aboutsummaryrefslogtreecommitdiff
path: root/fs/jbd
diff options
context:
space:
mode:
Diffstat (limited to 'fs/jbd')
-rw-r--r--fs/jbd/commit.c27
-rw-r--r--fs/jbd/journal.c47
-rw-r--r--fs/jbd/revoke.c12
-rw-r--r--fs/jbd/transaction.c32
4 files changed, 82 insertions, 36 deletions
diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c
index 86b39b167c2..bb217dcb41a 100644
--- a/fs/jbd/commit.c
+++ b/fs/jbd/commit.c
@@ -162,8 +162,17 @@ static void journal_do_submit_data(struct buffer_head **wbuf, int bufs,
for (i = 0; i < bufs; i++) {
wbuf[i]->b_end_io = end_buffer_write_sync;
- /* We use-up our safety reference in submit_bh() */
- submit_bh(write_op, wbuf[i]);
+ /*
+ * Here we write back pagecache data that may be mmaped. Since
+ * we cannot afford to clean the page and set PageWriteback
+ * here due to lock ordering (page lock ranks above transaction
+ * start), the data can change while IO is in flight. Tell the
+ * block layer it should bounce the bio pages if stable data
+ * during write is required.
+ *
+ * We use up our safety reference in submit_bh().
+ */
+ _submit_bh(write_op, wbuf[i], 1 << BIO_SNAP_STABLE);
}
}
@@ -331,13 +340,13 @@ void journal_commit_transaction(journal_t *journal)
J_ASSERT(journal->j_committing_transaction == NULL);
commit_transaction = journal->j_running_transaction;
- J_ASSERT(commit_transaction->t_state == T_RUNNING);
trace_jbd_start_commit(journal, commit_transaction);
jbd_debug(1, "JBD: starting commit of transaction %d\n",
commit_transaction->t_tid);
spin_lock(&journal->j_state_lock);
+ J_ASSERT(commit_transaction->t_state == T_RUNNING);
commit_transaction->t_state = T_LOCKED;
trace_jbd_commit_locking(journal, commit_transaction);
@@ -667,7 +676,17 @@ start_journal_io:
clear_buffer_dirty(bh);
set_buffer_uptodate(bh);
bh->b_end_io = journal_end_buffer_io_sync;
- submit_bh(write_op, bh);
+ /*
+ * In data=journal mode, here we can end up
+ * writing pagecache data that might be
+ * mmapped. Since we can't afford to clean the
+ * page and set PageWriteback (see the comment
+ * near the other use of _submit_bh()), the
+ * data can change while the write is in
+ * flight. Tell the block layer to bounce the
+ * bio pages if stable pages are required.
+ */
+ _submit_bh(write_op, bh, 1 << BIO_SNAP_STABLE);
}
cond_resched();
diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index 81cc7eaff86..06fe11e0abf 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -90,6 +90,24 @@ static int journal_convert_superblock_v1(journal_t *, journal_superblock_t *);
static void __journal_abort_soft (journal_t *journal, int errno);
static const char *journal_dev_name(journal_t *journal, char *buffer);
+#ifdef CONFIG_JBD_DEBUG
+void __jbd_debug(int level, const char *file, const char *func,
+ unsigned int line, const char *fmt, ...)
+{
+ struct va_format vaf;
+ va_list args;
+
+ if (level > journal_enable_debug)
+ return;
+ va_start(args, fmt);
+ vaf.fmt = fmt;
+ vaf.va = &args;
+ printk(KERN_DEBUG "%s: (%s, %u): %pV\n", file, func, line, &vaf);
+ va_end(args);
+}
+EXPORT_SYMBOL(__jbd_debug);
+#endif
+
/*
* Helper function used to manage commit timeouts
*/
@@ -310,8 +328,6 @@ int journal_write_metadata_buffer(transaction_t *transaction,
new_bh = alloc_buffer_head(GFP_NOFS|__GFP_NOFAIL);
/* keep subsequent assertions sane */
- new_bh->b_state = 0;
- init_buffer(new_bh, NULL, NULL);
atomic_set(&new_bh->b_count, 1);
new_jh = journal_add_journal_head(new_bh); /* This sleeps */
@@ -557,13 +573,23 @@ int log_wait_commit(journal_t *journal, tid_t tid)
#ifdef CONFIG_JBD_DEBUG
spin_lock(&journal->j_state_lock);
if (!tid_geq(journal->j_commit_request, tid)) {
- printk(KERN_EMERG
+ printk(KERN_ERR
"%s: error: j_commit_request=%d, tid=%d\n",
__func__, journal->j_commit_request, tid);
}
spin_unlock(&journal->j_state_lock);
#endif
spin_lock(&journal->j_state_lock);
+ /*
+ * Not running or committing trans? Must be already committed. This
+ * saves us from waiting for a *long* time when tid overflows.
