Commit e7746e31 authored by Eric Sandeen's avatar Eric Sandeen Committed by Ben Hutchings

ext4: disable synchronous transaction batching if max_batch_time==0

commit 5dd21424 upstream.

The mount manpage says of the max_batch_time option,

	This optimization can be turned off entirely
	by setting max_batch_time to 0.

But the code doesn't do that.  So fix the code to do
that.
Signed-off-by: default avatarEric Sandeen <sandeen@redhat.com>
Signed-off-by: default avatarTheodore Ts'o <tytso@mit.edu>
[bwh: Backported to 3.2: option parsing looks different]
Signed-off-by: default avatarBen Hutchings <ben@decadent.org.uk>
parent 5661f259
...@@ -1663,8 +1663,6 @@ static int parse_options(char *options, struct super_block *sb, ...@@ -1663,8 +1663,6 @@ static int parse_options(char *options, struct super_block *sb,
return 0; return 0;
if (option < 0) if (option < 0)
return 0; return 0;
if (option == 0)
option = EXT4_DEF_MAX_BATCH_TIME;
sbi->s_max_batch_time = option; sbi->s_max_batch_time = option;
break; break;
case Opt_min_batch_time: case Opt_min_batch_time:
......
...@@ -1388,9 +1388,12 @@ int jbd2_journal_stop(handle_t *handle) ...@@ -1388,9 +1388,12 @@ int jbd2_journal_stop(handle_t *handle)
* to perform a synchronous write. We do this to detect the * to perform a synchronous write. We do this to detect the
* case where a single process is doing a stream of sync * case where a single process is doing a stream of sync
* writes. No point in waiting for joiners in that case. * writes. No point in waiting for joiners in that case.
*
* Setting max_batch_time to 0 disables this completely.
*/ */
pid = current->pid; pid = current->pid;
if (handle->h_sync && journal->j_last_sync_writer != pid) { if (handle->h_sync && journal->j_last_sync_writer != pid &&
journal->j_max_batch_time) {
u64 commit_time, trans_time; u64 commit_time, trans_time;
journal->j_last_sync_writer = pid; journal->j_last_sync_writer = pid;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment