From: Tejun Heo <tj@kernel.org>
To: axboe@kernel.dk
Cc: linux-block@vger.kernel.org, cgroups@vger.kernel.org,
kernel-team@fb.com, linux-kernel@vger.kernel.org,
Tejun Heo <tj@kernel.org>
Subject: [PATCH 1/5] iocost: factor out ioc_forgive_debts()
Date: Thu, 17 Sep 2020 20:44:52 -0400 [thread overview]
Message-ID: <20200918004456.593983-2-tj@kernel.org> (raw)
In-Reply-To: <20200918004456.593983-1-tj@kernel.org>
Debt reduction logic is going to be improved and expanded. Factor it out
into ioc_forgive_debts() and generalize the comment a bit. No functional
change.
Signed-off-by: Tejun Heo <tj@kernel.org>
---
block/blk-iocost.c | 66 ++++++++++++++++++++++++----------------------
1 file changed, 35 insertions(+), 31 deletions(-)
diff --git a/block/blk-iocost.c b/block/blk-iocost.c
index ef9476fca1d8..bbf30bb06c07 100644
--- a/block/blk-iocost.c
+++ b/block/blk-iocost.c
@@ -1979,6 +1979,40 @@ static void transfer_surpluses(struct list_head *surpluses, struct ioc_now *now)
list_del_init(&iocg->walk_list);
}
+/*
+ * A low weight iocg can amass a large amount of debt, for example, when
+ * anonymous memory gets reclaimed aggressively. If the system has a lot of
+ * memory paired with a slow IO device, the debt can span multiple seconds or
+ * more. If there are no other subsequent IO issuers, the in-debt iocg may end
+ * up blocked paying its debt while the IO device is idle.
+ *
+ * The following protects against such cases. If the device has been
+ * sufficiently idle for a while, the debts are halved.
+ */
+static void ioc_forgive_debts(struct ioc *ioc, u64 usage_us_sum, int nr_debtors,
+ int nr_shortages, struct ioc_now *now)
+{
+ if (nr_shortages ||
+ div64_u64(100 * usage_us_sum, now->now - ioc->period_at) >=
+ DEBT_BUSY_USAGE_PCT)
+ ioc->debt_busy_at = now->now;
+
+ if (nr_debtors &&
+ now->now - ioc->debt_busy_at >= DEBT_REDUCTION_IDLE_DUR) {
+ struct ioc_gq *iocg;
+
+ list_for_each_entry(iocg, &ioc->active_iocgs, active_list) {
+ if (iocg->abs_vdebt) {
+ spin_lock(&iocg->waitq.lock);
+ iocg->abs_vdebt /= 2;
+ iocg_kick_waitq(iocg, true, now);
+ spin_unlock(&iocg->waitq.lock);
+ }
+ }
+ ioc->debt_busy_at = now->now;
+ }
+}
+
static void ioc_timer_fn(struct timer_list *timer)
{
struct ioc *ioc = container_of(timer, struct ioc, timer);
@@ -2171,37 +2205,7 @@ static void ioc_timer_fn(struct timer_list *timer)
list_for_each_entry_safe(iocg, tiocg, &surpluses, surplus_list)
list_del_init(&iocg->surplus_list);
- /*
- * A low weight iocg can amass a large amount of debt, for example, when
- * anonymous memory gets reclaimed aggressively. If the system has a lot
- * of memory paired with a slow IO device, the debt can span multiple
- * seconds or more. If there are no other subsequent IO issuers, the
- * in-debt iocg may end up blocked paying its debt while the IO device
- * is idle.
- *
- * The following protects against such pathological cases. If the device
- * has been sufficiently idle for a substantial amount of time, the
- * debts are halved. The criteria are on the conservative side as we
- * want to resolve the rare extreme cases without impacting regular
- * operation by forgiving debts too readily.
- */
- if (nr_shortages ||
- div64_u64(100 * usage_us_sum, now.now - ioc->period_at) >=
- DEBT_BUSY_USAGE_PCT)
- ioc->debt_busy_at = now.now;
-
- if (nr_debtors &&
- now.now - ioc->debt_busy_at >= DEBT_REDUCTION_IDLE_DUR) {
- list_for_each_entry(iocg, &ioc->active_iocgs, active_list) {
- if (iocg->abs_vdebt) {
- spin_lock(&iocg->waitq.lock);
- iocg->abs_vdebt /= 2;
- iocg_kick_waitq(iocg, true, &now);
- spin_unlock(&iocg->waitq.lock);
- }
- }
- ioc->debt_busy_at = now.now;
- }
+ ioc_forgive_debts(ioc, usage_us_sum, nr_debtors, nr_shortages, &now);
/*
* If q is getting clogged or we're missing too much, we're issuing
--
2.26.2
next prev parent reply other threads:[~2020-09-18 0:45 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-18 0:44 [PATCHSET for-5.10/block] iocost: improve debt forgiveness logic Tejun Heo
2020-09-18 0:44 ` Tejun Heo [this message]
2020-09-18 0:44 ` [PATCH 2/5] iocost: replace nr_shortages cond in ioc_forgive_debts() with busy_level one Tejun Heo
2020-09-18 0:44 ` [PATCH 3/5] iocost: recalculate delay after debt reduction Tejun Heo
2020-09-18 0:44 ` [PATCH 4/5] iocost: reimplement debt forgiveness using average usage Tejun Heo
2020-09-18 0:44 ` [PATCH 5/5] iocost: add iocg_forgive_debt tracepoint Tejun Heo
2020-09-18 18:41 ` [PATCH 6/5] iocost: consider iocgs with active delays for debt forgiveness Tejun Heo
2020-09-25 14:35 ` [PATCHSET for-5.10/block] iocost: improve debt forgiveness logic Jens Axboe
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200918004456.593983-2-tj@kernel.org \
--to=tj@kernel.org \
--cc=axboe@kernel.dk \
--cc=cgroups@vger.kernel.org \
--cc=kernel-team@fb.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).