+ */
+ if (!((journal->j_running_transaction &&
+ journal->j_running_transaction->t_tid == tid) ||
+ (journal->j_committing_transaction &&
+ journal->j_committing_transaction->t_tid == tid)))
+ goto out_unlock;
+
if (!tid_geq(journal->j_commit_waited, tid))
journal->j_commit_waited = tid;
while (tid_gt(tid, journal->j_commit_sequence)) {
@@ -575,12 +601,11 @@ int log_wait_commit(journal_t *journal, tid_t tid)
!tid_gt(tid, journal->j_commit_sequence));
spin_lock(&journal->j_state_lock);
}
+out_unlock:
spin_unlock(&journal->j_state_lock);
- if (unlikely(is_journal_aborted(journal))) {
- printk(KERN_EMERG "journal commit I/O error\n");
+ if (unlikely(is_journal_aborted(journal)))
err = -EIO;
- }
return err;
}
@@ -1845,7 +1870,7 @@ static struct journal_head *journal_alloc_journal_head(void)
#ifdef CONFIG_JBD_DEBUG
atomic_inc(&nr_journal_heads);
#endif
- ret = kmem_cache_alloc(journal_head_cache, GFP_NOFS);
+ ret = kmem_cache_zalloc(journal_head_cache, GFP_NOFS);
if (ret == NULL) {
jbd_debug(1, "out of memory for journal_head\n");
printk_ratelimited(KERN_NOTICE "ENOMEM in %s, retrying.\n",
@@ -1853,7 +1878,7 @@ static struct journal_head *journal_alloc_journal_head(void)
while (ret == NULL) {
yield();
- ret = kmem_cache_alloc(journal_head_cache, GFP_NOFS);
+ ret = kmem_cache_zalloc(journal_head_cache, GFP_NOFS);
}
}
return ret;
@@ -1915,10 +1940,8 @@ struct journal_head *journal_add_journal_head(struct buffer_head *bh)
struct journal_head *new_jh = NULL;
repeat:
- if (!buffer_jbd(bh)) {
+ if (!buffer_jbd(bh))
new_jh = journal_alloc_journal_head();
- memset(new_jh, 0, sizeof(*new_jh));
- }
jbd_lock_bh_journal_head(bh);
if (buffer_jbd(bh)) {
@@ -2111,7 +2134,7 @@ static void __exit journal_exit(void)
#ifdef CONFIG_JBD_DEBUG
int n = atomic_read(&nr_journal_heads);
if (n)
- printk(KERN_EMERG "JBD: leaked %d journal_heads!\n", n);
+ printk(KERN_ERR "JBD: leaked %d journal_heads!\n", n);
#endif
jbd_remove_debugfs_entry();
journal_destroy_caches();
diff --git a/fs/jbd/revoke.c b/fs/jbd/revoke.c
index 25c713e7071..8898bbd2b61 100644
--- a/fs/jbd/revoke.c
+++ b/fs/jbd/revoke.c
@@ -231,19 +231,15 @@ record_cache_failure:
static struct jbd_revoke_table_s *journal_init_revoke_table(int hash_size)
{
- int shift = 0;
- int tmp = hash_size;
+ int i;
struct jbd_revoke_table_s *table;
table = kmem_cache_alloc(revoke_table_cache, GFP_KERNEL);
if (!table)
goto out;
- while((tmp >>= 1UL) != 0UL)
- shift++;
-
table->hash_size = hash_size;
- table->hash_shift = shift;
+ table->hash_shift = ilog2(hash_size);
table->hash_table =
kmalloc(hash_size * sizeof(struct list_head), GFP_KERNEL);
if (!table->hash_table) {
@@ -252,8 +248,8 @@ static struct jbd_revoke_table_s *journal_init_revoke_table(int hash_size)
goto out;
}
- for (tmp = 0; tmp < hash_size; tmp++)
- INIT_LIST_HEAD(&table->hash_table[tmp]);
+ for (i = 0; i < hash_size; i++)
+ INIT_LIST_HEAD(&table->hash_table[i]);
out:
return table;
diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c
index 071d6905f0d..1695ba8334a 100644
--- a/fs/jbd/transaction.c
+++ b/fs/jbd/transaction.c
@@ -26,7 +26,6 @@
