blk-throttle: ignore idle cgroup limit

Last patch introduces a way to detect idle cgroup. We use it to make
upgrade/downgrade decision. And the new algorithm can detect completely
idle cgroup too, so we can delete the corresponding code.

Signed-off-by: Shaohua Li <shli@fb.com>
Signed-off-by: Jens Axboe <axboe@fb.com>

diff --git a/block/blk-throttle.c b/block/blk-throttle.c
index f03e158ee1977544f89476ee5634ba0a46bc582b..0ea8698b2b1f3df0052e207d0cd784ef42e6f584 100644 (file)
--- a/block/blk-throttle.c
+++ b/block/blk-throttle.c
@@ -152,8 +152,6 @@ struct throtl_grp {
 
        unsigned long last_check_time;
 
-       unsigned long last_dispatch_time[2];
-
        /* When did we start a new slice */
        unsigned long slice_start[2];
        unsigned long slice_end[2];
@@ -508,8 +506,6 @@ static void throtl_pd_online(struct blkg_policy_data *pd)
         * Update has_rules[] after a new group is brought online.
         */
        tg_update_has_rules(tg);
-       tg->last_dispatch_time[READ] = jiffies;
-       tg->last_dispatch_time[WRITE] = jiffies;
 }
 
 static void blk_throtl_update_limit_valid(struct throtl_data *td)
@@ -1708,9 +1704,8 @@ static bool throtl_tg_can_upgrade(struct throtl_grp *tg)
                return true;
 
        if (time_after_eq(jiffies,
-            tg->last_dispatch_time[READ] + tg->td->throtl_slice) &&
-           time_after_eq(jiffies,
-            tg->last_dispatch_time[WRITE] + tg->td->throtl_slice))
+               tg_last_low_overflow_time(tg) + tg->td->throtl_slice) &&
+           throtl_tg_is_idle(tg))
                return true;
        return false;
 }
@@ -1756,6 +1751,26 @@ static bool throtl_can_upgrade(struct throtl_data *td,
        return true;
 }
 
+static void throtl_upgrade_check(struct throtl_grp *tg)
+{
+       unsigned long now = jiffies;
+
+       if (tg->td->limit_index != LIMIT_LOW)
+               return;
+
+       if (time_after(tg->last_check_time + tg->td->throtl_slice, now))
+               return;
+
+       tg->last_check_time = now;
+
+       if (!time_after_eq(now,
+            __tg_last_low_overflow_time(tg) + tg->td->throtl_slice))
+               return;
+
+       if (throtl_can_upgrade(tg->td, NULL))
+               throtl_upgrade_state(tg->td);
+}
+
 static void throtl_upgrade_state(struct throtl_data *td)
 {
        struct cgroup_subsys_state *pos_css;
@@ -1797,18 +1812,15 @@ static bool throtl_tg_can_downgrade(struct throtl_grp *tg)
        struct throtl_data *td = tg->td;
        unsigned long now = jiffies;
 
-       if (time_after_eq(now, tg->last_dispatch_time[READ] +
-                                       td->throtl_slice) &&
-           time_after_eq(now, tg->last_dispatch_time[WRITE] +
-                                       td->throtl_slice))
-               return false;
        /*
         * If cgroup is below low limit, consider downgrade and throttle other
         * cgroups
         */
        if (time_after_eq(now, td->low_upgrade_time + td->throtl_slice) &&
            time_after_eq(now, tg_last_low_overflow_time(tg) +
-                                       td->throtl_slice))
+                                       td->throtl_slice) &&
+           (!throtl_tg_is_idle(tg) ||
+            !list_empty(&tg_to_blkg(tg)->blkcg->css.children)))
                return true;
        return false;
 }
@@ -1931,10 +1943,10 @@ bool blk_throtl_bio(struct request_queue *q, struct blkcg_gq *blkg,
 
 again:
        while (true) {
-               tg->last_dispatch_time[rw] = jiffies;
                if (tg->last_low_overflow_time[rw] == 0)
                        tg->last_low_overflow_time[rw] = jiffies;
                throtl_downgrade_check(tg);
+               throtl_upgrade_check(tg);
                /* throtl is FIFO - if bios are already queued, should queue */
                if (sq->nr_queued[rw])
                        break;