#include <linux/slab.h>
#include <linux/blkdev.h>
#include <linux/backing-dev.h>
+#include <linux/crc32.h>
#include "nilfs.h"
#include "segment.h"
#include "alloc.h"
#include "cpfile.h"
#include "sufile.h"
#include "dat.h"
-#include "seglist.h"
#include "segbuf.h"
+
+static LIST_HEAD(nilfs_objects);
+static DEFINE_SPINLOCK(nilfs_lock);
+
void nilfs_set_last_segment(struct the_nilfs *nilfs,
sector_t start_blocknr, u64 seq, __u64 cno)
{
* Return Value: On success, pointer to the_nilfs is returned.
* On error, NULL is returned.
*/
-struct the_nilfs *alloc_nilfs(struct block_device *bdev)
+static struct the_nilfs *alloc_nilfs(struct block_device *bdev)
{
struct the_nilfs *nilfs;
nilfs->ns_bdev = bdev;
atomic_set(&nilfs->ns_count, 1);
- atomic_set(&nilfs->ns_writer_refcount, -1);
atomic_set(&nilfs->ns_ndirtyblks, 0);
init_rwsem(&nilfs->ns_sem);
- mutex_init(&nilfs->ns_writer_mutex);
+ init_rwsem(&nilfs->ns_super_sem);
+ mutex_init(&nilfs->ns_mount_mutex);
+ init_rwsem(&nilfs->ns_writer_sem);
+ INIT_LIST_HEAD(&nilfs->ns_list);
INIT_LIST_HEAD(&nilfs->ns_supers);
spin_lock_init(&nilfs->ns_last_segment_lock);
nilfs->ns_gc_inodes_h = NULL;
- INIT_LIST_HEAD(&nilfs->ns_used_segments);
init_rwsem(&nilfs->ns_segctor_sem);
return nilfs;
}
/**
+ * find_or_create_nilfs - find or create nilfs object
+ * @bdev: block device to which the_nilfs is related
+ *
+ * find_nilfs() looks up an existent nilfs object created on the
+ * device and gets the reference count of the object. If no nilfs object
+ * is found on the device, a new nilfs object is allocated.
+ *
+ * Return Value: On success, pointer to the nilfs object is returned.
+ * On error, NULL is returned.
+ */
+struct the_nilfs *find_or_create_nilfs(struct block_device *bdev)
+{
+ struct the_nilfs *nilfs, *new = NULL;
+
+ retry:
+ spin_lock(&nilfs_lock);
+ list_for_each_entry(nilfs, &nilfs_objects, ns_list) {
+ if (nilfs->ns_bdev == bdev) {
+ get_nilfs(nilfs);
+ spin_unlock(&nilfs_lock);
+ if (new)
+ put_nilfs(new);
+ return nilfs; /* existing object */
+ }
+ }
+ if (new) {
+ list_add_tail(&new->ns_list, &nilfs_objects);
+ spin_unlock(&nilfs_lock);
+ return new; /* new object */
+ }
+ spin_unlock(&nilfs_lock);
+
+ new = alloc_nilfs(bdev);
+ if (new)
+ goto retry;
+ return NULL; /* insufficient memory */
+}
+
+/**
* put_nilfs - release a reference to the_nilfs
* @nilfs: the_nilfs structure to be released
*
*/
void put_nilfs(struct the_nilfs *nilfs)
{
- if (!atomic_dec_and_test(&nilfs->ns_count))
+ spin_lock(&nilfs_lock);
+ if (!atomic_dec_and_test(&nilfs->ns_count)) {
+ spin_unlock(&nilfs_lock);
return;
+ }
+ list_del_init(&nilfs->ns_list);
+ spin_unlock(&nilfs_lock);
+
/*
- * Increment of ns_count never occur below because the caller
+ * Increment of ns_count never occurs below because the caller
* of get_nilfs() holds at least one reference to the_nilfs.
* Thus its exclusion control is not required here.
*/
+
might_sleep();
if (nilfs_loaded(nilfs)) {
- nilfs_dispose_used_segments(nilfs);
nilfs_mdt_clear(nilfs->ns_sufile);
nilfs_mdt_destroy(nilfs->ns_sufile);
nilfs_mdt_clear(nilfs->ns_cpfile);
}
if (nilfs_init(nilfs)) {
nilfs_destroy_gccache(nilfs);
- brelse(nilfs->ns_sbh);
+ brelse(nilfs->ns_sbh[0]);
+ brelse(nilfs->ns_sbh[1]);
}
kfree(nilfs);
}
static int nilfs_load_super_root(struct the_nilfs *nilfs,
struct nilfs_sb_info *sbi, sector_t sr_block)
{
+ static struct lock_class_key dat_lock_key;
struct buffer_head *bh_sr;
struct nilfs_super_root *raw_sr;
+ struct nilfs_super_block **sbp = nilfs->ns_sbp;
unsigned dat_entry_size, segment_usage_size, checkpoint_size;
unsigned inode_size;
int err;
return err;
down_read(&nilfs->ns_sem);
- dat_entry_size = le16_to_cpu(nilfs->ns_sbp->s_dat_entry_size);
- checkpoint_size = le16_to_cpu(nilfs->ns_sbp->s_checkpoint_size);
- segment_usage_size = le16_to_cpu(nilfs->ns_sbp->s_segment_usage_size);
+ dat_entry_size = le16_to_cpu(sbp[0]->s_dat_entry_size);
+ checkpoint_size = le16_to_cpu(sbp[0]->s_checkpoint_size);
+ segment_usage_size = le16_to_cpu(sbp[0]->s_segment_usage_size);
up_read(&nilfs->ns_sem);
inode_size = nilfs->ns_inode_size;
err = -ENOMEM;
- nilfs->ns_dat = nilfs_mdt_new(
- nilfs, NULL, NILFS_DAT_INO, NILFS_DAT_GFP);
+ nilfs->ns_dat = nilfs_mdt_new(nilfs, NULL, NILFS_DAT_INO);
if (unlikely(!nilfs->ns_dat))
goto failed;
- nilfs->ns_gc_dat = nilfs_mdt_new(
- nilfs, NULL, NILFS_DAT_INO, NILFS_DAT_GFP);
+ nilfs->ns_gc_dat = nilfs_mdt_new(nilfs, NULL, NILFS_DAT_INO);
if (unlikely(!nilfs->ns_gc_dat))
goto failed_dat;
- nilfs->ns_cpfile = nilfs_mdt_new(
- nilfs, NULL, NILFS_CPFILE_INO, NILFS_CPFILE_GFP);
+ nilfs->ns_cpfile = nilfs_mdt_new(nilfs, NULL, NILFS_CPFILE_INO);
if (unlikely(!nilfs->ns_cpfile))
goto failed_gc_dat;
- nilfs->ns_sufile = nilfs_mdt_new(
- nilfs, NULL, NILFS_SUFILE_INO, NILFS_SUFILE_GFP);
+ nilfs->ns_sufile = nilfs_mdt_new(nilfs, NULL, NILFS_SUFILE_INO);
if (unlikely(!nilfs->ns_sufile))
goto failed_cpfile;
if (unlikely(err))
goto failed_sufile;
+ lockdep_set_class(&NILFS_MDT(nilfs->ns_dat)->mi_sem, &dat_lock_key);
+ lockdep_set_class(&NILFS_MDT(nilfs->ns_gc_dat)->mi_sem, &dat_lock_key);
+
nilfs_mdt_set_shadow(nilfs->ns_dat, nilfs->ns_gc_dat);
nilfs_mdt_set_entry_size(nilfs->ns_cpfile, checkpoint_size,
sizeof(struct nilfs_cpfile_header));
nilfs_mdt_destroy(nilfs->ns_dat);
goto failed;
}
- if (ri.ri_need_recovery == NILFS_RECOVERY_SR_UPDATED) {
- down_write(&nilfs->ns_sem);
- nilfs_update_last_segment(sbi, 0);
- up_write(&nilfs->ns_sem);
- }
+ if (ri.ri_need_recovery == NILFS_RECOVERY_SR_UPDATED)
+ sbi->s_super->s_dirt = 1;
}
set_nilfs_loaded(nilfs);
return res;
}
-static int
-nilfs_store_disk_layout(struct the_nilfs *nilfs, struct super_block *sb,
- struct nilfs_super_block *sbp)
+static int nilfs_store_disk_layout(struct the_nilfs *nilfs,
+ struct nilfs_super_block *sbp)
{
if (le32_to_cpu(sbp->s_rev_level) != NILFS_CURRENT_REV) {
printk(KERN_ERR "NILFS: revision mismatch "
NILFS_CURRENT_REV, NILFS_MINOR_REV);
return -EINVAL;
}
+ nilfs->ns_sbsize = le16_to_cpu(sbp->s_bytes);
+ if (nilfs->ns_sbsize > BLOCK_SIZE)
+ return -EINVAL;
+
nilfs->ns_inode_size = le16_to_cpu(sbp->s_inode_size);
nilfs->ns_first_ino = le32_to_cpu(sbp->s_first_ino);
return 0;
}
+static int nilfs_valid_sb(struct nilfs_super_block *sbp)
+{
+ static unsigned char sum[4];
+ const int sumoff = offsetof(struct nilfs_super_block, s_sum);
+ size_t bytes;
+ u32 crc;
+
+ if (!sbp || le16_to_cpu(sbp->s_magic) != NILFS_SUPER_MAGIC)
+ return 0;
+ bytes = le16_to_cpu(sbp->s_bytes);
+ if (bytes > BLOCK_SIZE)
+ return 0;
+ crc = crc32_le(le32_to_cpu(sbp->s_crc_seed), (unsigned char *)sbp,
+ sumoff);
+ crc = crc32_le(crc, sum, 4);
+ crc = crc32_le(crc, (unsigned char *)sbp + sumoff + 4,
+ bytes - sumoff - 4);
+ return crc == le32_to_cpu(sbp->s_sum);
+}
+
+static int nilfs_sb2_bad_offset(struct nilfs_super_block *sbp, u64 offset)
+{
+ return offset < ((le64_to_cpu(sbp->s_nsegments) *
+ le32_to_cpu(sbp->s_blocks_per_segment)) <<
+ (le32_to_cpu(sbp->s_log_block_size) + 10));
+}
+
+static void nilfs_release_super_block(struct the_nilfs *nilfs)
+{
+ int i;
+
+ for (i = 0; i < 2; i++) {
+ if (nilfs->ns_sbp[i]) {
+ brelse(nilfs->ns_sbh[i]);
+ nilfs->ns_sbh[i] = NULL;
+ nilfs->ns_sbp[i] = NULL;
+ }
+ }
+}
+
+void nilfs_fall_back_super_block(struct the_nilfs *nilfs)
+{
+ brelse(nilfs->ns_sbh[0]);
+ nilfs->ns_sbh[0] = nilfs->ns_sbh[1];
+ nilfs->ns_sbp[0] = nilfs->ns_sbp[1];
+ nilfs->ns_sbh[1] = NULL;
+ nilfs->ns_sbp[1] = NULL;
+}
+
+void nilfs_swap_super_block(struct the_nilfs *nilfs)
+{
+ struct buffer_head *tsbh = nilfs->ns_sbh[0];
+ struct nilfs_super_block *tsbp = nilfs->ns_sbp[0];
+
+ nilfs->ns_sbh[0] = nilfs->ns_sbh[1];
+ nilfs->ns_sbp[0] = nilfs->ns_sbp[1];
+ nilfs->ns_sbh[1] = tsbh;
+ nilfs->ns_sbp[1] = tsbp;
+}
+
+static int nilfs_load_super_block(struct the_nilfs *nilfs,
+ struct super_block *sb, int blocksize,
+ struct nilfs_super_block **sbpp)
+{
+ struct nilfs_super_block **sbp = nilfs->ns_sbp;
+ struct buffer_head **sbh = nilfs->ns_sbh;
+ u64 sb2off = NILFS_SB2_OFFSET_BYTES(nilfs->ns_bdev->bd_inode->i_size);
+ int valid[2], swp = 0;
+
+ sbp[0] = nilfs_read_super_block(sb, NILFS_SB_OFFSET_BYTES, blocksize,
+ &sbh[0]);
+ sbp[1] = nilfs_read_super_block(sb, sb2off, blocksize, &sbh[1]);
+
+ if (!sbp[0]) {
+ if (!sbp[1]) {
+ printk(KERN_ERR "NILFS: unable to read superblock\n");
+ return -EIO;
+ }
+ printk(KERN_WARNING
