Greeting,
FYI, we noticed a 126.2% improvement of fio.read_iops due to commit:
commit: 64ea9d6c5f473c29c5de97abaa697856db90fef7 ("iov_iter: Use IOCB/IOMAP_WRITE if available rather than iterator direction")
https://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs.git iov-extract
in testcase: fio-basic
on test machine: 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz (Cascade Lake) with 512G memory
with following parameters:
disk: 2pmem
fs: ext4
mount_option: dax
runtime: 200s
nr_task: 50%
time_based: tb
rw: randread
bs: 2M
ioengine: libaio
test_size: 200G
cpufreq_governor: performance
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
In addition to that, the commit also has significant impact on the following tests:
+------------------+------------------------------------------------------------------------------------------------+
| testcase: change | fio-basic: fio.write_iops 1942.1% improvement |
| test machine | 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz (Cascade Lake) with 512G memory |
| test parameters | bs=2M |
| | cpufreq_governor=performance |
| | disk=2pmem |
| | fs=ext4 |
| | ioengine=sync |
| | mount_option=dax |
| | nr_task=50% |
| | runtime=200s |
| | rw=rw |
| | test_size=200G |
| | time_based=tb |
+------------------+------------------------------------------------------------------------------------------------+
| testcase: change | reaim: reaim.jobs_per_min 89.7% improvement |
| test machine | 128 threads 2 sockets Intel(R) Xeon(R) Platinum 8358 CPU @ 2.60GHz (Ice Lake) with 128G memory |
| test parameters | cpufreq_governor=performance |
| | disk=1HDD |
| | fs=ext4 |
| | nr_task=100 |
| | runtime=300s |
| | test=disk |
+------------------+------------------------------------------------------------------------------------------------+
| testcase: change | fio-basic: fio.read_iops 127.0% improvement |
| test machine | 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz (Cascade Lake) with 512G memory |
| test parameters | bs=2M |
| | cpufreq_governor=performance |
| | disk=2pmem |
| | fs=ext4 |
| | ioengine=libaio |
| | mount_option=dax |
| | nr_task=50% |
| | runtime=200s |
| | rw=read |
| | test_size=200G |
| | time_based=tb |
+------------------+------------------------------------------------------------------------------------------------+
Details are as below:
=========================================================================================
bs/compiler/cpufreq_governor/disk/fs/ioengine/kconfig/mount_option/nr_task/rootfs/runtime/rw/tbox_group/test_size/testcase/time_based:
2M/gcc-11/performance/2pmem/ext4/libaio/x86_64-rhel-8.3/dax/50%/debian-11.1-x86_64-20220510.cgz/200s/randread/lkp-csl-2sp7/200G/fio-basic/tb
commit:
e6eadc0324 ("iov_iter: Use the direction in the iterator functions")
64ea9d6c5f ("iov_iter: Use IOCB/IOMAP_WRITE if available rather than iterator direction")
e6eadc0324e475e3 64ea9d6c5f473c29c5de97abaa6
---------------- ---------------------------
%stddev %change %stddev
\ | \
28.42 ? 12% +67.0 95.37 fio.latency_100ms%
0.01 +0.2 0.23 ? 58% fio.latency_10ms%
70.69 ? 5% -69.8 0.88 ? 18% fio.latency_250ms%
0.55 ? 45% -0.5 0.08 ? 30% fio.latency_500ms%
0.16 ? 73% +2.8 2.98 ? 7% fio.latency_50ms%
0.01 -0.0 0.00 fio.latency_50us%
25388 +126.2% 57424 fio.read_bw_MBps
1.407e+08 -63.3% 51642368 fio.read_clat_90%_us
1.428e+08 -63.5% 52166656 fio.read_clat_95%_us
1.957e+08 ? 3% -54.7% 88604672 ? 17% fio.read_clat_99%_us
1.171e+08 -55.8% 51798447 fio.read_clat_mean_us
27892624 ? 5% -51.6% 13508778 ? 7% fio.read_clat_stddev
12694 +126.2% 28712 fio.read_iops
3753253 -55.6% 1665798 fio.read_slat_mean_us
904821 ? 5% -50.3% 450059 ? 7% fio.read_slat_stddev
83484 -2.3% 81606 fio.time.minor_page_faults
82.60 ? 4% -46.1% 44.53 ? 19% fio.time.user_time
2538874 +126.2% 5742524 fio.workload
118966 ? 6% -5.6% 112317 ? 4% numa-meminfo.node1.Slab
0.60 -0.1 0.50 mpstat.cpu.all.irq%
0.51 ? 4% -0.2 0.32 ? 15% mpstat.cpu.all.usr%
335.01 ? 6% -15.6% 282.74 uptime.boot
21727 ? 9% -22.8% 16764 uptime.idle
365.40 ? 67% +341.9% 1614 ? 14% proc-vmstat.nr_written
36539 ? 4% +23.6% 45144 ? 6% proc-vmstat.numa_hint_faults
391222 +2.1% 399514 proc-vmstat.numa_hit
390931 +1.8% 397825 proc-vmstat.numa_local
1492 ? 65% +334.6% 6484 ? 14% proc-vmstat.pgpgout
45650 ? 12% -34.4% 29932 ? 16% turbostat.C1
0.15 -100.0% 0.00 turbostat.IPC
60.80 +3.3% 62.80 turbostat.PkgTmp
250.11 +11.2% 278.05 turbostat.PkgWatt
58.09 -6.3% 54.45 turbostat.RAMWatt
1115526 ? 6% +62.8% 1816010 ? 25% sched_debug.cpu.avg_idle.max
155834 ? 12% +39.7% 217674 ? 15% sched_debug.cpu.avg_idle.stddev
222305 ? 9% -23.2% 170765 sched_debug.cpu.clock.avg
222357 ? 9% -23.1% 170937 sched_debug.cpu.clock.max
222253 ? 9% -23.2% 170594 sched_debug.cpu.clock.min
29.89 ? 14% +238.5% 101.16 ? 3% sched_debug.cpu.clock.stddev
221210 ? 9% -23.3% 169766 sched_debug.cpu.clock_task.avg
222168 ? 9% -23.1% 170743 sched_debug.cpu.clock_task.max
200093 ? 10% -25.6% 148915 sched_debug.cpu.clock_task.min
539522 ? 4% +22.8% 662703 ? 8% sched_debug.cpu.max_idle_balance_cost.max
4764 ? 64% +456.7% 26527 ? 34% sched_debug.cpu.max_idle_balance_cost.stddev
0.00 ? 6% +163.8% 0.00 ? 3% sched_debug.cpu.next_balance.stddev
222251 ? 9% -23.2% 170591 sched_debug.cpu_clk
221620 ? 9% -23.3% 169962 sched_debug.ktime
222977 ? 9% -23.2% 171325 sched_debug.sched_clk
***************************************************************************************************
lkp-csl-2sp7: 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz (Cascade Lake) with 512G memory
=========================================================================================
bs/compiler/cpufreq_governor/disk/fs/ioengine/kconfig/mount_option/nr_task/rootfs/runtime/rw/tbox_group/test_size/testcase/time_based:
2M/gcc-11/performance/2pmem/ext4/sync/x86_64-rhel-8.3/dax/50%/debian-11.1-x86_64-20220510.cgz/200s/rw/lkp-csl-2sp7/200G/fio-basic/tb
commit:
e6eadc0324 ("iov_iter: Use the direction in the iterator functions")
