FYI, we noticed the below changes on git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master commit 3810631332465d967ba5e27ea2c7dff2c9afac6c ("PM / sleep: Re-implement suspend-to-idle handling") testbox/testcase/testparams: lkp-sb03/suspend/300s-30x-freeze 18320f2a6871aaf2 3810631332465d967ba5e27ea2 ---------------- -------------------------- %stddev %change %stddev \ | \ 212 ± 5% -23.3% 162 ± 6% numa-vmstat.node1.nr_kernel_stack 3418 ± 5% -23.6% 2610 ± 6% numa-meminfo.node1.KernelStack 32165483 ± 26% -31.3% 22108958 ± 7% cpuidle.C1E-SNB.time 9.401e+09 ± 1% +36.4% 1.283e+10 ± 0% cpuidle.C7-SNB.time 1137 ± 3% -7.0% 1057 ± 5% slabinfo.blkdev_requests.active_objs 1137 ± 3% -7.0% 1057 ± 5% slabinfo.blkdev_requests.num_objs 50.02 ± 1% -69.5% 15.26 ± 4% turbostat.CPU%c1 49.38 ± 1% +70.5% 84.19 ± 0% turbostat.CPU%c7 12.77 ± 3% -59.0% 5.24 ± 1% turbostat.CorWatt 7.83 ± 11% +545.0% 50.49 ± 0% turbostat.Pkg%pc6 38.91 ± 1% -48.0% 20.22 ± 0% turbostat.PkgWatt 90 ± 31% -34.7% 58 ± 20% sched_debug.cfs_rq[14]:/.tg_runnable_contrib 20 ± 46% -59.3% 8 ± 33% sched_debug.cfs_rq[14]:/.runnable_load_avg 4196 ± 31% -34.6% 2745 ± 20% sched_debug.cfs_rq[14]:/.avg->runnable_avg_sum 3489 ± 38% -33.2% 2329 ± 11% sched_debug.cfs_rq[15]:/.avg->runnable_avg_sum 2418 ± 43% +54.7% 3741 ± 21% sched_debug.cfs_rq[17]:/.avg->runnable_avg_sum 447 ± 28% +48.5% 663 ± 11% sched_debug.cfs_rq[17]:/.tg_load_contrib 434 ± 31% +45.7% 633 ± 13% sched_debug.cfs_rq[17]:/.blocked_load_avg 51 ± 44% +54.4% 79 ± 21% sched_debug.cfs_rq[17]:/.tg_runnable_contrib 9268 ± 26% -45.4% 5056 ± 46% sched_debug.cfs_rq[1]:/.min_vruntime 3653 ± 42% +96.0% 7162 ± 13% sched_debug.cfs_rq[24]:/.min_vruntime 4355 ± 14% +51.2% 6584 ± 22% sched_debug.cfs_rq[25]:/.min_vruntime 6 ± 36% -55.6% 3 ± 40% sched_debug.cfs_rq[27]:/.runnable_load_avg 589 ± 10% +24.4% 733 ± 12% sched_debug.cfs_rq[31]:/.tg_load_contrib 585 ± 10% +24.1% 727 ± 11% sched_debug.cfs_rq[31]:/.blocked_load_avg 2484 ± 30% +81.1% 4498 ± 20% sched_debug.cfs_rq[6]:/.avg->runnable_avg_sum 52 ± 30% +83.9% 97 ± 20% sched_debug.cfs_rq[6]:/.tg_runnable_contrib 14630 ± 41% -56.5% 6370 ± 40% sched_debug.cfs_rq[7]:/.min_vruntime 58 ± 8% -11.6% 51 ± 9% sched_debug.cfs_rq[9]:/.tg_runnable_contrib 2731 ± 8% -12.1% 2402 ± 8% sched_debug.cfs_rq[9]:/.avg->runnable_avg_sum 18774 ± 43% -60.3% 7460 ± 21% sched_debug.cpu#1.sched_count 16 ± 4% -25.4% 12 ± 18% sched_debug.cpu#10.cpu_load[3] 9 ± 8% -28.2% 7 ± 17% sched_debug.cpu#10.cpu_load[4] 32 ± 21% -33.3% 21 ± 28% sched_debug.cpu#13.cpu_load[1] 28 ± 4% -28.3% 20 ± 25% sched_debug.cpu#13.cpu_load[2] 29 ± 43% -58.0% 12 ± 20% sched_debug.cpu#14.cpu_load[3] 36 ± 26% -42.5% 21 ± 13% sched_debug.cpu#14.cpu_load[1] 215 ± 14% +53.6% 330 ± 9% sched_debug.cpu#14.curr->pid 34 ± 31% -47.8% 18 ± 18% sched_debug.cpu#14.cpu_load[2] 2431 ± 3% +25.3% 3046 ± 20% sched_debug.cpu#16.sched_count 1440 ± 15% +45.3% 2092 ± 34% sched_debug.cpu#16.ttwu_count 2233 ± 5% +12.5% 2512 ± 7% sched_debug.cpu#20.nr_switches 826 ± 16% +65.9% 1370 ± 31% sched_debug.cpu#21.ttwu_count 992 ± 6% +18.0% 1170 ± 4% sched_debug.cpu#23.sched_goidle 308 ± 13% +45.2% 448 ± 19% sched_debug.cpu#23.ttwu_local 2117 ± 5% +14.6% 2426 ± 4% sched_debug.cpu#23.sched_count 2072 ± 6% +14.6% 2373 ± 5% sched_debug.cpu#23.nr_switches 8 ± 44% +61.8% 13 ± 10% sched_debug.cpu#24.nr_uninterruptible 800 ± 17% -26.0% 592 ± 6% sched_debug.cpu#28.sched_goidle 5 ± 37% +195.0% 14 ± 46% sched_debug.cpu#30.cpu_load[1] 9 ± 33% +62.2% 15 ± 12% sched_debug.cpu#31.cpu_load[2] 28 ± 43% +64.9% 47 ± 42% sched_debug.cpu#8.cpu_load[1] 17 ± 8% -21.1% 14 ± 18% sched_debug.cpu#9.cpu_load[3] lkp-sb03: Sandy Bridge-EP Memory: 64G turbostat.CPU%c1 60 ++---------------------------------------------------------------------+ | .*..* | 50 *+*..* + .*.*..*. .*..*.*..*.*..*.*..*.*.*.. | | *. *.*..* * | | | 40 ++ | | | 30 ++ | | | 20 ++ | | O O O O O O O O O O O O O O | O O O O O O O O O O O O O O 10 ++ | | O O | 0 ++---------------------------------------------------------------------+ turbostat.CPU%c7 100 ++--------------------------------------------------------------------+ | O O | 90 ++ | O O O O | | O O O O O O O O O O O O O O O O O O O O O O O O 80 ++ | | | 70 ++ | | | 60 ++ | | | | *. .*. .*. | 50 *+*..*. + *..*.*. *. *. .*.*..*.*.*..*.*..*.* | | *..* *. | 40 ++--------------------------------------------------------------------+ turbostat.Pkg%pc6 55 ++---------------------------O-----------------------------------------+ 50 O+O O O O O O O O O O O O O O O O O O O O O O O O O O O | O | 45 ++ | 40 ++ | | | 35 ++ | 30 ++ | 25 ++ | | | 20 ++ | 15 ++ | | .*.. .* | 10 *+ .*.*..* *. .*.*.*. + .*..*.*.. .*..*.*.*..* | 5 ++*--------------*-----------*--*---------*----------------------------+ turbostat.PkgWatt 45 ++-----*---------------------------------------------------------------+ *. .* * .*. | 40 ++*. + .*.. .*. *.. | | *..*.*..*.* * *.*..*.*..*.*.*..* | | | 35 ++ | | | 30 ++ | | | 25 ++ | | | | O | 20 O+O O O O O O O O O O O O O O O O O O O O O O O O O O O O | | 15 ++---------------------------------------------------------------------+ turbostat.CorWatt 20 ++---------------------------------------------------------------------+ | .*.. | 18 ++ * | 16 *+ .. * *. | | * + .. *.. | 14 ++ *.. .*..*.*.*..*.* .*..*. .* | 12 ++ * *.*..* *.*. | | | 10 ++ | 8 ++ | | | 6 O+O O O O O O O O O O O O O O O O O O O O O | 4 ++ O O O O O O O | O | 2 ++---------------------------------------------------------------------+ [*] bisect-good sample [O] bisect-bad sample To reproduce: apt-get install ruby git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git cd lkp-tests bin/setup-local job.yaml # the job file attached in this email bin/run-local job.yaml Disclaimer: Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. Thanks, Ying Huang