aboutsummaryrefslogtreecommitdiffstats
path: root/fs/jbd2
diff options
context:
space:
mode:
Diffstat (limited to 'fs/jbd2')
-rw-r--r--fs/jbd2/checkpoint.c7
-rw-r--r--fs/jbd2/journal.c67
-rw-r--r--fs/jbd2/revoke.c15
-rw-r--r--fs/jbd2/transaction.c270
4 files changed, 205 insertions, 154 deletions
diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c
index 988b32ed4c87..4227dc4f7437 100644
--- a/fs/jbd2/checkpoint.c
+++ b/fs/jbd2/checkpoint.c
@@ -390,7 +390,7 @@ int jbd2_cleanup_journal_tail(journal_t *journal)
unsigned long blocknr;
if (is_journal_aborted(journal))
- return 1;
+ return -EIO;
if (!jbd2_journal_get_log_tail(journal, &first_tid, &blocknr))
return 1;
@@ -405,10 +405,9 @@ int jbd2_cleanup_journal_tail(journal_t *journal)
* jbd2_cleanup_journal_tail() doesn't get called all that often.
*/
if (journal->j_flags & JBD2_BARRIER)
- blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL);
+ blkdev_issue_flush(journal->j_fs_dev, GFP_NOFS, NULL);
- __jbd2_update_log_tail(journal, first_tid, blocknr);
- return 0;
+ return __jbd2_update_log_tail(journal, first_tid, blocknr);
}
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index b96bd8076b70..4ff3fad4e9e3 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -371,16 +371,7 @@ int jbd2_journal_write_metadata_buffer(transaction_t *transaction,
*/
J_ASSERT_BH(bh_in, buffer_jbddirty(bh_in));
-retry_alloc:
- new_bh = alloc_buffer_head(GFP_NOFS);
- if (!new_bh) {
- /*
- * Failure is not an option, but __GFP_NOFAIL is going
- * away; so we retry ourselves here.
- */
- congestion_wait(BLK_RW_ASYNC, HZ/50);
- goto retry_alloc;
- }
+ new_bh = alloc_buffer_head(GFP_NOFS|__GFP_NOFAIL);
/* keep subsequent assertions sane */
atomic_set(&new_bh->b_count, 1);
@@ -885,9 +876,10 @@ int jbd2_journal_get_log_tail(journal_t *journal, tid_t *tid,
*
* Requires j_checkpoint_mutex
*/
-void __jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block)
+int __jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block)
{
unsigned long freed;
+ int ret;
BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex));
@@ -897,7 +889,10 @@ void __jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block)
* space and if we lose sb update during power failure we'd replay
* old transaction with possibly newly overwritten data.
*/
- jbd2_journal_update_sb_log_tail(journal, tid, block, WRITE_FUA);
+ ret = jbd2_journal_update_sb_log_tail(journal, tid, block, WRITE_FUA);
+ if (ret)
+ goto out;
+
write_lock(&journal->j_state_lock);
freed = block - journal->j_tail;
if (block < journal->j_tail)
@@ -913,6 +908,9 @@ void __jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block)
journal->j_tail_sequence = tid;
journal->j_tail = block;
write_unlock(&journal->j_state_lock);
+
+out:
+ return ret;
}
/*
@@ -1137,7 +1135,6 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev,
{
journal_t *journal = journal_init_common();
struct buffer_head *bh;
- char *p;
int n;
if (!journal)