+ "NILFS warning: unable to read primary superblock\n");
+ } else if (!sbp[1])
+ printk(KERN_WARNING
+ "NILFS warning: unable to read secondary superblock\n");
+
+ valid[0] = nilfs_valid_sb(sbp[0]);
+ valid[1] = nilfs_valid_sb(sbp[1]);
+ swp = valid[1] &&
+ (!valid[0] ||
+ le64_to_cpu(sbp[1]->s_wtime) > le64_to_cpu(sbp[0]->s_wtime));
+
+ if (valid[swp] && nilfs_sb2_bad_offset(sbp[swp], sb2off)) {
+ brelse(sbh[1]);
+ sbh[1] = NULL;
+ sbp[1] = NULL;
+ swp = 0;
+ }
+ if (!valid[swp]) {
+ nilfs_release_super_block(nilfs);
+ printk(KERN_ERR "NILFS: Can't find nilfs on dev %s.\n",
+ sb->s_id);
+ return -EINVAL;
+ }
+
+ if (swp) {
+ printk(KERN_WARNING "NILFS warning: broken superblock. "
+ "using spare superblock.\n");
+ nilfs_swap_super_block(nilfs);
+ }
+
+ nilfs->ns_sbwtime[0] = le64_to_cpu(sbp[0]->s_wtime);
+ nilfs->ns_sbwtime[1] = valid[!swp] ? le64_to_cpu(sbp[1]->s_wtime) : 0;
+ nilfs->ns_prot_seq = le64_to_cpu(sbp[valid[1] & !swp]->s_last_seq);
+ *sbpp = sbp[0];
+ return 0;
+}
+
/**
* init_nilfs - initialize a NILFS instance.
* @nilfs: the_nilfs structure
int init_nilfs(struct the_nilfs *nilfs, struct nilfs_sb_info *sbi, char *data)
{
struct super_block *sb = sbi->s_super;
- struct buffer_head *sbh;
struct nilfs_super_block *sbp;
struct backing_dev_info *bdi;
int blocksize;
- int err = 0;
+ int err;
down_write(&nilfs->ns_sem);
if (nilfs_init(nilfs)) {
/* Load values from existing the_nilfs */
- sbp = nilfs->ns_sbp;
+ sbp = nilfs->ns_sbp[0];
err = nilfs_store_magic_and_option(sb, sbp, data);
if (err)
goto out;
goto out;
}
- sbp = nilfs_load_super_block(sb, &sbh);
- if (!sbp) {
+ blocksize = sb_min_blocksize(sb, BLOCK_SIZE);
+ if (!blocksize) {
+ printk(KERN_ERR "NILFS: unable to set blocksize\n");
err = -EINVAL;
goto out;
}
+ err = nilfs_load_super_block(nilfs, sb, blocksize, &sbp);
+ if (err)
+ goto out;
+
err = nilfs_store_magic_and_option(sb, sbp, data);
if (err)
goto failed_sbh;
blocksize = BLOCK_SIZE << le32_to_cpu(sbp->s_log_block_size);
if (sb->s_blocksize != blocksize) {
- sbp = nilfs_reload_super_block(sb, &sbh, blocksize);
- if (!sbp) {
+ int hw_blocksize = bdev_logical_block_size(sb->s_bdev);
+
+ if (blocksize < hw_blocksize) {
+ printk(KERN_ERR
+ "NILFS: blocksize %d too small for device "
+ "(sector-size = %d).\n",
+ blocksize, hw_blocksize);
err = -EINVAL;
+ goto failed_sbh;
+ }
+ nilfs_release_super_block(nilfs);
+ sb_set_blocksize(sb, blocksize);
+
+ err = nilfs_load_super_block(nilfs, sb, blocksize, &sbp);
+ if (err)
goto out;
/* not failed_sbh; sbh is released automatically
when reloading fails. */
- }
}
nilfs->ns_blocksize_bits = sb->s_blocksize_bits;
- err = nilfs_store_disk_layout(nilfs, sb, sbp);
+ err = nilfs_store_disk_layout(nilfs, sbp);
if (err)
goto failed_sbh;
sb->s_maxbytes = nilfs_max_size(sb->s_blocksize_bits);
nilfs->ns_mount_state = le16_to_cpu(sbp->s_state);
- nilfs->ns_sbh = sbh;
- nilfs->ns_sbp = sbp;
bdi = nilfs->ns_bdev->bd_inode_backing_dev_info;
if (!bdi)
return err;
failed_sbh:
- brelse(sbh);
+ nilfs_release_super_block(nilfs);
goto out;
}
return err;
}
-void nilfs_dispose_used_segments(struct the_nilfs *nilfs)
-{
- struct nilfs_segment_entry *ent, *n;
-
- /* nilfs->sem must be locked by the caller. */
- if (!nilfs_loaded(nilfs))
- return;
-
- list_for_each_entry_safe(ent, n, &nilfs->ns_used_segments, list) {
- list_del_init(&ent->list);
- nilfs_segment_usage_clear_volatile_active(ent->raw_su);
- nilfs_close_segment_entry(ent, nilfs->ns_sufile);
- nilfs_free_segment_entry(ent);
- }
-}
-
int nilfs_near_disk_full(struct the_nilfs *nilfs)
{
struct inode *sufile = nilfs->ns_sufile;
return ret;
}
+/**
+ * nilfs_find_sbinfo - find existing nilfs_sb_info structure
+ * @nilfs: nilfs object
+ * @rw_mount: mount type (non-zero value for read/write mount)
+ * @cno: checkpoint number (zero for read-only mount)
+ *
+ * nilfs_find_sbinfo() returns the nilfs_sb_info structure which
+ * @rw_mount and @cno (in case of snapshots) matched. If no instance
+ * was found, NULL is returned. Although the super block instance can
+ * be unmounted after this function returns, the nilfs_sb_info struct
+ * is kept on memory until nilfs_put_sbinfo() is called.
+ */
+struct nilfs_sb_info *nilfs_find_sbinfo(struct the_nilfs *nilfs,
+ int rw_mount, __u64 cno)
+{
+ struct nilfs_sb_info *sbi;
+
+ down_read(&nilfs->ns_super_sem);
+ /*
+ * The SNAPSHOT flag and sb->s_flags are supposed to be
+ * protected with nilfs->ns_super_sem.
+ */
+ sbi = nilfs->ns_current;
+ if (rw_mount) {
+ if (sbi && !(sbi->s_super->s_flags & MS_RDONLY))
+ goto found; /* read/write mount */
+ else
+ goto out;
+ } else if (cno == 0) {
+ if (sbi && (sbi->s_super->s_flags & MS_RDONLY))
+ goto found; /* read-only mount */
+ else
+ goto out;
+ }
+
+ list_for_each_entry(sbi, &nilfs->ns_supers, s_list) {
+ if (nilfs_test_opt(sbi, SNAPSHOT) &&
+ sbi->s_snapshot_cno == cno)
+ goto found; /* snapshot mount */
+ }
+ out:
+ up_read(&nilfs->ns_super_sem);
+ return NULL;
+
+ found:
+ atomic_inc(&sbi->s_count);
+ up_read(&nilfs->ns_super_sem);
+ return sbi;
+}
+
int nilfs_checkpoint_is_mounted(struct the_nilfs *nilfs, __u64 cno,
int snapshot_mount)
{
struct nilfs_sb_info *sbi;
int ret = 0;
- down_read(&nilfs->ns_sem);
+ down_read(&nilfs->ns_super_sem);
if (cno == 0 || cno > nilfs->ns_cno)
goto out_unlock;
ret++;
out_unlock:
- up_read(&nilfs->ns_sem);
+ up_read(&nilfs->ns_super_sem);
return ret;
}
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff