diff options
| author | Theodore Ts'o <tytso@mit.edu> | 2013-02-06 22:30:23 -0500 |
|---|---|---|
| committer | Theodore Ts'o <tytso@mit.edu> | 2013-02-06 22:30:23 -0500 |
| commit | 9fff24aa2c5c504aadead1ff9599e813604c2e53 (patch) | |
| tree | 24bbb1691ec816aed648a895882526ce361ec92f /include/linux/jbd2.h | |
| parent | 40ae3487628235e5f1eb27542cca0cdb6e5dbe16 (diff) | |
jbd2: track request delay statistics
Track the delay between when we first request that the commit begin
and when it actually begins, so we can see how much of a gap exists.
In theory, this should just be the remaining scheduling quantuum of
the thread which requested the commit (assuming it was not a
synchronous operation which triggered the commit request) plus
scheduling overhead; however, it's possible that real time processes
might get in the way of letting the kjournald thread from executing.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Diffstat (limited to 'include/linux/jbd2.h')
| -rw-r--r-- | include/linux/jbd2.h | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index e30b66346942..e0aafc46064f 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -581,6 +581,11 @@ struct transaction_s unsigned long t_start; /* + * When commit was requested + */ + unsigned long t_requested; + + /* * Checkpointing stats [j_checkpoint_sem] */ struct transaction_chp_stats_s t_chp_stats; @@ -637,6 +642,7 @@ struct transaction_s struct transaction_run_stats_s { unsigned long rs_wait; + unsigned long rs_request_delay; unsigned long rs_running; unsigned long rs_locked; unsigned long rs_flushing; @@ -649,6 +655,7 @@ struct transaction_run_stats_s { struct transaction_stats_s { unsigned long ts_tid; + unsigned long ts_requested; struct transaction_run_stats_s run; }; |