iget: stop JFS from using iget() and read_inode()
[linux-2.6.git] / fs / jfs / super.c
index 810a365..50ea654 100644 (file)
@@ -4,27 +4,32 @@
  *
  *   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 
+ *   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 
+ *   along with this program;  if not, write to the Free Software
  *   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
  */
 
 #include <linux/fs.h>
-#include <linux/config.h>
 #include <linux/module.h>
 #include <linux/parser.h>
 #include <linux/completion.h>
 #include <linux/vfs.h>
+#include <linux/mount.h>
 #include <linux/moduleparam.h>
+#include <linux/kthread.h>
+#include <linux/posix_acl.h>
+#include <linux/buffer_head.h>
+#include <linux/exportfs.h>
 #include <asm/uaccess.h>
+#include <linux/seq_file.h>
 
 #include "jfs_incore.h"
 #include "jfs_filsys.h"
@@ -40,10 +45,10 @@ MODULE_DESCRIPTION("The Journaled Filesystem (JFS)");
 MODULE_AUTHOR("Steve Best/Dave Kleikamp/Barry Arndt, IBM");
 MODULE_LICENSE("GPL");
 
-static kmem_cache_t * jfs_inode_cachep;
+static struct kmem_cache * jfs_inode_cachep;
 
-static struct super_operations jfs_super_operations;
-static struct export_operations jfs_export_operations;
+static const struct super_operations jfs_super_operations;
+static const struct export_operations jfs_export_operations;
 static struct file_system_type jfs_fs_type;
 
 #define MAX_COMMIT_THREADS 64
@@ -51,11 +56,9 @@ static int commit_threads = 0;
 module_param(commit_threads, int, 0);
 MODULE_PARM_DESC(commit_threads, "Number of commit threads");
 
-int jfs_stop_threads;
-static pid_t jfsIOthread;
-static pid_t jfsCommitThread[MAX_COMMIT_THREADS];
-static pid_t jfsSyncThread;
-DECLARE_COMPLETION(jfsIOwait);
+static struct task_struct *jfsCommitThread[MAX_COMMIT_THREADS];
+struct task_struct *jfsIOthread;
+struct task_struct *jfsSyncThread;
 
 #ifdef CONFIG_JFS_DEBUG
 int jfsloglevel = JFS_LOGLEVEL_WARN;
@@ -80,7 +83,7 @@ static void jfs_handle_error(struct super_block *sb)
                        "as read-only\n",
                        sb->s_id);
                sb->s_flags |= MS_RDONLY;
-       } 
+       }
 
        /* nothing is done for continue beyond marking the superblock dirty */
 }
@@ -91,7 +94,7 @@ void jfs_error(struct super_block *sb, const char * function, ...)
        va_list args;
 
        va_start(args, function);
-       vsprintf(error_buf, function, args);
+       vsnprintf(error_buf, sizeof(error_buf), function, args);
        va_end(args);
 
        printk(KERN_ERR "ERROR: (device %s): %s\n", sb->s_id, error_buf);
@@ -113,6 +116,8 @@ static void jfs_destroy_inode(struct inode *inode)
 {
        struct jfs_inode_info *ji = JFS_IP(inode);
 
+       BUG_ON(!list_empty(&ji->anon_inode_list));
+
        spin_lock_irq(&ji->ag_lock);
        if (ji->active_ag != -1) {
                struct bmap *bmap = JFS_SBI(inode->i_sb)->bmap;
@@ -135,9 +140,9 @@ static void jfs_destroy_inode(struct inode *inode)
        kmem_cache_free(jfs_inode_cachep, ji);
 }
 
-static int jfs_statfs(struct super_block *sb, struct kstatfs *buf)
+static int jfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
-       struct jfs_sb_info *sbi = JFS_SBI(sb);
+       struct jfs_sb_info *sbi = JFS_SBI(dentry->d_sb);
        s64 maxinodes;
        struct inomap *imap = JFS_IP(sbi->ipimap)->i_imap;
 
