
+ ret = osdfs_async_op(req, create_done, (void *)inode, oi->i_cred);
+ if (ret) {
+ atomic_dec(&inode->i_count);
+ free_osd_req(req);
+ return ERR_PTR(-EIO);
+ }
+ atomic_inc(&sbi->s_curr_pending);
+
+ return inode;
+}
+
diff --git a/fs/osdfs/namei.c b/fs/osdfs/namei.c
new file mode 100644
index 0000000..b747e90
--- /dev/null
+++ b/fs/osdfs/namei.c
@@ -0,0 +1,348 @@
+/*
+ * Copyright (C) 2005, 2006
+ * Avishay Traeger (avishay@xxxxxxxxx) (avishay@xxxxxxxxxx)
+ * Copyright (C) 2005, 2006
+ * International Business Machines
+ *
+ * Copyrights for code taken from ext2:
+ * Copyright (C) 1992, 1993, 1994, 1995
+ * Remy Card (card@xxxxxxxxxxx)
+ * Laboratoire MASI - Institut Blaise Pascal
+ * Universite Pierre et Marie Curie (Paris VI)
+ * from
+ * linux/fs/minix/inode.c
+ * Copyright (C) 1991, 1992 Linus Torvalds
+ *
+ * This file is part of osdfs.
+ *
+ * osdfs 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. Since it is based on ext2, and the only
+ * valid version of GPL for the Linux kernel is version 2, the only valid
+ * version of GPL for osdfs is version 2.
+ *
+ * osdfs 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 osdfs; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include "osdfs.h"
+
+static inline void osdfs_inc_count(struct inode *inode)
+{
+ inode->i_nlink++;
+ mark_inode_dirty(inode);
+}
+
+static inline void osdfs_dec_count(struct inode *inode)
+{
+ inode->i_nlink--;
+ mark_inode_dirty(inode);
+}
+
+static inline int osdfs_add_nondir(struct dentry *dentry, struct inode *inode)
+{
+ int err = osdfs_add_link(dentry, inode);
+ if (!err) {
+ d_instantiate(dentry, inode);
+ return 0;
+ }
+ osdfs_dec_count(inode);
+ iput(inode);
+ return err;
+}
+
+static struct dentry *osdfs_lookup(struct inode *dir, struct dentry *dentry,
+ struct nameidata *nd)
+{
+ struct inode *inode;
+ ino_t ino;
+
+ if (dentry->d_name.len > OSDFS_NAME_LEN)
+ return ERR_PTR(-ENAMETOOLONG);
+
+ ino = osdfs_inode_by_name(dir, dentry);
+ inode = NULL;
+ if (ino) {
+ inode = osdfs_iget(dir->i_sb, ino);
+ if (IS_ERR(inode))
+ return ERR_CAST(inode);
+ }
+ if (inode)
+ return d_splice_alias(inode, dentry);
+ d_add(dentry, inode);
+ return NULL;
+}
+
+static int osdfs_create(struct inode *dir, struct dentry *dentry, int mode,
+ struct nameidata *nd)
+{
+ struct inode *inode = osdfs_new_inode(dir, mode);
+ int err = PTR_ERR(inode);
+ if (!IS_ERR(inode)) {
+ inode->i_op = &osdfs_file_inode_operations;
+ inode->i_fop = &osdfs_file_operations;
+ inode->i_mapping->a_ops = &osdfs_aops;
+ mark_inode_dirty(inode);
+ err = osdfs_add_nondir(dentry, inode);
+ }
+ return err;
+}
+
+static int osdfs_mknod(struct inode *dir, struct dentry *dentry, int mode,
+ dev_t rdev)
+{
+ struct inode *inode;
+ int err;
+
+ if (!new_valid_dev(rdev))
+ return -EINVAL;
+
+ inode = osdfs_new_inode(dir, mode);
+ err = PTR_ERR(inode);
+ if (!IS_ERR(inode)) {
+ init_special_inode(inode, inode->i_mode, rdev);
+ mark_inode_dirty(inode);
+ err = osdfs_add_nondir(dentry, inode);
+ }
+ return err;
+}
+
+static int osdfs_symlink(struct inode *dir, struct dentry *dentry,
+ const char *symname)
+{
+ struct super_block *sb = dir->i_sb;
+ int err = -ENAMETOOLONG;
+ unsigned l = strlen(symname)+1;
+ struct inode *inode;
+
+ if (l > sb->s_blocksize)
+ goto out;
+
+ inode = osdfs_new_inode(dir, S_IFLNK | S_IRWXUGO);
+ err = PTR_ERR(inode);