64ea9d6c5f ("iov_iter: Use IOCB/IOMAP_WRITE if available rather than iterator direction")
e6eadc0324e475e3 64ea9d6c5f473c29c5de97abaa6
---------------- ---------------------------
%stddev %change %stddev
\ | \
0.12 ? 43% -0.1 0.02 ? 19% fio.latency_1000us%
9.55 ? 41% -9.5 0.01 ?100% fio.latency_10ms%
0.03 ? 60% +48.4 48.45 ? 39% fio.latency_250us%
20.78 ? 24% -20.7 0.04 ? 27% fio.latency_2ms%
69.23 ? 3% -69.2 0.02 ? 25% fio.latency_4ms%
0.13 ? 42% -0.1 0.05 ? 22% fio.latency_750us%
15167 ? 4% +1942.6% 309815 ? 12% fio.read_bw_MBps
4286054 ? 6% -96.5% 151296 ? 28% fio.read_clat_90%_us
4623564 ? 7% -96.5% 160597 ? 28% fio.read_clat_95%_us
5498470 ? 11% -96.7% 181162 ? 28% fio.read_clat_99%_us
3298308 ? 4% -96.5% 114311 ? 13% fio.read_clat_mean_us
767056 ? 18% -93.3% 51658 ? 26% fio.read_clat_stddev
7583 ? 4% +1942.6% 154907 ? 12% fio.read_iops
4754 -1.5% 4684 fio.time.percent_of_cpu_this_job_got
8493 -20.7% 6736 fio.time.system_time
1052 +153.3% 2665 ? 3% fio.time.user_time
22073 +12.6% 24848 ? 2% fio.time.voluntary_context_switches
3032891 ? 4% +1942.3% 61941623 ? 12% fio.workload
15160 ? 4% +1942.1% 309597 ? 12% fio.write_bw_MBps
3411148 ? 6% -95.7% 147029 ? 27% fio.write_clat_90%_us
3725721 ? 7% -95.8% 155733 ? 27% fio.write_clat_95%_us
4502323 ? 12% -96.1% 176384 ? 27% fio.write_clat_99%_us
2377925 ? 4% -95.3% 112602 ? 12% fio.write_clat_mean_us
772542 ? 17% -93.3% 51635 ? 24% fio.write_clat_stddev
7580 ? 4% +1942.1% 154798 ? 12% fio.write_iops
30.59 -1.2% 30.22 boot-time.dhcp
44.29 -17.6% 36.48 iostat.cpu.system
5.46 +149.4% 13.63 ? 3% iostat.cpu.user
226944 ? 10% +17.0% 265432 ? 7% numa-numastat.node0.local_node
229493 ? 11% +16.6% 267683 ? 6% numa-numastat.node0.numa_hit
327.94 ? 6% -13.6% 283.46 uptime.boot
20979 ? 9% -20.4% 16696 uptime.idle
5.00 +156.7% 12.83 ? 2% vmstat.cpu.us
107.20 ? 14% -70.9% 31.17 ? 43% vmstat.io.bo
1947 ? 8% +27.4% 2481 vmstat.system.cs
42803 ? 51% +99.7% 85467 ? 9% meminfo.Active
21377 ?103% +199.6% 64049 ? 13% meminfo.Active(anon)
1028752 ? 12% +14.7% 1180441 ? 2% meminfo.DirectMap4k
37174 ? 63% +120.7% 82042 ? 10% meminfo.Shmem
0.69 ? 79% +1.4 2.08 ? 2% mpstat.cpu.all.irq%
0.22 ? 5% -0.1 0.09 ? 7% mpstat.cpu.all.soft%
43.81 -9.2 34.66 mpstat.cpu.all.sys%
5.52 +8.2 13.76 ? 3% mpstat.cpu.all.usr%
31755 ? 93% +138.5% 75738 ? 21% numa-meminfo.node1.Active
19642 ?112% +214.5% 61783 ? 13% numa-meminfo.node1.Active(anon)
1297089 ?103% +127.4% 2950212 ? 2% numa-meminfo.node1.AnonHugePages
8554 ? 7% +9.0% 9320 ? 2% numa-meminfo.node1.KernelStack
8937 ? 48% +41.8% 12674 ? 3% numa-meminfo.node1.Mapped
29411 ? 86% +158.7% 76095 ? 11% numa-meminfo.node1.Shmem
229509 ? 11% +16.7% 267844 ? 6% numa-vmstat.node0.numa_hit
226960 ? 10% +17.0% 265593 ? 7% numa-vmstat.node0.numa_local
4884 ?112% +216.0% 15436 ? 13% numa-vmstat.node1.nr_active_anon
634.40 ?103% +127.1% 1440 ? 2% numa-vmstat.node1.nr_anon_transparent_hugepages
2248 ? 48% +41.6% 3184 ? 3% numa-vmstat.node1.nr_mapped
7338 ? 86% +159.2% 19022 ? 11% numa-vmstat.node1.nr_shmem
4884 ?112% +216.0% 15436 ? 13% numa-vmstat.node1.nr_zone_active_anon
2800 -1.6% 2755 turbostat.Bzy_MHz
6.80 ?123% +22.7 29.51 ? 59% turbostat.C1E%
0.09 ? 46% -80.1% 0.02 ? 20% turbostat.IPC
4217 ? 51% +14321.1% 608138 ?205% turbostat.POLL
63.20 ? 2% +4.2% 65.83 ? 2% turbostat.PkgTmp
259.23 +13.7% 294.79 turbostat.PkgWatt
53.95 -3.8% 51.88 ? 2% turbostat.RAMWatt
5349 ?103% +199.1% 16000 ? 13% proc-vmstat.nr_active_anon
837004 +1.1% 846381 proc-vmstat.nr_anon_pages
101.00 ? 32% -93.9% 6.17 ? 23% proc-vmstat.nr_dirtied
710747 +1.5% 721433 proc-vmstat.nr_file_pages
840884 +1.2% 850985 proc-vmstat.nr_inactive_anon
17652 +0.9% 17816 proc-vmstat.nr_kernel_stack
3469 +2.0% 3538 proc-vmstat.nr_page_table_pages
9308 ? 63% +120.3% 20503 ? 10% proc-vmstat.nr_shmem
48010 +1.1% 48555 proc-vmstat.nr_slab_unreclaimable
5349 ?103% +199.1% 16000 ? 13% proc-vmstat.nr_zone_active_anon
840884 +1.2% 850985 proc-vmstat.nr_zone_inactive_anon
445330 ? 16% +23.0% 547820 proc-vmstat.numa_hit
432726 ? 13% +19.4% 516623 proc-vmstat.numa_local
12623 ?120% +147.2% 31201 proc-vmstat.numa_other
7857 ?112% +223.3% 25404 ? 11% proc-vmstat.pgactivate
1252650 ? 6% +10.1% 1378998 proc-vmstat.pgalloc_normal
611086 ? 7% +9.5% 668925 proc-vmstat.pgfault
1232126 ? 5% +7.7% 1326857 proc-vmstat.pgfree
353.00 ? 6% +33.8% 472.19 ? 3% sched_debug.cfs_rq:/.util_est_enqueued.avg
354.05 ? 6% +11.0% 392.89 sched_debug.cfs_rq:/.util_est_enqueued.stddev
1227820 ? 7% -13.1% 1066553 ? 3% sched_debug.cpu.avg_idle.max
495552 ? 18% -48.0% 257439 ? 22% sched_debug.cpu.avg_idle.min
117660 ? 21% +48.8% 175072 ? 6% sched_debug.cpu.avg_idle.stddev
215198 ? 9% -20.7% 170684 sched_debug.cpu.clock.avg
215249 ? 9% -20.7% 170705 sched_debug.cpu.clock.max
215144 ? 9% -20.7% 170664 sched_debug.cpu.clock.min
30.06 ? 6% -60.5% 11.87 ? 13% sched_debug.cpu.clock.stddev
214020 ? 9% -21.3% 168388 sched_debug.cpu.clock_task.avg
214662 ? 9% -21.3% 168991 sched_debug.cpu.clock_task.max
192328 ? 9% -20.3% 153377 sched_debug.cpu.clock_task.min
2250 ? 22% -30.0% 1575 sched_debug.cpu.clock_task.stddev
602998 ? 8% -10.3% 541131 ? 3% sched_debug.cpu.max_idle_balance_cost.max
0.00 ? 23% -38.2% 0.00 ? 26% sched_debug.cpu.next_balance.stddev
4572 ? 2% +9.7% 5016 sched_debug.cpu.nr_switches.avg
3459 ? 7% +18.9% 4113 ? 9% sched_debug.cpu.nr_switches.stddev
215142 ? 9% -20.7% 170664 sched_debug.cpu_clk
214467 ? 9% -20.7% 170038 sched_debug.ktime
215909 ? 9% -20.6% 171391 sched_debug.sched_clk
19.99 ?122% +1288.4% 277.52 perf-stat.i.MPKI
0.07 ?122% +0.5 0.55 ? 6% perf-stat.i.branch-miss-rate%
1789007 ?122% +256.9% 6385596 ? 3% perf-stat.i.branch-misses
3.76e+08 ?122% +489.2% 2.215e+09 ? 12% perf-stat.i.cache-references
793.00 ?122% +192.3% 2317 perf-stat.i.context-switches
2.93 ?122% +476.2% 16.86 ? 11% perf-stat.i.cpi
126.67 ?122% +875.4% 1235 ? 11% perf-stat.i.cycles-between-cache-misses
0.00 ?126% +0.0 0.01 ? 61% perf-stat.i.dTLB-load-miss-rate%
0.00 ?125% +0.0 0.00 ? 21% perf-stat.i.dTLB-store-miss-rate%
15094 ?122% +359.9% 69419 ? 12% perf-stat.i.dTLB-store-misses
14.86 ?122% +29.2 44.09 ? 8% perf-stat.i.iTLB-load-miss-rate%
432930 ?122% +556.0% 2839854 ? 6% perf-stat.i.iTLB-load-misses
732747 ?122% +400.2% 3665217 ? 17% perf-stat.i.iTLB-loads
0.24 ?124% +0.9 1.09 ? 18% perf-stat.i.node-store-miss-rate%
108219 ?122% +433.1% 576866 ? 9% perf-stat.i.node-store-misses
20252133 ?122% +171.3% 54941059 ? 14% perf-stat.i.node-stores
20.03 ?122% +1278.5% 276.12 perf-stat.overall.MPKI
0.07 ?122% +0.5 0.59 ? 7% perf-stat.overall.branch-miss-rate%
2.93 ?122% +473.4% 16.79 ? 11% perf-stat.overall.cpi
125.84 ?122% +876.8% 1229 ? 11% perf-stat.overall.cycles-between-cache-misses
0.00 ?122% +0.0 0.01 ? 61% perf-stat.overall.dTLB-load-miss-rate%
0.00 ?122% +0.0 0.00 ? 22% perf-stat.overall.dTLB-store-miss-rate%
14.86 ?122% +29.2 44.07 ? 8% perf-stat.overall.iTLB-load-miss-rate%
0.21 ?123% +0.9 1.07 ? 19% perf-stat.overall.node-store-miss-rate%
1775193 ?122% +259.5% 6382224 ? 4% perf-stat.ps.branch-misses
3.743e+08 ?122% +488.6% 2.203e+09 ? 12% perf-stat.ps.cache-references
787.70 ?122% +192.3% 2302 perf-stat.ps.context-switches
14975 ?122% +361.3% 69085 ? 12% perf-stat.ps.dTLB-store-misses
430771 ?122% +555.7% 2824622 ? 6% perf-stat.ps.iTLB-load-misses
729029 ?122% +400.0% 3644932 ? 17% perf-stat.ps.iTLB-loads
976.71 ?122% +150.6% 2447 perf-stat.ps.minor-faults
108206 ?122% +430.4% 573881 ? 9% perf-stat.ps.node-store-misses
20157154 ?122% +171.1% 54649318 ? 14% perf-stat.ps.node-stores
976.87 ?122% +150.6% 2448 perf-stat.ps.page-faults
0.00 +4.2 4.24 ? 22% perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt.clear_user_erms.iov_iter_zero.dax_iomap_rw.ext4_dax_write_iter
0.00 +4.6 4.63 ? 23% perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt.clear_user_erms.iov_iter_zero.dax_iomap_rw.ext4_file_read_iter
2.83 ?122% +15.5 18.36 ? 11% perf-profile.calltrace.cycles-pp.get_io_u
11.19 ?123% +20.4 31.59 ? 20% perf-profile.calltrace.cycles-pp.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary
11.20 ?122% +20.5 31.70 ? 20% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle
11.19 ?122% +20.5 31.70 ? 20% perf-profile.calltrace.cycles-pp.mwait_idle_with_hints.intel_idle.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call
11.30 ?123% +20.8 32.14 ? 20% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry
0.00 +21.2 21.23 ? 8% perf-profile.calltrace.cycles-pp.clear_user_erms.iov_iter_zero.dax_iomap_rw.ext4_dax_write_iter.vfs_write
0.00 +22.0 21.95 ? 13% perf-profile.calltrace.cycles-pp.clear_user_erms.iov_iter_zero.dax_iomap_rw.ext4_file_read_iter.vfs_read
0.00 +23.3 23.26 ? 7% perf-profile.calltrace.cycles-pp.iov_iter_zero.dax_iomap_rw.ext4_dax_write_iter.vfs_write.ksys_write
0.00 +24.2 24.16 ? 11% perf-profile.calltrace.cycles-pp.iov_iter_zero.dax_iomap_rw.ext4_file_read_iter.vfs_read.ksys_read
0.02 ?125% +0.1 0.08 ? 19% perf-profile.children.cycles-pp.thread_main
0.00 +0.1 0.06 ? 11% perf-profile.children.cycles-pp.jbd2_transaction_committed
0.00 +0.1 0.06 ? 11% perf-profile.children.cycles-pp.ext4_set_iomap
0.03 ?123% +0.1 0.09 ? 18% perf-profile.children.cycles-pp.ext4_fill_raw_inode
0.00 +0.1 0.07 ? 18% perf-profile.children.cycles-pp.start_this_handle
0.03 ?124% +0.1 0.10 ? 16% perf-profile.children.cycles-pp.jbd2__journal_start
0.04 ?122% +0.1 0.12 ? 10% perf-profile.children.cycles-pp.ext4_map_blocks
0.02 ?122% +0.1 0.10 ? 39% perf-profile.children.cycles-pp.__libc_lseek64
0.03 ?122% +0.1 0.12 ? 18% perf-profile.children.cycles-pp.ext4_do_update_inode
0.03 ?122% +0.1 0.12 ? 15% perf-profile.children.cycles-pp.ext4_mark_iloc_dirty
0.00 +0.1 0.11 ? 11% perf-profile.children.cycles-pp.ext4_es_lookup_extent
0.08 ?122% +0.1 0.20 ? 13% perf-profile.children.cycles-pp.generic_update_time
0.05 ?125% +0.1 0.17 ? 13% perf-profile.children.cycles-pp.task_tick_fair
0.00 +0.1 0.13 ? 11% perf-profile.children.cycles-pp._raw_read_lock
0.07 ?122% +0.1 0.20 ? 18% perf-profile.children.cycles-pp.__ext4_mark_inode_dirty
0.09 ?122% +0.1 0.23 ? 15% perf-profile.children.cycles-pp.file_modified_flags
0.06 ?122% +0.2 0.22 ? 21% perf-profile.children.cycles-pp.touch_atime
0.05 ?122% +0.2 0.20 ? 11% perf-profile.children.cycles-pp.ext4_iomap_begin
0.07 ?124% +0.2 0.24 ? 15% perf-profile.children.cycles-pp.scheduler_tick
0.06 ?122% +0.2 0.23 ? 12% perf-profile.children.cycles-pp.iomap_iter
0.11 ?123% +0.2 0.33 ? 17% perf-profile.children.cycles-pp.ext4_dirty_inode
0.09 ?124% +0.3 0.35 ? 17% perf-profile.children.cycles-pp.update_process_times
0.13 ?123% +0.3 0.39 ? 16% perf-profile.children.cycles-pp.__mark_inode_dirty
0.09 ?124% +0.3 0.36 ? 16% perf-profile.children.cycles-pp.tick_sched_handle
0.11 ?125% +0.3 0.38 ? 14% perf-profile.children.cycles-pp.tick_sched_timer
0.13 ?124% +0.3 0.47 ? 13% perf-profile.children.cycles-pp.__hrtimer_run_queues
0.18 ?124% +0.4 0.57 ? 14% perf-profile.children.cycles-pp.hrtimer_interrupt
0.18 ?124% +0.4 0.58 ? 14% perf-profile.children.cycles-pp.__sysvec_apic_timer_interrupt
0.20 ?123% +0.5 0.66 ? 14% perf-profile.children.cycles-pp.sysvec_apic_timer_interrupt
0.18 ?128% +0.5 0.70 ? 33% perf-profile.children.cycles-pp.start_kernel
0.18 ?128% +0.5 0.70 ? 33% perf-profile.children.cycles-pp.arch_call_rest_init
0.18 ?128% +0.5 0.70 ? 33% perf-profile.children.cycles-pp.rest_init
0.29 ?123% +5.0 5.33 ? 19% perf-profile.children.cycles-pp.asm_sysvec_apic_timer_interrupt
2.84 ?122% +15.5 18.39 ? 11% perf-profile.children.cycles-pp.get_io_u
11.26 ?123% +20.6 31.84 ? 20% perf-profile.children.cycles-pp.intel_idle
11.37 ?123% +20.9 32.29 ? 20% perf-profile.children.cycles-pp.cpuidle_enter
11.37 ?123% +20.9 32.29 ? 20% perf-profile.children.cycles-pp.cpuidle_enter_state
11.27 ?123% +21.0 32.27 ? 20% perf-profile.children.cycles-pp.mwait_idle_with_hints
0.00 +47.4 47.40 ? 9% perf-profile.children.cycles-pp.clear_user_erms
0.00 +47.4 47.43 ? 9% perf-profile.children.cycles-pp.iov_iter_zero
0.02 ?123% +0.1 0.08 ? 18% perf-profile.self.cycles-pp.thread_main
0.00 +0.1 0.07 ? 14% perf-profile.self.cycles-pp.ext4_es_lookup_extent
0.00 +0.1 0.13 ? 11% perf-profile.self.cycles-pp._raw_read_lock
2.81 ?122% +15.4 18.18 ? 11% perf-profile.self.cycles-pp.get_io_u
11.27 ?123% +21.0 32.27 ? 20% perf-profile.self.cycles-pp.mwait_idle_with_hints
0.00 +46.9 46.91 ? 9% perf-profile.self.cycles-pp.clear_user_erms
***************************************************************************************************
lkp-icl-2sp6: 128 threads 2 sockets Intel(R) Xeon(R) Platinum 8358 CPU @ 2.60GHz (Ice Lake) with 128G memory
=========================================================================================
compiler/cpufreq_governor/disk/fs/kconfig/nr_task/rootfs/runtime/tbox_group/test/testcase:
gcc-11/performance/1HDD/ext4/x86_64-rhel-8.3/100/debian-11.1-x86_64-20220510.cgz/300s/lkp-icl-2sp6/disk/reaim
commit:
e6eadc0324 ("iov_iter: Use the direction in the iterator functions")
64ea9d6c5f ("iov_iter: Use IOCB/IOMAP_WRITE if available rather than iterator direction")
e6eadc0324e475e3 64ea9d6c5f473c29c5de97abaa6
---------------- ---------------------------
%stddev %change %stddev
\ | \
4946 +89.7% 9382 reaim.jobs_per_min
49.46 +89.7% 93.83 reaim.jobs_per_min_child
5025 +89.9% 9541 reaim.max_jobs_per_min
121.35 -47.3% 63.97 reaim.parent_time
0.97 ? 7% -43.2% 0.55 ? 3% reaim.std_dev_time
370.80 -10.7% 331.12 reaim.time.elapsed_time
370.80 -10.7% 331.12 reaim.time.elapsed_time.max
120024 +150.0% 300020 reaim.time.file_system_inputs
1338521 +59.1% 2129881 reaim.time.file_system_outputs
20309 ? 4% +37.4% 27914 ? 5% reaim.time.involuntary_context_switches
3514 ? 5% +60.7% 5646 ? 5% reaim.time.major_page_faults
4134743 +66.6% 6889989 reaim.time.minor_page_faults
22.50 ? 3% +86.7% 42.00 ? 2% reaim.time.percent_of_cpu_this_job_got
63.71 ? 2% +66.5% 106.10 reaim.time.system_time
21.05 +66.4% 35.02 reaim.time.user_time
676434 +65.9% 1122483 reaim.time.voluntary_context_switches
30000 +66.7% 50000 reaim.workload
47783 -9.3% 43362 uptime.idle
4.77e+10 -10.9% 4.248e+10 cpuidle..time
4928985 +48.8% 7335403 cpuidle..usage
15054 +42.6% 21463 meminfo.Active
10952 +60.1% 17533 meminfo.Active(file)
11536 +57.1% 18123 meminfo.Buffers
0.05 +0.0 0.08 mpstat.cpu.all.irq%
0.03 +0.0 0.05 mpstat.cpu.all.soft%
0.17 ? 2% +0.1 0.31 ? 2% mpstat.cpu.all.sys%
0.03 +0.0 0.05 mpstat.cpu.all.usr%
1590929 ? 2% +58.9% 2527853 ? 2% numa-numastat.node0.local_node
1590987 ? 2% +58.9% 2527899 ? 2% numa-numastat.node0.numa_hit
1804019 +46.9% 2649463 ? 2% numa-numastat.node1.local_node
1804160 +46.9% 2649614 ? 2% numa-numastat.node1.numa_hit
160.17 +179.6% 447.83 vmstat.io.bi
3103 ? 36% +76.0% 5462 ? 22% vmstat.io.bo
11573 +57.1% 18187 vmstat.memory.buff
6442 +57.9% 10170 vmstat.system.cs
11226 +66.2% 18653 vmstat.system.in
165299 ? 23% -36.4% 105103 ? 16% numa-meminfo.node0.AnonPages
173075 ? 22% -33.0% 115970 ? 18% numa-meminfo.node0.AnonPages.max
170839 ? 22% -35.3% 110555 ? 15% numa-meminfo.node0.Inactive
169315 ? 22% -35.4% 109428 ? 15% numa-meminfo.node0.Inactive(anon)
32962 ? 69% +161.0% 86037 ? 22% numa-meminfo.node1.AnonHugePages
82027 ? 46% +74.4% 143051 ? 12% numa-meminfo.node1.AnonPages
89346 ? 42% +69.9% 151775 ? 9% numa-meminfo.node1.AnonPages.max
84538 ? 44% +71.9% 145338 ? 12% numa-meminfo.node1.Inactive
83881 ? 44% +71.9% 144230 ? 12% numa-meminfo.node1.Inactive(anon)
2950 ? 21% -43.6% 1664 ? 37% sched_debug.cfs_rq:/.load.avg
21451 ? 19% -45.4% 11702 ? 43% sched_debug.cfs_rq:/.load.stddev
35609 ? 10% +56.3% 55645 ? 18% sched_debug.cfs_rq:/.min_vruntime.avg
74742 ? 8% +44.8% 108213 ? 16% sched_debug.cfs_rq:/.min_vruntime.max
21412 ? 11% +60.8% 34431 ? 12% sched_debug.cfs_rq:/.min_vruntime.min
83.47 ? 11% +35.5% 113.07 ? 11% sched_debug.cfs_rq:/.runnable_avg.avg
83.37 ? 11% +35.5% 112.95 ? 11% sched_debug.cfs_rq:/.util_avg.avg
371.30 ? 11% +82.0% 675.86 ? 29% sched_debug.cpu.curr->pid.avg
35562 ? 7% +44.9% 51542 ? 9% sched_debug.cpu.curr->pid.max
3407 ? 8% +57.5% 5365 ? 15% sched_debug.cpu.curr->pid.stddev
0.00 ? 22% -35.8% 0.00 ? 24% sched_debug.cpu.next_balance.stddev
10308 ? 6% +30.1% 13413 ? 8% sched_debug.cpu.nr_switches.avg
5078 ? 16% +49.2% 7576 ? 15% sched_debug.cpu.nr_switches.min
6.83 ? 5% +65.9% 11.33 ? 4% turbostat.Avg_MHz
0.43 +0.3 0.73 turbostat.Busy%
68252 ? 8% +10.5% 75386 ? 5% turbostat.C1
1148569 +48.6% 1706615 turbostat.C1E
2.05 +1.1 3.14 turbostat.C1E%
3552007 +52.2% 5405428 turbostat.C6
4.54 +54.5% 7.02 turbostat.CPU%c1
38.50 +8.7% 41.83 ? 3% turbostat.CoreTmp
4155915 +49.2% 6202633 turbostat.IRQ
21.47 ? 2% -10.6% 19.20 ? 2% turbostat.Pkg%pc2
38.80 -27.9% 27.97 ? 2% turbostat.Pkg%pc6
38.83 ? 2% +6.9% 41.50 ? 2% turbostat.PkgTmp
129.14 +13.3% 146.29 turbostat.PkgWatt
65.52 -1.5% 64.53 turbostat.RAMWatt
41326 ? 23% -36.4% 26278 ? 16% numa-vmstat.node0.nr_anon_pages
115634 ? 12% +57.0% 181517 ? 13% numa-vmstat.node0.nr_dirtied
42326 ? 22% -35.4% 27357 ? 15% numa-vmstat.node0.nr_inactive_anon
103.00 ? 10% -61.5% 39.67 ? 39% numa-vmstat.node0.nr_mlock
81722 ? 13% +57.3% 128519 ? 13% numa-vmstat.node0.nr_written
42326 ? 22% -35.4% 27357 ? 15% numa-vmstat.node0.nr_zone_inactive_anon
1591186 ? 2% +58.9% 2527861 ? 2% numa-vmstat.node0.numa_hit
1591128 ? 2% +58.9% 2527816 ? 2% numa-vmstat.node0.numa_local
20505 ? 46% +74.4% 35764 ? 12% numa-vmstat.node1.nr_anon_pages
53024 ? 27% +87.7% 99527 ? 24% numa-vmstat.node1.nr_dirtied
20966 ? 44% +72.0% 36055 ? 12% numa-vmstat.node1.nr_inactive_anon
36742 ? 30% +87.0% 68703 ? 25% numa-vmstat.node1.nr_written
20966 ? 44% +72.0% 36055 ? 12% numa-vmstat.node1.nr_zone_inactive_anon
1804367 +46.9% 2649736 ? 2% numa-vmstat.node1.numa_hit
1804226 +46.9% 2649586 ? 2% numa-vmstat.node1.numa_local
2738 +60.1% 4384 proc-vmstat.nr_active_file
168659 +66.6% 281044 proc-vmstat.nr_dirtied
108.17 ? 5% -55.2% 48.50 ? 10% proc-vmstat.nr_mlock
2518 -4.1% 2416 proc-vmstat.nr_shmem
31530 +1.3% 31939 proc-vmstat.nr_slab_reclaimable
118464 +66.5% 197222 proc-vmstat.nr_written
2738 +60.1% 4384 proc-vmstat.nr_zone_active_file
3398069 +52.4% 5179028 proc-vmstat.numa_hit
3396603 +52.5% 5178832 proc-vmstat.numa_local
19001 +63.8% 31122 proc-vmstat.pgactivate
3531992 +51.9% 5363392 proc-vmstat.pgalloc_normal
5059472 +52.9% 7736238 proc-vmstat.pgfault
3507816 +52.1% 5334638 proc-vmstat.pgfree
60412 +149.0% 150412 proc-vmstat.pgpgin
1167224 ? 37% +57.1% 1833427 ? 23% proc-vmstat.pgpgout
48082 -7.2% 44612 proc-vmstat.pgreuse
2824960 -9.5% 2556672 proc-vmstat.unevictable_pgs_scanned
***************************************************************************************************
lkp-csl-2sp7: 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz (Cascade Lake) with 512G memory
=========================================================================================
bs/compiler/cpufreq_governor/disk/fs/ioengine/kconfig/mount_option/nr_task/rootfs/runtime/rw/tbox_group/test_size/testcase/time_based:
2M/gcc-11/performance/2pmem/ext4/libaio/x86_64-rhel-8.3/dax/50%/debian-11.1-x86_64-20220510.cgz/200s/read/lkp-csl-2sp7/200G/fio-basic/tb
commit:
e6eadc0324 ("iov_iter: Use the direction in the iterator functions")
64ea9d6c5f ("iov_iter: Use IOCB/IOMAP_WRITE if available rather than iterator direction")
e6eadc0324e475e3 64ea9d6c5f473c29c5de97abaa6
---------------- ---------------------------
%stddev %change %stddev
\ | \
27.54 ? 9% +69.2 96.71 fio.latency_100ms%
0.01 -0.0 0.00 fio.latency_100us%
0.08 +0.1 0.17 fio.latency_10ms%
0.01 +0.0 0.02 fio.latency_10us%
0.13 ? 5% +0.2 0.29 fio.latency_20ms%
0.01 +0.0 0.03 fio.latency_20us%
71.47 ? 3% -71.4 0.10 fio.latency_250ms%
0.00 ?141% +0.0 0.05 fio.latency_2ms%
0.03 +0.0 0.05 fio.latency_4ms%
0.25 ? 45% -0.2 0.04 fio.latency_500ms%
0.43 ? 2% +2.1 2.55 fio.latency_50ms%
0.04 ? 2% -0.0 0.01 fio.latency_50us%
0.02 ? 38% -0.0 0.01 fio.latency_750ms%
25280 +127.0% 57386 fio.read_bw_MBps
1.402e+08 -62.8% 52166656 fio.read_clat_90%_us
1.409e+08 -63.0% 52166656 fio.read_clat_95%_us
1.744e+08 ? 2% -68.0% 55836672 fio.read_clat_99%_us
1.168e+08 -55.9% 51450892 fio.read_clat_mean_us
25859755 ? 3% -66.6% 8643794 fio.read_clat_stddev
12640 +127.0% 28693 fio.read_iops
3768596 -55.8% 1667016 fio.read_slat_mean_us
822091 ? 3% -64.9% 288388 fio.read_slat_stddev
14587 -10.2% 13097 fio.time.involuntary_context_switches
76.34 ? 12% -56.3% 33.35 fio.time.user_time
20547 +3.1% 21174 fio.time.voluntary_context_switches
2528054 +127.0% 5738728 fio.workload
9.493e+09 +40.3% 1.332e+10 cpuidle..time
19669558 +41.3% 27786024 cpuidle..usage
49.49 +15.9% 57.36 iostat.cpu.idle
50.01 -15.3% 42.37 iostat.cpu.system
49.01 +8.0 57.04 mpstat.cpu.all.idle%
2.43 +0.4 2.83 mpstat.cpu.all.irq%
47.86 -8.2 39.69 mpstat.cpu.all.sys%
0.50 ? 10% -0.2 0.27 mpstat.cpu.all.usr%
49.00 +16.3% 57.00 vmstat.cpu.id
103.33 ? 17% -93.2% 7.00 vmstat.io.bo
47.00 -17.0% 39.00 vmstat.procs.r
195642 -1.3% 193102 vmstat.system.in
207006 ? 12% +36.0% 281618 numa-numastat.node0.local_node
221613 ? 12% +41.1% 312658 numa-numastat.node0.numa_hit
14607 ? 86% +112.5% 31040 numa-numastat.node0.other_node
311816 ? 10% -17.1% 258424 numa-numastat.node1.numa_hit
16596 ? 76% -99.1% 150.00 numa-numastat.node1.other_node
60852 ? 6% -21.6% 47683 meminfo.Active
39736 ? 9% -33.2% 26556 meminfo.Active(anon)
3113621 -17.3% 2576069 meminfo.AnonHugePages
3386543 -16.2% 2836667 meminfo.AnonPages
5471585 -16.5% 4567705 meminfo.Committed_AS
10203136 ? 8% -32.7% 6868992 meminfo.DirectMap2M
3406759 -16.3% 2852721 meminfo.Inactive
3403984 -16.3% 2850189 meminfo.Inactive(anon)
14220 -13.8% 12254 meminfo.PageTables
56806 ? 7% -28.8% 40419 meminfo.Shmem
1446 -15.7% 1219 turbostat.Avg_MHz
51.78 -7.3 44.44 turbostat.Busy%
2799 -1.8% 2749 turbostat.Bzy_MHz
41071 ? 10% -20.7% 32578 turbostat.C1
17513765 ? 8% -19.0% 14194816 turbostat.C1E
39.55 ? 15% -27.3 12.28 turbostat.C1E%
2087050 ? 70% +547.6% 13516594 turbostat.C6
9.07 ? 69% +35.1 44.17 turbostat.C6%
0.20 ? 4% +1898.3% 3.93 turbostat.CPU%c6
61.33 -23.4% 47.00 turbostat.CoreTmp
0.04 -100.0% 0.00 turbostat.IPC
39925293 +19.3% 47644339 turbostat.IRQ
7201 ? 11% +239.4% 24439 turbostat.POLL
61.33 -21.7% 48.00 turbostat.PkgTmp
252.93 -3.3% 244.69 turbostat.PkgWatt
58.61 -12.0% 51.56 turbostat.RAMWatt
9967 ? 9% -33.4% 6639 proc-vmstat.nr_active_anon
845290 -16.1% 709177 proc-vmstat.nr_anon_pages
1517 -17.2% 1257 proc-vmstat.nr_anon_transparent_hugepages
849647 -16.1% 712557 proc-vmstat.nr_inactive_anon
691.67 -8.6% 632.00 proc-vmstat.nr_inactive_file
11249 -8.7% 10266 proc-vmstat.nr_mapped
3542 -13.5% 3063 proc-vmstat.nr_page_table_pages
14239 ? 7% -29.0% 10105 proc-vmstat.nr_shmem
49481 -1.3% 48815 proc-vmstat.nr_slab_unreclaimable
9967 ? 9% -33.4% 6639 proc-vmstat.nr_zone_active_anon
849647 -16.1% 712557 proc-vmstat.nr_zone_inactive_anon
691.67 -8.6% 632.00 proc-vmstat.nr_zone_inactive_file
27135 ? 5% -21.3% 21342 proc-vmstat.numa_hint_faults_local
535623 +7.1% 573676 proc-vmstat.numa_hit
23491 -18.3% 19198 proc-vmstat.numa_huge_pte_updates
504420 +7.5% 542486 proc-vmstat.numa_local
12075054 -18.3% 9860674 proc-vmstat.numa_pte_updates
16357 ? 6% -39.0% 9979 proc-vmstat.pgactivate
1743457 +2.5% 1786342 proc-vmstat.pgalloc_normal
698136 +9.1% 761457 proc-vmstat.pgfault
1705291 +2.8% 1753621 proc-vmstat.pgfree
39558 ? 3% +15.0% 45480 proc-vmstat.pgreuse
1560064 +19.5% 1864704 proc-vmstat.unevictable_pgs_scanned
8738 ?100% +164.0% 23070 numa-meminfo.node0.Active
1409 ? 14% +42.4% 2007 numa-meminfo.node0.Active(anon)
7328 ?118% +187.4% 21063 numa-meminfo.node0.Active(file)
1581062 ? 2% -19.3% 1275629 numa-meminfo.node0.AnonHugePages
1725552 ? 2% -17.5% 1423635 numa-meminfo.node0.AnonPages
2287942 ? 27% +42.8% 3267712 numa-meminfo.node0.AnonPages.max
1733797 -17.1% 1437793 numa-meminfo.node0.Inactive
1732845 -17.2% 1435355 numa-meminfo.node0.Inactive(anon)
79100 -16.5% 66071 numa-meminfo.node0.KReclaimable
79100 -16.5% 66071 numa-meminfo.node0.SReclaimable
111641 ? 3% -12.7% 97495 numa-meminfo.node0.SUnreclaim
190741 -14.2% 163567 numa-meminfo.node0.Slab
52238 ? 21% -52.9% 24620 numa-meminfo.node1.Active
38449 ? 10% -36.1% 24556 numa-meminfo.node1.Active(anon)
13787 ? 63% -99.5% 64.00 numa-meminfo.node1.Active(file)
1529392 -15.3% 1295492 numa-meminfo.node1.AnonHugePages
1657609 ? 2% -15.1% 1408075 numa-meminfo.node1.AnonPages
2677658 ? 22% -33.8% 1772936 numa-meminfo.node1.AnonPages.max
1669831 -15.6% 1409979 numa-meminfo.node1.Inactive
1668008 -15.5% 1409881 numa-meminfo.node1.Inactive(anon)
1823 ? 66% -94.7% 97.00 numa-meminfo.node1.Inactive(file)
31911 ? 4% +40.1% 44709 numa-meminfo.node1.KReclaimable
8625 ? 4% -8.7% 7872 numa-meminfo.node1.KernelStack
9770 ? 53% -85.5% 1416 numa-meminfo.node1.Mapped
8619 ? 50% -78.6% 1848 numa-meminfo.node1.PageTables
31911 ? 4% +40.1% 44709 numa-meminfo.node1.SReclaimable
86312 ? 3% +13.3% 97764 numa-meminfo.node1.SUnreclaim
48216 ? 14% -45.0% 26528 numa-meminfo.node1.Shmem
118223 +20.5% 142473 numa-meminfo.node1.Slab
106246 ? 69% +83.5% 194925 numa-meminfo.node1.Unevictable
352.33 ? 14% +42.5% 502.00 numa-vmstat.node0.nr_active_anon
1831 ?118% +187.5% 5265 numa-vmstat.node0.nr_active_file
431431 -17.5% 355907 numa-vmstat.node0.nr_anon_pages
772.00 ? 2% -19.4% 622.00 numa-vmstat.node0.nr_anon_transparent_hugepages
433200 -17.2% 358837 numa-vmstat.node0.nr_inactive_anon
19774 -16.5% 16517 numa-vmstat.node0.nr_slab_reclaimable
27907 ? 3% -12.7% 24373 numa-vmstat.node0.nr_slab_unreclaimable
68.33 ? 86% +324.4% 290.00 numa-vmstat.node0.nr_written
352.00 ? 14% +42.6% 502.00 numa-vmstat.node0.nr_zone_active_anon
1831 ?118% +187.5% 5265 numa-vmstat.node0.nr_zone_active_file
433200 -17.2% 358837 numa-vmstat.node0.nr_zone_inactive_anon
221313 ? 12% +41.4% 312932 numa-vmstat.node0.numa_hit
206706 ? 12% +36.4% 281892 numa-vmstat.node0.numa_local
14607 ? 86% +112.5% 31040 numa-vmstat.node0.numa_other
9602 ? 10% -36.1% 6140 numa-vmstat.node1.nr_active_anon
3446 ? 63% -99.5% 16.00 numa-vmstat.node1.nr_active_file
415090 ? 2% -15.2% 352032 numa-vmstat.node1.nr_anon_pages
747.67 -15.5% 632.00 numa-vmstat.node1.nr_anon_transparent_hugepages
417671 ? 2% -15.6% 352483 numa-vmstat.node1.nr_inactive_anon
455.00 ? 66% -94.7% 24.00 numa-vmstat.node1.nr_inactive_file
8624 ? 4% -8.6% 7879 numa-vmstat.node1.nr_kernel_stack
2476 ? 52% -85.6% 356.00 numa-vmstat.node1.nr_mapped
2150 ? 49% -78.5% 463.00 numa-vmstat.node1.nr_page_table_pages
12065 ? 14% -45.0% 6634 numa-vmstat.node1.nr_shmem
7974 ? 4% +40.2% 11177 numa-vmstat.node1.nr_slab_reclaimable
21576 ? 3% +13.3% 24441 numa-vmstat.node1.nr_slab_unreclaimable
26561 ? 69% +83.5% 48731 numa-vmstat.node1.nr_unevictable
232.67 ? 81% -96.6% 8.00 numa-vmstat.node1.nr_written
9602 ? 10% -36.1% 6140 numa-vmstat.node1.nr_zone_active_anon
3446 ? 63% -99.5% 16.00 numa-vmstat.node1.nr_zone_active_file
417671 ? 2% -15.6% 352483 numa-vmstat.node1.nr_zone_inactive_anon
455.00 ? 66% -94.7% 24.00 numa-vmstat.node1.nr_zone_inactive_file
26561 ? 69% +83.5% 48731 numa-vmstat.node1.nr_zone_unevictable
311712 ? 10% -17.0% 258614 numa-vmstat.node1.numa_hit
16596 ? 76% -99.1% 150.00 numa-vmstat.node1.numa_other
0.49 ? 9% -18.2% 0.40 sched_debug.cfs_rq:/.h_nr_running.avg
465919 ? 10% -18.9% 378087 sched_debug.cfs_rq:/.load.avg
365287 ? 3% -8.6% 333715 sched_debug.cfs_rq:/.load.stddev
481.31 ? 7% -20.5% 382.74 sched_debug.cfs_rq:/.load_avg.avg
1050 -10.5% 940.20 sched_debug.cfs_rq:/.load_avg.max
428.86 -12.5% 375.44 sched_debug.cfs_rq:/.load_avg.stddev
76204 ? 13% +31.9% 100504 sched_debug.cfs_rq:/.min_vruntime.avg
39740 ? 9% +18.1% 46947 sched_debug.cfs_rq:/.min_vruntime.min
17319 ? 21% +69.5% 29363 sched_debug.cfs_rq:/.min_vruntime.stddev
0.49 ? 9% -17.7% 0.40 sched_debug.cfs_rq:/.nr_running.avg
284.44 ? 14% -28.0% 204.80 sched_debug.cfs_rq:/.removed.load_avg.max
144.64 ? 13% -28.4% 103.60 sched_debug.cfs_rq:/.removed.runnable_avg.max
144.64 ? 13% -28.4% 103.60 sched_debug.cfs_rq:/.removed.util_avg.max
546.97 ? 6% -21.6% 428.59 sched_debug.cfs_rq:/.runnable_avg.avg
1155 ? 6% -18.5% 942.20 sched_debug.cfs_rq:/.runnable_avg.max
447.05 -13.9% 385.08 sched_debug.cfs_rq:/.runnable_avg.stddev
5587 ?372% +580.9% 38039 sched_debug.cfs_rq:/.spread0.avg
-30962 -49.9% -15523 sched_debug.cfs_rq:/.spread0.min
17327 ? 21% +69.5% 29364 sched_debug.cfs_rq:/.spread0.stddev
545.86 ? 6% -21.5% 428.25 sched_debug.cfs_rq:/.util_avg.avg
1119 ? 4% -15.8% 941.80 sched_debug.cfs_rq:/.util_avg.max
445.99 -13.7% 384.75 sched_debug.cfs_rq:/.util_avg.stddev
191.65 ? 7% -24.1% 145.38 sched_debug.cfs_rq:/.util_est_enqueued.avg
763.44 ? 11% -46.3% 409.80 sched_debug.cfs_rq:/.util_est_enqueued.max
184.14 -24.0% 140.00 sched_debug.cfs_rq:/.util_est_enqueued.stddev
936335 +11.3% 1042304 sched_debug.cpu.avg_idle.avg
1113428 +98.0% 2204219 sched_debug.cpu.avg_idle.max
484973 -32.9% 325215 sched_debug.cpu.avg_idle.min
105940 +168.2% 284137 sched_debug.cpu.avg_idle.stddev
25.77 ? 10% +218.7% 82.14 sched_debug.cpu.clock.stddev
2206 ? 4% -16.4% 1845 sched_debug.cpu.curr->pid.avg
7105 ? 5% +15.1% 8178 sched_debug.cpu.curr->pid.max
2394 ? 3% -9.5% 2167 sched_debug.cpu.curr->pid.stddev
553458 ? 2% +19.5% 661645 sched_debug.cpu.max_idle_balance_cost.max
8200 ? 43% +392.0% 40345 sched_debug.cpu.max_idle_balance_cost.stddev
0.00 ? 8% +168.0% 0.00 sched_debug.cpu.next_balance.stddev
0.41 ? 4% -17.7% 0.33 sched_debug.cpu.nr_running.avg
0.45 ? 2% -16.3% 0.38 sched_debug.cpu.nr_running.stddev
4341 ? 6% +17.3% 5092 sched_debug.cpu.nr_switches.avg
1292 -10.8% 1152 sched_debug.cpu.nr_switches.min
2633 ? 11% +32.1% 3479 sched_debug.cpu.nr_switches.stddev
42.19 +5.5% 44.52 perf-stat.i.MPKI
3.476e+09 -94.8% 1.817e+08 perf-stat.i.branch-instructions
0.13 ? 2% +3.3 3.40 perf-stat.i.branch-miss-rate%
4260734 +45.4% 6193178 perf-stat.i.branch-misses
94.70 -55.7 38.96 perf-stat.i.cache-miss-rate%
8.289e+08 -98.6% 11388051 perf-stat.i.cache-misses
8.747e+08 -95.7% 37441183 perf-stat.i.cache-references
6.62 +1894.1% 132.02 perf-stat.i.cpi
1.366e+11 -16.9% 1.136e+11 perf-stat.i.cpu-cycles
124.03 -8.7% 113.22 perf-stat.i.cpu-migrations
167.03 +5270.0% 8969 perf-stat.i.cycles-between-cache-misses
0.00 ? 6% +0.3 0.26 perf-stat.i.dTLB-load-miss-rate%
145663 ? 5% +300.9% 584004 perf-stat.i.dTLB-load-misses
3.55e+09 -92.9% 2.537e+08 perf-stat.i.dTLB-loads
0.00 ? 2% +0.0 0.05 perf-stat.i.dTLB-store-miss-rate%
40129 ? 2% +61.9% 64958 perf-stat.i.dTLB-store-misses
3.429e+09 -96.0% 1.355e+08 perf-stat.i.dTLB-stores
43.30 +17.4 60.74 perf-stat.i.iTLB-load-miss-rate%
1204294 ? 4% +175.2% 3314508 perf-stat.i.iTLB-load-misses
1575387 +24.0% 1953164 perf-stat.i.iTLB-loads
2.07e+10 -95.5% 9.308e+08 perf-stat.i.instructions
17218 ? 4% -97.8% 375.10 perf-stat.i.instructions-per-iTLB-miss
0.15 -81.2% 0.03 perf-stat.i.ipc
1.42 -16.9% 1.18 perf-stat.i.metric.GHz
938.97 ? 2% -46.1% 505.78 perf-stat.i.metric.K/sec
119.86 -95.1% 5.91 perf-stat.i.metric.M/sec
2609 ? 2% -8.4% 2390 perf-stat.i.minor-faults
50.80 +5.2 56.00 perf-stat.i.node-load-miss-rate%
44268862 -99.3% 289365 perf-stat.i.node-load-misses
42927271 ? 3% -99.3% 307807 perf-stat.i.node-loads
0.60 ? 9% +10.4 11.00 perf-stat.i.node-store-miss-rate%
791120 ? 13% -91.1% 70479 perf-stat.i.node-store-misses
1.79e+08 -95.2% 8530049 perf-stat.i.node-stores
2610 -8.4% 2390 perf-stat.i.page-faults
42.24 -7.3% 39.16 perf-stat.overall.MPKI
0.12 +3.2 3.32 perf-stat.overall.branch-miss-rate%
94.76 -63.5 31.27 perf-stat.overall.cache-miss-rate%
6.60 +1750.9% 122.19 perf-stat.overall.cpi
164.91 +5950.5% 9977 perf-stat.overall.cycles-between-cache-misses
0.00 ? 5% +0.2 0.22 perf-stat.overall.dTLB-load-miss-rate%
0.00 ? 2% +0.0 0.05 perf-stat.overall.dTLB-store-miss-rate%
43.32 +19.7 63.02 perf-stat.overall.iTLB-load-miss-rate%
17223 ? 4% -98.4% 281.41 perf-stat.overall.instructions-per-iTLB-miss
0.15 -94.6% 0.01 perf-stat.overall.ipc
50.78 -1.1 49.66 perf-stat.overall.node-load-miss-rate%
0.45 ? 13% +0.4 0.82 perf-stat.overall.node-store-miss-rate%
1646075 -97.6% 39879 perf-stat.overall.path-length
3.458e+09 -94.7% 1.832e+08 perf-stat.ps.branch-instructions
4239989 +43.4% 6081614 perf-stat.ps.branch-misses
8.245e+08 -98.6% 11479341 perf-stat.ps.cache-misses
8.7e+08 -95.8% 36710005 perf-stat.ps.cache-references
1.36e+11 -15.8% 1.145e+11 perf-stat.ps.cpu-cycles
122.93 -8.3% 112.76 perf-stat.ps.cpu-migrations
144896 ? 5% +282.8% 554706 perf-stat.ps.dTLB-load-misses
3.532e+09 -92.8% 2.552e+08 perf-stat.ps.dTLB-loads
39845 ? 2% +57.7% 62850 perf-stat.ps.dTLB-store-misses
3.41e+09 -96.0% 1.359e+08 perf-stat.ps.dTLB-stores
1197745 ? 4% +178.1% 3330985 perf-stat.ps.iTLB-load-misses
1566364 +24.8% 1954999 perf-stat.ps.iTLB-loads
2.059e+10 -95.4% 9.374e+08 perf-stat.ps.instructions
2594 ? 2% -7.1% 2408 perf-stat.ps.minor-faults
44022802 -99.3% 308451 perf-stat.ps.node-load-misses
42696925 ? 3% -99.3% 312663 perf-stat.ps.node-loads
807658 ? 14% -91.1% 71548 perf-stat.ps.node-store-misses
1.78e+08 -95.2% 8602052 perf-stat.ps.node-stores
2594 ? 2% -7.1% 2408 perf-stat.ps.page-faults
4.161e+12 -94.5% 2.289e+11 perf-stat.total.instructions
66.74 ? 9% -66.7 0.00 perf-profile.calltrace.cycles-pp.dax_iomap_iter.dax_iomap_rw.ext4_file_read_iter.aio_read.io_submit_one
66.68 ? 9% -66.7 0.00 perf-profile.calltrace.cycles-pp._copy_mc_to_iter.dax_iomap_iter.dax_iomap_rw.ext4_file_read_iter.aio_read
66.65 ? 9% -66.6 0.00 perf-profile.calltrace.cycles-pp.copyout_mc._copy_mc_to_iter.dax_iomap_iter.dax_iomap_rw.ext4_file_read_iter
66.65 ? 9% -66.6 0.00 perf-profile.calltrace.cycles-pp.copy_mc_to_user.copyout_mc._copy_mc_to_iter.dax_iomap_iter.dax_iomap_rw
66.44 ? 9% -66.4 0.00 perf-profile.calltrace.cycles-pp.copy_mc_fragile.copy_mc_to_user.copyout_mc._copy_mc_to_iter.dax_iomap_iter
67.34 ? 9% -11.5 55.88 perf-profile.calltrace.cycles-pp.syscall
67.32 ? 9% -11.4 55.87 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.syscall
67.31 ? 9% -11.4 55.87 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall
67.18 ? 9% -11.3 55.84 perf-profile.calltrace.cycles-pp.__x64_sys_io_submit.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall
67.17 ? 9% -11.3 55.84 perf-profile.calltrace.cycles-pp.io_submit_one.__x64_sys_io_submit.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall
67.02 ? 9% -11.2 55.82 perf-profile.calltrace.cycles-pp.aio_read.io_submit_one.__x64_sys_io_submit.do_syscall_64.entry_SYSCALL_64_after_hwframe
66.94 ? 9% -11.1 55.80 perf-profile.calltrace.cycles-pp.ext4_file_read_iter.aio_read.io_submit_one.__x64_sys_io_submit.do_syscall_64
66.88 ? 9% -11.1 55.77 perf-profile.calltrace.cycles-pp.dax_iomap_rw.ext4_file_read_iter.aio_read.io_submit_one.__x64_sys_io_submit
31.73 ? 20% +11.0 42.75 perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry
31.39 ? 21% +11.3 42.72 perf-profile.calltrace.cycles-pp.mwait_idle_with_hints.intel_idle.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call
31.39 ? 21% +11.3 42.72 perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle
31.75 ? 20% +11.5 43.23 perf-profile.calltrace.cycles-pp.secondary_startup_64_no_verify
31.12 ? 20% +11.6 42.76 perf-profile.calltrace.cycles-pp.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
31.11 ? 20% +11.6 42.75 perf-profile.calltrace.cycles-pp.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary
31.13 ? 20% +11.6 42.78 perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
31.13 ? 20% +11.6 42.78 perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
31.13 ? 20% +11.6 42.78 perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64_no_verify
0.00 +15.9 15.86 perf-profile.calltrace.cycles-pp.clear_user_erms.iov_iter_zero.dax_iomap_rw.ext4_file_read_iter.aio_read
0.00 +55.7 55.73 perf-profile.calltrace.cycles-pp.iov_iter_zero.dax_iomap_rw.ext4_file_read_iter.aio_read.io_submit_one
0.00 +80.1 80.09 perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt.clear_user_erms.iov_iter_zero.dax_iomap_rw.ext4_file_read_iter
66.74 ? 9% -66.7 0.00 perf-profile.children.cycles-pp.dax_iomap_iter
66.68 ? 9% -66.7 0.00 perf-profile.children.cycles-pp._copy_mc_to_iter
66.66 ? 9% -66.7 0.00 perf-profile.children.cycles-pp.copyout_mc
66.65 ? 9% -66.6 0.00 perf-profile.children.cycles-pp.copy_mc_fragile
66.65 ? 9% -66.6 0.00 perf-profile.children.cycles-pp.copy_mc_to_user
67.35 ? 9% -11.5 55.88 perf-profile.children.cycles-pp.syscall
67.18 ? 9% -11.3 55.84 perf-profile.children.cycles-pp.__x64_sys_io_submit
67.17 ? 9% -11.3 55.84 perf-profile.children.cycles-pp.io_submit_one
67.56 ? 9% -11.2 56.35 perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
67.55 ? 9% -11.2 56.35 perf-profile.children.cycles-pp.do_syscall_64
67.02 ? 9% -11.2 55.82 perf-profile.children.cycles-pp.aio_read
66.94 ? 9% -11.1 55.80 perf-profile.children.cycles-pp.ext4_file_read_iter
66.88 ? 9% -11.1 55.77 perf-profile.children.cycles-pp.dax_iomap_rw
0.83 ? 10% -0.4 0.47 perf-profile.children.cycles-pp.sysvec_apic_timer_interrupt
0.75 ? 8% -0.3 0.44 perf-profile.children.cycles-pp.__sysvec_apic_timer_interrupt
0.73 ? 7% -0.3 0.43 perf-profile.children.cycles-pp.hrtimer_interrupt
0.60 ? 2% -0.2 0.36 perf-profile.children.cycles-pp.__hrtimer_run_queues
0.43 ? 17% -0.2 0.24 perf-profile.children.cycles-pp.tick_sched_timer
0.40 ? 15% -0.2 0.21 perf-profile.children.cycles-pp.update_process_times
0.40 ? 16% -0.2 0.22 perf-profile.children.cycles-pp.tick_sched_handle
0.29 ? 13% -0.2 0.14 perf-profile.children.cycles-pp.scheduler_tick
0.61 ? 23% -0.2 0.46 perf-profile.children.cycles-pp.start_kernel
0.61 ? 23% -0.2 0.46 perf-profile.children.cycles-pp.arch_call_rest_init
0.61 ? 23% -0.2 0.46 perf-profile.children.cycles-pp.rest_init
0.22 ? 11% -0.1 0.11 perf-profile.children.cycles-pp.task_tick_fair
0.02 ?141% +0.0 0.06 perf-profile.children.cycles-pp.ksys_write
0.02 ?141% +0.0 0.06 perf-profile.children.cycles-pp.vfs_write
0.02 ?141% +0.0 0.06 perf-profile.children.cycles-pp.__intel_pmu_enable_all
0.00 +0.1 0.05 perf-profile.children.cycles-pp.kernel_clone
0.00 +0.1 0.05 perf-profile.children.cycles-pp.read
0.00 +0.1 0.05 perf-profile.children.cycles-pp.exec_binprm
0.00 +0.1 0.05 perf-profile.children.cycles-pp.search_binary_handler
0.00 +0.1 0.05 perf-profile.children.cycles-pp.load_elf_binary
0.00 +0.1 0.05 perf-profile.children.cycles-pp.record__pushfn
0.00 +0.1 0.05 perf-profile.children.cycles-pp.writen
0.00 +0.1 0.05 perf-profile.children.cycles-pp.__libc_write
0.00 +0.1 0.05 perf-profile.children.cycles-pp.generic_file_write_iter
0.00 +0.1 0.06 perf-profile.children.cycles-pp.irq_work_run_list
0.00 +0.1 0.06 perf-profile.children.cycles-pp.asm_sysvec_irq_work
0.00 +0.1 0.06 perf-profile.children.cycles-pp.sysvec_irq_work
0.00 +0.1 0.06 perf-profile.children.cycles-pp.__sysvec_irq_work
0.00 +0.1 0.06 perf-profile.children.cycles-pp.irq_work_single
0.00 +0.1 0.06 perf-profile.children.cycles-pp.irq_work_run
0.00 +0.1 0.06 perf-profile.children.cycles-pp._printk
0.00 +0.1 0.06 perf-profile.children.cycles-pp.vprintk_emit
0.00 +0.1 0.06 perf-profile.children.cycles-pp.console_unlock
0.00 +0.1 0.06 perf-profile.children.cycles-pp.console_flush_all
0.00 +0.1 0.06 perf-profile.children.cycles-pp.console_emit_next_record
0.00 +0.1 0.06 perf-profile.children.cycles-pp.ksys_read
0.00 +0.1 0.06 perf-profile.children.cycles-pp.memcpy_erms
0.00 +0.1 0.06 perf-profile.children.cycles-pp.serial8250_console_write
0.00 +0.1 0.06 perf-profile.children.cycles-pp.vfs_read
0.00 +0.1 0.06 perf-profile.children.cycles-pp.wait_for_lsr
0.04 ? 71% +0.1 0.10 perf-profile.children.cycles-pp.worker_thread
0.02 ?141% +0.1 0.08 perf-profile.children.cycles-pp.drm_fb_helper_damage_work
0.02 ?141% +0.1 0.08 perf-profile.children.cycles-pp.drm_fbdev_fb_dirty
0.02 ?141% +0.1 0.09 perf-profile.children.cycles-pp.process_one_work
0.00 +0.1 0.07 perf-profile.children.cycles-pp.bprm_execve
0.00 +0.1 0.07 perf-profile.children.cycles-pp.__handle_mm_fault
0.00 +0.1 0.07 perf-profile.children.cycles-pp.io_serial_in
0.07 ? 74% +0.1 0.14 perf-profile.children.cycles-pp.kthread
0.00 +0.1 0.08 perf-profile.children.cycles-pp.handle_mm_fault
0.07 ? 74% +0.1 0.15 perf-profile.children.cycles-pp.ret_from_fork
0.00 +0.1 0.09 perf-profile.children.cycles-pp.asm_exc_page_fault
0.00 +0.1 0.09 perf-profile.children.cycles-pp.exc_page_fault
0.00 +0.1 0.09 perf-profile.children.cycles-pp.do_user_addr_fault
0.02 ?141% +0.1 0.11 perf-profile.children.cycles-pp.execve
0.02 ?141% +0.1 0.11 perf-profile.children.cycles-pp.__x64_sys_execve
0.02 ?141% +0.1 0.11 perf-profile.children.cycles-pp.do_execveat_common
31.74 ? 20% +11.5 43.21 perf-profile.children.cycles-pp.cpuidle_idle_call
31.75 ? 20% +11.5 43.23 perf-profile.children.cycles-pp.secondary_startup_64_no_verify
31.75 ? 20% +11.5 43.23 perf-profile.children.cycles-pp.cpu_startup_entry
31.75 ? 20% +11.5 43.23 perf-profile.children.cycles-pp.do_idle
31.73 ? 20% +11.5 43.21 perf-profile.children.cycles-pp.cpuidle_enter
31.73 ? 20% +11.5 43.21 perf-profile.children.cycles-pp.cpuidle_enter_state
31.70 ? 20% +11.5 43.19 perf-profile.children.cycles-pp.mwait_idle_with_hints
31.55 ? 21% +11.6 43.17 perf-profile.children.cycles-pp.intel_idle
31.13 ? 20% +11.6 42.78 perf-profile.children.cycles-pp.start_secondary
1.20 ? 27% +39.1 40.32 perf-profile.children.cycles-pp.asm_sysvec_apic_timer_interrupt
0.00 +55.7 55.71 perf-profile.children.cycles-pp.clear_user_erms
0.00 +55.7 55.73 perf-profile.children.cycles-pp.iov_iter_zero
65.79 ? 9% -65.8 0.00 perf-profile.self.cycles-pp.copy_mc_fragile
0.02 ?141% +0.0 0.06 perf-profile.self.cycles-pp.__intel_pmu_enable_all
0.00 +0.1 0.05 perf-profile.self.cycles-pp.io_serial_in
31.70 ? 20% +11.5 43.19 perf-profile.self.cycles-pp.mwait_idle_with_hints
0.00 +55.2 55.20 perf-profile.self.cycles-pp.clear_user_erms
To reproduce:
git clone https://github.com/intel/lkp-tests.git
cd lkp-tests
sudo bin/lkp install job.yaml # job file is attached in this email
bin/lkp split-job --compatible job.yaml # generate the yaml file for lkp run
sudo bin/lkp run generated-yaml-file
# if come across any failure that blocks the test,
# please remove ~/.lkp and /lkp dir to run from a clean state.
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.
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests