lkml.org 
[lkml]   [2019]   [Sep]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
Subject[PATCH 1/3 for-5.4/block] iocost: better trace vrate changes
vrate_adj tracepoint traces vrate changes; however, it does so only
when busy_level is non-zero. busy_level turning to zero can sometimes
be as interesting an event. This patch also enables vrate_adj
tracepoint on other vrate related events - busy_level changes and
non-zero nr_lagging.

Signed-off-by: Tejun Heo <tj@kernel.org>
---
Hello, Jens.

I've encountered vrate regulation issues while testing on a hard disk
machine. These are three patches to improve vrate adj visibility and
fix the issue.

Thanks.

block/blk-iocost.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)

--- a/block/blk-iocost.c
+++ b/block/blk-iocost.c
@@ -1343,7 +1343,7 @@ static void ioc_timer_fn(struct timer_li
u32 ppm_wthr = MILLION - ioc->params.qos[QOS_WPPM];
u32 missed_ppm[2], rq_wait_pct;
u64 period_vtime;
- int i;
+ int prev_busy_level, i;

/* how were the latencies during the period? */
ioc_lat_stat(ioc, missed_ppm, &rq_wait_pct);
@@ -1531,6 +1531,7 @@ skip_surplus_transfers:
* and experiencing shortages but not surpluses, we're too stingy
* and should increase vtime rate.
*/
+ prev_busy_level = ioc->busy_level;
if (rq_wait_pct > RQ_WAIT_BUSY_PCT ||
missed_ppm[READ] > ppm_rthr ||
missed_ppm[WRITE] > ppm_wthr) {
@@ -1592,6 +1593,10 @@ skip_surplus_transfers:
atomic64_set(&ioc->vtime_rate, vrate);
ioc->inuse_margin_vtime = DIV64_U64_ROUND_UP(
ioc->period_us * vrate * INUSE_MARGIN_PCT, 100);
+ } else if (ioc->busy_level != prev_busy_level || nr_lagging) {
+ trace_iocost_ioc_vrate_adj(ioc, atomic64_read(&ioc->vtime_rate),
+ &missed_ppm, rq_wait_pct, nr_lagging,
+ nr_shortages, nr_surpluses);
}

ioc_refresh_params(ioc, false);
\
 
 \ /
  Last update: 2019-09-26 01:02    [W:0.040 / U:0.296 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site