| 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 linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <20200204091347.GI12867@shao2-debian>
Date: Tue, 4 Feb 2020 17:13:48 +0800
From: kernel test robot <rong.a.chen@...el.com>
To: Matthew Bobrowski <mbobrowski@...browski.org>
Cc: Theodore Ts'o <tytso@....edu>, Jan Kara <jack@...e.cz>,
Ritesh Harjani <riteshh@...ux.ibm.com>,
LKML <linux-kernel@...r.kernel.org>,
Linus Torvalds <torvalds@...ux-foundation.org>,
lkp@...ts.01.org
Subject: [ext4] 569342dc24: fio.read_bw_MBps 56.4% improvement
Greeting,
FYI, we noticed a 56.4% improvement of fio.read_bw_MBps due to commit:
commit: 569342dc2485392e95b6a626281708c25014ba37 ("ext4: move inode extension/truncate code out from ->iomap_end() callback")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
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: rw
bs: 4k
ioengine: libaio
test_size: 200G
cpufreq_governor: performance
ucode: 0x500002c
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_bw_MBps 54.0% improvement |
| test machine | 96 threads Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 256G memory |
| test parameters | bs=4k |
| | cpufreq_governor=performance |
| | disk=2pmem |
| | fs=ext4 |
| | ioengine=sync |
| | mount_option=dax |
| | nr_task=50% |
| | runtime=200s |
| | rw=randwrite |
| | test_size=200G |
| | time_based=tb |
| | ucode=0x500002c |
+------------------+----------------------------------------------------------------------+
| testcase: change | fio-basic: fio.write_bw_MBps 68.8% improvement |
| test machine | 96 threads Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 256G memory |
| test parameters | bs=4k |
| | cpufreq_governor=performance |
| | disk=2pmem |
| | fs=ext4 |
| | ioengine=libaio |
| | mount_option=dax |
| | nr_task=50% |
| | runtime=200s |
| | rw=randwrite |
| | test_size=200G |
| | time_based=tb |
| | ucode=0x500002c |
+------------------+----------------------------------------------------------------------+
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-7/performance/2pmem/ext4/libaio/x86_64-rhel-7.6/dax/50%/debian-x86_64-20191114.cgz/200s/rw/lkp-csl-2sp6/200G/fio-basic/tb/0x500002c
commit:
b1b4705d54 ("ext4: introduce direct I/O read using iomap infrastructure")
569342dc24 ("ext4: move inode extension/truncate code out from ->iomap_end() callback")
b1b4705d54abedfd 569342dc2485392e95b6a626281
---------------- ---------------------------
fail:runs %reproduction fail:runs
| | |
:4 25% 1:4 dmesg.WARNING:at#for_ip_swapgs_restore_regs_and_return_to_usermode/0x
:4 25% 1:4 dmesg.WARNING:at_ip__fsnotify_parent/0x
1:4 -25% :4 dmesg.WARNING:at_ip_aio_read/0x
:4 25% 1:4 dmesg.WARNING:stack_recursion
%stddev %change %stddev
\ | \
14.31 ± 30% -14.3 0.01 fio.latency_1000us%
3.05 ± 72% +20.2 23.21 ± 69% fio.latency_250us%
0.62 ± 83% -0.6 0.01 fio.latency_2ms%
39.71 ± 26% -31.5 8.24 ± 83% fio.latency_750us%
5456 +56.4% 8534 fio.read_bw_MBps
794624 ± 5% -40.5% 472576 ± 10% fio.read_clat_90%_us
849920 ± 6% -40.8% 502784 ± 10% fio.read_clat_95%_us
948224 ± 7% -41.1% 558592 ± 9% fio.read_clat_99%_us
532777 -36.1% 340678 fio.read_clat_mean_us
194664 ± 13% -49.1% 99057 ± 34% fio.read_clat_stddev
1396934 +56.4% 2184757 fio.read_iops
4719 ± 2% -5.4% 4464 fio.read_slat_mean_us
4020 ± 5% -17.9% 3300 ± 10% fio.read_slat_stddev
4567 ± 9% +11.3% 5084 ± 9% fio.time.involuntary_context_switches
9322 -1.7% 9160 fio.time.system_time
286.59 ± 2% +56.4% 448.31 ± 2% fio.time.user_time
5.588e+08 +56.4% 8.739e+08 fio.workload
5456 +56.4% 8534 fio.write_bw_MBps
796672 ± 4% -40.7% 472576 ± 10% fio.write_clat_90%_us
849920 ± 6% -40.7% 503808 ± 10% fio.write_clat_95%_us
948224 ± 7% -41.1% 558592 ± 9% fio.write_clat_99%_us
533381 -36.1% 341017 fio.write_clat_mean_us
194676 ± 13% -49.1% 99161 ± 34% fio.write_clat_stddev
1396910 +56.4% 2184882 fio.write_iops
28403 -42.8% 16259 fio.write_slat_mean_us
12283 ± 11% -45.7% 6666 ± 22% fio.write_slat_stddev
1.50 ± 2% +0.8 2.29 ± 4% mpstat.cpu.all.usr%
5899 ± 79% +151.9% 14862 ± 51% softirqs.CPU48.SCHED
47.80 -3.5% 46.11 ± 2% iostat.cpu.system
1.49 ± 2% +52.9% 2.27 ± 4% iostat.cpu.user
18587 -82.9% 3179 ±170% numa-numastat.node0.other_node
103.25 ± 33% +14978.9% 15569 ± 34% numa-numastat.node1.other_node
45512 ± 23% -59.2% 18576 ±144% sched_debug.cfs_rq:/.spread0.avg
15.56 ± 11% +33.3% 20.75 ± 7% sched_debug.cpu.nr_uninterruptible.max
38681 ± 3% +535.2% 245689 ±143% turbostat.C1
126.48 +4.2% 131.76 turbostat.RAMWatt
9311 ± 8% +20.5% 11221 ± 6% numa-vmstat.node0.nr_slab_reclaimable
19397 ± 2% -79.7% 3942 ±142% numa-vmstat.node0.numa_other
9935 ± 7% -19.6% 7983 ± 10% numa-vmstat.node1.nr_slab_reclaimable
156898 +9.7% 172186 ± 3% numa-vmstat.node1.numa_other
37246 ± 8% +20.5% 44887 ± 6% numa-meminfo.node0.KReclaimable
37246 ± 8% +20.5% 44887 ± 6% numa-meminfo.node0.SReclaimable
137281 ± 3% +7.8% 148012 ± 2% numa-meminfo.node0.Slab
39742 ± 7% -19.6% 31935 ± 10% numa-meminfo.node1.KReclaimable
39742 ± 7% -19.6% 31935 ± 10% numa-meminfo.node1.SReclaimable
47.25 ±137% -100.0% 0.00 interrupts.46:PCI-MSI.31981579-edge.i40e-eth0-TxRx-10
59.50 ± 61% -71.0% 17.25 ±109% interrupts.CPU10.TLB:TLB_shootdowns
83.25 ± 45% -70.9% 24.25 ± 81% interrupts.CPU11.TLB:TLB_shootdowns
70.50 ± 50% -76.6% 16.50 ± 90% interrupts.CPU16.TLB:TLB_shootdowns
87.00 ± 44% -73.9% 22.75 ± 93% interrupts.CPU17.TLB:TLB_shootdowns
75.75 ± 46% -59.1% 31.00 ± 98% interrupts.CPU20.TLB:TLB_shootdowns
78.00 ± 42% -72.4% 21.50 ± 67% interrupts.CPU22.TLB:TLB_shootdowns
7327 ± 12% -40.0% 4393 ± 46% interrupts.CPU26.NMI:Non-maskable_interrupts
7327 ± 12% -40.0% 4393 ± 46% interrupts.CPU26.PMI:Performance_monitoring_interrupts
7.25 ± 65% +1220.7% 95.75 ± 82% interrupts.CPU3.RES:Rescheduling_interrupts
1965 ± 7% -9.4% 1781 ± 9% interrupts.CPU30.CAL:Function_call_interrupts
1951 ± 6% -10.0% 1756 ± 7% interrupts.CPU31.CAL:Function_call_interrupts
1957 ± 7% -10.2% 1757 ± 7% interrupts.CPU32.CAL:Function_call_interrupts
1853 ± 13% +25.4% 2324 ± 16% interrupts.CPU43.CAL:Function_call_interrupts
7367 ± 12% -45.3% 4027 ± 58% interrupts.CPU44.NMI:Non-maskable_interrupts
7367 ± 12% -45.3% 4027 ± 58% interrupts.CPU44.PMI:Performance_monitoring_interrupts
13.25 ± 35% +300.0% 53.00 ± 83% interrupts.CPU48.RES:Rescheduling_interrupts
64.25 ± 64% -60.3% 25.50 ± 54% interrupts.CPU55.TLB:TLB_shootdowns
74.00 ± 57% -74.7% 18.75 ± 81% interrupts.CPU7.TLB:TLB_shootdowns
40.50 ± 50% +108.0% 84.25 ± 21% interrupts.CPU75.TLB:TLB_shootdowns
5558 ± 32% -52.9% 2617 ± 21% interrupts.CPU8.NMI:Non-maskable_interrupts
5558 ± 32% -52.9% 2617 ± 21% interrupts.CPU8.PMI:Performance_monitoring_interrupts
5.50 ± 91% +1440.9% 84.75 ± 34% interrupts.CPU8.RES:Rescheduling_interrupts
40.25 ± 38% +91.3% 77.00 ± 33% interrupts.CPU83.TLB:TLB_shootdowns
41.00 ± 46% +95.7% 80.25 ± 22% interrupts.CPU90.TLB:TLB_shootdowns
4.867e+09 +46.6% 7.133e+09 ± 3% perf-stat.i.branch-instructions
35659892 ± 2% +40.1% 49959300 perf-stat.i.branch-misses
48.56 ± 2% +6.7 55.23 ± 2% perf-stat.i.cache-miss-rate%
1.806e+08 ± 3% +24.0% 2.239e+08 ± 3% perf-stat.i.cache-misses
3.709e+08 ± 2% +7.7% 3.993e+08 ± 2% perf-stat.i.cache-references
5.12 -31.6% 3.50 ± 3% perf-stat.i.cpi
750.82 ± 3% -16.1% 630.03 ± 9% perf-stat.i.cycles-between-cache-misses
7.13e+09 ± 9% +47.4% 1.051e+10 ± 8% perf-stat.i.dTLB-loads
4.827e+09 ± 8% +45.6% 7.027e+09 ± 3% perf-stat.i.dTLB-stores
38590154 ± 4% +23.4% 47627762 ± 9% perf-stat.i.iTLB-load-misses
2.618e+10 +46.5% 3.835e+10 ± 3% perf-stat.i.instructions
686.01 ± 3% +19.6% 820.13 ± 7% perf-stat.i.instructions-per-iTLB-miss
0.20 +47.3% 0.29 perf-stat.i.ipc
1537507 ± 4% +17.0% 1798882 ± 7% perf-stat.i.node-loads
13902646 ± 3% -8.9% 12667482 ± 3% perf-stat.i.node-store-misses
17738 ± 3% +16.9% 20736 ± 4% perf-stat.i.node-stores
14.17 ± 2% -26.5% 10.41 perf-stat.overall.MPKI
0.73 -0.0 0.70 perf-stat.overall.branch-miss-rate%
48.71 ± 2% +7.4 56.08 perf-stat.overall.cache-miss-rate%
5.13 -32.9% 3.44 perf-stat.overall.cpi
744.79 ± 3% -20.8% 589.98 perf-stat.overall.cycles-between-cache-misses
86.06 +1.7 87.73 perf-stat.overall.iTLB-load-miss-rate%
679.44 ± 3% +19.2% 810.22 ± 6% perf-stat.overall.instructions-per-iTLB-miss
0.19 +49.0% 0.29 perf-stat.overall.ipc
9405 -4.4% 8987 perf-stat.overall.path-length
4.842e+09 +46.6% 7.097e+09 ± 3% perf-stat.ps.branch-instructions
35478652 ± 2% +40.1% 49709486 perf-stat.ps.branch-misses
1.797e+08 ± 3% +24.0% 2.228e+08 ± 3% perf-stat.ps.cache-misses
3.69e+08 ± 2% +7.7% 3.973e+08 ± 2% perf-stat.ps.cache-references
7.093e+09 ± 9% +47.4% 1.046e+10 ± 8% perf-stat.ps.dTLB-loads
4.803e+09 ± 8% +45.6% 6.992e+09 ± 3% perf-stat.ps.dTLB-stores
38394613 ± 4% +23.4% 47389655 ± 9% perf-stat.ps.iTLB-load-misses
2.605e+10 +46.5% 3.816e+10 ± 3% perf-stat.ps.instructions
1529726 ± 4% +17.0% 1789899 ± 7% perf-stat.ps.node-loads
13832194 ± 3% -8.9% 12604236 ± 3% perf-stat.ps.node-store-misses
17673 ± 3% +16.9% 20659 ± 4% perf-stat.ps.node-stores
5.255e+12 +49.5% 7.854e+12 perf-stat.total.instructions
20.14 ± 9% -20.1 0.00 perf-profile.calltrace.cycles-pp.ext4_iomap_end.iomap_apply.dax_iomap_rw.ext4_file_write_iter.aio_write
