#include <linux/log2.h>
#include <linux/vmalloc.h>
#include <linux/backing-dev.h>
+#include <linux/bitops.h>
+#include <linux/ratelimit.h>
#define CREATE_TRACE_POINTS
#include <trace/events/jbd2.h>
#include <asm/uaccess.h>
#include <asm/page.h>
+#include <asm/system.h>
EXPORT_SYMBOL(jbd2_journal_extend);
EXPORT_SYMBOL(jbd2_journal_stop);
EXPORT_SYMBOL(jbd2_journal_init_jbd_inode);
EXPORT_SYMBOL(jbd2_journal_release_jbd_inode);
EXPORT_SYMBOL(jbd2_journal_begin_ordered_truncate);
+EXPORT_SYMBOL(jbd2_inode_cache);
static int journal_convert_superblock_v1(journal_t *, journal_superblock_t *);
static void __journal_abort_soft (journal_t *journal, int errno);
}
/*
- * Called under j_state_lock. Returns true if a transaction commit was started.
+ * Called with j_state_lock locked for writing.
+ * Returns true if a transaction commit was started.
*/
int __jbd2_log_start_commit(journal_t *journal, tid_t target)
{
/*
- * Are we already doing a recent enough commit?
+ * The only transaction we can possibly wait upon is the
+ * currently running transaction (if it exists). Otherwise,
+ * the target tid must be an old one.
*/
- if (!tid_geq(journal->j_commit_request, target)) {
+ if (journal->j_running_transaction &&
+ journal->j_running_transaction->t_tid == target) {
/*
- * We want a new commit: OK, mark the request and wakup the
+ * We want a new commit: OK, mark the request and wakeup the
* commit thread. We do _not_ do the commit ourselves.
*/
journal->j_commit_sequence);
wake_up(&journal->j_wait_commit);
return 1;
- }
+ } else if (!tid_geq(journal->j_commit_request, target))
+ /* This should never happen, but if it does, preserve
+ the evidence before kjournald goes into a loop and
+ increments j_commit_sequence beyond all recognition. */
+ WARN_ONCE(1, "jbd: bad log_start_commit: %u %u %u %u\n",
+ journal->j_commit_request,
+ journal->j_commit_sequence,
+ target, journal->j_running_transaction ?
+ journal->j_running_transaction->t_tid : 0);
return 0;
}
{
transaction_t *transaction = NULL;
tid_t tid;
+ int need_to_start = 0;
read_lock(&journal->j_state_lock);
if (journal->j_running_transaction && !current->journal_info) {
transaction = journal->j_running_transaction;
- __jbd2_log_start_commit(journal, transaction->t_tid);
+ if (!tid_geq(journal->j_commit_request, transaction->t_tid))
+ need_to_start = 1;
} else if (journal->j_committing_transaction)
transaction = journal->j_committing_transaction;
tid = transaction->t_tid;
read_unlock(&journal->j_state_lock);
+ if (need_to_start)
+ jbd2_log_start_commit(journal, tid);
jbd2_log_wait_commit(journal, tid);
return 1;
}
return ret;
}
+/*
+ * Return 1 if a given transaction has not yet sent barrier request
+ * connected with a transaction commit. If 0 is returned, transaction
+ * may or may not have sent the barrier. Used to avoid sending barrier
+ * twice in common cases.
+ */
+int jbd2_trans_will_send_data_barrier(journal_t *journal, tid_t tid)
+{
+ int ret = 0;
+ transaction_t *commit_trans;
+
+ if (!(journal->j_flags & JBD2_BARRIER))
+ return 0;
+ read_lock(&journal->j_state_lock);
+ /* Transaction already committed? */
+ if (tid_geq(journal->j_commit_sequence, tid))
+ goto out;
+ commit_trans = journal->j_committing_transaction;
+ if (!commit_trans || commit_trans->t_tid != tid) {
+ ret = 1;
+ goto out;
+ }
+ /*
+ * Transaction is being committed and we already proceeded to
+ * submitting a flush to fs partition?
+ */
+ if (journal->j_fs_dev != journal->j_dev) {
+ if (!commit_trans->t_need_data_flush ||
+ commit_trans->t_state >= T_COMMIT_DFLUSH)
+ goto out;
+ } else {
+ if (commit_trans->t_state >= T_COMMIT_JFLUSH)
+ goto out;
+ }
+ ret = 1;
+out:
+ read_unlock(&journal->j_state_lock);
+ return ret;
+}
+EXPORT_SYMBOL(jbd2_trans_will_send_data_barrier);
+
/*
* Wait for a specified commit to complete.
* The caller may not hold the journal lock.
journal = kzalloc(sizeof(*journal), GFP_KERNEL);
if (!journal)
- goto fail;
+ return NULL;
init_waitqueue_head(&journal->j_wait_transaction_locked);
init_waitqueue_head(&journal->j_wait_logspace);
err = jbd2_journal_init_revoke(journal, JOURNAL_REVOKE_DEFAULT_HASH);
if (err) {
kfree(journal);
- goto fail;
+ return NULL;
}
spin_lock_init(&journal->j_history_lock);
return journal;
-fail:
- return NULL;
}
/* jbd2_journal_init_dev and jbd2_journal_init_inode:
/* journal descriptor can store up to n blocks -bzzz */
journal->j_blocksize = blocksize;
- jbd2_stats_proc_init(journal);
- n = journal->j_blocksize / sizeof(journal_block_tag_t);
- journal->j_wbufsize = n;
- journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL);
- if (!journal->j_wbuf) {
- printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n",
- __func__);
- goto out_err;
- }
journal->j_dev = bdev;
journal->j_fs_dev = fs_dev;
journal->j_blk_offset = start;
p = journal->j_devname;
while ((p = strchr(p, '/')))
*p = '!';
+ jbd2_stats_proc_init(journal);
+ n = journal->j_blocksize / sizeof(journal_block_tag_t);
+ journal->j_wbufsize = n;
+ journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL);
+ if (!journal->j_wbuf) {
+ printk(KERN_ERR "%s: Can't allocate bhs for commit thread\n",
+ __func__);
+ goto out_err;
+ }
bh = __getblk(journal->j_dev, start, journal->j_blocksize);
if (!bh) {
journal->j_wbufsize = n;
journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL);
if (!journal->j_wbuf) {
- printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n",
+ printk(KERN_ERR "%s: Can't allocate bhs for commit thread\n",
__func__);
goto out_err;
}
err = jbd2_journal_bmap(journal, 0, &blocknr);
/* If that failed, give up */
if (err) {
- printk(KERN_ERR "%s: Cannnot locate journal superblock\n",
+ printk(KERN_ERR "%s: Cannot locate journal superblock\n",
__func__);
goto out_err;
}
goto out;
}
+ if (be32_to_cpu(sb->s_first) == 0 ||
+ be32_to_cpu(sb->s_first) >= journal->j_maxlen) {
+ printk(KERN_WARNING
+ "JBD2: Invalid start block of journal: %u\n",
+ be32_to_cpu(sb->s_first));
+ goto out;
+ }
+
return 0;
out:
if (!compat && !ro && !incompat)
return 1;
+ /* Load journal superblock if it is not loaded yet. */
+ if (journal->j_format_version == 0 &&
+ journal_get_superblock(journal) != 0)
+ return 0;
if (journal->j_format_version == 1)
return 0;
*/
#define JBD2_MAX_SLABS 8
static struct kmem_cache *jbd2_slab[JBD2_MAX_SLABS];
-static DECLARE_MUTEX(jbd2_slab_create_sem);
static const char *jbd2_slab_names[JBD2_MAX_SLABS] = {
"jbd2_1k", "jbd2_2k", "jbd2_4k", "jbd2_8k",
static int jbd2_journal_create_slab(size_t size)
{
+ static DEFINE_MUTEX(jbd2_slab_create_mutex);
int i = order_base_2(size) - 10;
size_t slab_size;
if (unlikely(i < 0))
i = 0;
- down(&jbd2_slab_create_sem);
+ mutex_lock(&jbd2_slab_create_mutex);
if (jbd2_slab[i]) {
- up(&jbd2_slab_create_sem);
+ mutex_unlock(&jbd2_slab_create_mutex);
return 0; /* Already created */
}
slab_size = 1 << (i+10);
jbd2_slab[i] = kmem_cache_create(jbd2_slab_names[i], slab_size,
slab_size, 0, NULL);
- up(&jbd2_slab_create_sem);
+ mutex_unlock(&jbd2_slab_create_mutex);
if (!jbd2_slab[i]) {
printk(KERN_EMERG "JBD2: no memory for jbd2_slab cache\n");
return -ENOMEM;
static struct journal_head *journal_alloc_journal_head(void)
{
struct journal_head *ret;
- static unsigned long last_warning;
#ifdef CONFIG_JBD2_DEBUG
atomic_inc(&nr_journal_heads);
ret = kmem_cache_alloc(jbd2_journal_head_cache, GFP_NOFS);
if (!ret) {
jbd_debug(1, "out of memory for journal_head\n");
- if (time_after(jiffies, last_warning + 5*HZ)) {
- printk(KERN_NOTICE "ENOMEM in %s, retrying.\n",
- __func__);
- last_warning = jiffies;
- }
+ pr_notice_ratelimited("ENOMEM in %s, retrying.\n", __func__);
while (!ret) {
yield();
ret = kmem_cache_alloc(jbd2_journal_head_cache, GFP_NOFS);
* When a buffer has its BH_JBD bit set it is immune from being released by
* core kernel code, mainly via ->b_count.
*
- * A journal_head may be detached from its buffer_head when the journal_head's
- * b_transaction, b_cp_transaction and b_next_transaction pointers are NULL.
- * Various places in JBD call jbd2_journal_remove_journal_head() to indicate that the
- * journal_head can be dropped if needed.
+ * A journal_head is detached from its buffer_head when the journal_head's
+ * b_jcount reaches zero. Running transaction (b_transaction) and checkpoint
+ * transaction (b_cp_transaction) hold their references to b_jcount.
*
* Various places in the kernel want to attach a journal_head to a buffer_head
* _before_ attaching the journal_head to a transaction. To protect the
* (Attach a journal_head if needed. Increments b_jcount)
* struct journal_head *jh = jbd2_journal_add_journal_head(bh);
* ...
+ * (Get another reference for transaction)
+ * jbd2_journal_grab_journal_head(bh);
* jh->b_transaction = xxx;
+ * (Put original reference)
* jbd2_journal_put_journal_head(jh);
- *
- * Now, the journal_head's b_jcount is zero, but it is safe from being released
- * because it has a non-zero b_transaction.
*/
/*
* Give a buffer_head a journal_head.
*
- * Doesn't need the journal lock.
* May sleep.
*/
struct journal_head *jbd2_journal_add_journal_head(struct buffer_head *bh)
struct journal_head *jh = bh2jh(bh);
J_ASSERT_JH(jh, jh->b_jcount >= 0);
-
- get_bh(bh);
- if (jh->b_jcount == 0) {
- if (jh->b_transaction == NULL &&
- jh->b_next_transaction == NULL &&
- jh->b_cp_transaction == NULL) {
- J_ASSERT_JH(jh, jh->b_jlist == BJ_None);
- J_ASSERT_BH(bh, buffer_jbd(bh));
- J_ASSERT_BH(bh, jh2bh(jh) == bh);
- BUFFER_TRACE(bh, "remove journal_head");
- if (jh->b_frozen_data) {
- printk(KERN_WARNING "%s: freeing "
- "b_frozen_data\n",
- __func__);
- jbd2_free(jh->b_frozen_data, bh->b_size);
- }
- if (jh->b_committed_data) {
- printk(KERN_WARNING "%s: freeing "
- "b_committed_data\n",
- __func__);
- jbd2_free(jh->b_committed_data, bh->b_size);
- }
- bh->b_private = NULL;
- jh->b_bh = NULL; /* debug, really */
- clear_buffer_jbd(bh);
- __brelse(bh);
- journal_free_journal_head(jh);
- } else {
- BUFFER_TRACE(bh, "journal_head was locked");
- }
+ J_ASSERT_JH(jh, jh->b_transaction == NULL);
+ J_ASSERT_JH(jh, jh->b_next_transaction == NULL);
+ J_ASSERT_JH(jh, jh->b_cp_transaction == NULL);
+ J_ASSERT_JH(jh, jh->b_jlist == BJ_None);
+ J_ASSERT_BH(bh, buffer_jbd(bh));
+ J_ASSERT_BH(bh, jh2bh(jh) == bh);
+ BUFFER_TRACE(bh, "remove journal_head");
+ if (jh->b_frozen_data) {
+ printk(KERN_WARNING "%s: freeing b_frozen_data\n", __func__);
+ jbd2_free(jh->b_frozen_data, bh->b_size);
}
+ if (jh->b_committed_data) {
+ printk(KERN_WARNING "%s: freeing b_committed_data\n", __func__);
+ jbd2_free(jh->b_committed_data, bh->b_size);
+ }
+ bh->b_private = NULL;
+ jh->b_bh = NULL; /* debug, really */
+ clear_buffer_jbd(bh);
+ journal_free_journal_head(jh);
}
/*
- * jbd2_journal_remove_journal_head(): if the buffer isn't attached to a transaction
- * and has a zero b_jcount then remove and release its journal_head. If we did
- * see that the buffer is not used by any transaction we also "logically"
- * decrement ->b_count.
- *
- * We in fact take an additional increment on ->b_count as a convenience,
- * because the caller usually wants to do additional things with the bh
- * after calling here.
- * The caller of jbd2_journal_remove_journal_head() *must* run __brelse(bh) at some
- * time. Once the caller has run __brelse(), the buffer is eligible for
- * reaping by try_to_free_buffers().
- */
-void jbd2_journal_remove_journal_head(struct buffer_head *bh)
-{
- jbd_lock_bh_journal_head(bh);
- __journal_remove_journal_head(bh);
- jbd_unlock_bh_journal_head(bh);
-}
-
-/*
- * Drop a reference on the passed journal_head. If it fell to zero then try to
+ * Drop a reference on the passed journal_head. If it fell to zero then
* release the journal_head from the buffer_head.
*/
void jbd2_journal_put_journal_head(struct journal_head *jh)
jbd_lock_bh_journal_head(bh);
J_ASSERT_JH(jh, jh->b_jcount > 0);
--jh->b_jcount;
- if (!jh->b_jcount && !jh->b_transaction) {
+ if (!jh->b_jcount) {
__journal_remove_journal_head(bh);
+ jbd_unlock_bh_journal_head(bh);
__brelse(bh);
- }
- jbd_unlock_bh_journal_head(bh);
+ } else
+ jbd_unlock_bh_journal_head(bh);
}
/*
restart:
spin_lock(&journal->j_list_lock);
/* Is commit writing out inode - we have to wait */
- if (jinode->i_flags & JI_COMMIT_RUNNING) {
+ if (test_bit(__JI_COMMIT_RUNNING, &jinode->i_flags)) {
wait_queue_head_t *wq;
DEFINE_WAIT_BIT(wait, &jinode->i_flags, __JI_COMMIT_RUNNING);
wq = bit_waitqueue(&jinode->i_flags, __JI_COMMIT_RUNNING);
#endif
-struct kmem_cache *jbd2_handle_cache;
+struct kmem_cache *jbd2_handle_cache, *jbd2_inode_cache;
static int __init journal_init_handle_cache(void)
{
- jbd2_handle_cache = kmem_cache_create("jbd2_journal_handle",
- sizeof(handle_t),
- 0, /* offset */
- SLAB_TEMPORARY, /* flags */
- NULL); /* ctor */
+ jbd2_handle_cache = KMEM_CACHE(jbd2_journal_handle, SLAB_TEMPORARY);
if (jbd2_handle_cache == NULL) {
- printk(KERN_EMERG "JBD: failed to create handle cache\n");
+ printk(KERN_EMERG "JBD2: failed to create handle cache\n");
+ return -ENOMEM;
+ }
+ jbd2_inode_cache = KMEM_CACHE(jbd2_inode, 0);
+ if (jbd2_inode_cache == NULL) {
+ printk(KERN_EMERG "JBD2: failed to create inode cache\n");
+ kmem_cache_destroy(jbd2_handle_cache);
return -ENOMEM;
}
return 0;
{
if (jbd2_handle_cache)
kmem_cache_destroy(jbd2_handle_cache);
+ if (jbd2_inode_cache)
+ kmem_cache_destroy(jbd2_inode_cache);
+
}
/*
new_dev = kmalloc(sizeof(struct devname_cache), GFP_KERNEL);
if (!new_dev)
return "NODEV-ALLOCFAILURE"; /* Something non-NULL */
+ bd = bdget(device);
spin_lock(&devname_cache_lock);
if (devcache[i]) {
if (devcache[i]->device == device) {
kfree(new_dev);
+ bdput(bd);
ret = devcache[i]->devname;
spin_unlock(&devname_cache_lock);
return ret;
}
devcache[i] = new_dev;
devcache[i]->device = device;
- bd = bdget(device);
if (bd) {
bdevname(bd, devcache[i]->devname);
bdput(bd);