Greeting,
There is no primary kpi change in this test, below is the data collected through multiple
monitors running background just for your information.
commit: e64855c6cfaa0a80c1b71c5f647cb792dc436668 ("fs: Add helper to clean bdev
aliases under a bh and use it")
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master
in testcase: fio-basic
on test machine: 16 threads Intel(R) Xeon(R) CPU D-1541 @ 2.10GHz with 8G memory
with following parameters:
runtime: 300s
disk: 1SSD
fs: xfs
nr_task: 4
rw: randwrite
bs: 4M
ioengine: sync
test_size: 400g
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
Details are as below:
-------------------------------------------------------------------------------------------------->
To reproduce:
git clone
git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
=========================================================================================
tbox_group/testcase/rootfs/kconfig/compiler/runtime/disk/fs/nr_task/rw/bs/ioengine/test_size/cpufreq_governor:
lkp-bdw-de1/fio-basic/debian-x86_64-2016-08-31.cgz/x86_64-rhel-7.2/gcc-6/300s/1SSD/xfs/4/randwrite/4M/sync/400g/performance
commit:
69a9bea146b185be8ec50e80eaecd8e487e689f8
e64855c6cfaa0a80c1b71c5f647cb792dc436668
69a9bea146b185be e64855c6cfaa0a80c1b71c5f64
---------------- --------------------------
%stddev %change %stddev
\ | \
64.95 ± 0% +18.7% 77.13 ± 0% fio.time.system_time
23.00 ± 0% +18.8% 27.33 ± 1% fio.time.percent_of_cpu_this_job_got
7.63 ± 4% -27.4% 5.54 ± 16% fio.latency_100ms%
91.53 ± 0% +2.3% 93.61 ± 0% fio.latency_50ms%
94514 ± 2% -10.8% 84334 ± 1% softirqs.RCU
10498 ± 7% +12.4% 11798 ± 4% slabinfo.anon_vma_chain.active_objs
10606 ± 6% +11.2% 11798 ± 4% slabinfo.anon_vma_chain.num_objs
23.00 ± 0% +18.8% 27.33 ± 1% time.percent_of_cpu_this_job_got
64.95 ± 0% +18.7% 77.13 ± 0% time.system_time
1984 ±173% +1307.0% 27918 ± 71%
latency_stats.avg.perf_event_alloc.SYSC_perf_event_open.SyS_perf_event_open.entry_SYSCALL_64_fastpath
1984 ±173% +1307.0% 27918 ± 71%
latency_stats.max.perf_event_alloc.SYSC_perf_event_open.SyS_perf_event_open.entry_SYSCALL_64_fastpath
1984 ±173% +1307.0% 27918 ± 71%
latency_stats.sum.perf_event_alloc.SYSC_perf_event_open.SyS_perf_event_open.entry_SYSCALL_64_fastpath
6894619 ± 5% +15.5% 7961219 ± 1% cpuidle.C1-BDW.time
84207 ± 5% +12.3% 94570 ± 3% cpuidle.C1-BDW.usage
35423874 ± 14% +24.1% 43971600 ± 5% cpuidle.POLL.time
3743 ± 7% +17.0% 4381 ± 3% cpuidle.POLL.usage
6.09 ± 1% +3.9% 6.33 ± 1% turbostat.%Busy
123.75 ± 2% +8.0% 133.67 ± 1% turbostat.Avg_MHz
19.55 ± 0% +1.4% 19.83 ± 0% turbostat.PkgWatt
10.90 ± 0% +1.5% 11.07 ± 0% turbostat.RAMWatt
908.10 ± 31% +58.2% 1436 ± 14% sched_debug.cfs_rq:/.exec_clock.min
242.71 ± 3% -8.6% 221.94 ± 6% sched_debug.cfs_rq:/.load_avg.min
8612 ± 2% +12.3% 9671 ± 3% sched_debug.cfs_rq:/.min_vruntime.avg
4872 ± 10% +29.6% 6312 ± 3% sched_debug.cfs_rq:/.min_vruntime.min
332.71 ±144% -87.5% 41.50 ± 8% sched_debug.cfs_rq:/.runnable_load_avg.max
-9518 ± -5% -8.4% -8721 ± -6% sched_debug.cfs_rq:/.spread0.min
115.00 ± 5% -10.9% 102.50 ± 2% sched_debug.cfs_rq:/.util_avg.min
328.42 ±147% -90.2% 32.17 ± 13% sched_debug.cpu.cpu_load[0].max
79.95 ±147% -89.1% 8.69 ± 17% sched_debug.cpu.cpu_load[0].stddev
311.85 ± 5% -12.2% 273.88 ± 12% sched_debug.cpu.curr->pid.avg