+ if (IS_ERR(inode))
+ goto out;
+
+ if (l > sizeof(OSDFS_I(inode)->i_data)) {
+ /* slow symlink */
+ inode->i_op = &osdfs_symlink_inode_operations;
+ inode->i_mapping->a_ops = &osdfs_aops;
+ err = page_symlink(inode, symname, l);
+ memset((char *)(OSDFS_I(inode)->i_data), 0, OSDFS_IDATA);
+ if (err)
+ goto out_fail;
+ } else {
+ /* fast symlink */
+ inode->i_op = &osdfs_fast_symlink_inode_operations;
+ memcpy((char *)(OSDFS_I(inode)->i_data), symname, l);
+ inode->i_size = l-1;
+ }
+ mark_inode_dirty(inode);
+
+ err = osdfs_add_nondir(dentry, inode);
+out:
+ return err;
+
+out_fail:
+ osdfs_dec_count(inode);
+ iput(inode);
+ goto out;
+}
+
+static int osdfs_link(struct dentry *old_dentry, struct inode *dir,
+ struct dentry *dentry)
+{
+ struct inode *inode = old_dentry->d_inode;
+
+ if (inode->i_nlink >= OSDFS_LINK_MAX)
+ return -EMLINK;
+
+ inode->i_ctime = CURRENT_TIME;
+ osdfs_inc_count(inode);
+ atomic_inc(&inode->i_count);
+
+ return osdfs_add_nondir(dentry, inode);
+}
+
+static int osdfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
+{
+ struct inode *inode;
+ int err = -EMLINK;
+
+ if (dir->i_nlink >= OSDFS_LINK_MAX)
+ goto out;
+
+ osdfs_inc_count(dir);
+
+ inode = osdfs_new_inode(dir, S_IFDIR | mode);
+ err = PTR_ERR(inode);
+ if (IS_ERR(inode))
+ goto out_dir;
+
+ inode->i_op = &osdfs_dir_inode_operations;
+ inode->i_fop = &osdfs_dir_operations;
+ inode->i_mapping->a_ops = &osdfs_aops;
+
+ osdfs_inc_count(inode);
+
+ err = osdfs_make_empty(inode, dir);
+ if (err)
+ goto out_fail;
+
+ err = osdfs_add_link(dentry, inode);
+ if (err)
+ goto out_fail;
+
+ d_instantiate(dentry, inode);
+out:
+ return err;
+
+out_fail:
+ osdfs_dec_count(inode);
+ osdfs_dec_count(inode);
+ iput(inode);
+out_dir:
+ osdfs_dec_count(dir);
+ goto out;
+}
+
+static int osdfs_unlink(struct inode *dir, struct dentry *dentry)
+{
+ struct inode *inode = dentry->d_inode;
+ struct osdfs_dir_entry *de;
+ struct page *page;
+ int err = -ENOENT;
+
+ de = osdfs_find_entry(dir, dentry, &page);
+ if (!de)
+ goto out;
+
+ err = osdfs_delete_entry(de, page);
+ if (err)
+ goto out;
+
+ inode->i_ctime = dir->i_ctime;
+ osdfs_dec_count(inode);
+ err = 0;
+out:
+ return err;
+}
+
+static int osdfs_rmdir(struct inode *dir, struct dentry *dentry)
+{
+ struct inode *inode = dentry->d_inode;
+ int err = -ENOTEMPTY;
+
+ if (osdfs_empty_dir(inode)) {
+ err = osdfs_unlink(dir, dentry);
+ if (!err) {
+ inode->i_size = 0;
+ osdfs_dec_count(inode);
+ osdfs_dec_count(dir);
+ }
+ }
+ return err;
+}
+
+static int osdfs_rename(struct inode *old_dir, struct dentry *old_dentry,
+ struct inode *new_dir, struct dentry *new_dentry)
+{
+ struct inode *old_inode = old_dentry->d_inode;
+ struct inode *new_inode = new_dentry->d_inode;
+ struct page *dir_page = NULL;
+ struct osdfs_dir_entry *dir_de = NULL;
+ struct page *old_page;
+ struct osdfs_dir_entry *old_de;
+ int err = -ENOENT;
+
+ old_de = osdfs_find_entry(old_dir, old_dentry, &old_page);
+ if (!old_de)
+ goto out;
+
+ if (S_ISDIR(old_inode->i_mode)) {
+ err = -EIO;
+ dir_de = osdfs_dotdot(old_inode, &dir_page);
+ if (!dir_de)
+ goto out_old;
+ }
+
+ if (new_inode) {
+ struct page *new_page;
+ struct osdfs_dir_entry *new_de;
+
+ err = -ENOTEMPTY;
+ if (dir_de && !osdfs_empty_dir(new_inode))
+ goto out_dir;
+
+ err = -ENOENT;
+ new_de = osdfs_find_entry(new_dir, new_dentry, &new_page);