14.28 ± 9% -14.3 0.00 perf-profile.calltrace.cycles-pp.jbd2__journal_start.ext4_iomap_end.iomap_apply.dax_iomap_rw.ext4_file_write_iter
14.20 ± 9% -14.2 0.00 perf-profile.calltrace.cycles-pp.start_this_handle.jbd2__journal_start.ext4_iomap_end.iomap_apply.dax_iomap_rw
0.60 ± 10% +0.4 1.05 ± 7% perf-profile.calltrace.cycles-pp.do_io_getevents.__x64_sys_io_getevents.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.64 ± 10% +0.5 1.11 ± 7% perf-profile.calltrace.cycles-pp.__x64_sys_io_getevents.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.80 ± 12% +0.6 1.39 ± 10% perf-profile.calltrace.cycles-pp.__srcu_read_unlock.dax_iomap_actor.iomap_apply.dax_iomap_rw.ext4_file_write_iter
0.12 ±173% +0.6 0.75 ± 7% perf-profile.calltrace.cycles-pp.read_events.do_io_getevents.__x64_sys_io_getevents.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 +0.7 0.68 ± 7% perf-profile.calltrace.cycles-pp.aio_read_events.read_events.do_io_getevents.__x64_sys_io_getevents.do_syscall_64
1.02 ± 11% +0.7 1.74 ± 4% perf-profile.calltrace.cycles-pp.__copy_user_nocache.__copy_user_flushcache._copy_from_iter_flushcache.dax_iomap_actor.iomap_apply
1.02 ± 12% +0.7 1.75 ± 4% perf-profile.calltrace.cycles-pp.__copy_user_flushcache._copy_from_iter_flushcache.dax_iomap_actor.iomap_apply.dax_iomap_rw
1.04 ± 11% +0.7 1.78 ± 4% perf-profile.calltrace.cycles-pp._copy_from_iter_flushcache.dax_iomap_actor.iomap_apply.dax_iomap_rw.ext4_file_write_iter
1.09 ± 7% +0.8 1.87 ± 7% perf-profile.calltrace.cycles-pp.__memcpy_mcsafe.copyout_mcsafe._copy_to_iter_mcsafe.dax_iomap_actor.iomap_apply
1.11 ± 8% +0.8 1.91 ± 8% perf-profile.calltrace.cycles-pp.copyout_mcsafe._copy_to_iter_mcsafe.dax_iomap_actor.iomap_apply.dax_iomap_rw
1.16 ± 8% +0.8 1.98 ± 8% perf-profile.calltrace.cycles-pp._copy_to_iter_mcsafe.dax_iomap_actor.iomap_apply.dax_iomap_rw.ext4_file_read_iter
1.24 ± 8% +0.9 2.14 ± 7% perf-profile.calltrace.cycles-pp.dax_iomap_actor.iomap_apply.dax_iomap_rw.ext4_file_read_iter.aio_read
2.00 ± 7% +1.4 3.38 ± 7% perf-profile.calltrace.cycles-pp.ext4_journal_check_start.__ext4_journal_start_sb.ext4_iomap_begin.iomap_apply.dax_iomap_rw
1.91 ± 8% +1.4 3.29 ± 6% perf-profile.calltrace.cycles-pp.dax_iomap_actor.iomap_apply.dax_iomap_rw.ext4_file_write_iter.aio_write
2.01 ± 7% +1.4 3.40 ± 7% perf-profile.calltrace.cycles-pp.__ext4_journal_start_sb.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter
2.81 ± 9% +1.5 4.34 ± 6% perf-profile.calltrace.cycles-pp._raw_read_lock.start_this_handle.jbd2__journal_start.ext4_iomap_begin.iomap_apply
3.50 ± 9% +2.1 5.63 ± 6% perf-profile.calltrace.cycles-pp.add_transaction_credits.start_this_handle.jbd2__journal_start.ext4_iomap_begin.iomap_apply
4.08 ± 8% +2.9 6.94 ± 7% perf-profile.calltrace.cycles-pp._raw_read_lock.jbd2_transaction_committed.ext4_set_iomap.ext4_iomap_begin.iomap_apply
7.23 ± 9% +3.6 10.86 ± 5% perf-profile.calltrace.cycles-pp.jbd2_journal_stop.__ext4_journal_stop.ext4_iomap_begin.iomap_apply.dax_iomap_rw
5.24 ± 8% +3.8 8.99 ± 7% perf-profile.calltrace.cycles-pp.ext4_set_iomap.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter
5.21 ± 7% +3.8 8.98 ± 8% perf-profile.calltrace.cycles-pp.ext4_set_iomap.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_read_iter
7.33 ± 9% +3.8 11.12 ± 5% perf-profile.calltrace.cycles-pp.__ext4_journal_stop.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter
5.36 ± 8% +3.9 9.24 ± 8% perf-profile.calltrace.cycles-pp.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_read_iter.aio_read
6.67 ± 7% +4.8 11.48 ± 8% perf-profile.calltrace.cycles-pp.iomap_apply.dax_iomap_rw.ext4_file_read_iter.aio_read.io_submit_one
6.68 ± 7% +4.8 11.51 ± 8% perf-profile.calltrace.cycles-pp.dax_iomap_rw.ext4_file_read_iter.aio_read.io_submit_one.__x64_sys_io_submit
7.75 ± 8% +5.0 12.72 ± 8% perf-profile.calltrace.cycles-pp.ext4_file_read_iter.aio_read.io_submit_one.__x64_sys_io_submit.do_syscall_64
7.98 ± 8% +5.1 13.13 ± 8% perf-profile.calltrace.cycles-pp.aio_read.io_submit_one.__x64_sys_io_submit.do_syscall_64.entry_SYSCALL_64_after_hwframe
10.41 ± 8% +7.5 17.87 ± 7% perf-profile.calltrace.cycles-pp.jbd2_transaction_committed.ext4_set_iomap.ext4_iomap_begin.iomap_apply.dax_iomap_rw
14.19 ± 9% +8.7 22.93 ± 7% perf-profile.calltrace.cycles-pp.start_this_handle.jbd2__journal_start.ext4_iomap_begin.iomap_apply.dax_iomap_rw
14.26 ± 9% +8.8 23.05 ± 7% perf-profile.calltrace.cycles-pp.jbd2__journal_start.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter
29.03 ± 8% +17.9 46.90 ± 6% perf-profile.calltrace.cycles-pp.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter.aio_write
20.16 ± 9% -20.2 0.00 perf-profile.children.cycles-pp.ext4_iomap_end
29.63 ± 9% -5.5 24.17 ± 7% perf-profile.children.cycles-pp.jbd2__journal_start
29.48 ± 9% -5.4 24.04 ± 7% perf-profile.children.cycles-pp.start_this_handle
7.30 ± 9% -1.4 5.90 ± 7% perf-profile.children.cycles-pp.add_transaction_credits
4.17 ± 7% -0.6 3.54 ± 7% perf-profile.children.cycles-pp.ext4_journal_check_start
4.19 ± 7% -0.6 3.56 ± 7% perf-profile.children.cycles-pp.__ext4_journal_start_sb
0.06 ± 17% +0.0 0.09 ± 9% perf-profile.children.cycles-pp._raw_spin_unlock_irqrestore
0.08 ± 15% +0.0 0.11 ± 4% perf-profile.children.cycles-pp.aio_complete_rw
0.06 ± 6% +0.0 0.10 ± 4% perf-profile.children.cycles-pp._cond_resched
0.05 ± 8% +0.0 0.09 ± 8% perf-profile.children.cycles-pp.put_reqs_available
0.04 ± 58% +0.0 0.08 ± 10% perf-profile.children.cycles-pp.__get_user_8
0.09 ± 13% +0.0 0.13 ± 6% perf-profile.children.cycles-pp._raw_spin_lock_irqsave
0.07 ± 10% +0.0 0.12 ± 7% perf-profile.children.cycles-pp.__check_object_size
0.04 ± 58% +0.0 0.09 ± 7% perf-profile.children.cycles-pp.__put_user_4
0.07 ± 14% +0.1 0.12 ± 12% perf-profile.children.cycles-pp.__fget
0.06 ± 11% +0.1 0.11 ± 11% perf-profile.children.cycles-pp.mutex_lock
0.01 ±173% +0.1 0.06 ± 13% perf-profile.children.cycles-pp.fsnotify
0.01 ±173% +0.1 0.07 ± 7% perf-profile.children.cycles-pp.fput_many
0.01 ±173% +0.1 0.07 ± 13% perf-profile.children.cycles-pp.__virt_addr_valid
0.00 +0.1 0.05 ± 8% perf-profile.children.cycles-pp.__pmem_direct_access
0.18 ± 8% +0.1 0.23 ± 8% perf-profile.children.cycles-pp.kmem_cache_alloc
0.04 ± 57% +0.1 0.09 ± 4% perf-profile.children.cycles-pp.__get_reqs_available
0.01 ±173% +0.1 0.07 ± 6% perf-profile.children.cycles-pp.mutex_unlock
0.00 +0.1 0.06 ± 7% perf-profile.children.cycles-pp.refill_reqs_available
0.01 ±173% +0.1 0.07 ± 10% perf-profile.children.cycles-pp.__srcu_read_lock
0.01 ±173% +0.1 0.07 ± 5% perf-profile.children.cycles-pp.__inode_security_revalidate
0.00 +0.1 0.06 ± 13% perf-profile.children.cycles-pp.ext4_write_checks
0.00 +0.1 0.07 ± 17% perf-profile.children.cycles-pp.ext4_data_block_valid
0.08 ± 13% +0.1 0.15 ± 17% perf-profile.children.cycles-pp.aio_prep_rw
0.03 ±100% +0.1 0.09 ± 15% perf-profile.children.cycles-pp.__check_block_validity
0.01 ±173% +0.1 0.08 ± 10% perf-profile.children.cycles-pp.copy_user_generic_unrolled
0.00 +0.1 0.07 ± 16% perf-profile.children.cycles-pp.__sb_start_write
0.00 +0.1 0.07 ± 17% perf-profile.children.cycles-pp.current_time
0.12 ± 10% +0.1 0.19 ± 9% perf-profile.children.cycles-pp.__might_sleep
0.11 ± 7% +0.1 0.18 ± 8% perf-profile.children.cycles-pp.__get_user_4
0.02 ±173% +0.1 0.10 ± 17% perf-profile.children.cycles-pp.ext4_meta_trans_blocks
0.11 ± 7% +0.1 0.19 ± 6% perf-profile.children.cycles-pp.copy_user_enhanced_fast_string
0.00 +0.1 0.08 ± 5% perf-profile.children.cycles-pp.dax_direct_access
0.11 ± 9% +0.1 0.19 ± 9% perf-profile.children.cycles-pp.selinux_file_permission
0.10 ± 12% +0.1 0.18 ± 10% perf-profile.children.cycles-pp._copy_to_user
0.12 ± 7% +0.1 0.20 ± 7% perf-profile.children.cycles-pp.entry_SYSCALL_64
0.12 ± 8% +0.1 0.22 ± 8% perf-profile.children.cycles-pp._copy_from_user
0.15 ± 8% +0.1 0.26 ± 8% perf-profile.children.cycles-pp.syscall_return_via_sysret
0.13 ± 10% +0.1 0.24 ± 8% perf-profile.children.cycles-pp.ext4_es_lookup_extent
0.15 ± 11% +0.1 0.26 ± 8% perf-profile.children.cycles-pp.security_file_permission
0.18 ± 12% +0.1 0.30 ± 7% perf-profile.children.cycles-pp.___might_sleep
0.18 ± 11% +0.1 0.32 ± 9% perf-profile.children.cycles-pp.__might_fault
0.20 ± 11% +0.2 0.38 ± 10% perf-profile.children.cycles-pp.ext4_map_blocks
0.30 ± 10% +0.2 0.52 ± 7% perf-profile.children.cycles-pp.lookup_ioctx
0.39 ± 10% +0.3 0.69 ± 7% perf-profile.children.cycles-pp.aio_read_events
0.43 ± 10% +0.3 0.76 ± 7% perf-profile.children.cycles-pp.read_events
0.61 ± 10% +0.4 1.05 ± 7% perf-profile.children.cycles-pp.do_io_getevents
0.64 ± 10% +0.5 1.12 ± 7% perf-profile.children.cycles-pp.__x64_sys_io_getevents
0.81 ± 12% +0.6 1.42 ± 10% perf-profile.children.cycles-pp.__srcu_read_unlock
1.02 ± 11% +0.7 1.74 ± 4% perf-profile.children.cycles-pp.__copy_user_nocache
1.03 ± 12% +0.7 1.75 ± 4% perf-profile.children.cycles-pp.__copy_user_flushcache
1.04 ± 11% +0.7 1.78 ± 4% perf-profile.children.cycles-pp._copy_from_iter_flushcache
1.09 ± 7% +0.8 1.88 ± 7% perf-profile.children.cycles-pp.__memcpy_mcsafe
1.12 ± 7% +0.8 1.92 ± 7% perf-profile.children.cycles-pp.copyout_mcsafe