#include <linux/mm.h>
#include <linux/highmem.h>
#include <linux/hrtimer.h>
-#include <linux/backing-dev.h>
static void __journal_temp_unlink_buffer(struct journal_head *jh);
@@ -100,10 +99,11 @@ static int start_this_handle(journal_t *journal, handle_t *handle)
alloc_transaction:
if (!journal->j_running_transaction) {
- new_transaction = kzalloc(sizeof(*new_transaction), GFP_NOFS);
+ new_transaction = kzalloc(sizeof(*new_transaction),
+ GFP_NOFS|__GFP_NOFAIL);
if (!new_transaction) {
- congestion_wait(BLK_RW_ASYNC, HZ/50);
- goto alloc_transaction;
+ ret = -ENOMEM;
+ goto out;
}
}
@@ -245,7 +245,6 @@ static handle_t *new_handle(int nblocks)
handle_t *handle = jbd_alloc_handle(GFP_NOFS);
if (!handle)
return NULL;
- memset(handle, 0, sizeof(*handle));
handle->h_buffer_credits = nblocks;
handle->h_ref = 1;
@@ -676,7 +675,7 @@ repeat:
jbd_alloc(jh2bh(jh)->b_size,
GFP_NOFS);
if (!frozen_buffer) {
- printk(KERN_EMERG
+ printk(KERN_ERR
"%s: OOM for frozen_buffer\n",
__func__);
JBUFFER_TRACE(jh, "oom!");
@@ -899,7 +898,7 @@ repeat:
if (!jh->b_committed_data) {
committed_data = jbd_alloc(jh2bh(jh)->b_size, GFP_NOFS);
if (!committed_data) {
- printk(KERN_EMERG "%s: No memory for committed data\n",
+ printk(KERN_ERR "%s: No memory for committed data\n",
__func__);
err = -ENOMEM;
goto out;
@@ -2020,16 +2019,20 @@ zap_buffer_unlocked:
* void journal_invalidatepage() - invalidate a journal page
* @journal: journal to use for flush
* @page: page to flush
- * @offset: length of page to invalidate.
+ * @offset: offset of the range to invalidate
+ * @length: length of the range to invalidate
*
- * Reap page buffers containing data after offset in page.
+ * Reap page buffers containing data in specified range in page.
*/
void journal_invalidatepage(journal_t *journal,
struct page *page,
- unsigned long offset)
+ unsigned int offset,
+ unsigned int length)
{
struct buffer_head *head, *bh, *next;
+ unsigned int stop = offset + length;
unsigned int curr_off = 0;
+ int partial_page = (offset || length < PAGE_CACHE_SIZE);
int may_free = 1;
if (!PageLocked(page))
@@ -2037,6 +2040,8 @@ void journal_invalidatepage(journal_t *journal,
if (!page_has_buffers(page))
return;
+ BUG_ON(stop > PAGE_CACHE_SIZE || stop < length);
+
/* We will potentially be playing with lists other than just the
* data lists (especially for journaled data mode), so be
* cautious in our locking. */
@@ -2046,11 +2051,14 @@ void journal_invalidatepage(journal_t *journal,
unsigned int next_off = curr_off + bh->b_size;
next = bh->b_this_page;
+ if (next_off > stop)
+ return;
+
if (offset <= curr_off) {
/* This block is wholly outside the truncation point */
lock_buffer(bh);
may_free &= journal_unmap_buffer(journal, bh,
- offset > 0);
+ partial_page);
unlock_buffer(bh);
}
curr_off = next_off;
@@ -2058,7 +2066,7 @@ void journal_invalidatepage(journal_t *journal,
} while (bh != head);
- if (!offset) {
+ if (!partial_page) {
if (may_free && try_to_free_buffers(page))
J_ASSERT(!page_has_buffers(page));
}
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-11-09 20:52:01 +0000