@@ -189,7 +194,8 @@ static void jfs_put_super(struct super_block *sb)
 
 enum {
        Opt_integrity, Opt_nointegrity, Opt_iocharset, Opt_resize,
-       Opt_resize_nosize, Opt_errors, Opt_ignore, Opt_err,
+       Opt_resize_nosize, Opt_errors, Opt_ignore, Opt_err, Opt_quota,
+       Opt_usrquota, Opt_grpquota, Opt_uid, Opt_gid, Opt_umask
 };
 
 static match_table_t tokens = {
@@ -201,8 +207,11 @@ static match_table_t tokens = {
        {Opt_errors, "errors=%s"},
        {Opt_ignore, "noquota"},
        {Opt_ignore, "quota"},
-       {Opt_ignore, "usrquota"},
-       {Opt_ignore, "grpquota"},
+       {Opt_usrquota, "usrquota"},
+       {Opt_grpquota, "grpquota"},
+       {Opt_uid, "uid=%u"},
+       {Opt_gid, "gid=%u"},
+       {Opt_umask, "umask=%u"},
        {Opt_err, NULL}
 };
 
@@ -290,6 +299,46 @@ static int parse_options(char *options, struct super_block *sb, s64 *newLVSize,
                        }
                        break;
                }
+
+#ifdef CONFIG_QUOTA
+               case Opt_quota:
+               case Opt_usrquota:
+                       *flag |= JFS_USRQUOTA;
+                       break;
+               case Opt_grpquota:
+                       *flag |= JFS_GRPQUOTA;
+                       break;
+#else
+               case Opt_usrquota:
+               case Opt_grpquota:
+               case Opt_quota:
+                       printk(KERN_ERR
+                              "JFS: quota operations not supported\n");
+                       break;
+#endif
+               case Opt_uid:
+               {
+                       char *uid = args[0].from;
+                       sbi->uid = simple_strtoul(uid, &uid, 0);
+                       break;
+               }
+               case Opt_gid:
+               {
+                       char *gid = args[0].from;
+                       sbi->gid = simple_strtoul(gid, &gid, 0);
+                       break;
+               }
+               case Opt_umask:
+               {
+                       char *umask = args[0].from;
+                       sbi->umask = simple_strtoul(umask, &umask, 8);
+                       if (sbi->umask & ~0777) {
+                               printk(KERN_ERR
+                                      "JFS: Invalid value of umask\n");
+                               goto cleanup;
+                       }
+                       break;
+               }
                default:
                        printk("jfs: Unrecognized mount option \"%s\" "
                                        " or missing value\n", p);
@@ -365,19 +414,19 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent)
        struct inode *inode;
        int rc;
        s64 newLVSize = 0;
-       int flag;
+       int flag, ret = -EINVAL;
 
        jfs_info("In jfs_read_super: s_flags=0x%lx", sb->s_flags);
 
        if (!new_valid_dev(sb->s_bdev->bd_dev))
                return -EOVERFLOW;
 
-       sbi = kmalloc(sizeof (struct jfs_sb_info), GFP_KERNEL);
+       sbi = kzalloc(sizeof (struct jfs_sb_info), GFP_KERNEL);
        if (!sbi)
-               return -ENOSPC;
-       memset(sbi, 0, sizeof (struct jfs_sb_info));
+               return -ENOMEM;
        sb->s_fs_info = sbi;
        sbi->sb = sb;
+       sbi->uid = sbi->gid = sbi->umask = -1;
 