+ if (!new_de)
+ goto out_dir;
+ osdfs_inc_count(old_inode);
+ osdfs_set_link(new_dir, new_de, new_page, old_inode);
+ new_inode->i_ctime = CURRENT_TIME;
+ if (dir_de)
+ new_inode->i_nlink--;
+ osdfs_dec_count(new_inode);
+ } else {
+ if (dir_de) {
+ err = -EMLINK;
+ if (new_dir->i_nlink >= OSDFS_LINK_MAX)
+ goto out_dir;
+ }
+ osdfs_inc_count(old_inode);
+ err = osdfs_add_link(new_dentry, old_inode);
+ if (err) {
+ osdfs_dec_count(old_inode);
+ goto out_dir;
+ }
+ if (dir_de)
+ osdfs_inc_count(new_dir);
+ }
+
+ old_inode->i_ctime = CURRENT_TIME;
+
+ osdfs_delete_entry(old_de, old_page);
+ osdfs_dec_count(old_inode);
+
+ if (dir_de) {
+ osdfs_set_link(old_inode, dir_de, dir_page, new_dir);
+ osdfs_dec_count(old_dir);
+ }
+ return 0;
+
+
+out_dir:
+ if (dir_de) {
+ kunmap(dir_page);
+ page_cache_release(dir_page);
+ }
+out_old:
+ kunmap(old_page);
+ page_cache_release(old_page);
+out:
+ return err;
+}
+
+struct inode_operations osdfs_dir_inode_operations = {
+ .create = osdfs_create,
+ .lookup = osdfs_lookup,
+ .link = osdfs_link,
+ .unlink = osdfs_unlink,
+ .symlink = osdfs_symlink,
+ .mkdir = osdfs_mkdir,
+ .rmdir = osdfs_rmdir,
+ .mknod = osdfs_mknod,
+ .rename = osdfs_rename,
+ .setattr = osdfs_setattr,
+};
+
+struct inode_operations osdfs_special_inode_operations = {
+ .setattr = osdfs_setattr,
+};
diff --git a/fs/osdfs/osdfs.h b/fs/osdfs/osdfs.h
index 29e7d7b..00c89f7 100644
--- a/fs/osdfs/osdfs.h
+++ b/fs/osdfs/osdfs.h
@@ -106,6 +106,11 @@ static inline struct osdfs_i_info *OSDFS_I(struct inode *inode)
return container_of(inode, struct osdfs_i_info, vfs_inode);
}
+/*
+ * Maximum count of links to a file
+ */
+#define OSDFS_LINK_MAX 32000
+
/*************************
* function declarations *
*************************/
@@ -179,11 +184,28 @@ void free_osd_req(struct osd_request *req);
/* inode.c */
void osdfs_truncate(struct inode *inode);
+extern struct inode *osdfs_iget(struct super_block *, unsigned long);
+struct inode *osdfs_new_inode(struct inode *, int);
int osdfs_setattr(struct dentry *, struct iattr *);
+/* dir.c: */
+int osdfs_add_link(struct dentry *, struct inode *);
+ino_t osdfs_inode_by_name(struct inode *, struct dentry *);
+int osdfs_delete_entry(struct osdfs_dir_entry *, struct page *);
+int osdfs_make_empty(struct inode *, struct inode *);
+struct osdfs_dir_entry *osdfs_find_entry(struct inode *, struct dentry *,
+ struct page **);
+int osdfs_empty_dir(struct inode *);
+struct osdfs_dir_entry *osdfs_dotdot(struct inode *, struct page **);
+void osdfs_set_link(struct inode *, struct osdfs_dir_entry *, struct page *,
+ struct inode *);
+
/*********************
* operation vectors *
*********************/
+/* dir.c: */
+extern struct file_operations osdfs_dir_operations;
+
/* file.c */
extern struct inode_operations osdfs_file_inode_operations;
extern struct file_operations osdfs_file_operations;
@@ -191,6 +213,10 @@ extern struct file_operations osdfs_file_operations;
/* inode.c */
extern struct address_space_operations osdfs_aops;
+/* namei.c */
+extern struct inode_operations osdfs_dir_inode_operations;
+extern struct inode_operations osdfs_special_inode_operations;
+
/* symlink.c */
extern struct inode_operations osdfs_symlink_inode_operations;
extern struct inode_operations osdfs_fast_symlink_inode_operations;
--
1.6.0.1
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
