iocost: improve nr_lagging handling

Some IOs may span multiple periods.  As latencies are collected on
completion, the inbetween periods won't register them and may
incorrectly decide to increase vrate.  nr_lagging tracks these IOs to
avoid those situations.  Currently, whenever there are IOs which are
spanning from the previous period, busy_level is reset to 0 if
negative thus suppressing vrate increase.

This has the following two problems.

* When latency target percentiles aren't set, vrate adjustment should
  only be governed by queue depth depletion; however, the current code
  keeps nr_lagging active which pulls in latency results and can keep
  down vrate unexpectedly.

* When lagging condition is detected, it resets the entire negative
  busy_level.  This turned out to be way too aggressive on some
  devices which sometimes experience extended latencies on a small
  subset of commands.  In addition, a lagging IO will be accounted as
  latency target miss on completion anyway and resetting busy_level
  amplifies its impact unnecessarily.

This patch fixes the above two problems by disabling nr_lagging
counting when latency target percentiles aren't set and blocking vrate
increases when there are lagging IOs while leaving busy_level as-is.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

diff --git a/block/blk-iocost.c b/block/blk-iocost.c
index 32d4d6d86a99404633ebf27b8ddcf2030afedab8..10160de62e3b48b77c30430f38c35c658b49c5b0 100644 (file)
--- a/block/blk-iocost.c
+++ b/block/blk-iocost.c
@@ -1407,7 +1407,8 @@ static void ioc_timer_fn(struct timer_list *timer)
                 * comparing vdone against period start.  If lagging behind
                 * IOs from past periods, don't increase vrate.
                 */
-               if (!atomic_read(&iocg_to_blkg(iocg)->use_delay) &&
+               if ((ppm_rthr != MILLION || ppm_wthr != MILLION) &&
+                   !atomic_read(&iocg_to_blkg(iocg)->use_delay) &&
                    time_after64(vtime, vdone) &&
                    time_after64(vtime, now.vnow -
                                 MAX_LAGGING_PERIODS * period_vtime) &&
@@ -1537,21 +1538,23 @@ skip_surplus_transfers:
            missed_ppm[WRITE] > ppm_wthr) {
                ioc->busy_level = max(ioc->busy_level, 0);
                ioc->busy_level++;
-       } else if (nr_lagging) {
-               ioc->busy_level = max(ioc->busy_level, 0);
-       } else if (nr_shortages && !nr_surpluses &&
-                  rq_wait_pct <= RQ_WAIT_BUSY_PCT * UNBUSY_THR_PCT / 100 &&
+       } else if (rq_wait_pct <= RQ_WAIT_BUSY_PCT * UNBUSY_THR_PCT / 100 &&
                   missed_ppm[READ] <= ppm_rthr * UNBUSY_THR_PCT / 100 &&
                   missed_ppm[WRITE] <= ppm_wthr * UNBUSY_THR_PCT / 100) {
-               ioc->busy_level = min(ioc->busy_level, 0);
-               ioc->busy_level--;
+               /* take action iff there is contention */
+               if (nr_shortages && !nr_lagging) {
+                       ioc->busy_level = min(ioc->busy_level, 0);
+                       /* redistribute surpluses first */
+                       if (!nr_surpluses)
+                               ioc->busy_level--;
+               }
        } else {
                ioc->busy_level = 0;
        }
 
        ioc->busy_level = clamp(ioc->busy_level, -1000, 1000);
 
-       if (ioc->busy_level) {
+       if (ioc->busy_level > 0 || (ioc->busy_level < 0 && !nr_lagging)) {
                u64 vrate = atomic64_read(&ioc->vtime_rate);
                u64 vrate_min = ioc->vrate_min, vrate_max = ioc->vrate_max;