@@ -1150,9 +1147,7 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev,
journal->j_blk_offset = start;
journal->j_maxlen = len;
bdevname(journal->j_dev, journal->j_devname);
- p = journal->j_devname;
- while ((p = strchr(p, '/')))
- *p = '!';
+ strreplace(journal->j_devname, '/', '!');
jbd2_stats_proc_init(journal);
n = journal->j_blocksize / sizeof(journal_block_tag_t);
journal->j_wbufsize = n;
@@ -1204,10 +1199,7 @@ journal_t * jbd2_journal_init_inode (struct inode *inode)
journal->j_dev = journal->j_fs_dev = inode->i_sb->s_bdev;
journal->j_inode = inode;
bdevname(journal->j_dev, journal->j_devname);
- p = journal->j_devname;
- while ((p = strchr(p, '/')))
- *p = '!';
- p = journal->j_devname + strlen(journal->j_devname);
+ p = strreplace(journal->j_devname, '/', '!');
sprintf(p, "-%lu", journal->j_inode->i_ino);
jbd_debug(1,
"journal %p: inode %s/%ld, size %Ld, bits %d, blksize %ld\n",
@@ -1331,7 +1323,7 @@ static int journal_reset(journal_t *journal)
return jbd2_journal_start_thread(journal);
}
-static void jbd2_write_superblock(journal_t *journal, int write_op)
+static int jbd2_write_superblock(journal_t *journal, int write_op)
{
struct buffer_head *bh = journal->j_sb_buffer;
journal_superblock_t *sb = journal->j_superblock;
@@ -1370,7 +1362,10 @@ static void jbd2_write_superblock(journal_t *journal, int write_op)
printk(KERN_ERR "JBD2: Error %d detected when updating "
"journal superblock for %s.\n", ret,
journal->j_devname);
+ jbd2_journal_abort(journal, ret);
}
+
+ return ret;
}
/**
@@ -1383,10 +1378,11 @@ static void jbd2_write_superblock(journal_t *journal, int write_op)
* Update a journal's superblock information about log tail and write it to
* disk, waiting for the IO to complete.
*/
-void jbd2_journal_update_sb_log_tail(journal_t *journal, tid_t tail_tid,
+int jbd2_journal_update_sb_log_tail(journal_t *journal, tid_t tail_tid,
unsigned long tail_block, int write_op)
{
journal_superblock_t *sb = journal->j_superblock;
+ int ret;
BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex));
jbd_debug(1, "JBD2: updating superblock (start %lu, seq %u)\n",
@@ -1395,13 +1391,18 @@ void jbd2_journal_update_sb_log_tail(journal_t *journal, tid_t tail_tid,
sb->s_sequence = cpu_to_be32(tail_tid);
sb->s_start = cpu_to_be32(tail_block);
- jbd2_write_superblock(journal, write_op);
+ ret = jbd2_write_superblock(journal, write_op);
+ if (ret)
+ goto out;
/* Log is no longer empty */
write_lock(&journal->j_state_lock);
WARN_ON(!sb->s_sequence);
journal->j_flags &= ~JBD2_FLUSHED;
write_unlock(&journal->j_state_lock);
+
+out:
+ return ret;
}
/**
@@ -1950,7 +1951,14 @@ int jbd2_journal_flush(journal_t *journal)
return -EIO;
mutex_lock(&journal->j_checkpoint_mutex);
- jbd2_cleanup_journal_tail(journal);
+ if (!err) {
+ err = jbd2_cleanup_journal_tail(journal);
+ if (err < 0) {
+ mutex_unlock(&journal->j_checkpoint_mutex);
+ goto out;
+ }
+ err = 0;
+ }
/* Finally, mark the journal as really needing no recovery.
* This sets s_start==0 in the underlying superblock, which is
@@ -1966,7 +1974,8 @@ int jbd2_journal_flush(journal_t *journal)
J_ASSERT(journal->j_head == journal->j_tail);
J_ASSERT(journal->j_tail_sequence == journal->j_transaction_sequence);
write_unlock(&journal->j_state_lock);
- return 0;
+out:
+ return err;
}
/**
@@ -2330,7 +2339,7 @@ static int jbd2_journal_init_journal_head_cache(void)
jbd2_journal_head_cache = kmem_cache_create("jbd2_journal_head",
sizeof(struct journal_head),
0, /* offset */
- SLAB_TEMPORARY, /* flags */
+ SLAB_TEMPORARY | SLAB_DESTROY_BY_RCU,
NULL); /* ctor */
retval = 0;
if (!jbd2_journal_head_cache) {
@@ -2362,10 +2371,8 @@ static struct journal_head *journal_alloc_journal_head(void)
if (!ret) {
jbd_debug(1, "out of memory for journal_head\n");
pr_notice_ratelimited("ENOMEM in %s, retrying.\n", __func__);
- while (!ret) {
- yield();
- ret = kmem_cache_zalloc(jbd2_journal_head_cache, GFP_NOFS);
- }
+ ret = kmem_cache_zalloc(jbd2_journal_head_cache,
+ GFP_NOFS | __GFP_NOFAIL);
}
return ret;
}
diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c
index 14214da80eb8..0abf2e7f725b 100644
--- a/fs/jbd2/revoke.c
+++ b/fs/jbd2/revoke.c
@@ -141,11 +141,13 @@ static int insert_revoke_hash(journal_t *journal, unsigned long long blocknr,
{
struct list_head *hash_list;
struct jbd2_revoke_record_s *record;
+ gfp_t gfp_mask = GFP_NOFS;
-repeat:
- record = kmem_cache_alloc(jbd2_revoke_record_cache, GFP_NOFS);
+ if (journal_oom_retry)
+ gfp_mask |= __GFP_NOFAIL;
+ record = kmem_cache_alloc(jbd2_revoke_record_cache, gfp_mask);
if (!record)
- goto oom;
+ return -ENOMEM;
record->sequence = seq;
record->blocknr = blocknr;
@@ -154,13 +156,6 @@ repeat:
list_add(&record->hash, hash_list);
spin_unlock(&journal->j_revoke_lock);
return 0;
-
-oom:
- if (!journal_oom_retry)
- return -ENOMEM;
- jbd_debug(1, "ENOMEM in %s, retrying\n", __func__);
- yield();
- goto repeat;
}
/* Find a revoke record in the journal's hash table. */
diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index ff2f2e6ad311..f3d06174b051 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -278,22 +278,16 @@ static int start_this_handle(journal_t *journal, handle_t *handle,
alloc_transaction:
if (!journal->j_running_transaction) {
+ /*
+ * If __GFP_FS is not present, then we may be being called from
+ * inside the fs writeback layer, so we MUST NOT fail.
+ */
+ if ((gfp_mask & __GFP_FS) == 0)
+ gfp_mask |= __GFP_NOFAIL;
new_transaction = kmem_cache_zalloc(transaction_cache,
gfp_mask);
- if (!new_transaction) {
- /*
- * If __GFP_FS is not present, then we may be
- * being called from inside the fs writeback
- * layer, so we MUST NOT fail. Since
- * __GFP_NOFAIL is going away, we will arrange
- * to retry the allocation ourselves.
- */
- if ((gfp_mask & __GFP_FS) == 0) {
- congestion_wait(BLK_RW_ASYNC, HZ/50);
- goto alloc_transaction;
- }
+ if (!new_transaction)
return -ENOMEM;
- }
}
jbd_debug(3, "New handle %p going live.\n", handle);
@@ -761,6 +755,30 @@ static void warn_dirty_buffer(struct buffer_head *bh)
bdevname(bh->b_bdev, b), (unsigned long long)bh->b_blocknr);
}
+/* Call t_frozen trigger and copy buffer data into jh->b_frozen_data. */
+static void jbd2_freeze_jh_data(struct journal_head *jh)
+{
+ struct page *page;
+ int offset;
+ char *source;
+ struct buffer_head *bh = jh2bh(jh);
+
+ J_EXPECT_JH(jh, buffer_uptodate(bh), "Possible IO failure.\n");
+ page = bh->b_page;
+ offset = offset_in_page(bh->b_data);
+ source = kmap_atomic(page);
+ /* Fire data frozen trigger just before we copy the data */
+ jbd2_buffer_frozen_trigger(jh, source + offset, jh->b_triggers);
+ memcpy(jh->b_frozen_data, source + offset, bh->b_size);
+ kunmap_atomic(source);
+
+ /*
+ * Now that the frozen data is saved off, we need to store any matching
+ * triggers.
+ */
+ jh->b_frozen_triggers = jh->b_triggers;
+}
+
/*
* If the buffer is already part of the current transaction, then there
* is nothing we need to do. If it is already part of a prior
@@ -780,7 +798,6 @@ do_get_write_access(handle_t *handle, struct journal_head *jh,
journal_t *journal;
int error;
char *frozen_buffer = NULL;
- int need_copy = 0;
unsigned long start_lock, time_lock;
if (is_handle_aborted(handle))
@@ -867,119 +884,96 @@ repeat:
jh->b_modified = 0;
/*
+ * If the buffer is not journaled right now, we need to make sure it
+ * doesn't get written to disk before the caller actually commits the
+ * new data
+ */
+ if (!jh->b_transaction) {
+ JBUFFER_TRACE(jh, "no transaction");
+ J_ASSERT_JH(jh, !jh->b_next_transaction);
+ JBUFFER_TRACE(jh, "file as BJ_Reserved");
+ /*
+ * Make sure all stores to jh (b_modified, b_frozen_data) are
+ * visible before attaching it to the running transaction.
+ * Paired with barrier in jbd2_write_access_granted()
+ */
+ smp_wmb();
+ spin_lock(&journal->j_list_lock);
+ __jbd2_journal_file_buffer(jh, transaction, BJ_Reserved);
+ spin_unlock(&journal->j_list_lock);
+ goto done;
+ }
+ /*
* If there is already a copy-out version of this buffer, then we don't
* need to make another one
*/
if (jh->b_frozen_data) {
JBUFFER_TRACE(jh, "has frozen data");
J_ASSERT_JH(jh, jh->b_next_transaction == NULL);
- jh->b_next_transaction = transaction;
- goto done;
+ goto attach_next;
}
- /* Is there data here we need to preserve? */
+ JBUFFER_TRACE(jh, "owned by older transaction");
+ J_ASSERT_JH(jh, jh->b_next_transaction == NULL);
+ J_ASSERT_JH(jh, jh->b_transaction == journal->j_committing_transaction);
- if (jh->b_transaction && jh->b_transaction != transaction) {
- JBUFFER_TRACE(jh, "owned by older transaction");
- J_ASSERT_JH(jh, jh->b_next_transaction == NULL);
- J_ASSERT_JH(jh, jh->b_transaction ==
- journal->j_committing_transaction);
+ /*
+ * There is one case we have to be very careful about. If the
+ * committing transaction is currently writing this buffer out to disk
+ * and has NOT made a copy-out, then we cannot modify the buffer
+ * contents at all right now. The essence of copy-out is that it is
+ * the extra copy, not the primary copy, which gets journaled. If the
+ * primary copy is already going to disk then we cannot do copy-out
+ * here.
+ */
+ if (buffer_shadow(bh)) {
+ JBUFFER_TRACE(jh, "on shadow: sleep");
+ jbd_unlock_bh_state(bh);
+ wait_on_bit_io(&bh->b_state, BH_Shadow, TASK_UNINTERRUPTIBLE);
+ goto repeat;
+ }
- /* There is one case we have to be very careful about.
- * If the committing transaction is currently writing
- * this buffer out to disk and has NOT made a copy-out,
- * then we cannot modify the buffer contents at all
- * right now. The essence of copy-out is that it is the
- * extra copy, not the primary copy, which gets
- * journaled. If the primary copy is already going to
- * disk then we cannot do copy-out here. */
-
- if (buffer_shadow(bh)) {
- JBUFFER_TRACE(jh, "on shadow: sleep");
+ /*
+ * Only do the copy if the currently-owning transaction still needs it.
+ * If buffer isn't on BJ_Metadata list, the committing transaction is
+ * past that stage (here we use the fact that BH_Shadow is set under
+ * bh_state lock together with refiling to BJ_Shadow list and at this
+ * point we know the buffer doesn't have BH_Shadow set).
+ *
+ * Subtle point, though: if this is a get_undo_access, then we will be
+ * relying on the frozen_data to contain the new value of the
+ * committed_data record after the transaction, so we HAVE to force the
+ * frozen_data copy in that case.
+ */
+ if (jh->b_jlist == BJ_Metadata || force_copy) {
+ JBUFFER_TRACE(jh, "generate frozen data");
+ if (!frozen_buffer) {
+ JBUFFER_TRACE(jh, "allocate memory for buffer");
jbd_unlock_bh_state(bh);
- wait_on_bit_io(&bh->b_state, BH_Shadow,
- TASK_UNINTERRUPTIBLE);
- goto repeat;
- }
-
- /*
- * Only do the copy if the currently-owning transaction still
- * needs it. If buffer isn't on BJ_Metadata list, the
- * committing transaction is past that stage (here we use the
- * fact that BH_Shadow is set under bh_state lock together with
- * refiling to BJ_Shadow list and at this point we know the
- * buffer doesn't have BH_Shadow set).
- *
- * Subtle point, though: if this is a get_undo_access,
- * then we will be relying on the frozen_data to contain
- * the new value of the committed_data record after the
- * transaction, so we HAVE to force the frozen_data copy
- * in that case.
- */
- if (jh->b_jlist == BJ_Metadata || force_copy) {
- JBUFFER_TRACE(jh, "generate frozen data");
+ frozen_buffer = jbd2_alloc(jh2bh(jh)->b_size, GFP_NOFS);
if (!frozen_buffer) {
- JBUFFER_TRACE(jh, "allocate memory for buffer");
- jbd_unlock_bh_state(bh);
- frozen_buffer =
- jbd2_alloc(jh2bh(jh)->b_size,
- GFP_NOFS);
- if (!frozen_buffer) {
- printk(KERN_ERR
- "%s: OOM for frozen_buffer\n",
- __func__);
- JBUFFER_TRACE(jh, "oom!");
- error = -ENOMEM;
- jbd_lock_bh_state(bh);
- goto done;
- }
- goto repeat;
+ printk(KERN_ERR "%s: OOM for frozen_buffer\n",
+ __func__);
+ JBUFFER_TRACE(jh, "oom!");
+ error = -ENOMEM;
+ goto out;
}
- jh->b_frozen_data = frozen_buffer;
- frozen_buffer = NULL;
- need_copy = 1;
+ goto repeat;
}
- jh->b_next_transaction = transaction;
+ jh->b_frozen_data = frozen_buffer;
+ frozen_buffer = NULL;
+ jbd2_freeze_jh_data(jh);
}
-
-
+attach_next:
/*
- * Finally, if the buffer is not journaled right now, we need to make
- * sure it doesn't get written to disk before the caller actually
- * commits the new data
+ * Make sure all stores to jh (b_modified, b_frozen_data) are visible
+ * before attaching it to the running transaction. Paired with barrier
+ * in jbd2_write_access_granted()
*/
- if (!jh->b_transaction) {
- JBUFFER_TRACE(jh, "no transaction");
- J_ASSERT_JH(jh, !jh->b_next_transaction);
- JBUFFER_TRACE(jh, "file as BJ_Reserved");
- spin_lock(&journal->j_list_lock);
- __jbd2_journal_file_buffer(jh, transaction, BJ_Reserved);
- spin_unlock(&journal->j_list_lock);
- }
+ smp_wmb();
+ jh->b_next_transaction = transaction;
done:
- if (need_copy) {
- struct page *page;
- int offset;
- char *source;
-
- J_EXPECT_JH(jh, buffer_uptodate(jh2bh(jh)),
- "Possible IO failure.\n");
- page = jh2bh(jh)->b_page;
- offset = offset_in_page(jh2bh(jh)->b_data);
- source = kmap_atomic(page);
- /* Fire data frozen trigger just before we copy the data */
- jbd2_buffer_frozen_trigger(jh, source + offset,
- jh->b_triggers);
- memcpy(jh->b_frozen_data, source+offset, jh2bh(jh)->b_size);
- kunmap_atomic(source);
-
- /*
- * Now that the frozen data is saved off, we need to store
- * any matching triggers.
- */
- jh->b_frozen_triggers = jh->b_triggers;
- }
jbd_unlock_bh_state(bh);
/*
@@ -996,6 +990,55 @@ out:
return error;
}
+/* Fast check whether buffer is already attached to the required transaction */
+static bool jbd2_write_access_granted(handle_t *handle, struct buffer_head *bh)
+{
+ struct journal_head *jh;
+ bool ret = false;
+
+ /* Dirty buffers require special handling... */
+ if (buffer_dirty(bh))
+ return false;
+
+ /*
+ * RCU protects us from dereferencing freed pages. So the checks we do
+ * are guaranteed not to oops. However the jh slab object can get freed
+ * & reallocated while we work with it. So we have to be careful. When
+ * we see jh attached to the running transaction, we know it must stay
+ * so until the transaction is committed. Thus jh won't be freed and
+ * will be attached to the same bh while we run. However it can
+ * happen jh gets freed, reallocated, and attached to the transaction
+ * just after we get pointer to it from bh. So we have to be careful
+ * and recheck jh still belongs to our bh before we return success.
+ */
+ rcu_read_lock();
+ if (!buffer_jbd(bh))
+ goto out;
+ /* This should be bh2jh() but that doesn't work with inline functions */
+ jh = READ_ONCE(bh->b_private);
+ if (!jh)
+ goto out;
+ if (jh->b_transaction != handle->h_transaction &&
+ jh->b_next_transaction != handle->h_transaction)
+ goto out;
+ /*
+ * There are two reasons for the barrier here:
+ * 1) Make sure to fetch b_bh after we did previous checks so that we
+ * detect when jh went through free, realloc, attach to transaction
+ * while we were checking. Paired with implicit barrier in that path.
+ * 2) So that access to bh done after jbd2_write_access_granted()
+ * doesn't get reordered and see inconsistent state of concurrent
+ * do_get_write_access().
+ */
+ smp_mb();
+ if (unlikely(jh->b_bh != bh))
+ goto out;
+ ret = true;
+out:
+ rcu_read_unlock();
+ return ret;
+}
+
/**
* int jbd2_journal_get_write_access() - notify intent to modify a buffer for metadata (not data) update.
* @handle: transaction to add buffer modifications to
@@ -1009,9 +1052,13 @@ out:
int jbd2_journal_get_write_access(handle_t *handle, struct buffer_head *bh)
{
- struct journal_head *jh = jbd2_journal_add_journal_head(bh);
+ struct journal_head *jh;
int rc;
+ if (jbd2_write_access_granted(handle, bh))
+ return 0;
+
+ jh = jbd2_journal_add_journal_head(bh);
/* We do not want to get caught playing with fields which the
* log thread also manipulates. Make sure that the buffer
* completes any outstanding IO before proceeding. */
@@ -1141,11 +1188,14 @@ out:
int jbd2_journal_get_undo_access(handle_t *handle, struct buffer_head *bh)
{
int err;
- struct journal_head *jh = jbd2_journal_add_journal_head(bh);
+ struct journal_head *jh;
char *committed_data = NULL;
JBUFFER_TRACE(jh, "entry");
+ if (jbd2_write_access_granted(handle, bh))
+ return 0;
+ jh = jbd2_journal_add_journal_head(bh);
/*
* Do this first --- it can drop the journal lock, so we want to
* make sure that obtaining the committed_data is done
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.