aboutsummaryrefslogtreecommitdiffstats
path: root/fs/jbd2
diff options
context:
space:
mode:
Diffstat (limited to 'fs/jbd2')
-rw-r--r--fs/jbd2/commit.c44
-rw-r--r--fs/jbd2/journal.c13
-rw-r--r--fs/jbd2/recovery.c7
-rw-r--r--fs/jbd2/transaction.c94
4 files changed, 100 insertions, 58 deletions
diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index 216f4299f65..3091d42992f 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -349,12 +349,12 @@ static void jbd2_block_tag_csum_set(journal_t *j, journal_block_tag_t *tag,
return;
sequence = cpu_to_be32(sequence);
- addr = kmap_atomic(page, KM_USER0);
+ addr = kmap_atomic(page);
csum = jbd2_chksum(j, j->j_csum_seed, (__u8 *)&sequence,
sizeof(sequence));
csum = jbd2_chksum(j, csum, addr + offset_in_page(bh->b_data),
bh->b_size);
- kunmap_atomic(addr, KM_USER0);
+ kunmap_atomic(addr);
tag->t_checksum = cpu_to_be32(csum);
}
@@ -1014,17 +1014,35 @@ restart_loop:
* there's no point in keeping a checkpoint record for
* it. */
- /* A buffer which has been freed while still being
- * journaled by a previous transaction may end up still
- * being dirty here, but we want to avoid writing back
- * that buffer in the future after the "add to orphan"
- * operation been committed, That's not only a performance
- * gain, it also stops aliasing problems if the buffer is
- * left behind for writeback and gets reallocated for another
- * use in a different page. */
- if (buffer_freed(bh) && !jh->b_next_transaction) {
- clear_buffer_freed(bh);
- clear_buffer_jbddirty(bh);
+ /*
+ * A buffer which has been freed while still being journaled by
+ * a previous transaction.
+ */
+ if (buffer_freed(bh)) {
+ /*
+ * If the running transaction is the one containing
+ * "add to orphan" operation (b_next_transaction !=
+ * NULL), we have to wait for that transaction to
+ * commit before we can really get rid of the buffer.
+ * So just clear b_modified to not confuse transaction
+ * credit accounting and refile the buffer to
+ * BJ_Forget of the running transaction. If the just
+ * committed transaction contains "add to orphan"
+ * operation, we can completely invalidate the buffer
+ * now. We are rather through in that since the
+ * buffer may be still accessible when blocksize <
+ * pagesize and it is attached to the last partial
+ * page.
+ */
+ jh->b_modified = 0;
+ if (!jh->b_next_transaction) {
+ clear_buffer_freed(bh);
+ clear_buffer_jbddirty(bh);
+ clear_buffer_mapped(bh);
+ clear_buffer_new(bh);
+ clear_buffer_req(bh);
+ bh->b_bdev = NULL;
+ }
}
if (buffer_jbddirty(bh)) {
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index e9a3c4c8559..dbf41f9452d 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -60,7 +60,6 @@ EXPORT_SYMBOL(jbd2_journal_get_create_access);
EXPORT_SYMBOL(jbd2_journal_get_undo_access);
EXPORT_SYMBOL(jbd2_journal_set_triggers);
EXPORT_SYMBOL(jbd2_journal_dirty_metadata);
-EXPORT_SYMBOL(jbd2_journal_release_buffer);
EXPORT_SYMBOL(jbd2_journal_forget);
#if 0
EXPORT_SYMBOL(journal_sync_buffer);
@@ -612,8 +611,8 @@ int jbd2_journal_start_commit(journal_t *journal, tid_t *ptid)
ret = 1;
} else if (journal->j_committing_transaction) {
/*
- * If ext3_write_super() recently started a commit, then we
- * have to wait for completion of that transaction
+ * If commit has been started, then we have to wait for
+ * completion of that transaction.
*/
if (ptid)
*ptid = journal->j_committing_transaction->t_tid;
@@ -1354,6 +1353,11 @@ static void jbd2_mark_journal_empty(journal_t *journal)
BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex));
read_lock(&journal->j_state_lock);
+ /* Is it already empty? */
+ if (sb->s_start == 0) {
+ read_unlock(&journal->j_state_lock);
+ return;
+ }
jbd_debug(1, "JBD2: Marking journal as empty (seq %d)\n",
journal->j_tail_sequence);
@@ -1377,7 +1381,7 @@ static void jbd2_mark_journal_empty(journal_t *journal)
* Update a journal's errno. Write updated superblock to disk waiting for IO
* to complete.
*/
-static void jbd2_journal_update_sb_errno(journal_t *journal)
+void jbd2_journal_update_sb_errno(journal_t *journal)
{
journal_superblock_t *sb = journal->j_superblock;
@@ -1390,6 +1394,7 @@ static void jbd2_journal_update_sb_errno(journal_t *journal)
jbd2_write_superblock(journal, WRITE_SYNC);
}
+EXPORT_SYMBOL(jbd2_journal_update_sb_errno);
/*
* Read the superblock for a given journal, performing initial
diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c
index 0131e436253..626846bac32 100644
--- a/fs/jbd2/recovery.c
+++ b/fs/jbd2/recovery.c
@@ -289,8 +289,11 @@ int jbd2_journal_recover(journal_t *journal)
if (!err)
err = err2;
/* Make sure all replayed data is on permanent storage */
- if (journal->j_flags & JBD2_BARRIER)
- blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL);
+ if (journal->j_flags & JBD2_BARRIER) {
+ err2 = blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL);
+ if (!err)
+ err = err2;
+ }
return err;
}
diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index fb1ab9533b6..df9f29760ef 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -209,7 +209,8 @@ repeat:
if (!new_transaction)
goto alloc_transaction;
write_lock(&journal->j_state_lock);
- if (!journal->j_running_transaction) {
+ if (!journal->j_running_transaction &&
+ !journal->j_barrier_count) {
jbd2_get_transaction(journal, new_transaction);
new_transaction = NULL;
}
@@ -1207,17 +1208,6 @@ out:
return ret;
}
-/*
- * jbd2_journal_release_buffer: undo a get_write_access without any buffer
- * updates, if the update decided in the end that it didn't need access.
- *
- */
-void
-jbd2_journal_release_buffer(handle_t *handle, struct buffer_head *bh)
-{
- BUFFER_TRACE(bh, "entry");
-}
-
/**
* void jbd2_journal_forget() - bforget() for potentially-journaled buffers.
* @handle: transaction handle
@@ -1261,7 +1251,7 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
goto not_jbd;
}
- /* keep track of wether or not this transaction modified us */
+ /* keep track of whether or not this transaction modified us */
was_modified = jh->b_modified;
/*
@@ -1841,12 +1831,12 @@ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction)
* We're outside-transaction here. Either or both of j_running_transaction
* and j_committing_transaction may be NULL.
*/
-static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
+static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh,
+ int partial_page)
{
transaction_t *transaction;
struct journal_head *jh;
int may_free = 1;
- int ret;
BUFFER_TRACE(bh, "entry");
@@ -1878,10 +1868,18 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
* clear the buffer dirty bit at latest at the moment when the
* transaction marking the buffer as freed in the filesystem
* structures is committed because from that moment on the
- * buffer can be reallocated and used by a different page.
+ * block can be reallocated and used by a different page.
* Since the block hasn't been freed yet but the inode has
* already been added to orphan list, it is safe for us to add
* the buffer to BJ_Forget list of the newest transaction.
+ *
+ * Also we have to clear buffer_mapped flag of a truncated buffer
+ * because the buffer_head may be attached to the page straddling
+ * i_size (can happen only when blocksize < pagesize) and thus the
+ * buffer_head can be reused when the file is extended again. So we end
+ * up keeping around invalidated buffers attached to transactions'
+ * BJ_Forget list just to stop checkpointing code from cleaning up
+ * the transaction this buffer was modified in.
*/
transaction = jh->b_transaction;
if (transaction == NULL) {
@@ -1908,13 +1906,9 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
* committed, the buffer won't be needed any
* longer. */
JBUFFER_TRACE(jh, "checkpointed: add to BJ_Forget");
- ret = __dispose_buffer(jh,
+ may_free = __dispose_buffer(jh,
journal->j_running_transaction);
- jbd2_journal_put_journal_head(jh);
- spin_unlock(&journal->j_list_lock);
- jbd_unlock_bh_state(bh);
- write_unlock(&journal->j_state_lock);
- return ret;
+ goto zap_buffer;
} else {
/* There is no currently-running transaction. So the
* orphan record which we wrote for this file must have
@@ -1922,13 +1916,9 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
* the committing transaction, if it exists. */
if (journal->j_committing_transaction) {
JBUFFER_TRACE(jh, "give to committing trans");
- ret = __dispose_buffer(jh,
+ may_free = __dispose_buffer(jh,
journal->j_committing_transaction);
- jbd2_journal_put_journal_head(jh);
- spin_unlock(&journal->j_list_lock);
- jbd_unlock_bh_state(bh);
- write_unlock(&journal->j_state_lock);
- return ret;
+ goto zap_buffer;
} else {
/* The orphan record's transaction has
* committed. We can cleanse this buffer */
@@ -1940,10 +1930,21 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
JBUFFER_TRACE(jh, "on committing transaction");
/*
* The buffer is committing, we simply cannot touch
- * it. So we just set j_next_transaction to the
- * running transaction (if there is one) and mark
- * buffer as freed so that commit code knows it should
- * clear dirty bits when it is done with the buffer.
+ * it. If the page is straddling i_size we have to wait
+ * for commit and try again.
+ */
+ if (partial_page) {
+ jbd2_journal_put_journal_head(jh);
+ spin_unlock(&journal->j_list_lock);
+ jbd_unlock_bh_state(bh);
+ write_unlock(&journal->j_state_lock);
+ return -EBUSY;
+ }
+ /*
+ * OK, buffer won't be reachable after truncate. We just set
+ * j_next_transaction to the running transaction (if there is
+ * one) and mark buffer as freed so that commit code knows it
+ * should clear dirty bits when it is done with the buffer.
*/
set_buffer_freed(bh);
if (journal->j_running_transaction && buffer_jbddirty(bh))
@@ -1966,6 +1967,15 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
}
zap_buffer:
+ /*
+ * This is tricky. Although the buffer is truncated, it may be reused
+ * if blocksize < pagesize and it is attached to the page straddling
+ * EOF. Since the buffer might have been added to BJ_Forget list of the
+ * running transaction, journal_get_write_access() won't clear
+ * b_modified and credit accounting gets confused. So clear b_modified
+ * here.
+ */
+ jh->b_modified = 0;
jbd2_journal_put_journal_head(jh);
zap_buffer_no_jh:
spin_unlock(&journal->j_list_lock);
@@ -1989,21 +1999,23 @@ zap_buffer_unlocked:
* @page: page to flush
* @offset: length of page to invalidate.
*
- * Reap page buffers containing data after offset in page.
- *
+ * Reap page buffers containing data after offset in page. Can return -EBUSY
+ * if buffers are part of the committing transaction and the page is straddling
+ * i_size. Caller then has to wait for current commit and try again.
*/
-void jbd2_journal_invalidatepage(journal_t *journal,
- struct page *page,
- unsigned long offset)
+int jbd2_journal_invalidatepage(journal_t *journal,
+ struct page *page,
+ unsigned long offset)
{
struct buffer_head *head, *bh, *next;
unsigned int curr_off = 0;
int may_free = 1;
+ int ret = 0;
if (!PageLocked(page))
BUG();
if (!page_has_buffers(page))
- return;
+ return 0;
/* We will potentially be playing with lists other than just the
* data lists (especially for journaled data mode), so be
@@ -2017,8 +2029,11 @@ void jbd2_journal_invalidatepage(journal_t *journal,
if (offset <= curr_off) {
/* This block is wholly outside the truncation point */
lock_buffer(bh);
- may_free &= journal_unmap_buffer(journal, bh);
+ ret = journal_unmap_buffer(journal, bh, offset > 0);
unlock_buffer(bh);
+ if (ret < 0)
+ return ret;
+ may_free &= ret;
}
curr_off = next_off;
bh = next;
@@ -2029,6 +2044,7 @@ void jbd2_journal_invalidatepage(journal_t *journal,
if (may_free && try_to_free_buffers(page))
J_ASSERT(!page_has_buffers(page));
}
+ return 0;
}
/*