        /* initialize the mount flag and determine the default error handler */
        flag = JFS_ERR_REMOUNT_RO;
@@ -412,12 +461,15 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent)
         * Initialize direct-mapping inode/address-space
         */
        inode = new_inode(sb);
-       if (inode == NULL)
+       if (inode == NULL) {
+               ret = -ENOMEM;
                goto out_kfree;
+       }
        inode->i_ino = 0;
        inode->i_nlink = 1;
        inode->i_size = sb->s_bdev->bd_inode->i_size;
        inode->i_mapping->a_ops = &jfs_metapage_aops;
+       insert_inode_hash(inode);
        mapping_set_gfp_mask(inode->i_mapping, GFP_NOFS);
 
        sbi->direct_inode = inode;
@@ -444,9 +496,11 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent)
 
        sb->s_magic = JFS_SUPER_MAGIC;
 
-       inode = iget(sb, ROOT_I);
-       if (!inode || is_bad_inode(inode))
+       inode = jfs_iget(sb, ROOT_I);
+       if (IS_ERR(inode)) {
+               ret = PTR_ERR(inode);
                goto out_no_root;
+       }
        sb->s_root = d_alloc_root(inode);
        if (!sb->s_root)
                goto out_no_root;
@@ -477,8 +531,7 @@ out_no_rw:
                jfs_err("jfs_umount failed with return code %d", rc);
        }
 out_mount_failed:
-       filemap_fdatawrite(sbi->direct_inode->i_mapping);
-       filemap_fdatawait(sbi->direct_inode->i_mapping);
+       filemap_write_and_wait(sbi->direct_inode->i_mapping);
        truncate_inode_pages(sbi->direct_inode->i_mapping, 0);
        make_bad_inode(sbi->direct_inode);
        iput(sbi->direct_inode);
@@ -487,7 +540,7 @@ out_kfree:
        if (sbi->nls_tab)
                unload_nls(sbi->nls_tab);
        kfree(sbi);
-       return -EINVAL;
+       return ret;
 }
 
 static void jfs_write_super_lockfs(struct super_block *sb)
@@ -517,10 +570,11 @@ static void jfs_unlockfs(struct super_block *sb)
        }
 }
 
-static struct super_block *jfs_get_sb(struct file_system_type *fs_type, 
-       int flags, const char *dev_name, void *data)
+static int jfs_get_sb(struct file_system_type *fs_type,
+       int flags, const char *dev_name, void *data, struct vfsmount *mnt)
 {
-       return get_sb_bdev(fs_type, flags, dev_name, data, jfs_fill_super);
+       return get_sb_bdev(fs_type, flags, dev_name, data, jfs_fill_super,
+                          mnt);
 }
 
 static int jfs_sync_fs(struct super_block *sb, int wait)
@@ -530,16 +584,152 @@ static int jfs_sync_fs(struct super_block *sb, int wait)
        /* log == NULL indicates read-only mount */
        if (log) {
                jfs_flush_journal(log, wait);
-               jfs_syncpt(log);
+               jfs_syncpt(log, 0);
        }
 
        return 0;
 }
 
-static struct super_operations jfs_super_operations = {
+static int jfs_show_options(struct seq_file *seq, struct vfsmount *vfs)
+{
+       struct jfs_sb_info *sbi = JFS_SBI(vfs->mnt_sb);
+
+       if (sbi->uid != -1)
+               seq_printf(seq, ",uid=%d", sbi->uid);
+       if (sbi->gid != -1)
+               seq_printf(seq, ",gid=%d", sbi->gid);
+       if (sbi->umask != -1)
+               seq_printf(seq, ",umask=%03o", sbi->umask);
+       if (sbi->flag & JFS_NOINTEGRITY)
+               seq_puts(seq, ",nointegrity");
+       if (sbi->nls_tab)
+               seq_printf(seq, ",iocharset=%s", sbi->nls_tab->charset);
+       if (sbi->flag & JFS_ERR_CONTINUE)
+               seq_printf(seq, ",errors=continue");
+       if (sbi->flag & JFS_ERR_PANIC)
+               seq_printf(seq, ",errors=panic");
+
+#ifdef CONFIG_QUOTA
+       if (sbi->flag & JFS_USRQUOTA)
+               seq_puts(seq, ",usrquota");
+
+       if (sbi->flag & JFS_GRPQUOTA)
+               seq_puts(seq, ",grpquota");
+#endif
+
+       return 0;
+}
+
+#ifdef CONFIG_QUOTA
+
+/* Read data from quotafile - avoid pagecache and such because we cannot afford
+ * acquiring the locks... As quota files are never truncated and quota code
+ * itself serializes the operations (and noone else should touch the files)
+ * we don't have to be afraid of races */
+static ssize_t jfs_quota_read(struct super_block *sb, int type, char *data,
+                             size_t len, loff_t off)
+{
+       struct inode *inode = sb_dqopt(sb)->files[type];
+       sector_t blk = off >> sb->s_blocksize_bits;
+       int err = 0;
+       int offset = off & (sb->s_blocksize - 1);
+       int tocopy;
+       size_t toread;
+       struct buffer_head tmp_bh;
+       struct buffer_head *bh;
+       loff_t i_size = i_size_read(inode);
+
+       if (off > i_size)
+               return 0;
+       if (off+len > i_size)
+               len = i_size-off;
+       toread = len;
+       while (toread > 0) {
+               tocopy = sb->s_blocksize - offset < toread ?
+                               sb->s_blocksize - offset : toread;
+
+               tmp_bh.b_state = 0;
+               tmp_bh.b_size = 1 << inode->i_blkbits;
+               err = jfs_get_block(inode, blk, &tmp_bh, 0);
+               if (err)
+                       return err;
+               if (!buffer_mapped(&tmp_bh))    /* A hole? */
+                       memset(data, 0, tocopy);
+               else {
+                       bh = sb_bread(sb, tmp_bh.b_blocknr);
+                       if (!bh)
+                               return -EIO;
+                       memcpy(data, bh->b_data+offset, tocopy);
+                       brelse(bh);
+               }
+               offset = 0;
+               toread -= tocopy;
+               data += tocopy;
+               blk++;
+       }
+       return len;
+}
+
+/* Write to quotafile */
+static ssize_t jfs_quota_write(struct super_block *sb, int type,
+                              const char *data, size_t len, loff_t off)
+{
+       struct inode *inode = sb_dqopt(sb)->files[type];
+       sector_t blk = off >> sb->s_blocksize_bits;
+       int err = 0;
+       int offset = off & (sb->s_blocksize - 1);
+       int tocopy;
+       size_t towrite = len;
+       struct buffer_head tmp_bh;
+       struct buffer_head *bh;
+
+       mutex_lock(&inode->i_mutex);
+       while (towrite > 0) {
+               tocopy = sb->s_blocksize - offset < towrite ?
+                               sb->s_blocksize - offset : towrite;
+
+               tmp_bh.b_state = 0;
+               tmp_bh.b_size = 1 << inode->i_blkbits;
+               err = jfs_get_block(inode, blk, &tmp_bh, 1);
+               if (err)
+                       goto out;
+               if (offset || tocopy != sb->s_blocksize)
+                       bh = sb_bread(sb, tmp_bh.b_blocknr);
+               else
+                       bh = sb_getblk(sb, tmp_bh.b_blocknr);
+               if (!bh) {
+                       err = -EIO;
+                       goto out;
+               }
+               lock_buffer(bh);
+               memcpy(bh->b_data+offset, data, tocopy);
+               flush_dcache_page(bh->b_page);
+               set_buffer_uptodate(bh);
+               mark_buffer_dirty(bh);
+               unlock_buffer(bh);
+               brelse(bh);
+               offset = 0;
+               towrite -= tocopy;
+               data += tocopy;
+               blk++;
+       }
+out:
+       if (len == towrite)
+               return err;
+       if (inode->i_size < off+len-towrite)
+               i_size_write(inode, off+len-towrite);
+       inode->i_version++;
+       inode->i_mtime = inode->i_ctime = CURRENT_TIME;
+       mark_inode_dirty(inode);
+       mutex_unlock(&inode->i_mutex);
+       return len - towrite;
+}
+
+#endif
+
+static const struct super_operations jfs_super_operations = {
        .alloc_inode    = jfs_alloc_inode,
        .destroy_inode  = jfs_destroy_inode,
-       .read_inode     = jfs_read_inode,
        .dirty_inode    = jfs_dirty_inode,
        .write_inode    = jfs_write_inode,
        .delete_inode   = jfs_delete_inode,
@@ -549,9 +739,16 @@ static struct super_operations jfs_super_operations = {
        .unlockfs       = jfs_unlockfs,
        .statfs         = jfs_statfs,
        .remount_fs     = jfs_remount,
+       .show_options   = jfs_show_options,
+#ifdef CONFIG_QUOTA
+       .quota_read     = jfs_quota_read,
+       .quota_write    = jfs_quota_write,
+#endif
 };
 
