lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening PHC | |
Open Source and information security mailing list archives
| ||
|
Date: Sun, 4 Oct 2020 21:32:31 +0800 From: kernel test robot <rong.a.chen@...el.com> To: Ritesh Harjani <riteshh@...ux.ibm.com> Cc: Theodore Ts'o <tytso@....edu>, Dan Williams <dan.j.williams@...el.com>, Jan Kara <jack@...e.cz>, LKML <linux-kernel@...r.kernel.org>, linux-ext4@...r.kernel.org, lkp@...ts.01.org, lkp@...el.com, ying.huang@...el.com, feng.tang@...el.com, zhengjun.xing@...el.com Subject: [ext4] 061113efe9: fio.write_iops 364.6% improvement Greeting, FYI, we noticed a 364.6% improvement of fio.write_iops due to commit: commit: 061113efe99b24ac026db5aa5a72844e16318bd7 ("ext4: optimize file overwrites") https://git.kernel.org/cgit/linux/kernel/git/tytso/ext4.git dev in testcase: fio-basic on test machine: 96 threads Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 256G memory with following parameters: disk: 2pmem fs: ext4 mount_option: dax runtime: 200s nr_task: 50% time_based: tb rw: randwrite bs: 4k ioengine: sync test_size: 200G cpufreq_governor: performance ucode: 0x5002f01 test-description: Fio is a tool that will spawn a number of threads or processes doing a particular type of I/O action as specified by the user. test-url: https://github.com/axboe/fio Details are as below: --------------------------------------------------------------------------------------------------> To reproduce: git clone https://github.com/intel/lkp-tests.git cd lkp-tests bin/lkp install job.yaml # job file is attached in this email bin/lkp run job.yaml ========================================================================================= bs/compiler/cpufreq_governor/disk/fs/ioengine/kconfig/mount_option/nr_task/rootfs/runtime/rw/tbox_group/test_size/testcase/time_based/ucode: 4k/gcc-9/performance/2pmem/ext4/sync/x86_64-rhel-8.3/dax/50%/debian-10.4-x86_64-20200603.cgz/200s/randwrite/lkp-csl-2sp6/200G/fio-basic/tb/0x5002f01 commit: 9ffd5728cc ("ext4: remove unused including <linux/version.h>") 061113efe9 ("ext4: optimize file overwrites") 9ffd5728cca71e4f 061113efe99b24ac026db5aa5a7 ---------------- --------------------------- %stddev %change %stddev \ | \ 0.16 ± 33% -0.2 0.01 fio.latency_100us% 48.50 ± 31% -47.2 1.28 ± 43% fio.latency_20us% 0.01 +32.7 32.76 ± 51% fio.latency_2us% 0.21 ± 89% +47.9 48.10 ± 17% fio.latency_4us% 25.01 ± 23% -24.6 0.37 ± 51% fio.latency_50us% 15563 ± 4% -10.4% 13952 ± 4% fio.time.involuntary_context_switches 13539 ± 5% -13.9% 11659 ± 9% fio.time.minor_page_faults 9206 -11.5% 8144 ± 3% fio.time.system_time 316.40 ± 5% +334.8% 1375 ± 18% fio.time.user_time 24203 -7.2% 22455 fio.time.voluntary_context_switches 5.93e+08 ± 2% +364.6% 2.755e+09 ± 18% fio.workload 11581 ± 2% +364.6% 53803 ± 18% fio.write_bw_MBps 25088 ± 8% -80.7% 4832 ± 24% fio.write_clat_90%_us 28480 ± 8% -78.7% 6072 ± 23% fio.write_clat_95%_us 37248 ± 6% -65.5% 12848 ± 21% fio.write_clat_99%_us 15694 ± 3% -80.2% 3100 ± 22% fio.write_clat_mean_us 9722 ± 15% -72.8% 2648 ± 18% fio.write_clat_stddev 2964935 ± 2% +364.6% 13773690 ± 18% fio.write_iops 1858 +4.2% 1936 vmstat.system.cs 47.96 -11.3% 42.52 ± 3% iostat.cpu.system 1.69 ± 5% +321.6% 7.11 ± 18% iostat.cpu.user 1.918e+09 ± 36% +226.2% 6.257e+09 ± 41% cpuidle.C1E.time 10452864 ± 3% +42.6% 14909339 ± 25% cpuidle.C1E.usage 7.753e+09 ± 8% -55.6% 3.44e+09 ± 74% cpuidle.C6.time 39387 ± 2% +15.2% 45385 meminfo.Active(anon) 7410 ± 10% +84.6% 13677 ± 4% meminfo.Dirty 122418 -9.6% 110640 meminfo.KReclaimable 122418 -9.6% 110640 meminfo.SReclaimable 0.01 ± 16% -0.0 0.00 ± 37% mpstat.cpu.all.iowait% 0.03 ± 3% +0.0 0.03 ± 4% mpstat.cpu.all.soft% 47.59 -5.5 42.07 ± 3% mpstat.cpu.all.sys% 1.70 ± 5% +5.5 7.17 ± 18% mpstat.cpu.all.usr% 741453 ± 7% -58.1% 310382 ± 30% numa-numastat.node0.local_node 756956 ± 6% -57.6% 321292 ± 33% numa-numastat.node0.numa_hit 247432 ± 24% +153.0% 626065 ± 15% numa-numastat.node1.local_node 263275 ± 19% +145.6% 646527 ± 17% numa-numastat.node1.numa_hit 3830 ± 5% -7.1% 3558 ± 2% slabinfo.dmaengine-unmap-16.active_objs 3830 ± 5% -7.1% 3558 ± 2% slabinfo.dmaengine-unmap-16.num_objs 421181 ± 5% -32.1% 286015 ± 7% slabinfo.ext4_extent_status.active_objs 6662 ± 6% -42.1% 3855 ± 8% slabinfo.ext4_extent_status.active_slabs 679573 ± 6% -42.1% 393340 ± 8% slabinfo.ext4_extent_status.num_objs 6662 ± 6% -42.1% 3855 ± 8% slabinfo.ext4_extent_status.num_slabs 9849 ± 2% +14.9% 11317 proc-vmstat.nr_active_anon 1812 ± 10% +87.2% 3392 ± 3% proc-vmstat.nr_dirty 2123 +1.3% 2151 proc-vmstat.nr_page_table_pages 30595 -9.6% 27662 proc-vmstat.nr_slab_reclaimable 9849 ± 2% +14.9% 11317 proc-vmstat.nr_zone_active_anon 1813 ± 10% +87.1% 3391 ± 3% proc-vmstat.nr_zone_write_pending 1064225 -5.0% 1011161 proc-vmstat.numa_hit 1032828 -5.1% 979762 proc-vmstat.numa_local 1104273 -5.0% 1049441 proc-vmstat.pgalloc_normal 898139 ± 14% -18.3% 733903 proc-vmstat.pgfree 6689 ± 9% +78.3% 11924 ± 15% numa-meminfo.node0.Dirty 1787289 -62.9% 662714 ± 3% numa-meminfo.node0.FilePages 1308617 ± 2% -88.8% 146900 ± 15% numa-meminfo.node0.Inactive 