diff options
| author | Kemeng Shi <shikemeng@huawei.com> | 2022-10-18 15:19:30 +0300 |
|---|---|---|
| committer | Jens Axboe <axboe@kernel.dk> | 2022-12-01 17:44:12 +0300 |
| commit | 63c9eac4b6d75859703f5820414986edefb01210 (patch) | |
| tree | e4db3eb364bf6f664f99547825707b5cf5b7513b /block/blk-iocost.c | |
| parent | c6d2efdd38b4e45741e103b6338f08e5ff2aae03 (diff) | |
| download | linux-63c9eac4b6d75859703f5820414986edefb01210.tar.xz | |
blk-iocost: Trace vtime_base_rate instead of vtime_rate
Since commit ac33e91e2daca ("blk-iocost: implement vtime loss
compensation") rename original vtime_rate to vtime_base_rate
and current vtime_rate is original vtime_rate with compensation.
The current rate showed in tracepoint is mixed with vtime_rate
and vtime_base_rate:
1) In function ioc_adjust_base_vrate, the first trace_iocost_ioc_vrate_adj
shows vtime_rate, the second trace_iocost_ioc_vrate_adj shows
vtime_base_rate.
2) In function iocg_activate shows vtime_rate by calling
TRACE_IOCG_PATH(iocg_activate...
3) In function ioc_check_iocgs shows vtime_rate by calling
TRACE_IOCG_PATH(iocg_idle...
Trace vtime_base_rate instead of vtime_rate as:
1) Before commit ac33e91e2daca ("blk-iocost: implement vtime loss
compensation"), the traced rate is without compensation, so still
show rate without compensation.
2) The vtime_base_rate is more stable while vtime_rate heavily depends on
excess budeget on current period which may change abruptly in next period.
Signed-off-by: Kemeng Shi <shikemeng@huawei.com>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/r/20221018121932.10792-4-shikemeng@huawei.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Diffstat (limited to 'block/blk-iocost.c')
| -rw-r--r-- | block/blk-iocost.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/block/blk-iocost.c b/block/blk-iocost.c index c3945d470779..f57a84ff0244 100644 --- a/block/blk-iocost.c +++ b/block/blk-iocost.c @@ -977,7 +977,7 @@ static void ioc_adjust_base_vrate(struct ioc *ioc, u32 rq_wait_pct, if (!ioc->busy_level || (ioc->busy_level < 0 && nr_lagging)) { if (ioc->busy_level != prev_busy_level || nr_lagging) - trace_iocost_ioc_vrate_adj(ioc, atomic64_read(&ioc->vtime_rate), + trace_iocost_ioc_vrate_adj(ioc, vrate, missed_ppm, rq_wait_pct, nr_lagging, nr_shortages); |