-static struct export_operations jfs_export_operations = {
+static const struct export_operations jfs_export_operations = {
+       .fh_to_dentry   = jfs_fh_to_dentry,
+       .fh_to_parent   = jfs_fh_to_parent,
        .get_parent     = jfs_get_parent,
 };
 
@@ -563,25 +760,22 @@ static struct file_system_type jfs_fs_type = {
        .fs_flags       = FS_REQUIRES_DEV,
 };
 
-static void init_once(void *foo, kmem_cache_t * cachep, unsigned long flags)
+static void init_once(struct kmem_cache *cachep, void *foo)
 {
        struct jfs_inode_info *jfs_ip = (struct jfs_inode_info *) foo;
 
-       if ((flags & (SLAB_CTOR_VERIFY | SLAB_CTOR_CONSTRUCTOR)) ==
-           SLAB_CTOR_CONSTRUCTOR) {
-               memset(jfs_ip, 0, sizeof(struct jfs_inode_info));
-               INIT_LIST_HEAD(&jfs_ip->anon_inode_list);
-               init_rwsem(&jfs_ip->rdwrlock);
-               init_MUTEX(&jfs_ip->commit_sem);
-               init_rwsem(&jfs_ip->xattr_sem);
-               spin_lock_init(&jfs_ip->ag_lock);
-               jfs_ip->active_ag = -1;
+       memset(jfs_ip, 0, sizeof(struct jfs_inode_info));
+       INIT_LIST_HEAD(&jfs_ip->anon_inode_list);
+       init_rwsem(&jfs_ip->rdwrlock);
+       mutex_init(&jfs_ip->commit_mutex);
+       init_rwsem(&jfs_ip->xattr_sem);
+       spin_lock_init(&jfs_ip->ag_lock);
+       jfs_ip->active_ag = -1;
 #ifdef CONFIG_JFS_POSIX_ACL
-               jfs_ip->i_acl = JFS_ACL_NOT_CACHED;
-               jfs_ip->i_default_acl = JFS_ACL_NOT_CACHED;
+       jfs_ip->i_acl = JFS_ACL_NOT_CACHED;
+       jfs_ip->i_default_acl = JFS_ACL_NOT_CACHED;
 #endif
-               inode_init_once(&jfs_ip->vfs_inode);
-       }
+       inode_init_once(&jfs_ip->vfs_inode);
 }
 
 static int __init init_jfs_fs(void)
@@ -590,8 +784,9 @@ static int __init init_jfs_fs(void)
        int rc;
 
        jfs_inode_cachep =
-           kmem_cache_create("jfs_ip", sizeof(struct jfs_inode_info), 0, 
-                           SLAB_RECLAIM_ACCOUNT, init_once, NULL);
+           kmem_cache_create("jfs_ip", sizeof(struct jfs_inode_info), 0,
+                           SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD,
+                           init_once);
        if (jfs_inode_cachep == NULL)
                return -ENOMEM;
 
@@ -616,12 +811,12 @@ static int __init init_jfs_fs(void)
        /*
         * I/O completion thread (endio)
         */
-       jfsIOthread = kernel_thread(jfsIOWait, NULL, CLONE_KERNEL);
-       if (jfsIOthread < 0) {
-               jfs_err("init_jfs_fs: fork failed w/rc = %d", jfsIOthread);
+       jfsIOthread = kthread_run(jfsIOWait, NULL, "jfsIO");
+       if (IS_ERR(jfsIOthread)) {
+               rc = PTR_ERR(jfsIOthread);
+               jfs_err("init_jfs_fs: fork failed w/rc = %d", rc);
                goto end_txmngr;
        }
-       wait_for_completion(&jfsIOwait);        /* Wait until thread starts */
 
        if (commit_threads < 1)
                commit_threads = num_online_cpus();
@@ -629,24 +824,21 @@ static int __init init_jfs_fs(void)
                commit_threads = MAX_COMMIT_THREADS;
 
        for (i = 0; i < commit_threads; i++) {
-               jfsCommitThread[i] = kernel_thread(jfs_lazycommit, NULL,
-                                                  CLONE_KERNEL);
-               if (jfsCommitThread[i] < 0) {
-                       jfs_err("init_jfs_fs: fork failed w/rc = %d",
-                               jfsCommitThread[i]);
+               jfsCommitThread[i] = kthread_run(jfs_lazycommit, NULL, "jfsCommit");
+               if (IS_ERR(jfsCommitThread[i])) {
+                       rc = PTR_ERR(jfsCommitThread[i]);
+                       jfs_err("init_jfs_fs: fork failed w/rc = %d", rc);
                        commit_threads = i;
                        goto kill_committask;
                }
-               /* Wait until thread starts */
-               wait_for_completion(&jfsIOwait);
        }
 
-       jfsSyncThread = kernel_thread(jfs_sync, NULL, CLONE_KERNEL);
-       if (jfsSyncThread < 0) {
-               jfs_err("init_jfs_fs: fork failed w/rc = %d", jfsSyncThread);
+       jfsSyncThread = kthread_run(jfs_sync, NULL, "jfsSync");
+       if (IS_ERR(jfsSyncThread)) {
+               rc = PTR_ERR(jfsSyncThread);
+               jfs_err("init_jfs_fs: fork failed w/rc = %d", rc);
                goto kill_committask;
        }
-       wait_for_completion(&jfsIOwait);        /* Wait until thread starts */
 
 #ifdef PROC_FS_JFS
        jfs_proc_init();
@@ -655,13 +847,9 @@ static int __init init_jfs_fs(void)
        return register_filesystem(&jfs_fs_type);
 
 kill_committask:
-       jfs_stop_threads = 1;
-       wake_up_all(&jfs_commit_thread_wait);
        for (i = 0; i < commit_threads; i++)
-               wait_for_completion(&jfsIOwait);
-
-       wake_up(&jfs_IO_thread_wait);
-       wait_for_completion(&jfsIOwait);        /* Wait for thread exit */
+               kthread_stop(jfsCommitThread[i]);
+       kthread_stop(jfsIOthread);
 end_txmngr:
        txExit();
 free_metapage:
@@ -677,16 +865,13 @@ static void __exit exit_jfs_fs(void)
 
        jfs_info("exit_jfs_fs called");
 
-       jfs_stop_threads = 1;
        txExit();
        metapage_exit();
-       wake_up(&jfs_IO_thread_wait);
-       wait_for_completion(&jfsIOwait);        /* Wait until IO thread exits */
-       wake_up_all(&jfs_commit_thread_wait);
+
+       kthread_stop(jfsIOthread);
        for (i = 0; i < commit_threads; i++)
-               wait_for_completion(&jfsIOwait);
-       wake_up(&jfs_sync_thread_wait);
-       wait_for_completion(&jfsIOwait);        /* Wait until Sync thread exits */
+               kthread_stop(jfsCommitThread[i]);
+       kthread_stop(jfsSyncThread);
 #ifdef PROC_FS_JFS
        jfs_proc_clean();
 #endif