1.16 ± 7% +0.8 1.98 ± 8% perf-profile.children.cycles-pp._copy_to_iter_mcsafe
9.94 ± 8% +1.7 11.60 ± 7% perf-profile.children.cycles-pp._raw_read_lock
3.15 ± 8% +2.3 5.44 ± 7% perf-profile.children.cycles-pp.dax_iomap_actor
7.75 ± 8% +5.0 12.72 ± 8% perf-profile.children.cycles-pp.ext4_file_read_iter
7.98 ± 8% +5.2 13.14 ± 8% perf-profile.children.cycles-pp.aio_read
10.41 ± 8% +7.5 17.88 ± 7% perf-profile.children.cycles-pp.jbd2_transaction_committed
10.46 ± 8% +7.5 17.98 ± 7% perf-profile.children.cycles-pp.ext4_set_iomap
34.40 ± 8% +21.8 56.16 ± 7% perf-profile.children.cycles-pp.ext4_iomap_begin
16.32 ± 9% -2.8 13.53 ± 8% perf-profile.self.cycles-pp.start_this_handle
7.28 ± 9% -1.4 5.87 ± 6% perf-profile.self.cycles-pp.add_transaction_credits
4.12 ± 7% -0.6 3.50 ± 7% perf-profile.self.cycles-pp.ext4_journal_check_start
0.07 ± 12% +0.0 0.10 ± 5% perf-profile.self.cycles-pp.aio_complete_rw
0.06 ± 14% +0.0 0.09 ± 10% perf-profile.self.cycles-pp._raw_spin_unlock_irqrestore
0.08 ± 13% +0.0 0.12 ± 5% perf-profile.self.cycles-pp.kmem_cache_alloc
0.05 ± 8% +0.0 0.09 ± 8% perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe
0.05 ± 8% +0.0 0.09 ± 8% perf-profile.self.cycles-pp.put_reqs_available
0.09 ± 13% +0.0 0.13 ± 9% perf-profile.self.cycles-pp._raw_spin_lock_irqsave
0.07 ± 13% +0.0 0.11 ± 7% perf-profile.self.cycles-pp.__x64_sys_io_submit
0.04 ± 58% +0.0 0.08 ± 5% perf-profile.self.cycles-pp.__put_user_4
0.04 ± 57% +0.0 0.08 ± 10% perf-profile.self.cycles-pp.do_syscall_64
0.04 ± 57% +0.0 0.08 ± 10% perf-profile.self.cycles-pp.__get_user_8
0.08 ± 15% +0.0 0.13 ± 8% perf-profile.self.cycles-pp.ext4_iomap_begin
0.07 ± 12% +0.0 0.12 ± 9% perf-profile.self.cycles-pp.selinux_file_permission
0.01 ±173% +0.0 0.06 ± 11% perf-profile.self.cycles-pp.fsnotify
0.01 ±173% +0.0 0.06 ± 11% perf-profile.self.cycles-pp.kmem_cache_free
0.04 ± 58% +0.0 0.09 ± 7% perf-profile.self.cycles-pp.ext4_es_lookup_extent
0.07 ± 14% +0.1 0.12 ± 12% perf-profile.self.cycles-pp.__fget
0.00 +0.1 0.05 perf-profile.self.cycles-pp.copyout_mcsafe
0.04 ± 57% +0.1 0.09 ± 7% perf-profile.self.cycles-pp.__get_reqs_available
0.03 ±100% +0.1 0.08 ± 8% perf-profile.self.cycles-pp.dax_iomap_actor
0.01 ±173% +0.1 0.07 ± 7% perf-profile.self.cycles-pp.__srcu_read_lock
0.01 ±173% +0.1 0.07 ± 13% perf-profile.self.cycles-pp.__virt_addr_valid
0.00 +0.1 0.06 ± 9% perf-profile.self.cycles-pp.aio_write
0.00 +0.1 0.06 ± 9% perf-profile.self.cycles-pp.ext4_map_blocks
0.01 ±173% +0.1 0.07 ± 6% perf-profile.self.cycles-pp.mutex_unlock
0.00 +0.1 0.06 ± 7% perf-profile.self.cycles-pp.refill_reqs_available
0.00 +0.1 0.06 ± 11% perf-profile.self.cycles-pp.ext4_file_read_iter
0.01 ±173% +0.1 0.07 ± 11% perf-profile.self.cycles-pp.mutex_lock
0.04 ± 57% +0.1 0.10 ± 25% perf-profile.self.cycles-pp.ext4_set_iomap
0.00 +0.1 0.06 ± 6% perf-profile.self.cycles-pp.fput_many
0.09 ± 11% +0.1 0.16 ± 7% perf-profile.self.cycles-pp.iomap_apply
0.01 ±173% +0.1 0.08 ± 15% perf-profile.self.cycles-pp.copy_user_generic_unrolled
0.11 ± 7% +0.1 0.17 ± 8% perf-profile.self.cycles-pp.__might_sleep
0.00 +0.1 0.07 ± 12% perf-profile.self.cycles-pp.__x64_sys_io_getevents
0.00 +0.1 0.07 ± 12% perf-profile.self.cycles-pp.__might_fault
0.08 ± 15% +0.1 0.15 ± 17% perf-profile.self.cycles-pp.aio_prep_rw
0.11 ± 7% +0.1 0.18 ± 6% perf-profile.self.cycles-pp.__get_user_4
0.00 +0.1 0.07 ± 20% perf-profile.self.cycles-pp.ext4_file_write_iter
0.11 ± 7% +0.1 0.19 ± 6% perf-profile.self.cycles-pp.copy_user_enhanced_fast_string
0.10 ± 4% +0.1 0.18 ± 6% perf-profile.self.cycles-pp.aio_read_events
0.12 ± 7% +0.1 0.20 ± 7% perf-profile.self.cycles-pp.entry_SYSCALL_64
0.13 ± 14% +0.1 0.22 ± 7% perf-profile.self.cycles-pp.lookup_ioctx
0.15 ± 8% +0.1 0.26 ± 8% perf-profile.self.cycles-pp.syscall_return_via_sysret
0.18 ± 11% +0.1 0.29 ± 8% perf-profile.self.cycles-pp.___might_sleep
0.19 ± 6% +0.2 0.35 ± 7% perf-profile.self.cycles-pp.io_submit_one
0.81 ± 12% +0.6 1.41 ± 10% perf-profile.self.cycles-pp.__srcu_read_unlock
1.02 ± 11% +0.7 1.73 ± 4% perf-profile.self.cycles-pp.__copy_user_nocache
1.08 ± 7% +0.8 1.86 ± 8% perf-profile.self.cycles-pp.__memcpy_mcsafe
9.89 ± 9% +1.6 11.54 ± 7% perf-profile.self.cycles-pp._raw_read_lock
6.29 ± 8% +4.6 10.88 ± 7% perf-profile.self.cycles-pp.jbd2_transaction_committed
fio.read_bw_MBps
10000 +-+-----------------------------------------------------------------+
9000 O-OO O O O O O O OO |
| O O O OO O O O O O O OO |
8000 +-+ O O |
7000 +-+ |
| |
6000 +-+ .+ .+.++.+.++.+.+.++.+.++.+.++.+.+.++.+.++.+.++.+.+. +.+. +.|
5000 +-++.+.+ + + + |
4000 +-+ |
| |
3000 +-+ |
2000 +-+ |
| |
1000 +-+ |
0 +-+------------O----------------------------------------------------+
fio.read_iops
2.5e+06 +-+---------------------------------------------------------------+
O OO O OO O OO OO OO O OO O |
| OO O OO O OO O |
2e+06 +-+ |
| |
| |
1.5e+06 +-++.+. +. .++.+.++.++.+.++.+.++.+.++.+.++.+.++.+.++.++.+. +.+.++.|
|.+ + + + |
1e+06 +-+ |
| |
| |
500000 +-+ |
| |
| |
0 +-+------------O--------------------------------------------------+
fio.read_clat_mean_us
600000 +-+----------------------------------------------------------------+
|.++.+ +.+. + : .+ |
500000 +-+ ++.+.++.+.++.+.++.+.+.++.+.++.+.++.+.++.+.++.+ +.+ +.|
| |
| |
400000 +-+ |
| O O O O O O O OO O O OO O OO O O |
300000 O-OO O O O O O |
| |
200000 +-+ |
| |
| |
100000 +-+ |
| |
0 +-+------------O---------------------------------------------------+
fio.write_bw_MBps
10000 +-+-----------------------------------------------------------------+
9000 O-OO O O O O O O OO |
| O O O OO O O O O O O OO |
8000 +-+ O O |
7000 +-+ |
| |
6000 +-+ .+ .+.++.+.++.+.+.++.+.++.+.++.+.+.++.+.++.+.++.+.+. +.+. +.|
5000 +-++.+.+ + + + |
4000 +-+ |
| |
3000 +-+ |
2000 +-+ |
| |
1000 +-+ |
0 +-+------------O----------------------------------------------------+
fio.write_iops
2.5e+06 +-+---------------------------------------------------------------+
O OO O OO O OO OO OO O OO O |
| OO O OO O OO O |
2e+06 +-+ |
| |
| |
1.5e+06 +-++.+. +. .++.+.++.++.+.++.+.++.+.++.+.++.+.++.+.++.++.+. +.+.++.|
|.+ + + + |
1e+06 +-+ |
| |
| |
500000 +-+ |
| |
| |
0 +-+------------O--------------------------------------------------+
fio.write_clat_mean_us
600000 +-+----------------------------------------------------------------+
|.++.+ + + + : .+ |
500000 +-+ ++.+.++.+.++.+.++.+.+.++.+.++.+.++.+.++.+.++.+ +.+ +.|
| |
| |
400000 +-+ |
| O O O O O O O OO O O OO O OO O O |
300000 O-OO O O O O O |
| |
200000 +-+ |
| |
| |
100000 +-+ |
| |
0 +-+------------O---------------------------------------------------+
fio.write_slat_mean_us
35000 +-+-----------------------------------------------------------------+
| |
30000 +-+ .+. +. .+ .+ |
| +.+ + +.++.+.++.+.+.++.+.++.+.++.+.+.++. .++.+.++.+.+ +.+ +.|
25000 +-+ + |
| |
20000 +-+ |
| O O OO O O |
15000 O-OO O O OO O O O OO O O OO O O O O |
| |
10000 +-+ |
| |
5000 +-+ |
| |
0 +-+------------O----------------------------------------------------+
fio.workload
1e+09 +-+-----------------------------------------------------------------+
9e+08 O-OO O O OO O O O OO O O OO O |
| OO O OO O O OO |
8e+08 +-+ |
7e+08 +-+ |
| |
6e+08 +-++.+. .+ .+.++.+.++.+.+.++.+.++.+.++.+.+.++.+.++.+.++.+.+. +.+.++.|
5e+08 +-+ + + + |
4e+08 +-+ |
| |
3e+08 +-+ |
2e+08 +-+ |
| |
1e+08 +-+ |
0 +-+------------O----------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
***************************************************************************************************
lkp-csl-2sp6: 96 threads Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 256G memory
=========================================================================================
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-7/performance/2pmem/ext4/sync/x86_64-rhel-7.6/dax/50%/debian-x86_64-20191114.cgz/200s/randwrite/lkp-csl-2sp6/200G/fio-basic/tb/0x500002c
commit:
b1b4705d54 ("ext4: introduce direct I/O read using iomap infrastructure")
569342dc24 ("ext4: move inode extension/truncate code out from ->iomap_end() callback")
b1b4705d54abedfd 569342dc2485392e95b6a626281
---------------- ---------------------------
fail:runs %reproduction fail:runs
| | |
:4 25% 1:4 dmesg.WARNING:at#for_ip_interrupt_entry/0x
%stddev %change %stddev
\ | \
4.08 ± 25% -3.9 0.15 ± 34% fio.latency_100us%
0.81 ± 45% +15.6 16.45 ± 47% fio.latency_10us%
63.62 ± 3% -27.6 36.07 ± 17% fio.latency_50us%
175.73 +50.8% 265.02 ± 2% fio.time.user_time
3.401e+08 ± 2% +54.0% 5.237e+08 fio.workload
6642 ± 2% +54.0% 10229 fio.write_bw_MBps
43392 ± 4% -36.1% 27712 ± 5% fio.write_clat_90%_us
48384 ± 4% -35.7% 31104 ± 5% fio.write_clat_95%_us
59136 ± 3% -33.3% 39424 ± 5% fio.write_clat_99%_us
27760 ± 2% -35.7% 17839 fio.write_clat_mean_us
12867 ± 9% -33.9% 8506 ± 18% fio.write_clat_stddev
1700412 ± 2% +54.0% 2618693 fio.write_iops
2011 +1.2% 2034 boot-time.idle
0.94 +48.6% 1.39 ± 2% iostat.cpu.user
2972 ± 3% +21.0% 3597 meminfo.Dirty
66317 ± 2% +24.3% 82442 ± 20% softirqs.CPU48.TIMER
0.00 ± 9% -0.0 0.00 ± 24% mpstat.cpu.all.iowait%
0.94 +0.5 1.40 ± 2% mpstat.cpu.all.usr%
13981 ± 56% -99.2% 109.75 ± 95% numa-numastat.node0.other_node
4771 ±167% +292.0% 18700 numa-numastat.node1.other_node
745.50 ± 3% +21.1% 903.00 proc-vmstat.nr_dirty
743.75 ± 3% +21.5% 903.75 proc-vmstat.nr_zone_write_pending
238649 ± 16% +28.0% 305583 ± 7% numa-meminfo.node0.Active
208040 ± 24% +37.6% 286262 ± 5% numa-meminfo.node0.Active(anon)
229500 ± 26% -36.3% 146157 ± 7% numa-meminfo.node1.Active(anon)
19606483 -18.4% 16003967 ± 20% turbostat.C1E
49.52 -12.5 37.06 ± 30% turbostat.C1E%
124.19 +3.3% 128.34 turbostat.RAMWatt
9.6e+09 -25.2% 7.183e+09 ± 30% cpuidle.C1E.time
19607054 -18.4% 16005122 ± 20% cpuidle.C1E.usage
42226648 ± 44% +5712.0% 2.454e+09 ± 88% cpuidle.C6.time
40856 ± 44% +8800.0% 3636237 ± 90% cpuidle.C6.usage
51953 ± 24% +37.6% 71503 ± 5% numa-vmstat.node0.nr_active_anon
51953 ± 24% +37.6% 71503 ± 5% numa-vmstat.node0.nr_zone_active_anon
14639 ± 52% -95.4% 678.00 ± 85% numa-vmstat.node0.numa_other
57365 ± 26% -36.3% 36564 ± 7% numa-vmstat.node1.nr_active_anon
57365 ± 26% -36.3% 36564 ± 7% numa-vmstat.node1.nr_zone_active_anon
161195 ± 4% +9.1% 175918 numa-vmstat.node1.numa_other
1476 ± 3% -14.7% 1259 ± 4% slabinfo.ext4_extent_status.active_slabs
150614 ± 3% -14.7% 128502 ± 4% slabinfo.ext4_extent_status.num_objs
1476 ± 3% -14.7% 1259 ± 4% slabinfo.ext4_extent_status.num_slabs
1096 ± 7% -14.8% 934.25 ± 7% slabinfo.khugepaged_mm_slot.active_objs
1096 ± 7% -14.8% 934.25 ± 7% slabinfo.khugepaged_mm_slot.num_objs
400.00 ± 13% +28.2% 513.00 ± 8% slabinfo.skbuff_fclone_cache.active_objs
400.00 ± 13% +28.2% 513.00 ± 8% slabinfo.skbuff_fclone_cache.num_objs
2785 ± 7% +14.5% 3188 ± 7% slabinfo.sock_inode_cache.active_objs
2785 ± 7% +14.5% 3188 ± 7% slabinfo.sock_inode_cache.num_objs
2.393e+09 ± 2% +35.2% 3.235e+09 perf-stat.i.branch-instructions
0.98 -0.0 0.93 ± 2% perf-stat.i.branch-miss-rate%
23967630 ± 2% +28.1% 30704745 ± 3% perf-stat.i.branch-misses
10.10 ± 2% -26.6% 7.41 perf-stat.i.cpi
4.179e+09 ± 4% +38.3% 5.779e+09 ± 8% perf-stat.i.dTLB-loads
2.663e+09 +36.1% 3.625e+09 ± 7% perf-stat.i.dTLB-stores
79.97 +3.0 82.93 perf-stat.i.iTLB-load-miss-rate%
26249647 +31.2% 34434951 ± 4% perf-stat.i.iTLB-load-misses
6552634 +8.0% 7076093 ± 3% perf-stat.i.iTLB-loads
1.347e+10 ± 2% +35.8% 1.83e+10 perf-stat.i.instructions
0.10 ± 2% +34.3% 0.14 ± 2% perf-stat.i.ipc
1.00 -0.1 0.95 perf-stat.overall.branch-miss-rate%
9.99 ± 2% -26.4% 7.35 perf-stat.overall.cpi
0.00 ± 21% -0.0 0.00 ± 26% perf-stat.overall.dTLB-store-miss-rate%
80.02 +2.9 82.94 perf-stat.overall.iTLB-load-miss-rate%
0.10 ± 2% +35.9% 0.14 perf-stat.overall.ipc
7939 -11.9% 6995 perf-stat.overall.path-length
2.381e+09 ± 2% +35.2% 3.218e+09 perf-stat.ps.branch-instructions
23847364 ± 2% +28.1% 30544365 ± 3% perf-stat.ps.branch-misses
4.157e+09 ± 4% +38.3% 5.749e+09 ± 8% perf-stat.ps.dTLB-loads
2.649e+09 +36.1% 3.606e+09 ± 7% perf-stat.ps.dTLB-stores
26117405 +31.2% 34259896 ± 4% perf-stat.ps.iTLB-load-misses
6519655 +8.0% 7039502 ± 3% perf-stat.ps.iTLB-loads
1.341e+10 ± 2% +35.8% 1.821e+10 perf-stat.ps.instructions
2.7e+12 ± 2% +35.7% 3.664e+12 perf-stat.total.instructions
57.00 ± 88% -96.5% 2.00 ±173% interrupts.55:PCI-MSI.31981588-edge.i40e-eth0-TxRx-19
1760 ± 4% +17.3% 2065 ± 4% interrupts.CPU0.CAL:Function_call_interrupts
5733 -24.1% 4353 ± 21% interrupts.CPU0.NMI:Non-maskable_interrupts
5733 -24.1% 4353 ± 21% interrupts.CPU0.PMI:Performance_monitoring_interrupts
1761 ± 5% +26.4% 2226 ± 10% interrupts.CPU1.CAL:Function_call_interrupts
14.75 ± 49% +2852.5% 435.50 ±111% interrupts.CPU1.RES:Rescheduling_interrupts
1734 ± 5% +17.9% 2044 ± 13% interrupts.CPU12.CAL:Function_call_interrupts
1681 ± 6% +23.1% 2068 ± 8% interrupts.CPU15.CAL:Function_call_interrupts
1757 ± 5% +15.9% 2037 ± 2% interrupts.CPU2.CAL:Function_call_interrupts
1829 ± 9% +19.8% 2191 ± 7% interrupts.CPU31.CAL:Function_call_interrupts
1757 ± 4% +19.6% 2101 ± 9% interrupts.CPU4.CAL:Function_call_interrupts
1761 ± 4% +17.6% 2070 ± 4% interrupts.CPU5.CAL:Function_call_interrupts
1753 ± 4% +18.8% 2082 ± 3% interrupts.CPU6.CAL:Function_call_interrupts
6375 ± 25% -39.4% 3861 ± 30% interrupts.CPU60.NMI:Non-maskable_interrupts
6375 ± 25% -39.4% 3861 ± 30% interrupts.CPU60.PMI:Performance_monitoring_interrupts
6.50 ± 93% +807.7% 59.00 ± 90% interrupts.CPU65.RES:Rescheduling_interrupts
6333 ± 25% -31.0% 4370 ± 46% interrupts.CPU66.NMI:Non-maskable_interrupts
6333 ± 25% -31.0% 4370 ± 46% interrupts.CPU66.PMI:Performance_monitoring_interrupts
1754 ± 4% +16.8% 2050 ± 3% interrupts.CPU7.CAL:Function_call_interrupts
6364 ± 25% -30.9% 4399 ± 46% interrupts.CPU71.NMI:Non-maskable_interrupts
6364 ± 25% -30.9% 4399 ± 46% interrupts.CPU71.PMI:Performance_monitoring_interrupts
61.50 ± 77% -68.3% 19.50 ±108% interrupts.CPU77.RES:Rescheduling_interrupts
1750 ± 4% +17.6% 2058 ± 4% interrupts.CPU8.CAL:Function_call_interrupts
6785 ± 15% -36.5% 4311 ± 23% interrupts.CPU85.NMI:Non-maskable_interrupts
6785 ± 15% -36.5% 4311 ± 23% interrupts.CPU85.PMI:Performance_monitoring_interrupts
6262 ± 14% -27.2% 4557 ± 15% interrupts.CPU86.NMI:Non-maskable_interrupts
6262 ± 14% -27.2% 4557 ± 15% interrupts.CPU86.PMI:Performance_monitoring_interrupts
1752 ± 4% +17.8% 2064 ± 4% interrupts.CPU9.CAL:Function_call_interrupts
21.45 ± 8% -21.4 0.00 perf-profile.calltrace.cycles-pp.ext4_iomap_end.iomap_apply.dax_iomap_rw.ext4_file_write_iter.new_sync_write
14.50 ± 8% -14.5 0.00 perf-profile.calltrace.cycles-pp.jbd2__journal_start.ext4_iomap_end.iomap_apply.dax_iomap_rw.ext4_file_write_iter
14.42 ± 8% -14.4 0.00 perf-profile.calltrace.cycles-pp.start_this_handle.jbd2__journal_start.ext4_iomap_end.iomap_apply.dax_iomap_rw
45.47 ± 9% -8.0 37.44 ± 16% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry
45.61 ± 9% -8.0 37.66 ± 16% perf-profile.calltrace.cycles-pp.secondary_startup_64
44.50 ± 9% -7.6 36.90 ± 16% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary
44.50 ± 9% -7.6 36.91 ± 16% perf-profile.calltrace.cycles-pp.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
44.58 ± 9% -7.6 37.00 ± 16% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
44.58 ± 9% -7.6 37.00 ± 16% perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64
44.58 ± 9% -7.6 37.00 ± 16% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64
4.87 ± 8% -4.9 0.00 perf-profile.calltrace.cycles-pp.__ext4_journal_stop.ext4_iomap_end.iomap_apply.dax_iomap_rw.ext4_file_write_iter
4.74 ± 8% -4.7 0.00 perf-profile.calltrace.cycles-pp.jbd2_journal_stop.__ext4_journal_stop.ext4_iomap_end.iomap_apply.dax_iomap_rw
0.77 ± 7% +0.5 1.31 ± 5% perf-profile.calltrace.cycles-pp.__srcu_read_unlock.dax_iomap_actor.iomap_apply.dax_iomap_rw.ext4_file_write_iter
0.80 ± 12% +0.9 1.70 ± 26% perf-profile.calltrace.cycles-pp.__copy_user_nocache.__copy_user_flushcache._copy_from_iter_flushcache.dax_iomap_actor.iomap_apply
0.81 ± 12% +0.9 1.71 ± 26% perf-profile.calltrace.cycles-pp.__copy_user_flushcache._copy_from_iter_flushcache.dax_iomap_actor.iomap_apply.dax_iomap_rw
0.83 ± 12% +0.9 1.74 ± 25% perf-profile.calltrace.cycles-pp._copy_from_iter_flushcache.dax_iomap_actor.iomap_apply.dax_iomap_rw.ext4_file_write_iter
1.66 ± 6% +1.5 3.18 ± 16% perf-profile.calltrace.cycles-pp.dax_iomap_actor.iomap_apply.dax_iomap_rw.ext4_file_write_iter.new_sync_write
2.29 ± 9% +1.7 3.98 ± 13% perf-profile.calltrace.cycles-pp._raw_read_lock.jbd2_transaction_committed.ext4_set_iomap.ext4_iomap_begin.iomap_apply
2.00 ± 11% +2.0 3.96 ± 12% perf-profile.calltrace.cycles-pp.ext4_journal_check_start.__ext4_journal_start_sb.ext4_iomap_begin.iomap_apply.dax_iomap_rw
2.01 ± 11% +2.0 4.03 ± 12% perf-profile.calltrace.cycles-pp.__ext4_journal_start_sb.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter
2.94 ± 9% +2.5 5.40 ± 9% perf-profile.calltrace.cycles-pp._raw_read_lock.start_this_handle.jbd2__journal_start.ext4_iomap_begin.iomap_apply
3.23 ± 9% +3.6 6.81 ± 14% perf-profile.calltrace.cycles-pp.add_transaction_credits.start_this_handle.jbd2__journal_start.ext4_iomap_begin.iomap_apply
5.56 ± 9% +4.8 10.40 ± 14% perf-profile.calltrace.cycles-pp.jbd2_transaction_committed.ext4_set_iomap.ext4_iomap_begin.iomap_apply.dax_iomap_rw
5.58 ± 9% +5.0 10.54 ± 13% perf-profile.calltrace.cycles-pp.ext4_set_iomap.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter
7.53 ± 9% +6.4 13.92 ± 6% perf-profile.calltrace.cycles-pp.jbd2_journal_stop.__ext4_journal_stop.ext4_iomap_begin.iomap_apply.dax_iomap_rw
7.63 ± 9% +6.5 14.15 ± 6% perf-profile.calltrace.cycles-pp.__ext4_journal_stop.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter
54.07 ± 8% +7.8 61.91 ± 10% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe
54.10 ± 8% +7.9 61.97 ± 10% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe
13.99 ± 8% +13.3 27.25 ± 10% perf-profile.calltrace.cycles-pp.start_this_handle.jbd2__journal_start.ext4_iomap_begin.iomap_apply.dax_iomap_rw
14.08 ± 8% +13.3 27.40 ± 10% perf-profile.calltrace.cycles-pp.jbd2__journal_start.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter
29.51 ± 8% +27.1 56.59 ± 9% perf-profile.calltrace.cycles-pp.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter.new_sync_write
21.45 ± 8% -21.4 0.00 perf-profile.children.cycles-pp.ext4_iomap_end
45.47 ± 9% -8.0 37.44 ± 16% perf-profile.children.cycles-pp.intel_idle
45.53 ± 9% -8.0 37.55 ± 16% perf-profile.children.cycles-pp.cpuidle_enter
45.53 ± 9% -8.0 37.55 ± 16% perf-profile.children.cycles-pp.cpuidle_enter_state
45.61 ± 9% -8.0 37.66 ± 16% perf-profile.children.cycles-pp.do_idle
45.61 ± 9% -8.0 37.66 ± 16% perf-profile.children.cycles-pp.secondary_startup_64
45.61 ± 9% -8.0 37.66 ± 16% perf-profile.children.cycles-pp.cpu_startup_entry
44.58 ± 9% -7.6 37.00 ± 16% perf-profile.children.cycles-pp.start_secondary
0.07 ± 5% +0.0 0.09 ± 14% perf-profile.children.cycles-pp.update_process_times
0.07 ± 5% +0.0 0.10 ± 15% perf-profile.children.cycles-pp.tick_sched_handle
0.09 ± 4% +0.0 0.12 ± 15% perf-profile.children.cycles-pp.tick_sched_timer
0.06 ± 11% +0.0 0.10 ± 8% perf-profile.children.cycles-pp.selinux_file_permission
0.05 ± 8% +0.0 0.09 ± 11% perf-profile.children.cycles-pp.__fdget_pos
0.07 ± 12% +0.0 0.11 ± 11% perf-profile.children.cycles-pp.security_file_permission
0.04 ± 58% +0.1 0.10 ± 18% perf-profile.children.cycles-pp.ext4_llseek
0.04 ± 58% +0.1 0.10 ± 9% perf-profile.children.cycles-pp.entry_SYSCALL_64
0.09 ± 8% +0.1 0.15 ± 8% perf-profile.children.cycles-pp.syscall_return_via_sysret
0.01 ±173% +0.1 0.07 ± 10% perf-profile.children.cycles-pp.__fget_light
0.06 ± 58% +0.1 0.12 ± 36% perf-profile.children.cycles-pp.ktime_get
0.00 +0.1 0.08 ± 19% perf-profile.children.cycles-pp.__sb_start_write
0.10 ± 4% +0.1 0.18 ± 11% perf-profile.children.cycles-pp.ksys_lseek
0.24 ± 13% +0.1 0.34 ± 15% perf-profile.children.cycles-pp.hrtimer_interrupt
0.06 ± 6% +0.1 0.16 ± 45% perf-profile.children.cycles-pp.ext4_es_lookup_extent
0.28 ± 11% +0.1 0.39 ± 23% perf-profile.children.cycles-pp.smp_apic_timer_interrupt
0.31 ± 9% +0.1 0.44 ± 21% perf-profile.children.cycles-pp.apic_timer_interrupt
0.11 ± 17% +0.2 0.29 ± 52% perf-profile.children.cycles-pp.ext4_map_blocks
0.77 ± 7% +0.5 1.31 ± 5% perf-profile.children.cycles-pp.__srcu_read_unlock
0.80 ± 12% +0.9 1.71 ± 26% perf-profile.children.cycles-pp.__copy_user_nocache
0.81 ± 12% +0.9 1.71 ± 26% perf-profile.children.cycles-pp.__copy_user_flushcache
0.83 ± 12% +0.9 1.74 ± 25% perf-profile.children.cycles-pp._copy_from_iter_flushcache
1.66 ± 6% +1.5 3.19 ± 16% perf-profile.children.cycles-pp.dax_iomap_actor
12.68 ± 8% +1.6 14.32 ± 6% perf-profile.children.cycles-pp.__ext4_journal_stop
12.46 ± 8% +1.6 14.10 ± 6% perf-profile.children.cycles-pp.jbd2_journal_stop
5.56 ± 9% +4.9 10.41 ± 14% perf-profile.children.cycles-pp.jbd2_transaction_committed
5.58 ± 9% +5.0 10.54 ± 13% perf-profile.children.cycles-pp.ext4_set_iomap
54.18 ± 8% +7.8 62.02 ± 10% perf-profile.children.cycles-pp.do_syscall_64
54.21 ± 8% +7.9 62.07 ± 10% perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
29.51 ± 8% +27.1 56.60 ± 9% perf-profile.children.cycles-pp.ext4_iomap_begin
45.47 ± 9% -8.0 37.44 ± 16% perf-profile.self.cycles-pp.intel_idle
0.01 ±173% +0.1 0.06 ± 17% perf-profile.self.cycles-pp.___might_sleep
0.00 +0.1 0.05 ± 8% perf-profile.self.cycles-pp.do_syscall_64
0.04 ± 58% +0.1 0.10 ± 9% perf-profile.self.cycles-pp.entry_SYSCALL_64
0.09 ± 8% +0.1 0.15 ± 8% perf-profile.self.cycles-pp.syscall_return_via_sysret
0.00 +0.1 0.06 ± 14% perf-profile.self.cycles-pp.selinux_file_permission
0.01 ±173% +0.1 0.07 ± 10% perf-profile.self.cycles-pp.__fget_light
0.05 ± 58% +0.1 0.11 ± 33% perf-profile.self.cycles-pp.ktime_get
0.00 +0.1 0.06 ± 20% perf-profile.self.cycles-pp.ext4_es_lookup_extent
0.03 ±100% +0.1 0.09 ± 16% perf-profile.self.cycles-pp.iomap_apply
0.03 ±100% +0.1 0.10 ± 18% perf-profile.self.cycles-pp.ext4_llseek
0.00 +0.1 0.07 ± 15% perf-profile.self.cycles-pp.ext4_iomap_begin
0.02 ±173% +0.1 0.14 ± 56% perf-profile.self.cycles-pp.ext4_file_write_iter
0.76 ± 7% +0.5 1.30 ± 6% perf-profile.self.cycles-pp.__srcu_read_unlock
0.80 ± 12% +0.9 1.69 ± 26% perf-profile.self.cycles-pp.__copy_user_nocache
3.25 ± 9% +3.1 6.40 ± 14% perf-profile.self.cycles-pp.jbd2_transaction_committed
***************************************************************************************************
lkp-csl-2sp6: 96 threads Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 256G memory
=========================================================================================
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-7/performance/2pmem/ext4/libaio/x86_64-rhel-7.6/dax/50%/debian-x86_64-20191114.cgz/200s/randwrite/lkp-csl-2sp6/200G/fio-basic/tb/0x500002c
commit:
b1b4705d54 ("ext4: introduce direct I/O read using iomap infrastructure")
569342dc24 ("ext4: move inode extension/truncate code out from ->iomap_end() callback")
b1b4705d54abedfd 569342dc2485392e95b6a626281
---------------- ---------------------------
fail:runs %reproduction fail:runs
| | |
1:4 -25% :4 dmesg.WARNING:at_ip__x64_sys_io_submit/0x
:4 25% 1:4 dmesg.WARNING:at_ip_aio_write/0x
%stddev %change %stddev
\ | \
22.50 ± 85% -16.5 6.00 ±150% fio.latency_1000us%
0.01 +3.3 3.29 ±158% fio.latency_250us%
43.58 ± 27% -43.5 0.04 ±104% fio.latency_2ms%
22.38 ± 16% +40.0 62.34 ± 24% fio.latency_750us%
9422 -1.2% 9304 fio.time.system_time
185.76 ± 3% +63.1% 303.05 ± 2% fio.time.user_time
3.191e+08 +68.8% 5.384e+08 fio.workload
6231 +68.8% 10516 fio.write_bw_MBps
1327104 ± 7% -46.3% 712704 ± 10% fio.write_clat_90%_us
1380352 ± 6% -46.6% 737280 ± 10% fio.write_clat_95%_us
1503232 ± 7% -47.4% 790528 ± 10% fio.write_clat_99%_us
933218 -40.7% 553062 fio.write_clat_mean_us
310780 ± 29% -57.8% 131049 ± 33% fio.write_clat_stddev
1595317 +68.8% 2692113 fio.write_iops
29413 -41.6% 17164 fio.write_slat_mean_us
14463 ± 30% -52.9% 6814 ± 13% fio.write_slat_stddev
7159296 ± 12% -11.5% 6334976 meminfo.DirectMap2M
21181 -1.0% 20962 proc-vmstat.nr_slab_reclaimable
48.55 -1.4% 47.88 iostat.cpu.system
0.99 ± 3% +60.9% 1.59 ± 2% iostat.cpu.user
0.00 ± 17% -0.0 0.00 ± 24% mpstat.cpu.all.iowait%
0.99 ± 3% +0.6 1.60 ± 2% mpstat.cpu.all.usr%
383524 ± 37% -42.5% 220431 ± 57% numa-numastat.node1.local_node
392957 ± 35% -40.3% 234425 ± 50% numa-numastat.node1.numa_hit
248.01 +2.5% 254.11 turbostat.PkgWatt
124.91 +3.3% 129.03 turbostat.RAMWatt
40962 ± 10% -16.3% 34274 ± 3% numa-vmstat.node0.nr_anon_pages
26868 ± 16% +25.4% 33684 ± 3% numa-vmstat.node1.nr_anon_pages
8600 ± 12% +17.7% 10124 ± 10% numa-vmstat.node1.nr_slab_reclaimable
163884 ± 10% -16.3% 137230 ± 3% numa-meminfo.node0.AnonPages
138488 ± 14% +22.8% 170029 numa-meminfo.node1.Active
67242 ± 22% +25.5% 84358 ± 6% numa-meminfo.node1.AnonHugePages
107464 ± 16% +25.2% 134574 ± 3% numa-meminfo.node1.AnonPages
34404 ± 12% +17.7% 40494 ± 10% numa-meminfo.node1.KReclaimable
34404 ± 12% +17.7% 40494 ± 10% numa-meminfo.node1.SReclaimable
70842 ± 6% -7.7% 65369 ± 3% softirqs.CPU3.TIMER
8500 ±105% +128.3% 19405 ± 45% softirqs.CPU35.SCHED
8786 ±103% +123.1% 19604 ± 46% softirqs.CPU40.SCHED
8981 ± 98% +116.0% 19402 ± 44% softirqs.CPU42.SCHED
68254 ± 5% +6.0% 72342 ± 4% softirqs.CPU44.TIMER
65323 ± 3% +10.5% 72152 ± 2% softirqs.CPU58.TIMER
19561 ± 49% -71.6% 5557 ± 91% softirqs.CPU82.SCHED
1744 ± 6% +23.5% 2154 ± 3% slabinfo.UNIX.active_objs
1744 ± 6% +23.5% 2154 ± 3% slabinfo.UNIX.num_objs
111978 ± 4% -17.4% 92440 ± 4% slabinfo.ext4_extent_status.active_objs
1416 ± 4% -22.0% 1104 ± 4% slabinfo.ext4_extent_status.active_slabs
144495 ± 4% -22.0% 112722 ± 4% slabinfo.ext4_extent_status.num_objs
1416 ± 4% -22.0% 1104 ± 4% slabinfo.ext4_extent_status.num_slabs
2737 ± 4% +20.5% 3298 ± 4% slabinfo.sock_inode_cache.active_objs
2737 ± 4% +20.5% 3298 ± 4% slabinfo.sock_inode_cache.num_objs
535573 ± 2% +8.1% 579164 ± 3% sched_debug.cfs_rq:/.load.avg
516.79 ± 2% +7.8% 557.21 ± 3% sched_debug.cfs_rq:/.runnable_load_avg.avg
535510 ± 2% +8.1% 579108 ± 3% sched_debug.cfs_rq:/.runnable_weight.avg
0.01 ± 24% -50.0% 0.00 ± 24% sched_debug.cpu.nr_uninterruptible.avg
14089 ± 4% -21.0% 11124 ± 12% sched_debug.cpu.sched_count.max
2552 ± 9% -13.5% 2208 ± 12% sched_debug.cpu.sched_count.stddev
6997 ± 5% -21.1% 5521 ± 12% sched_debug.cpu.sched_goidle.max
1261 ± 8% -13.7% 1089 ± 11% sched_debug.cpu.sched_goidle.stddev
5749 ± 5% -23.7% 4386 ± 13% sched_debug.cpu.ttwu_local.max
23.50 ± 66% +293.6% 92.50 ± 39% interrupts.CPU12.TLB:TLB_shootdowns
4.75 ± 63% +3384.2% 165.50 ±162% interrupts.CPU13.RES:Rescheduling_interrupts
2382 ± 14% -24.1% 1809 ± 10% interrupts.CPU32.CAL:Function_call_interrupts
2351 ± 11% -19.1% 1901 ± 9% interrupts.CPU36.CAL:Function_call_interrupts
2259 ± 11% -16.2% 1894 ± 8% interrupts.CPU38.CAL:Function_call_interrupts
2428 ± 12% -17.6% 2001 ± 12% interrupts.CPU60.CAL:Function_call_interrupts
5774 ± 25% -48.5% 2973 ± 56% interrupts.CPU7.NMI:Non-maskable_interrupts
5774 ± 25% -48.5% 2973 ± 56% interrupts.CPU7.PMI:Performance_monitoring_interrupts
80.50 ± 38% +60.2% 129.00 ± 20% interrupts.CPU73.TLB:TLB_shootdowns
54.50 ± 84% -83.9% 8.75 ±100% interrupts.CPU83.RES:Rescheduling_interrupts
91.75 ±100% -86.4% 12.50 ±122% interrupts.CPU89.RES:Rescheduling_interrupts
18.16 ± 13% -36.0% 11.62 ± 5% perf-stat.i.MPKI
2.855e+09 +51.8% 4.333e+09 perf-stat.i.branch-instructions
0.80 -0.1 0.73 perf-stat.i.branch-miss-rate%
23116045 +37.2% 31705911 perf-stat.i.branch-misses
8.56 -34.6% 5.59 perf-stat.i.cpi
4.579e+09 +66.3% 7.613e+09 ± 10% perf-stat.i.dTLB-loads
3.192e+09 ± 2% +45.8% 4.654e+09 ± 8% perf-stat.i.dTLB-stores
81.32 +2.0 83.28 perf-stat.i.iTLB-load-miss-rate%
28448462 ± 4% +18.0% 33581121 ± 3% perf-stat.i.iTLB-load-misses
1.588e+10 +52.2% 2.417e+10 perf-stat.i.instructions
573.58 ± 3% +27.9% 733.82 ± 4% perf-stat.i.instructions-per-iTLB-miss
0.12 +50.6% 0.18 perf-stat.i.ipc
14520205 -10.4% 13007330 ± 13% perf-stat.i.node-store-misses
17811 ± 3% +21.2% 21596 ± 3% perf-stat.i.node-stores
18.03 ± 13% -35.8% 11.57 ± 5% perf-stat.overall.MPKI
0.81 -0.1 0.73 perf-stat.overall.branch-miss-rate%
8.48 -34.4% 5.56 perf-stat.overall.cpi
81.36 +2.0 83.34 perf-stat.overall.iTLB-load-miss-rate%
559.06 ± 3% +29.0% 721.05 ± 4% perf-stat.overall.instructions-per-iTLB-miss
0.12 +52.4% 0.18 perf-stat.overall.ipc
9977 -9.8% 8996 perf-stat.overall.path-length
2.84e+09 +51.8% 4.311e+09 perf-stat.ps.branch-instructions
23003415 +37.1% 31544372 perf-stat.ps.branch-misses
4.556e+09 +66.2% 7.574e+09 ± 10% perf-stat.ps.dTLB-loads
3.176e+09 ± 2% +45.8% 4.63e+09 ± 8% perf-stat.ps.dTLB-stores
28304001 ± 4% +18.0% 33410816 ± 3% perf-stat.ps.iTLB-load-misses
1.58e+10 +52.2% 2.405e+10 perf-stat.ps.instructions
14446128 -10.4% 12941633 ± 13% perf-stat.ps.node-store-misses
17748 ± 3% +21.0% 21481 ± 3% perf-stat.ps.node-stores
3.184e+12 +52.1% 4.844e+12 perf-stat.total.instructions
24.52 ± 9% -24.5 0.00 perf-profile.calltrace.cycles-pp.ext4_iomap_end.iomap_apply.dax_iomap_rw.ext4_file_write_iter.aio_write
17.04 ± 10% -17.0 0.00 perf-profile.calltrace.cycles-pp.jbd2__journal_start.ext4_iomap_end.iomap_apply.dax_iomap_rw.ext4_file_write_iter
16.95 ± 10% -16.9 0.00 perf-profile.calltrace.cycles-pp.start_this_handle.jbd2__journal_start.ext4_iomap_end.iomap_apply.dax_iomap_rw
5.24 ± 7% -5.2 0.00 perf-profile.calltrace.cycles-pp.__ext4_journal_stop.ext4_iomap_end.iomap_apply.dax_iomap_rw.ext4_file_write_iter
5.09 ± 7% -5.1 0.00 perf-profile.calltrace.cycles-pp.jbd2_journal_stop.__ext4_journal_stop.ext4_iomap_end.iomap_apply.dax_iomap_rw
0.00 +0.6 0.58 ± 8% perf-profile.calltrace.cycles-pp.do_io_getevents.__x64_sys_io_getevents.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 +0.6 0.62 ± 8% perf-profile.calltrace.cycles-pp.__x64_sys_io_getevents.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.97 ± 5% +0.6 1.60 ± 22% perf-profile.calltrace.cycles-pp.__srcu_read_unlock.dax_iomap_actor.iomap_apply.dax_iomap_rw.ext4_file_write_iter
1.13 ± 12% +1.0 2.08 ± 12% perf-profile.calltrace.cycles-pp.__copy_user_nocache.__copy_user_flushcache._copy_from_iter_flushcache.dax_iomap_actor.iomap_apply
1.14 ± 12% +1.0 2.09 ± 12% perf-profile.calltrace.cycles-pp.__copy_user_flushcache._copy_from_iter_flushcache.dax_iomap_actor.iomap_apply.dax_iomap_rw
1.16 ± 12% +1.0 2.12 ± 12% perf-profile.calltrace.cycles-pp._copy_from_iter_flushcache.dax_iomap_actor.iomap_apply.dax_iomap_rw.ext4_file_write_iter
2.20 ± 8% +1.7 3.86 ± 7% perf-profile.calltrace.cycles-pp.dax_iomap_actor.iomap_apply.dax_iomap_rw.ext4_file_write_iter.aio_write
2.18 ± 12% +1.7 3.90 ± 7% perf-profile.calltrace.cycles-pp._raw_read_lock.jbd2_transaction_committed.ext4_set_iomap.ext4_iomap_begin.iomap_apply
2.17 ± 12% +1.7 3.89 ± 7% perf-profile.calltrace.cycles-pp.ext4_journal_check_start.__ext4_journal_start_sb.ext4_iomap_begin.iomap_apply.dax_iomap_rw
2.19 ± 12% +1.7 3.93 ± 7% perf-profile.calltrace.cycles-pp.__ext4_journal_start_sb.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter
3.46 ± 10% +1.9 5.32 ± 11% perf-profile.calltrace.cycles-pp._raw_read_lock.start_this_handle.jbd2__journal_start.ext4_iomap_begin.iomap_apply
4.08 ± 10% +2.6 6.69 ± 7% perf-profile.calltrace.cycles-pp.add_transaction_credits.start_this_handle.jbd2__journal_start.ext4_iomap_begin.iomap_apply
9.21 ± 9% +4.5 13.74 ± 12% perf-profile.calltrace.cycles-pp.jbd2_journal_stop.__ext4_journal_stop.ext4_iomap_begin.iomap_apply.dax_iomap_rw
5.56 ± 12% +4.6 10.18 ± 7% perf-profile.calltrace.cycles-pp.jbd2_transaction_committed.ext4_set_iomap.ext4_iomap_begin.iomap_apply.dax_iomap_rw
5.59 ± 12% +4.7 10.25 ± 8% perf-profile.calltrace.cycles-pp.ext4_set_iomap.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter
9.32 ± 9% +4.7 14.00 ± 12% perf-profile.calltrace.cycles-pp.__ext4_journal_stop.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter
16.80 ± 10% +10.3 27.15 ± 10% perf-profile.calltrace.cycles-pp.start_this_handle.jbd2__journal_start.ext4_iomap_begin.iomap_apply.dax_iomap_rw
16.89 ± 10% +10.4 27.30 ± 10% perf-profile.calltrace.cycles-pp.jbd2__journal_start.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter
34.20 ± 10% +21.7 55.87 ± 10% perf-profile.calltrace.cycles-pp.ext4_iomap_begin.iomap_apply.dax_iomap_rw.ext4_file_write_iter.aio_write
24.52 ± 9% -24.5 0.00 perf-profile.children.cycles-pp.ext4_iomap_end
34.47 ± 10% -6.7 27.80 ± 10% perf-profile.children.cycles-pp.jbd2__journal_start
34.30 ± 10% -6.7 27.65 ± 10% perf-profile.children.cycles-pp.start_this_handle
8.29 ± 11% -1.5 6.81 ± 7% perf-profile.children.cycles-pp.add_transaction_credits
0.05 ± 8% +0.0 0.07 ± 12% perf-profile.children.cycles-pp._cond_resched
0.07 ± 7% +0.0 0.10 ± 13% perf-profile.children.cycles-pp.selinux_file_permission
0.08 ± 10% +0.0 0.12 ± 7% perf-profile.children.cycles-pp.__might_sleep
0.07 ± 14% +0.0 0.11 ± 11% perf-profile.children.cycles-pp.security_file_permission
0.03 ±100% +0.0 0.07 ± 12% perf-profile.children.cycles-pp.__fget
0.06 ± 14% +0.0 0.10 ± 7% perf-profile.children.cycles-pp._copy_to_user
0.07 ± 7% +0.0 0.11 ± 6% perf-profile.children.cycles-pp.copy_user_enhanced_fast_string
0.06 ± 13% +0.0 0.11 ± 7% perf-profile.children.cycles-pp.__get_user_4
0.04 ± 57% +0.0 0.09 ± 8% perf-profile.children.cycles-pp.aio_prep_rw
0.07 ± 7% +0.0 0.11 ± 9% perf-profile.children.cycles-pp.entry_SYSCALL_64
0.03 ±100% +0.1 0.08 ± 14% perf-profile.children.cycles-pp.__sb_start_write
0.00 +0.1 0.05 ± 9% perf-profile.children.cycles-pp.put_reqs_available
0.00 +0.1 0.05 ± 9% perf-profile.children.cycles-pp._raw_spin_unlock_irqrestore
0.04 ± 58% +0.1 0.10 ± 18% perf-profile.children.cycles-pp.ext4_meta_trans_blocks
0.01 ±173% +0.1 0.07 ± 12% perf-profile.children.cycles-pp.__check_object_size
0.07 ± 11% +0.1 0.13 ± 9% perf-profile.children.cycles-pp._copy_from_user
0.09 ± 13% +0.1 0.14 ± 13% perf-profile.children.cycles-pp.syscall_return_via_sysret
0.07 ± 17% +0.1 0.12 ± 13% perf-profile.children.cycles-pp.ext4_es_lookup_extent
0.12 ± 11% +0.1 0.18 ± 7% perf-profile.children.cycles-pp.___might_sleep
0.00 +0.1 0.06 ± 13% perf-profile.children.cycles-pp.mutex_lock
0.10 ± 11% +0.1 0.17 ± 9% perf-profile.children.cycles-pp.__might_fault
0.11 ± 17% +0.1 0.21 ± 17% perf-profile.children.cycles-pp.ext4_map_blocks
0.18 ± 13% +0.1 0.30 ± 9% perf-profile.children.cycles-pp.lookup_ioctx
0.22 ± 9% +0.2 0.38 ± 8% perf-profile.children.cycles-pp.aio_read_events
0.25 ± 9% +0.2 0.42 ± 8% perf-profile.children.cycles-pp.read_events
0.35 ± 9% +0.2 0.58 ± 8% perf-profile.children.cycles-pp.do_io_getevents
0.37 ± 10% +0.3 0.63 ± 8% perf-profile.children.cycles-pp.__x64_sys_io_getevents
0.97 ± 5% +0.6 1.60 ± 22% perf-profile.children.cycles-pp.__srcu_read_unlock
1.14 ± 12% +1.0 2.09 ± 12% perf-profile.children.cycles-pp.__copy_user_flushcache
1.13 ± 12% +1.0 2.08 ± 12% perf-profile.children.cycles-pp.__copy_user_nocache
1.16 ± 12% +1.0 2.12 ± 12% perf-profile.children.cycles-pp._copy_from_iter_flushcache
2.20 ± 8% +1.7 3.87 ± 7% perf-profile.children.cycles-pp.dax_iomap_actor
5.56 ± 12% +4.6 10.19 ± 7% perf-profile.children.cycles-pp.jbd2_transaction_committed
5.59 ± 12% +4.7 10.26 ± 8% perf-profile.children.cycles-pp.ext4_set_iomap
34.20 ± 10% +21.7 55.87 ± 10% perf-profile.children.cycles-pp.ext4_iomap_begin
8.26 ± 10% -1.5 6.79 ± 7% perf-profile.self.cycles-pp.add_transaction_credits
0.08 ± 6% +0.0 0.11 ± 4% perf-profile.self.cycles-pp.__might_sleep
0.05 +0.0 0.09 ± 9% perf-profile.self.cycles-pp.iomap_apply
0.06 ± 13% +0.0 0.10 ± 10% perf-profile.self.cycles-pp.__get_user_4
0.03 ±100% +0.0 0.07 ± 12% perf-profile.self.cycles-pp.__fget
0.07 ± 7% +0.0 0.11 ± 6% perf-profile.self.cycles-pp.copy_user_enhanced_fast_string
0.03 ±100% +0.0 0.07 ± 14% perf-profile.self.cycles-pp.__x64_sys_io_submit
0.05 ± 9% +0.0 0.10 ± 7% perf-profile.self.cycles-pp.aio_read_events
0.08 ± 13% +0.0 0.13 ± 12% perf-profile.self.cycles-pp.lookup_ioctx
0.04 ± 57% +0.0 0.09 ± 8% perf-profile.self.cycles-pp.aio_prep_rw
0.07 ± 7% +0.0 0.11 ± 9% perf-profile.self.cycles-pp.entry_SYSCALL_64
0.00 +0.1 0.05 ± 9% perf-profile.self.cycles-pp.aio_write
0.00 +0.1 0.05 ± 9% perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe
0.00 +0.1 0.05 ± 9% perf-profile.self.cycles-pp.put_reqs_available
0.09 ± 13% +0.1 0.14 ± 13% perf-profile.self.cycles-pp.syscall_return_via_sysret
0.12 ± 11% +0.1 0.18 ± 8% perf-profile.self.cycles-pp.___might_sleep
0.01 ±173% +0.1 0.07 ± 14% perf-profile.self.cycles-pp.ext4_iomap_begin
0.11 ± 10% +0.1 0.18 ± 8% perf-profile.self.cycles-pp.io_submit_one
0.03 ±102% +0.1 0.11 ± 17% perf-profile.self.cycles-pp.ext4_file_write_iter
0.97 ± 5% +0.6 1.59 ± 22% perf-profile.self.cycles-pp.__srcu_read_unlock
1.13 ± 12% +0.9 2.07 ± 12% perf-profile.self.cycles-pp.__copy_user_nocache
3.37 ± 12% +2.9 6.25 ± 8% perf-profile.self.cycles-pp.jbd2_transaction_committed
***************************************************************************************************
lkp-hsw-d01: 8 threads Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz with 8G memory
=========================================================================================
compiler/cpufreq_governor/disk/fs/kconfig/rootfs/tbox_group/test/testcase/ucode:
gcc-7/performance/1HDD/ext4/x86_64-rhel-7.6/debian-x86_64-2019-11-14.cgz/lkp-hsw-d01/fivestreamreaddirect.f/filebench/0x27
commit:
b1b4705d54 ("ext4: introduce direct I/O read using iomap infrastructure")
569342dc24 ("ext4: move inode extension/truncate code out from ->iomap_end() callback")
b1b4705d54abedfd 569342dc2485392e95b6a626281
---------------- ---------------------------
fail:runs %reproduction fail:runs
| | |
1:4 -25% :2 dmesg.WARNING:at#for_ip_swapgs_restore_regs_and_return_to_usermode/0x
1:4 -25% :2 dmesg.WARNING:stack_recursion
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:da
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:db
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:dc
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:dd
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:de
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:df
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:ea
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:eb
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:ec
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:ed
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:ee
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:ef
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:fa
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:fb
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:fc
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:fd
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:fe
1:4 -25% :2 kmsg.addr:#vm_flags:#anon_vma:#mapping:#index:ff
1:4 -25% :2 kmsg.file:libnss_files-#so_fault:filemap_fault_mmap:generic_file_mmap_readpage:simple_readpage
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#a07fffffa0ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#a17fffffa1ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#a27fffffa2ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#a37fffffa3ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#a47fffffa4ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#a57fffffa5ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#a67fffffa6ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#a77fffffa7ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#a7fffff8aff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#a7fffff9aff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#a87fffffa8ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#a97fffffa9ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#aa7fffffaaff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#ab7fffffabff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#ac7fffffacff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#ad7fffffadff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#ae7fffffaeff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#af7fffffafff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#b07fffffb0ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#b17fffffb1ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#b27fffffb2ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#b37fffffb3ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#b47fffffb4ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#b57fffffb5ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#b67fffffb6ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#b77fffffb7ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#b7fffff8bff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#b7fffff9bff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#b87fffffb8ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#b97fffffb9ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#ba7fffffbaff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#bb7fffffbbff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#bc7fffffbcff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#bd7fffffbdff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#be7fffffbeff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#bf7fffffbfff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#c07fffffc0ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#c17fffffc1ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#c27fffffc2ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#c37fffffc3ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#c47fffffc4ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#c57fffffc5ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#c67fffffc6ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#c77fffffc7ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#c7fffff8cff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#c7fffff9cff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#c87fffffc8ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#c97fffffc9ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#ca7fffffcaff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#cb7fffffcbff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#cc7fffffccff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#cd7fffffcdff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#ce7fffffceff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#cf7fffffcfff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#d07fffffd0ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#d17fffffd1ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#d27fffffd2ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#d37fffffd3ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#d47fffffd4ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#d57fffffd5ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#d67fffffd6ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#d77fffffd7ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#d7fffff8dff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#d7fffff9dff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#d87fffffd8ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#d97fffffd9ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#da7fffffdaff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#db7fffffdbff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#dc7fffffdcff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#dd7fffffddff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#de7fffffdeff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#df7fffffdfff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#e07fffffe0ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#e17fffffe1ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#e27fffffe2ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#e37fffffe3ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#e47fffffe4ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#e57fffffe5ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#e67fffffe6ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#e77fffffe7ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#e7fffff8eff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#e7fffff9eff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#e87fffffe8ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#e97fffffe9ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#ea7fffffeaff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#eb7fffffebff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#ec7fffffecff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#ed7fffffedff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#ee7fffffeeff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#ef7fffffefff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#f07ffffff0ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#f17ffffff1ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#f27ffffff2ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#f37ffffff3ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#f47ffffff4ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#f57ffffff5ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#f67ffffff6ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#f77ffffff7ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#f7fffff8fff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#f7fffff9fff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#f87ffffff8ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#f97ffffff9ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fa7ffffffaff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fb7ffffffbff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fc7ffffffcff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fd7ffffffdff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fe7ffffffeff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#ff7fffffffff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff80ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff81ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff82ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff83ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff84ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff85ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff86ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff87ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff88ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff89ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff90ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff91ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff92ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff93ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff94ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff95ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff96ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff97ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff98ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffff99ff
1:4 -25% :2 kmsg.swap_info_get:Bad_swap_file_entry#fffffffffff
%stddev %change %stddev
\ | \
219770 ± 63% +70.4% 374538 ± 3% cpuidle.C7s.usage
219674 ± 63% +70.5% 374616 ± 3% turbostat.C7s
49.50 ± 3% +9.1% 54.00 vmstat.cpu.id
47.75 ± 2% -7.9% 44.00 ± 2% vmstat.cpu.wa
50.18 ± 3% +8.5% 54.43 iostat.cpu.idle
48.35 ± 2% -7.1% 44.91 ± 2% iostat.cpu.iowait
0.87 ±137% -79.7% 0.18 iostat.cpu.user
6679 ± 14% -29.6% 4701 ± 4% slabinfo.anon_vma_chain.active_objs
7047 ± 12% -23.5% 5387 ± 2% slabinfo.anon_vma_chain.num_objs
2794 ± 4% +9.1% 3049 ± 2% slabinfo.filp.active_objs
332426 ± 16% +21.2% 402873 ± 3% interrupts.CPU0.LOC:Local_timer_interrupts
169.50 ± 15% +134.8% 398.00 ± 45% interrupts.CPU0.NMI:Non-maskable_interrupts
169.50 ± 15% +134.8% 398.00 ± 45% interrupts.CPU0.PMI:Performance_monitoring_interrupts
204.75 ± 21% -48.7% 105.00 interrupts.CPU0.RES:Rescheduling_interrupts
307421 ± 3% +13.9% 350098 ± 2% interrupts.CPU2.LOC:Local_timer_interrupts
158.25 ± 54% +137.6% 376.00 ± 61% interrupts.CPU4.RES:Rescheduling_interrupts
786.00 ± 30% +26.4% 993.50 ± 2% interrupts.CPU5.CAL:Function_call_interrupts
1588 ± 14% +30.7% 2076 interrupts.RES:Rescheduling_interrupts
331.32 ± 13% -27.3% 240.78 sched_debug.cfs_rq:/.exec_clock.min
4972 ± 5% -13.2% 4317 ± 3% sched_debug.cfs_rq:/.min_vruntime.min
0.45 ± 17% -51.7% 0.22 ± 42% sched_debug.cfs_rq:/.nr_spread_over.avg
2.00 -50.0% 1.00 sched_debug.cfs_rq:/.nr_spread_over.max
0.68 ± 4% -44.4% 0.38 ± 12% sched_debug.cfs_rq:/.nr_spread_over.stddev
564589 ± 4% +4.8% 591417 ± 3% sched_debug.cpu.avg_idle.avg
13.12 ± 19% +21.9% 16.00 ± 9% sched_debug.cpu.nr_uninterruptible.max
133870 ± 6% -16.3% 112088 ± 10% sched_debug.cpu.ttwu_count.max
15029 ± 8% -33.4% 10005 ± 7% sched_debug.cpu.ttwu_count.min
132827 ± 6% -16.0% 111524 ± 10% sched_debug.cpu.ttwu_local.max
10470 ± 9% -27.5% 7589 ± 14% sched_debug.cpu.ttwu_local.min
10.17 ± 2% +0.3 10.48 perf-stat.i.cache-miss-rate%
558901 ± 2% +6.4% 594670 ± 3% perf-stat.i.cache-misses
5172535 +1.8% 5265269 perf-stat.i.cache-references
2.40 +3.8% 2.49 ± 2% perf-stat.i.cpi
14.90 ± 6% +9.7% 16.35 ± 3% perf-stat.i.cpu-migrations
1.115e+08 ± 2% +3.3% 1.152e+08 perf-stat.i.dTLB-loads
62749 ± 3% +8.0% 67738 ± 4% perf-stat.i.dTLB-store-misses
10.81 ± 2% +0.5 11.29 ± 3% perf-stat.overall.cache-miss-rate%
1.82 ± 2% +4.2% 1.89 perf-stat.overall.cpi
0.07 ± 3% +0.0 0.08 ± 3% perf-stat.overall.dTLB-store-miss-rate%
0.55 ± 2% -4.1% 0.53 perf-stat.overall.ipc
553610 ± 2% +6.4% 589043 ± 3% perf-stat.ps.cache-misses
5123552 +1.8% 5215383 perf-stat.ps.cache-references
14.76 ± 6% +9.7% 16.20 ± 3% perf-stat.ps.cpu-migrations
1.104e+08 ± 2% +3.3% 1.141e+08 perf-stat.ps.dTLB-loads
62154 ± 3% +8.0% 67096 ± 4% perf-stat.ps.dTLB-store-misses
0.68 ±104% +0.8 1.53 ± 26% perf-profile.calltrace.cycles-pp.free_unref_page_list.release_pages.__pagevec_release.truncate_inode_pages_range.ext4_evict_inode
0.12 ± 17% -0.1 0.04 ±100% perf-profile.children.cycles-pp.__delete_from_page_cache
0.07 ± 14% +0.0 0.11 ± 18% perf-profile.children.cycles-pp.irq_work_needs_cpu
0.01 ±173% +0.1 0.07 ± 7% perf-profile.children.cycles-pp.kmem_cache_alloc
0.01 ±173% +0.1 0.07 ± 7% perf-profile.children.cycles-pp.__unwind_start
0.00 +0.1 0.07 ± 7% perf-profile.children.cycles-pp.truncate_exceptional_pvec_entries
0.00 +0.1 0.07 ± 7% perf-profile.children.cycles-pp.rcu_core_si
0.04 ±106% +0.1 0.11 ± 18% perf-profile.children.cycles-pp.drm_fb_helper_dirty
0.04 ±102% +0.1 0.11 ± 18% perf-profile.children.cycles-pp.ext4_invalidatepage
0.04 ±102% +0.1 0.11 ± 18% perf-profile.children.cycles-pp.xas_find
0.02 ±173% +0.1 0.10 ± 36% perf-profile.children.cycles-pp.__wake_up_common
0.02 ±173% +0.1 0.10 ± 30% perf-profile.children.cycles-pp.intel_pmu_disable_all
0.00 +0.1 0.08 ± 12% perf-profile.children.cycles-pp.timerqueue_iterate_next
0.04 ±110% +0.1 0.13 perf-profile.children.cycles-pp.free_unref_page_prepare
0.04 ±102% +0.1 0.13 perf-profile.children.cycles-pp.xas_start
0.00 +0.1 0.10 ± 36% perf-profile.children.cycles-pp.autoremove_wake_function
0.17 ± 51% +0.1 0.28 ± 20% perf-profile.children.cycles-pp.__blk_mq_delay_run_hw_queue
0.11 ± 44% +0.1 0.21 ± 4% perf-profile.children.cycles-pp.ext4_io_submit
0.03 ±173% +0.1 0.15 ± 37% perf-profile.children.cycles-pp.update_rt_rq_load_avg
0.22 ± 45% +0.1 0.34 ± 2% perf-profile.children.cycles-pp.rb_next
0.11 ± 81% +0.1 0.23 ± 17% perf-profile.children.cycles-pp.cpuidle_reflect
0.32 ± 11% +0.1 0.46 perf-profile.children.cycles-pp.__x86_indirect_thunk_rax
0.46 ± 25% +0.2 0.62 ± 15% perf-profile.children.cycles-pp._raw_spin_lock
0.14 ± 93% +0.2 0.31 ± 48% perf-profile.children.cycles-pp.workingset_update_node
0.44 ± 23% +0.2 0.61 ± 18% perf-profile.children.cycles-pp.sched_clock
0.09 ± 59% +0.2 0.26 ± 3% perf-profile.children.cycles-pp.__mod_zone_page_state
0.20 ± 19% +0.2 0.37 ± 45% perf-profile.children.cycles-pp.pm_qos_read_value
0.41 ± 24% +0.2 0.58 ± 24% perf-profile.children.cycles-pp.native_sched_clock
0.15 ± 53% +0.2 0.34 ± 35% perf-profile.children.cycles-pp.unaccount_page_cache_page
0.20 ± 42% +0.2 0.42 ± 20% perf-profile.children.cycles-pp.nr_iowait_cpu
0.25 ± 55% +0.3 0.52 ± 22% perf-profile.children.cycles-pp.update_ts_time_stats
0.56 ± 16% +0.3 0.83 ± 20% perf-profile.children.cycles-pp.sched_clock_cpu
0.95 ± 58% +0.7 1.61 ± 27% perf-profile.children.cycles-pp.free_unref_page_list
0.16 ± 23% +0.0 0.21 ± 4% perf-profile.self.cycles-pp.menu_reflect
0.05 ± 60% +0.1 0.10 ± 30% perf-profile.self.cycles-pp.leave_mm
0.04 ±110% +0.1 0.10 ± 30% perf-profile.self.cycles-pp.free_unref_page_prepare
0.01 ±173% +0.1 0.08 ± 12% perf-profile.self.cycles-pp.jbd2_journal_try_to_free_buffers
0.00 +0.1 0.07 ± 7% perf-profile.self.cycles-pp.timerqueue_iterate_next
0.00 +0.1 0.07 ± 7% perf-profile.self.cycles-pp.truncate_exceptional_pvec_entries
0.00 +0.1 0.07 ± 7% perf-profile.self.cycles-pp.rcu_core_si
0.04 ±102% +0.1 0.11 ± 18% perf-profile.self.cycles-pp.ext4_invalidatepage
0.15 ± 24% +0.1 0.24 ± 10% perf-profile.self.cycles-pp.get_next_timer_interrupt
0.09 ± 67% +0.1 0.20 ± 35% perf-profile.self.cycles-pp.try_to_free_buffers
0.00 +0.1 0.11 ± 18% perf-profile.self.cycles-pp.xas_start
0.03 ±173% +0.1 0.15 ± 37% perf-profile.self.cycles-pp.update_rt_rq_load_avg
0.11 ± 81% +0.1 0.23 ± 17% perf-profile.self.cycles-pp.cpuidle_reflect
0.26 ± 20% +0.1 0.39 ± 14% perf-profile.self.cycles-pp.__x86_indirect_thunk_rax
0.39 ± 22% +0.1 0.52 ± 15% perf-profile.self.cycles-pp.native_sched_clock
0.10 ±109% +0.1 0.23 ± 17% perf-profile.self.cycles-pp.update_ts_time_stats
0.10 ± 57% +0.1 0.24 ± 10% perf-profile.self.cycles-pp.sched_clock_cpu
0.07 ± 66% +0.2 0.22 ± 25% perf-profile.self.cycles-pp.free_unref_page_commit
0.46 ± 25% +0.2 0.62 ± 15% perf-profile.self.cycles-pp._raw_spin_lock
0.03 ±173% +0.2 0.20 ± 35% perf-profile.self.cycles-pp.unaccount_page_cache_page
0.09 ± 59% +0.2 0.26 ± 3% perf-profile.self.cycles-pp.__mod_zone_page_state
0.20 ± 19% +0.2 0.37 ± 45% perf-profile.self.cycles-pp.pm_qos_read_value
0.18 ± 40% +0.2 0.37 ± 27% perf-profile.self.cycles-pp.timerqueue_del
0.38 ± 35% +0.2 0.58 ± 13% perf-profile.self.cycles-pp.__hrtimer_run_queues
0.19 ± 36% +0.2 0.42 ± 20% perf-profile.self.cycles-pp.nr_iowait_cpu
0.00 +0.2 0.23 ± 43% perf-profile.self.cycles-pp.free_unref_page_list
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-5.4.0-rc3-00048-g569342dc24853" of type "text/plain" (200550 bytes)
View attachment "job-script" of type "text/plain" (8266 bytes)
View attachment "job.yaml" of type "text/plain" (5796 bytes)
View attachment "reproduce" of type "text/plain" (900 bytes)
Powered by blists - more mailing lists