1308337 ± 2% -88.8% 146757 ± 15% numa-meminfo.node0.Inactive(anon) 1131509 -98.6% 16081 ± 14% numa-meminfo.node0.Mapped 4350653 -28.2% 3121909 ± 3% numa-meminfo.node0.MemUsed 7561 ± 3% -37.8% 4704 ± 12% numa-meminfo.node0.PageTables 1120696 -99.6% 4832 ± 79% numa-meminfo.node0.Shmem 37743 +15.4% 43552 numa-meminfo.node1.Active(anon) 539135 +209.5% 1668823 numa-meminfo.node1.FilePages 119869 ± 30% +969.4% 1281909 numa-meminfo.node1.Inactive 119766 ± 30% +970.2% 1281701 numa-meminfo.node1.Inactive(anon) 17210 ± 16% +6484.3% 1133150 numa-meminfo.node1.Mapped 2835108 ± 2% +43.3% 4061792 ± 2% numa-meminfo.node1.MemUsed 941.50 ± 19% +314.5% 3902 ± 14% numa-meminfo.node1.PageTables 44833 ± 10% +2502.5% 1166807 numa-meminfo.node1.Shmem 1601 ± 7% +85.5% 2970 ± 15% numa-vmstat.node0.nr_dirty 446809 -62.9% 165679 ± 3% numa-vmstat.node0.nr_file_pages 327071 ± 2% -88.8% 36681 ± 15% numa-vmstat.node0.nr_inactive_anon 282962 -98.5% 4157 ± 14% numa-vmstat.node0.nr_mapped 1890 ± 3% -37.9% 1174 ± 12% numa-vmstat.node0.nr_page_table_pages 280164 -99.6% 1208 ± 79% numa-vmstat.node0.nr_shmem 327071 ± 2% -88.8% 36681 ± 15% numa-vmstat.node0.nr_zone_inactive_anon 1601 ± 7% +85.5% 2970 ± 15% numa-vmstat.node0.nr_zone_write_pending 1917342 -22.9% 1477585 ± 7% numa-vmstat.node0.numa_hit 1869725 ± 3% -21.9% 1461109 ± 7% numa-vmstat.node0.numa_local 9474 +15.3% 10920 numa-vmstat.node1.nr_active_anon 134747 +208.6% 415874 numa-vmstat.node1.nr_file_pages 29850 ± 30% +968.9% 319054 numa-vmstat.node1.nr_inactive_anon 4207 ± 16% +6601.2% 281918 numa-vmstat.node1.nr_mapped 235.25 ± 18% +312.8% 971.00 ± 14% numa-vmstat.node1.nr_page_table_pages 11172 ± 10% +2499.0% 290369 numa-vmstat.node1.nr_shmem 9474 +15.3% 10920 numa-vmstat.node1.nr_zone_active_anon 29850 ± 30% +968.9% 319054 numa-vmstat.node1.nr_zone_inactive_anon 687971 ± 3% +56.8% 1078550 ± 10% numa-vmstat.node1.numa_hit 557691 ± 11% +64.5% 917161 ± 12% numa-vmstat.node1.numa_local 28318 ± 22% +31.8% 37317 ± 16% sched_debug.cfs_rq:/.exec_clock.stddev 385.22 ± 8% +9.7% 422.63 ± 9% sched_debug.cfs_rq:/.load_avg.avg 59833 ± 14% +24.5% 74463 ± 10% sched_debug.cfs_rq:/.min_vruntime.avg 103784 ± 12% +19.3% 123853 ± 11% sched_debug.cfs_rq:/.min_vruntime.max 29685 ± 20% +28.7% 38216 ± 16% sched_debug.cfs_rq:/.min_vruntime.stddev 29685 ± 20% +28.8% 38221 ± 16% sched_debug.cfs_rq:/.spread0.stddev 673766 ± 5% +28.1% 862888 ± 2% sched_debug.cpu.avg_idle.avg 316122 ± 5% -41.5% 185066 ± 8% sched_debug.cpu.avg_idle.stddev 2928 ± 8% +10.9% 3248 ± 7% sched_debug.cpu.nr_switches.avg 18266 ± 24% +53.1% 27974 ± 16% sched_debug.cpu.nr_switches.max 2854 ± 20% +37.8% 3934 ± 15% sched_debug.cpu.nr_switches.stddev 28.58 ± 24% +59.5% 45.58 ± 9% sched_debug.cpu.nr_uninterruptible.max 14485 ± 33% +69.6% 24568 ± 21% sched_debug.cpu.sched_count.max 2206 ± 27% +57.9% 3483 ± 19% sched_debug.cpu.sched_count.stddev 605.82 ± 19% +25.4% 759.56 ± 15% sched_debug.cpu.sched_goidle.avg 7179 ± 33% +70.3% 12229 ± 21% sched_debug.cpu.sched_goidle.max 34.46 ± 10% -42.4% 19.83 ± 10% sched_debug.cpu.sched_goidle.min 1117 ± 27% +57.4% 1759 ± 19% sched_debug.cpu.sched_goidle.stddev 7717 ± 27% +53.7% 11861 ± 20% sched_debug.cpu.ttwu_count.max 1125 ± 25% +49.2% 1679 ± 18% sched_debug.cpu.ttwu_count.stddev 5318 ± 36% +75.8% 9350 ± 23% sched_debug.cpu.ttwu_local.max 83.33 +33.3% 111.06 ± 14% sched_debug.cpu.ttwu_local.min 750.49 ± 30% +61.3% 1210 ± 21% sched_debug.cpu.ttwu_local.stddev 14.37 -79.1% 3.01 ± 55% perf-stat.i.MPKI 4.045e+09 ± 2% +276.2% 1.522e+10 ± 17% perf-stat.i.branch-instructions 1.01 -0.1 0.93 ± 2% perf-stat.i.branch-miss-rate% 44395538 +205.6% 1.357e+08 ± 14% perf-stat.i.branch-misses 35.66 ± 4% -8.0 27.62 ± 24% perf-stat.i.cache-miss-rate% 1814 +4.4% 1894 perf-stat.i.context-switches 6.02 ± 2% -73.5% 1.59 ± 19% perf-stat.i.cpi 0.00 ± 27% -0.0 0.00 ± 37% perf-stat.i.dTLB-load-miss-rate% 7.109e+09 ± 2% +276.9% 2.679e+10 ± 17% perf-stat.i.dTLB-loads 134633 ± 23% +202.7% 407497 ± 13% perf-stat.i.dTLB-store-misses 4.516e+09 ± 2% +291.9% 1.77e+10 ± 17% perf-stat.i.dTLB-stores 88.04 +6.5 94.54 perf-stat.i.iTLB-load-miss-rate% 36917229 ± 11% +180.6% 1.036e+08 ± 13% perf-stat.i.iTLB-load-misses 4880282 +16.3% 5678123 perf-stat.i.iTLB-loads 2.308e+10 ± 2% +279.2% 8.752e+10 ± 17% perf-stat.i.instructions 635.06 ± 8% +34.3% 852.98 ± 7% perf-stat.i.instructions-per-iTLB-miss 0.17 ± 2% +276.4% 0.65 ± 17% perf-stat.i.ipc 0.39 ± 61% -66.9% 0.13 ± 5% perf-stat.i.metric.K/sec 166.97 ± 2% +274.2% 624.73 ± 17% perf-stat.i.metric.M/sec 84.49 +3.4 87.88 ± 2% perf-stat.i.node-load-miss-rate% 10345827 ± 2% -61.8% 3956301 ± 22% perf-stat.i.node-store-misses 13.96 -78.8% 2.96 ± 56% perf-stat.overall.MPKI 1.10 -0.2 0.90 ± 3% perf-stat.overall.branch-miss-rate% 35.81 ± 4% -7.7 28.08 ± 24% perf-stat.overall.cache-miss-rate% 5.84 ± 2% -72.9% 1.58 ± 19% perf-stat.overall.cpi 0.00 ± 36% -0.0 0.00 ± 49% perf-stat.overall.dTLB-load-miss-rate% 0.00 ± 24% -0.0 0.00 ± 8% perf-stat.overall.dTLB-store-miss-rate% 88.21 +6.5 94.73 perf-stat.overall.iTLB-load-miss-rate% 631.78 ± 8% +33.0% 840.08 ± 7% perf-stat.overall.instructions-per-iTLB-miss 0.17 ± 2% +281.6% 0.65 ± 17% perf-stat.overall.ipc 7818 -18.3% 6384 perf-stat.overall.path-length 4.025e+09 ± 2% +276.0% 1.514e+10 ± 17% perf-stat.ps.branch-instructions 44199217 +205.6% 1.351e+08 ± 14% perf-stat.ps.branch-misses 1805 +4.4% 1885 perf-stat.ps.context-switches 7.075e+09 ± 2% +276.8% 2.665e+10 ± 17% perf-stat.ps.dTLB-loads 134275 ± 23% +202.0% 405555 ± 13% perf-stat.ps.dTLB-store-misses 4.494e+09 ± 2% +291.8% 1.76e+10 ± 17% perf-stat.ps.dTLB-stores 36727460 ± 11% +180.5% 1.03e+08 ± 13% perf-stat.ps.iTLB-load-misses 4854565 +16.3% 5647668 perf-stat.ps.iTLB-loads 2.297e+10 ± 2% +279.1% 8.706e+10 ± 17% perf-stat.ps.instructions 10291805 ± 2% -61.7% 3937017 ± 22% perf-stat.ps.node-store-misses 4.636e+12 ± 2% +279.0% 1.757e+13 ± 17% perf-stat.total.instructions 3980 ± 8% +122.3% 8848 ± 35% softirqs.CPU1.RCU 3187 ± 10% +159.2% 8263 ± 34% softirqs.CPU12.RCU 3430 ± 16% +142.1% 8305 ± 35% softirqs.CPU13.RCU 3187 ± 3% +155.3% 8137 ± 26% softirqs.CPU14.RCU 3655 ± 7% +153.6% 9271 ± 21% softirqs.CPU2.RCU 3302 ± 4% +139.9% 7923 ± 24% softirqs.CPU23.RCU 3629 ± 12% +147.4% 8979 ± 12% softirqs.CPU24.RCU 5417 ± 25% +72.7% 9355 ± 15% softirqs.CPU25.RCU 3915 ± 19% +141.6% 9459 ± 17% softirqs.CPU26.RCU 4003 ± 13% +113.5% 8549 ± 21% softirqs.CPU27.RCU 3553 ± 14% +159.2% 9210 ± 19% softirqs.CPU28.RCU 3413 ± 10% +174.0% 9352 ± 16% softirqs.CPU29.RCU 3784 ± 8% +124.7% 8505 ± 34% softirqs.CPU3.RCU 3802 ± 7% +146.5% 9372 ± 16% softirqs.CPU30.RCU 3767 ± 8% +146.8% 9295 ± 16% softirqs.CPU31.RCU 3939 ± 9% +130.9% 9097 ± 20% softirqs.CPU32.RCU 3935 ± 10% +125.3% 8867 ± 24% softirqs.CPU33.RCU 3991 +131.0% 9218 ± 19% softirqs.CPU34.RCU 3792 ± 16% +132.0% 8798 ± 21% softirqs.CPU35.RCU 3089 ± 8% +235.6% 10367 ± 19% softirqs.CPU36.RCU 14423 ± 24% -77.4% 3261 ± 10% softirqs.CPU36.SCHED 3965 ± 15% +134.5% 9298 ± 19% softirqs.CPU37.RCU 3548 ± 15% +195.0% 10468 ± 21% softirqs.CPU38.RCU 3596 ± 14% +198.2% 10723 ± 21% softirqs.CPU39.RCU 3815 ± 6% +169.3% 10275 ± 19% softirqs.CPU40.RCU 3468 ± 6% +170.9% 9396 ± 27% softirqs.CPU41.RCU 3712 ± 12% +179.7% 10381 ± 18% softirqs.CPU42.RCU 3573 ± 10% +166.6% 9527 ± 21% softirqs.CPU43.RCU 3866 +164.1% 10210 ± 18% softirqs.CPU44.RCU 3902 ± 2% +142.7% 9470 ± 26% softirqs.CPU45.RCU 3529 ± 16% +157.6% 9091 ± 25% softirqs.CPU46.RCU 3669 ± 13% +162.9% 9647 ± 20% softirqs.CPU47.RCU 3060 ± 17% +211.6% 9538 ± 16% softirqs.CPU48.RCU 3776 ± 14% +160.0% 9815 ± 18% softirqs.CPU49.RCU 3525 ± 13% +143.2% 8574 ± 38% softirqs.CPU5.RCU 4257 ± 3% +135.3% 10014 ± 34% softirqs.CPU50.RCU 4323 ± 4% +124.0% 9684 ± 16% softirqs.CPU51.RCU 4218 ± 16% +156.8% 10830 ± 18% softirqs.CPU52.RCU 4049 ± 11% +148.7% 10072 ± 17% softirqs.CPU53.RCU 4376 ± 20% +153.4% 11088 ± 16% softirqs.CPU54.RCU 8728 ±116% -69.9% 2625 ± 5% softirqs.CPU54.SCHED 4460 +122.1% 9903 ± 19% softirqs.CPU55.RCU 4369 ± 7% +124.8% 9823 ± 17% softirqs.CPU56.RCU 5078 ± 45% +112.4% 10788 ± 18% softirqs.CPU58.RCU 4510 ± 15% +143.1% 10964 ± 18% softirqs.CPU59.RCU 4504 ± 8% +118.8% 9854 ± 17% softirqs.CPU60.RCU 4252 ± 4% +121.7% 9428 ± 17% softirqs.CPU61.RCU 4241 ± 2% +152.2% 10695 ± 18% softirqs.CPU62.RCU 4388 ± 6% +132.2% 10189 ± 25% softirqs.CPU63.RCU 4937 ± 13% +122.2% 10971 ± 20% softirqs.CPU64.RCU 4411 ± 2% +155.2% 11259 ± 20% softirqs.CPU65.RCU 4435 ± 10% +148.9% 11038 ± 20% softirqs.CPU66.RCU 4047 ± 17% +179.7% 11320 ± 20% softirqs.CPU67.RCU 4238 +166.5% 11294 ± 23% softirqs.CPU68.RCU 4241 ± 4% +170.5% 11473 ± 21% softirqs.CPU69.RCU 3567 ± 8% +132.1% 8278 ± 37% softirqs.CPU7.RCU 4301 ± 2% +156.8% 11043 ± 19% softirqs.CPU70.RCU 4360 ± 2% +132.7% 10146 ± 30% softirqs.CPU71.RCU 3460 ± 11% +149.4% 8631 ± 34% softirqs.CPU72.RCU 3248 ± 11% +143.1% 7897 ± 22% softirqs.CPU73.RCU 3484 ± 17% +140.0% 8362 ± 36% softirqs.CPU74.RCU 3223 ± 11% +167.8% 8634 ± 32% softirqs.CPU75.RCU 3120 ± 16% +130.4% 7189 ± 33% softirqs.CPU76.RCU 3493 ± 5% +135.0% 8210 ± 40% softirqs.CPU77.RCU 2908 ± 12% +165.5% 7720 ± 42% softirqs.CPU78.RCU 3348 ± 17% +135.8% 7897 ± 42% softirqs.CPU79.RCU 3380 ± 4% +146.4% 8329 ± 35% softirqs.CPU8.RCU 3136 ± 12% +131.3% 7255 ± 37% softirqs.CPU80.RCU 2708 ± 7% +164.8% 7172 ± 45% softirqs.CPU81.RCU 2888 ± 8% +163.8% 7619 ± 34% softirqs.CPU82.RCU 2795 ± 16% +171.9% 7599 ± 33% softirqs.CPU83.RCU 10700 ± 27% +144.6% 26171 ± 3% softirqs.CPU84.SCHED 2686 ± 8% +177.9% 7465 ± 37% softirqs.CPU85.RCU 2640 ± 10% +174.3% 7242 ± 35% softirqs.CPU88.RCU 2957 ± 15% +142.1% 7160 ± 36% softirqs.CPU91.RCU 354901 +137.8% 843880 ± 17% softirqs.RCU 34287 ± 53% +55.0% 53147 interrupts.CAL:Function_call_interrupts 84.00 ± 92% -90.2% 8.25 ± 68% interrupts.CPU10.RES:Rescheduling_interrupts 83.25 ± 72% -88.9% 9.25 ± 31% interrupts.CPU11.RES:Rescheduling_interrupts 64.50 ± 37% -56.6% 28.00 ± 33% interrupts.CPU14.TLB:TLB_shootdowns 69.00 ± 33% -79.7% 14.00 ± 44% interrupts.CPU18.TLB:TLB_shootdowns 262.00 ± 73% +105.9% 539.50 ± 16% interrupts.CPU24.CAL:Function_call_interrupts 210.75 ± 70% +127.9% 480.25 ± 3% interrupts.CPU25.CAL:Function_call_interrupts 12.75 ±133% +639.2% 94.25 ± 31% interrupts.CPU25.TLB:TLB_shootdowns 13.50 ± 69% +525.9% 84.50 ± 29% interrupts.CPU27.TLB:TLB_shootdowns 28.00 ± 53% +223.2% 90.50 ± 19% interrupts.CPU28.TLB:TLB_shootdowns 16.00 ± 62% +393.8% 79.00 ± 27% interrupts.CPU29.TLB:TLB_shootdowns 340.00 ± 70% +217.4% 1079 ± 39% interrupts.CPU3.CAL:Function_call_interrupts 242.25 ± 79% +96.4% 475.75 ± 4% interrupts.CPU30.CAL:Function_call_interrupts 12.75 ± 80% +609.8% 90.50 ± 38% interrupts.CPU30.TLB:TLB_shootdowns 251.25 ± 75% +95.9% 492.25 ± 7% interrupts.CPU31.CAL:Function_call_interrupts 23.75 ± 28% +263.2% 86.25 ± 33% interrupts.CPU31.TLB:TLB_shootdowns 174.50 ± 22% +62.2% 283.00 ± 47% interrupts.CPU32.RES:Rescheduling_interrupts 19.75 ± 63% +432.9% 105.25 ± 28% interrupts.CPU32.TLB:TLB_shootdowns 246.75 ± 77% +130.7% 569.25 ± 29% interrupts.CPU33.CAL:Function_call_interrupts 8.75 ± 79% +900.0% 87.50 ± 32% interrupts.CPU33.TLB:TLB_shootdowns 244.75 ± 79% +99.6% 488.50 ± 3% interrupts.CPU35.CAL:Function_call_interrupts 2580 ± 18% +166.2% 6870 ± 24% interrupts.CPU36.NMI:Non-maskable_interrupts 2580 ± 18% +166.2% 6870 ± 24% interrupts.CPU36.PMI:Performance_monitoring_interrupts 84.50 ± 53% +133.7% 197.50 interrupts.CPU36.RES:Rescheduling_interrupts 22.25 ± 65% +1022.5% 249.75 ±103% interrupts.CPU36.TLB:TLB_shootdowns 17.50 ± 66% +578.6% 118.75 ± 33% interrupts.CPU38.TLB:TLB_shootdowns 241.25 ± 80% +101.2% 485.50 ± 3% interrupts.CPU39.CAL:Function_call_interrupts 21.00 ± 58% +385.7% 102.00 ± 28% interrupts.CPU39.TLB:TLB_shootdowns 19.75 ± 35% +370.9% 93.00 ± 32% interrupts.CPU40.TLB:TLB_shootdowns 27.50 ± 48% +278.2% 104.00 ± 9% interrupts.CPU42.TLB:TLB_shootdowns 23.00 ± 32% +241.3% 78.50 ± 24% interrupts.CPU43.TLB:TLB_shootdowns 246.00 ± 79% +94.9% 479.50 interrupts.CPU44.CAL:Function_call_interrupts 21.75 ± 30% +388.5% 106.25 ± 21% interrupts.CPU44.TLB:TLB_shootdowns 24.75 ± 24% +168.7% 66.50 ± 51% interrupts.CPU45.TLB:TLB_shootdowns 29.00 ± 33% +200.0% 87.00 ± 39% interrupts.CPU46.TLB:TLB_shootdowns 22.00 ± 34% +352.3% 99.50 ± 24% interrupts.CPU47.TLB:TLB_shootdowns 71.50 ± 25% -62.9% 26.50 ± 61% interrupts.CPU48.TLB:TLB_shootdowns 72.00 ± 28% -70.5% 21.25 ± 60% interrupts.CPU54.TLB:TLB_shootdowns 70.50 ± 24% -59.6% 28.50 ± 43% interrupts.CPU62.TLB:TLB_shootdowns 255.75 ± 78% +89.6% 485.00 interrupts.CPU72.CAL:Function_call_interrupts 23.75 ± 24% +307.4% 96.75 ± 13% interrupts.CPU72.TLB:TLB_shootdowns 19.25 ± 61% +366.2% 89.75 ± 12% interrupts.CPU73.TLB:TLB_shootdowns 246.75 ± 79% +105.6% 507.25 ± 8% interrupts.CPU74.CAL:Function_call_interrupts 39.75 ± 73% +190.6% 115.50 ± 30% interrupts.CPU74.TLB:TLB_shootdowns 251.25 ± 79% +193.7% 738.00 ± 53% interrupts.CPU75.CAL:Function_call_interrupts 27.00 ± 33% +408.3% 137.25 ± 55% interrupts.CPU75.TLB:TLB_shootdowns 249.00 ± 78% +243.1% 854.25 ± 42% interrupts.CPU76.CAL:Function_call_interrupts 28.00 ± 24% +248.2% 97.50 ± 21% interrupts.CPU76.TLB:TLB_shootdowns 5753 ± 36% -28.5% 4112 ± 52% interrupts.CPU77.NMI:Non-maskable_interrupts 5753 ± 36% -28.5% 4112 ± 52% interrupts.CPU77.PMI:Performance_monitoring_interrupts 26.00 ± 27% +253.8% 92.00 ± 18% interrupts.CPU77.TLB:TLB_shootdowns 248.25 ± 77% +91.9% 476.50 interrupts.CPU78.CAL:Function_call_interrupts 21.75 ± 15% +302.3% 87.50 ± 27% interrupts.CPU78.TLB:TLB_shootdowns 22.25 ± 61% +370.8% 104.75 ± 18% interrupts.CPU79.TLB:TLB_shootdowns 30.50 ± 33% +235.2% 102.25 ± 8% interrupts.CPU80.TLB:TLB_shootdowns 251.25 ± 78% +200.6% 755.25 ± 60% interrupts.CPU81.CAL:Function_call_interrupts 30.25 ± 54% +237.2% 102.00 ± 24% interrupts.CPU81.TLB:TLB_shootdowns 32.25 ± 23% +186.8% 92.50 ± 13% interrupts.CPU82.TLB:TLB_shootdowns 26.00 ± 16% +243.3% 89.25 ± 23% interrupts.CPU83.TLB:TLB_shootdowns 7732 ± 2% -67.6% 2505 ± 24% interrupts.CPU84.NMI:Non-maskable_interrupts 7732 ± 2% -67.6% 2505 ± 24% interrupts.CPU84.PMI:Performance_monitoring_interrupts 20.50 ± 39% +393.9% 101.25 ± 16% interrupts.CPU84.TLB:TLB_shootdowns 32.00 ± 31% +176.6% 88.50 ± 22% interrupts.CPU85.TLB:TLB_shootdowns 253.00 ± 78% +105.9% 521.00 ± 7% interrupts.CPU86.CAL:Function_call_interrupts 82.75 ± 82% -81.0% 15.75 ± 89% interrupts.CPU86.RES:Rescheduling_interrupts 26.75 ± 31% +238.3% 90.50 ± 28% interrupts.CPU86.TLB:TLB_shootdowns 270.75 ± 80% +389.6% 1325 ± 62% interrupts.CPU87.CAL:Function_call_interrupts 29.75 ± 33% +236.1% 100.00 ± 31% interrupts.CPU87.TLB:TLB_shootdowns 28.50 ± 16% +249.1% 99.50 ± 13% interrupts.CPU88.TLB:TLB_shootdowns 28.00 ± 55% +225.9% 91.25 ± 23% interrupts.CPU89.TLB:TLB_shootdowns 411.00 ± 81% -86.0% 57.50 ±148% interrupts.CPU9.RES:Rescheduling_interrupts 257.00 ± 77% +125.5% 579.50 ± 22% interrupts.CPU90.CAL:Function_call_interrupts 34.00 ± 29% +197.8% 101.25 ± 14% interrupts.CPU90.TLB:TLB_shootdowns 256.00 ± 76% +218.2% 814.50 ± 40% interrupts.CPU91.CAL:Function_call_interrupts 30.75 ± 28% +216.3% 97.25 ± 12% interrupts.CPU91.TLB:TLB_shootdowns 252.00 ± 77% +197.7% 750.25 ± 34% interrupts.CPU92.CAL:Function_call_interrupts 30.50 ± 27% +245.1% 105.25 ± 9% interrupts.CPU92.TLB:TLB_shootdowns 10.50 ± 51% +690.5% 83.00 ± 87% interrupts.CPU93.RES:Rescheduling_interrupts 33.00 ± 24% +215.9% 104.25 ± 16% interrupts.CPU93.TLB:TLB_shootdowns 96.75 ± 7% +35.1% 130.75 interrupts.IWI:IRQ_work_interrupts 4367 ± 18% +69.1% 7387 ± 31% interrupts.TLB:TLB_shootdowns 41.48 ± 2% -41.5 0.00 perf-profile.calltrace.cycles-pp.__ext4_journal_start_sb.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter 39.78 -39.8 0.00 perf-profile.calltrace.cycles-pp.jbd2__journal_start.__ext4_journal_start_sb.ext4_iomap_begin.iomap_apply.dax_iomap_rw 39.54 -39.5 0.00 perf-profile.calltrace.cycles-pp.start_this_handle.jbd2__journal_start.__ext4_journal_start_sb.ext4_iomap_begin.iomap_apply 65.49 ± 2% -34.1 31.39 ± 9% perf-profile.calltrace.cycles-pp.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter.new_sync_write 69.67 -11.6 58.02 ± 7% perf-profile.calltrace.cycles-pp.iomap_apply.dax_iomap_rw.ext4_file_write_iter.new_sync_write.vfs_write 69.69 -11.5 58.19 ± 7% perf-profile.calltrace.cycles-pp.dax_iomap_rw.ext4_file_write_iter.new_sync_write.vfs_write.ksys_write 9.33 -9.3 0.00 perf-profile.calltrace.cycles-pp.add_transaction_credits.start_this_handle.jbd2__journal_start.__ext4_journal_start_sb.ext4_iomap_begin 8.94 ± 7% -8.9 0.00 perf-profile.calltrace.cycles-pp.__ext4_journal_stop.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter 8.64 ± 7% -8.6 0.00 perf-profile.calltrace.cycles-pp.jbd2_journal_stop.__ext4_journal_stop.ext4_iomap_begin.iomap_apply.dax_iomap_rw 71.18 -8.3 62.88 ± 3% perf-profile.calltrace.cycles-pp.ext4_file_write_iter.new_sync_write.vfs_write.ksys_write.do_syscall_64 71.23 -7.5 63.72 perf-profile.calltrace.cycles-pp.new_sync_write.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe 6.62 ± 3% -6.6 0.00 perf-profile.calltrace.cycles-pp._raw_read_lock.start_this_handle.jbd2__journal_start.__ext4_journal_start_sb.ext4_iomap_begin 6.41 ± 5% -6.4 0.00 perf-profile.calltrace.cycles-pp.stop_this_handle.jbd2_journal_stop.__ext4_journal_stop.ext4_iomap_begin.iomap_apply 1.06 ± 5% +1.0 2.03 ± 55% perf-profile.calltrace.cycles-pp.file_update_time.ext4_write_checks.ext4_file_write_iter.new_sync_write.vfs_write 0.00 +1.0 1.04 ± 52% perf-profile.calltrace.cycles-pp.ext4_inode_block_valid.__check_block_validity.ext4_map_blocks.ext4_iomap_begin.iomap_apply 0.00 +1.4 1.45 ± 53% perf-profile.calltrace.cycles-pp.ksys_lseek.do_syscall_64.entry_SYSCALL_64_after_hwframe 0.00 +1.5 1.48 ± 9% perf-profile.calltrace.cycles-pp.ext4_es_lookup_extent.ext4_map_blocks.ext4_iomap_begin.iomap_apply.dax_iomap_rw 0.00 +1.6 1.64 ± 28% perf-profile.calltrace.cycles-pp.__check_block_validity.ext4_map_blocks.ext4_iomap_begin.iomap_apply.dax_iomap_rw 1.13 ± 6% +1.7 2.86 ± 67% perf-profile.calltrace.cycles-pp.ext4_write_checks.ext4_file_write_iter.new_sync_write.vfs_write.ksys_write 26.94 ± 2% +2.0 28.91 ± 2% perf-profile.calltrace.cycles-pp.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64 26.94 ± 2% +2.0 28.91 ± 2% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary 26.94 ± 2% +2.0 28.91 ± 2% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64 26.94 ± 2% +2.0 28.91 ± 2% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64 26.94 ± 2% +2.0 28.91 ± 2% perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64 27.09 ± 3% +2.3 29.41 ± 3% perf-profile.calltrace.cycles-pp.secondary_startup_64 27.08 ± 3% +2.3 29.41 ± 3% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry 0.00 +3.3 3.33 ± 11% perf-profile.calltrace.cycles-pp.ext4_map_blocks.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter 1.64 ± 8% +10.0 11.63 ± 25% perf-profile.calltrace.cycles-pp.__srcu_read_unlock.dax_iomap_actor.iomap_apply.dax_iomap_rw.ext4_file_write_iter 2.18 ± 15% +10.9 13.09 ± 31% perf-profile.calltrace.cycles-pp.__copy_user_nocache.__copy_user_flushcache._copy_from_iter_flushcache.dax_iomap_actor.iomap_apply 2.19 ± 15% +11.0 13.16 ± 31% perf-profile.calltrace.cycles-pp.__copy_user_flushcache._copy_from_iter_flushcache.dax_iomap_actor.iomap_apply.dax_iomap_rw 2.23 ± 14% +11.2 13.41 ± 31% perf-profile.calltrace.cycles-pp._copy_from_iter_flushcache.dax_iomap_actor.iomap_apply.dax_iomap_rw.ext4_file_write_iter 14.12 +11.6 25.71 ± 10% perf-profile.calltrace.cycles-pp.jbd2_transaction_committed.ext4_set_iomap.ext4_iomap_begin.iomap_apply.dax_iomap_rw 7.77 ± 4% +12.4 20.20 ± 10% perf-profile.calltrace.cycles-pp._raw_read_lock.jbd2_transaction_committed.ext4_set_iomap.ext4_iomap_begin.iomap_apply 14.31 +13.3 27.60 ± 10% perf-profile.calltrace.cycles-pp.ext4_set_iomap.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter 4.04 ± 6% +21.9 25.93 ± 22% perf-profile.calltrace.cycles-pp.dax_iomap_actor.iomap_apply.dax_iomap_rw.ext4_file_write_iter.new_sync_write 42.24 ± 2% -41.5 0.77 ± 6% perf-profile.children.cycles-pp.__ext4_journal_start_sb 40.51 -39.8 0.73 ± 5% perf-profile.children.cycles-pp.jbd2__journal_start 40.28 -39.6 0.72 ± 5% perf-profile.children.cycles-pp.start_this_handle 65.50 ± 2% -34.1 31.40 ± 9% perf-profile.children.cycles-pp.ext4_iomap_begin 69.67 -11.6 58.03 ± 7% perf-profile.children.cycles-pp.iomap_apply 69.70 -11.5 58.20 ± 7% perf-profile.children.cycles-pp.dax_iomap_rw 9.50 -9.3 0.18 ± 6% perf-profile.children.cycles-pp.add_transaction_credits 9.02 ± 7% -9.0 0.04 ± 57% perf-profile.children.cycles-pp.__ext4_journal_stop 8.72 ± 7% -8.7 0.04 ± 58% perf-profile.children.cycles-pp.jbd2_journal_stop 71.18 -8.3 62.91 ± 3% perf-profile.children.cycles-pp.ext4_file_write_iter 71.25 -7.5 63.75 perf-profile.children.cycles-pp.new_sync_write 6.43 ± 5% -6.4 0.00 perf-profile.children.cycles-pp.stop_this_handle 71.51 -5.8 65.70 perf-profile.children.cycles-pp.vfs_write 71.58 -5.5 66.11 perf-profile.children.cycles-pp.ksys_write 71.93 -3.8 68.15 perf-profile.children.cycles-pp.do_syscall_64 72.05 -3.2 68.81 perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe 0.08 +0.0 0.10 ± 4% perf-profile.children.cycles-pp.scheduler_tick 0.05 +0.0 0.07 ± 10% perf-profile.children.cycles-pp.task_tick_fair 0.06 ± 6% +0.0 0.09 ± 14% perf-profile.children.cycles-pp.ext4_reserve_inode_write 0.00 +0.1 0.06 ± 11% perf-profile.children.cycles-pp.rcu_all_qs 0.15 ± 8% +0.1 0.21 ± 12% perf-profile.children.cycles-pp.__ext4_mark_inode_dirty 0.00 +0.1 0.07 ± 17% perf-profile.children.cycles-pp.rw_verify_area 0.00 +0.1 0.08 ± 24% perf-profile.children.cycles-pp.pmem_dax_direct_access 0.04 ± 57% +0.1 0.12 ± 18% perf-profile.children.cycles-pp._cond_resched 0.00 +0.1 0.08 ± 8% perf-profile.children.cycles-pp.ktime_get_coarse_real_ts64 0.00 +0.1 0.09 ± 26% perf-profile.children.cycles-pp.generic_file_llseek_size 0.00 +0.1 0.09 ± 20% perf-profile.children.cycles-pp.file_modified 0.00 +0.1 0.09 ± 15% perf-profile.children.cycles-pp.aa_file_perm 0.00 +0.1 0.10 ± 37% perf-profile.children.cycles-pp.apparmor_file_permission 0.00 +0.1 0.11 ± 20% perf-profile.children.cycles-pp.__pmem_direct_access 0.03 ±100% +0.1 0.14 ± 28% perf-profile.children.cycles-pp.__might_sleep 0.04 ± 57% +0.1 0.16 ± 28% perf-profile.children.cycles-pp.___might_sleep 0.00 +0.1 0.15 ± 20% perf-profile.children.cycles-pp.__x86_indirect_thunk_rax 0.03 ±100% +0.2 0.21 ± 22% perf-profile.children.cycles-pp.up_write 0.00 +0.2 0.19 ±102% perf-profile.children.cycles-pp.timestamp_truncate 0.01 ±173% +0.2 0.21 ± 22% perf-profile.children.cycles-pp.exit_to_user_mode_prepare 0.03 ±100% +0.2 0.22 ± 20% perf-profile.children.cycles-pp.dax_direct_access 0.00 +0.2 0.22 ±134% perf-profile.children.cycles-pp.__sb_end_write 0.05 ± 9% +0.2 0.29 ± 18% perf-profile.children.cycles-pp.__srcu_read_lock 0.06 ± 7% +0.3 0.32 ± 21% perf-profile.children.cycles-pp.syscall_exit_to_user_mode 0.04 ± 58% +0.3 0.34 ± 54% perf-profile.children.cycles-pp.current_time 0.07 ± 11% +0.3 0.39 ± 22% perf-profile.children.cycles-pp.down_write 0.00 +0.3 0.32 ± 83% perf-profile.children.cycles-pp.__fsnotify_parent 0.09 ± 7% +0.3 0.42 ± 19% perf-profile.children.cycles-pp.common_file_perm 0.07 ± 6% +0.3 0.40 ± 22% perf-profile.children.cycles-pp.syscall_enter_from_user_mode 0.10 ± 8% +0.4 0.55 ± 21% perf-profile.children.cycles-pp.__fget_light 0.11 ± 9% +0.5 0.58 ± 21% perf-profile.children.cycles-pp.security_file_permission 0.05 ± 9% +0.5 0.52 ± 46% perf-profile.children.cycles-pp.__sb_start_write 0.06 ± 28% +0.5 0.57 ± 15% perf-profile.children.cycles-pp.percpu_counter_add_batch 0.16 ± 30% +0.5 0.68 ± 20% perf-profile.children.cycles-pp.entry_SYSCALL_64 0.12 ± 6% +0.5 0.67 ± 20% perf-profile.children.cycles-pp.__fdget_pos 0.00 +0.5 0.55 ±154% perf-profile.children.cycles-pp.generic_write_check_limits 0.00 +0.6 0.64 ±129% perf-profile.children.cycles-pp.generic_write_checks 0.01 ±173% +0.7 0.70 ±117% perf-profile.children.cycles-pp.ext4_generic_write_checks 0.10 ± 10% +0.8 0.91 ± 97% perf-profile.children.cycles-pp.ext4_llseek 0.19 ± 6% +0.8 1.00 ± 20% perf-profile.children.cycles-pp.syscall_return_via_sysret 0.12 ± 42% +0.9 1.04 ± 52% perf-profile.children.cycles-pp.ext4_inode_block_valid 1.06 ± 4% +1.0 2.04 ± 55% perf-profile.children.cycles-pp.file_update_time 0.21 ± 8% +1.3 1.46 ± 52% perf-profile.children.cycles-pp.ksys_lseek 0.21 ± 10% +1.3 1.50 ± 9% perf-profile.children.cycles-pp.ext4_es_lookup_extent 0.17 ± 22% +1.5 1.65 ± 28% perf-profile.children.cycles-pp.__check_block_validity 1.13 ± 6% +1.7 2.87 ± 67% perf-profile.children.cycles-pp.ext4_write_checks 26.94 ± 2% +2.0 28.91 ± 2% perf-profile.children.cycles-pp.start_secondary 27.09 ± 3% +2.3 29.41 ± 3% perf-profile.children.cycles-pp.secondary_startup_64 27.09 ± 3% +2.3 29.41 ± 3% perf-profile.children.cycles-pp.cpu_startup_entry 27.09 ± 3% +2.3 29.41 ± 3% perf-profile.children.cycles-pp.do_idle 27.09 ± 3% +2.3 29.41 ± 3% perf-profile.children.cycles-pp.cpuidle_enter 27.09 ± 3% +2.3 29.41 ± 3% perf-profile.children.cycles-pp.cpuidle_enter_state 27.09 ± 3% +2.3 29.41 ± 3% perf-profile.children.cycles-pp.intel_idle 0.41 ± 11% +2.9 3.36 ± 11% perf-profile.children.cycles-pp.ext4_map_blocks 14.60 ± 2% +6.1 20.66 ± 10% perf-profile.children.cycles-pp._raw_read_lock 1.64 ± 8% +10.0 11.64 ± 25% perf-profile.children.cycles-pp.__srcu_read_unlock 2.18 ± 15% +10.9 13.12 ± 31% perf-profile.children.cycles-pp.__copy_user_nocache 2.19 ± 15% +11.0 13.17 ± 31% perf-profile.children.cycles-pp.__copy_user_flushcache 2.24 ± 15% +11.2 13.41 ± 31% perf-profile.children.cycles-pp._copy_from_iter_flushcache 14.13 +11.6 25.73 ± 10% perf-profile.children.cycles-pp.jbd2_transaction_committed 14.31 +13.3 27.62 ± 10% perf-profile.children.cycles-pp.ext4_set_iomap 4.04 ± 6% +21.9 25.97 ± 22% perf-profile.children.cycles-pp.dax_iomap_actor 23.93 -23.5 0.42 ± 6% perf-profile.self.cycles-pp.start_this_handle 9.46 -9.3 0.18 ± 6% perf-profile.self.cycles-pp.add_transaction_credits 6.40 ± 5% -6.4 0.00 perf-profile.self.cycles-pp.stop_this_handle 0.00 +0.1 0.08 ± 19% perf-profile.self.cycles-pp.do_syscall_64 0.00 +0.1 0.08 ± 24% perf-profile.self.cycles-pp.current_time 0.00 +0.1 0.08 ± 8% perf-profile.self.cycles-pp.ktime_get_coarse_real_ts64 0.00 +0.1 0.08 ± 17% perf-profile.self.cycles-pp.aa_file_perm 0.00 +0.1 0.08 ± 24% perf-profile.self.cycles-pp.generic_file_llseek_size 0.00 +0.1 0.09 ± 40% perf-profile.self.cycles-pp.apparmor_file_permission 0.00 +0.1 0.09 ± 23% perf-profile.self.cycles-pp.generic_write_checks 0.00 +0.1 0.10 ± 18% perf-profile.self.cycles-pp.__x86_indirect_thunk_rax 0.01 ±173% +0.1 0.12 ± 30% perf-profile.self.cycles-pp.__might_sleep 0.00 +0.1 0.11 ± 20% perf-profile.self.cycles-pp.__pmem_direct_access 0.00 +0.1 0.11 ± 20% perf-profile.self.cycles-pp.ksys_lseek 0.00 +0.1 0.11 ± 19% perf-profile.self.cycles-pp.ksys_write 0.00 +0.1 0.12 ± 18% perf-profile.self.cycles-pp.syscall_exit_to_user_mode 0.00 +0.1 0.13 ± 22% perf-profile.self.cycles-pp.__fdget_pos 0.03 ±100% +0.1 0.16 ± 28% perf-profile.self.cycles-pp.___might_sleep 0.00 +0.2 0.16 ± 22% perf-profile.self.cycles-pp.dax_iomap_rw 0.00 +0.2 0.18 ±110% perf-profile.self.cycles-pp.timestamp_truncate 0.00 +0.2 0.18 ± 24% perf-profile.self.cycles-pp.exit_to_user_mode_prepare 0.03 ±100% +0.2 0.21 ± 22% perf-profile.self.cycles-pp.up_write 0.03 ±100% +0.2 0.22 ± 22% perf-profile.self.cycles-pp.down_write 0.00 +0.2 0.22 ± 28% perf-profile.self.cycles-pp.ext4_map_blocks 0.00 +0.2 0.22 ±133% perf-profile.self.cycles-pp.__sb_end_write 0.05 ± 9% +0.2 0.29 ± 20% perf-profile.self.cycles-pp.dax_iomap_actor 0.04 ± 58% +0.2 0.28 ± 17% perf-profile.self.cycles-pp.__srcu_read_lock 0.00 +0.2 0.24 ± 25% perf-profile.self.cycles-pp._copy_from_iter_flushcache 0.03 ±100% +0.2 0.27 ± 27% perf-profile.self.cycles-pp.vfs_write 0.08 ± 5% +0.2 0.33 ± 20% perf-profile.self.cycles-pp.common_file_perm 0.07 ± 10% +0.3 0.36 ± 21% perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe 0.00 +0.3 0.30 ± 95% perf-profile.self.cycles-pp.__sb_start_write 0.00 +0.3 0.31 ± 88% perf-profile.self.cycles-pp.__fsnotify_parent 0.07 ± 7% +0.3 0.39 ± 22% perf-profile.self.cycles-pp.syscall_enter_from_user_mode 0.10 ± 8% +0.3 0.43 ± 26% perf-profile.self.cycles-pp.ext4_iomap_begin 0.10 ± 4% +0.4 0.53 ± 20% perf-profile.self.cycles-pp.__fget_light 0.04 ± 63% +0.5 0.53 ± 15% perf-profile.self.cycles-pp.percpu_counter_add_batch 0.11 ± 4% +0.5 0.62 ± 21% perf-profile.self.cycles-pp.iomap_apply 0.07 ± 10% +0.5 0.61 ± 19% perf-profile.self.cycles-pp.ext4_es_lookup_extent 0.00 +0.5 0.55 ±154% perf-profile.self.cycles-pp.generic_write_check_limits 0.13 ± 6% +0.6 0.68 ± 20% perf-profile.self.cycles-pp.entry_SYSCALL_64 0.04 ± 60% +0.6 0.60 ± 21% perf-profile.self.cycles-pp.__check_block_validity 0.00 +0.7 0.66 ±136% perf-profile.self.cycles-pp.file_update_time 0.01 ±173% +0.8 0.79 ±116% perf-profile.self.cycles-pp.new_sync_write 0.10 ± 8% +0.8 0.89 ± 99% perf-profile.self.cycles-pp.ext4_llseek 0.19 ± 6% +0.8 1.00 ± 20% perf-profile.self.cycles-pp.syscall_return_via_sysret 0.11 ± 39% +0.9 1.04 ± 52% perf-profile.self.cycles-pp.ext4_inode_block_valid 0.23 ± 18% +1.0 1.26 ± 38% perf-profile.self.cycles-pp.ext4_file_write_iter 0.18 ± 16% +1.7 1.88 ± 67% perf-profile.self.cycles-pp.ext4_set_iomap 27.09 ± 3% +2.3 29.41 ± 3% perf-profile.self.cycles-pp.intel_idle 14.53 ± 2% +6.0 20.54 ± 10% perf-profile.self.cycles-pp._raw_read_lock 1.63 ± 8% +10.0 11.58 ± 25% perf-profile.self.cycles-pp.__srcu_read_unlock 2.18 ± 15% +10.9 13.05 ± 31% perf-profile.self.cycles-pp.__copy_user_nocache fio.write_bw_MBps 70000 +-------------------------------------------------------------------+ | O | 60000 |-O O | | O O | | O O O | 50000 |-+ O O O O O O O O O | | O O | 40000 |-+ O | | | 30000 |-+ | | | | | 20000 |-+ | | .+. | 10000 +-------------------------------------------------------------------+ fio.write_iops 1.8e+07 +-----------------------------------------------------------------+ | O | 1.6e+07 |-O O | 1.4e+07 |-+ O O | | O O O | 1.2e+07 |-+ O O O O O O O O O | | O O | 1e+07 |-+ O | | | 8e+06 |-+ | 6e+06 |-+ | | | 4e+06 |-+ | |.+.+.+.+.+.+.+.++.+.+.+.+.+.+.+.+.+.+.+.+.+.+.+.++.+.+.+.+.+.+.+.| 2e+06 +-----------------------------------------------------------------+ fio.write_clat_mean_us 18000 +-------------------------------------------------------------------+ | .+. .+. .+. .+. +.| 16000 |.+.+ +.+.+.+.+ +.+.+.+.+.+.+.+.+.+.+.+ +.+.+.+ + +.+. + | 14000 |-+ + + + | | + | 12000 |-+ | | | 10000 |-+ | | | 8000 |-+ | 6000 |-+ | | | 4000 |-+ O O O | | O O O O O O O O O O O O O O O | 2000 +-------------------------------------------------------------------+ fio.write_clat_stddev 16000 +-------------------------------------------------------------------+ | + | 14000 |-+ :: | 12000 |-+ : : + | | .+ : :.+ : : | 10000 |.+. .+. + + : + + .+. .+. .+. .+. : : | | + +. + + +.+.+. .+ +.+ + +.+ + +.+.+ +.+.| 8000 |-+ + + + + | | + | 6000 |-+ | 4000 |-+ O | | O O O O O O | 2000 |-O O O O O O O O O O O O O | | | 0 +-------------------------------------------------------------------+ fio.write_clat_90__us 30000 +-------------------------------------------------------------------+ | .+. + .+. .+. +. | 25000 |.+.+ +. : : +.+.+ .+.+.+.+.+.+.+ +.+.+.+ +.+. + +.+ | | +. : : : + .+ + +| | + +. : + | 20000 |-+ + | | | 15000 |-+ | | | 10000 |-+ | | | | O O O O O O O O O | 5000 |-O O O O O O O O O O O | | | 0 +-------------------------------------------------------------------+ fio.write_clat_95__us 35000 +-------------------------------------------------------------------+ | + | 30000 |.+.+.+.+ : : +.+. +.+.+.+.+.+.+.+.+.+.+.+.+.+. .+.+. | | + : : : +. + +.+ +.| 25000 |-+ +.+ +. : +.+ | | + | 20000 |-+ | | | 15000 |-+ | | | 10000 |-+ | | O O O O O O O O O O O | 5000 |-O O O O O O O O O | | | 0 +-------------------------------------------------------------------+ fio.write_clat_99__us 45000 +-------------------------------------------------------------------+ | + | 40000 |.+. .+. :: .+. +. .+.+. .+. .+. .+. .+. .+.+. | 35000 |-+ + +. : : + +. + + + + +.+ + +.+.+ +.| | +. : : : +.+ | 30000 |-+ + +. : | | + | 25000 |-+ | | | 20000 |-+ | 15000 |-+ O O O | | O O O O O O O O O O O O O O | 10000 |-O O O | | | 5000 +-------------------------------------------------------------------+ fio.latency_4us_ 70 +----------------------------------------------------------------------+ | O O | 60 |-+ O O | | O O O O O O O O | 50 |-+ O O O O O | | O | 40 |-+ O | | O | 30 |-+ | | | 20 |-+ | | | 10 |-+ | | | 0 +----------------------------------------------------------------------+ fio.latency_20us_ 80 +----------------------------------------------------------------------+ | .+ +. +. | 70 |-+ +. : : + : + | 60 |-+ + : : : : : :| |. .+ : : : + : +. .+.+. .+.. .+ :| 50 |-+.+.+ : : + : : + +.+.+ +.+.+ + : : | | : : + : : .+.+ : + | 40 |-+ :: + +.+. +. + | | : + | 30 |-+ + | 20 |-+ | | | 10 |-+ | | | 0 +----------------------------------------------------------------------+ fio.latency_50us_ 35 +----------------------------------------------------------------------+ | + | 30 |-+ + : + + +. | | : + : : +.+.. : + : + .+ : +.+ | 25 |.+. : + : : +.+. : +.+. .+. : +. .+. : +. + : : | | + : : : : + : + + + + + :| 20 |-+ : : : : + : :| | +..+ : : +.+ | 15 |-+ + : | | : : | 10 |-+ :: | | + | 5 |-+ | | | 0 +----------------------------------------------------------------------+ fio.latency_100us_ 0.3 +--------------------------------------------------------------------+ | | 0.25 |-+ + + + + + | | :: : + :: :: : | | : : :: + + + : : : : : : | 0.2 |-+. : : : : :+ + + +. : : +. : : : : | |+ + + : : .+ : + + + + +. + + +.+.: : | 0.15 |-+ + : : + + : + + + +.+ | | + : : : +. .+ :| 0.1 |-+ +: + : + :| | + + : | | + | 0.05 |-+ | | | 0 +--------------------------------------------------------------------+ fio.workload 3.5e+09 +-----------------------------------------------------------------+ | O | 3e+09 |-O O | | O O O | | O O O O | 2.5e+09 |-+ O O O O O O O | | O O | 2e+09 |-+ O | | | 1.5e+09 |-+ | | | | | 1e+09 |-+ | | +. .+. | 5e+08 +-----------------------------------------------------------------+ fio.time.user_time 1800 +--------------------------------------------------------------------+ | O O | 1600 |-+ O | 1400 |-+ O O | | O O O O | 1200 |-+ O O O O O O O O | | O O | 1000 |-+ O | | | 800 |-+ | 600 |-+ | | | 400 |-+ | |.+.+.+.+.+.+.+.+.+.+.+.+.+.+.+.+..+.+.+.+.+.+.+.+.+.+.+.+.+.+.+.+.+.| 200 +--------------------------------------------------------------------+ fio.time.system_time 9400 +--------------------------------------------------------------------+ | .+. .+. | 9200 |.+.+.+.+.+.+.+ +.+.+.+.+.+.+.+..+.+.+.+.+.+.+.+.+.+.+.+ +.+.+.+.| 9000 |-+ | | | 8800 |-+ | | | 8600 |-+ | | O O | 8400 |-+ O | 8200 |-+ O O O O O O O O O O O | | O | 8000 |-+ O O O | | O | 7800 +--------------------------------------------------------------------+ fio.time.voluntary_context_switches 25000 +-------------------------------------------------------------------+ | +. | 24500 |-+ .+.+.+ + + | |.+.+. .+ + .+ + .+. .+. .+.+. .+.+. .+.+.+. | | +.+ + + +.+ +.+ +.+.+ +.+ +.+.| 24000 |-+ | | | 23500 |-+ | | | 23000 |-+ | | | | O O O O | 22500 |-O O O | | O O O O O O O O O O O O O | 22000 +-------------------------------------------------------------------+ [*] bisect-good sample [O] bisect-bad sample 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, Rong Chen View attachment "config-5.9.0-rc7-00013-g061113efe99b2" of type "text/plain" (170148 bytes) View attachment "job-script" of type "text/plain" (8348 bytes) View attachment "job.yaml" of type "text/plain" (5848 bytes) View attachment "reproduce" of type "text/plain" (931 bytes)
Powered by blists - more mailing lists