0.46 ± 8% -17.1% 0.38 ± 7% sched_debug.cpu.nr_running.stddev
7.013e+10 ± 1% +34.8% 9.454e+10 ± 1% perf-stat.branch-instructions
0.73 ± 2% -15.7% 0.61 ± 3% perf-stat.branch-miss-rate%
5.109e+08 ± 2% +13.6% 5.802e+08 ± 3% perf-stat.branch-misses
5.566e+11 ± 0% +9.9% 6.117e+11 ± 1% perf-stat.cpu-cycles
9913 ± 3% -4.0% 9521 ± 3% perf-stat.cpu-migrations
0.17 ± 3% -10.9% 0.15 ± 0% perf-stat.dTLB-load-miss-rate%
1.072e+11 ± 3% +11.1% 1.191e+11 ± 2% perf-stat.dTLB-loads
69.70 ± 0% +2.5% 71.47 ± 0% perf-stat.iTLB-load-miss-rate%
48688123 ± 0% -1.2% 48109913 ± 0% perf-stat.iTLB-load-misses
21172777 ± 2% -9.3% 19208560 ± 1% perf-stat.iTLB-loads
3.912e+11 ± 1% +22.9% 4.807e+11 ± 1% perf-stat.instructions
8035 ± 1% +24.4% 9993 ± 2% perf-stat.instructions-per-iTLB-miss
0.70 ± 2% +11.8% 0.79 ± 1% perf-stat.ipc
2.14 ± 16% +218.8% 6.82 ± 1%
perf-profile.calltrace.cycles-pp.__block_write_begin_int.iomap_write_begin.iomap_write_actor.iomap_apply.iomap_file_buffered_write
4.44 ± 7% -11.5% 3.93 ± 8%
perf-profile.calltrace.cycles-pp.__irqentry_text_start.cpuidle_enter.call_cpuidle.cpu_startup_entry.start_secondary
0.94 ± 14% -100.0% 0.00 ± -1%
perf-profile.calltrace.cycles-pp.__lru_cache_add.lru_cache_add.add_to_page_cache_lru.pagecache_get_page.grab_cache_page_write_begin
0.00 ± -1% +Inf% 1.57 ± 5%
perf-profile.calltrace.cycles-pp.__pagevec_release.clean_bdev_aliases.__block_write_begin_int.iomap_write_begin.iomap_write_actor
2.43 ± 10% -31.1% 1.67 ± 9%
perf-profile.calltrace.cycles-pp.add_to_page_cache_lru.pagecache_get_page.grab_cache_page_write_begin.iomap_write_begin.iomap_write_actor
0.97 ± 5% -22.3% 0.75 ± 8%
perf-profile.calltrace.cycles-pp.ast_imageblit.soft_cursor.bit_cursor.fb_flashcursor.process_one_work
0.97 ± 5% -22.3% 0.75 ± 8%
perf-profile.calltrace.cycles-pp.bit_cursor.fb_flashcursor.process_one_work.worker_thread.kthread
1.84 ± 18% -40.7% 1.09 ± 33%
perf-profile.calltrace.cycles-pp.call_console_drivers.console_unlock.vprintk_emit.vprintk_default.printk
0.00 ± -1% +Inf% 5.12 ± 1%
perf-profile.calltrace.cycles-pp.clean_bdev_aliases.__block_write_begin_int.iomap_write_begin.iomap_write_actor.iomap_apply
0.74 ± 21% -24.3% 0.56 ± 12%
perf-profile.calltrace.cycles-pp.clear_page_dirty_for_io.write_cache_pages.xfs_vm_writepages.do_writepages.__writeback_single_inode
1.84 ± 18% -40.7% 1.09 ± 33%
perf-profile.calltrace.cycles-pp.console_unlock.vprintk_emit.vprintk_default.printk.perf_duration_warn
0.97 ± 5% -22.3% 0.75 ± 8%
perf-profile.calltrace.cycles-pp.fb_flashcursor.process_one_work.worker_thread.kthread.ret_from_fork
0.00 ± -1% +Inf% 3.35 ± 3%
perf-profile.calltrace.cycles-pp.find_get_pages.pagevec_lookup.clean_bdev_aliases.__block_write_begin_int.iomap_write_begin
7.20 ± 13% +52.9% 11.01 ± 3%
perf-profile.calltrace.cycles-pp.iomap_write_begin.iomap_write_actor.iomap_apply.iomap_file_buffered_write.xfs_file_buffered_aio_write
1.74 ± 6% -13.1% 1.52 ± 3%
perf-profile.calltrace.cycles-pp.irq_exit.smp_apic_timer_interrupt.__irqentry_text_start.cpuidle_enter.call_cpuidle
0.00 ± -1% +Inf% 1.36 ± 9%
perf-profile.calltrace.cycles-pp.lru_add_drain_cpu.__pagevec_release.clean_bdev_aliases.__block_write_begin_int.iomap_write_begin
0.95 ± 13% -100.0% 0.00 ± -1%
perf-profile.calltrace.cycles-pp.lru_cache_add.add_to_page_cache_lru.pagecache_get_page.grab_cache_page_write_begin.iomap_write_begin
0.94 ± 4% -20.4% 0.75 ± 8%
perf-profile.calltrace.cycles-pp.memcpy_erms.ast_imageblit.soft_cursor.bit_cursor.fb_flashcursor
0.00 ± -1% +Inf% 3.36 ± 3%
perf-profile.calltrace.cycles-pp.pagevec_lookup.clean_bdev_aliases.__block_write_begin_int.iomap_write_begin.iomap_write_actor
0.78 ± 17% -100.0% 0.00 ± -1%
perf-profile.calltrace.cycles-pp.pagevec_lru_move_fn.__lru_cache_add.lru_cache_add.add_to_page_cache_lru.pagecache_get_page
0.00 ± -1% +Inf% 1.29 ± 8%
perf-profile.calltrace.cycles-pp.pagevec_lru_move_fn.lru_add_drain_cpu.__pagevec_release.clean_bdev_aliases.__block_write_begin_int
1.84 ± 18% -40.7% 1.09 ± 33%
perf-profile.calltrace.cycles-pp.perf_duration_warn.irq_work_run_list.irq_work_run.smp_irq_work_interrupt.irq_work_interrupt
6.71 ± 20% +80.3% 12.10 ± 39%
perf-profile.calltrace.cycles-pp.poll_idle.cpuidle_enter_state.cpuidle_enter.call_cpuidle.cpu_startup_entry
1.84 ± 18% -40.7% 1.09 ± 33%
perf-profile.calltrace.cycles-pp.printk.perf_duration_warn.irq_work_run_list.irq_work_run.smp_irq_work_interrupt
0.00 ± -1% +Inf% 1.81 ± 8%
perf-profile.calltrace.cycles-pp.radix_tree_next_chunk.find_get_pages.pagevec_lookup.clean_bdev_aliases.__block_write_begin_int
1.72 ± 17% -41.4% 1.01 ± 33%
perf-profile.calltrace.cycles-pp.serial8250_console_putchar.uart_console_write.serial8250_console_write.univ8250_console_write.call_console_drivers
1.76 ± 17% -41.1% 1.04 ± 33%
perf-profile.calltrace.cycles-pp.serial8250_console_write.univ8250_console_write.call_console_drivers.console_unlock.vprintk_emit
4.34 ± 7% -12.1% 3.81 ± 7%
perf-profile.calltrace.cycles-pp.smp_apic_timer_interrupt.__irqentry_text_start.cpuidle_enter.call_cpuidle.cpu_startup_entry
0.97 ± 5% -22.3% 0.75 ± 8%
perf-profile.calltrace.cycles-pp.soft_cursor.bit_cursor.fb_flashcursor.process_one_work.worker_thread
1.06 ± 7% -16.2% 0.89 ± 9%
perf-profile.calltrace.cycles-pp.tick_nohz_stop_sched_tick.__tick_nohz_idle_enter.tick_nohz_irq_exit.irq_exit.smp_apic_timer_interrupt
1.72 ± 17% -41.4% 1.01 ± 33%
perf-profile.calltrace.cycles-pp.uart_console_write.serial8250_console_write.univ8250_console_write.call_console_drivers.console_unlock
1.76 ± 17% -41.1% 1.04 ± 33%
perf-profile.calltrace.cycles-pp.univ8250_console_write.call_console_drivers.console_unlock.vprintk_emit.vprintk_default
1.84 ± 18% -40.7% 1.09 ± 33%
perf-profile.calltrace.cycles-pp.vprintk_default.printk.perf_duration_warn.irq_work_run_list.irq_work_run
1.84 ± 18% -40.7% 1.09 ± 33%
perf-profile.calltrace.cycles-pp.vprintk_emit.vprintk_default.printk.perf_duration_warn.irq_work_run_list
1.56 ± 10% -38.2% 0.97 ± 33%
perf-profile.calltrace.cycles-pp.wait_for_xmitr.serial8250_console_putchar.uart_console_write.serial8250_console_write.univ8250_console_write
2.15 ± 16% +218.7% 6.84 ± 1%
perf-profile.children.cycles-pp.__block_write_begin_int
0.77 ± 23% -37.9% 0.48 ± 13%
perf-profile.children.cycles-pp.__const_udelay
4.83 ± 4% -10.1% 4.34 ± 7%
perf-profile.children.cycles-pp.__irqentry_text_start
0.94 ± 14% -83.4% 0.16 ± 15%
perf-profile.children.cycles-pp.__lru_cache_add
0.12 ± 30% +1288.9% 1.67 ± 5%
perf-profile.children.cycles-pp.__pagevec_release
2.44 ± 9% -31.4% 1.68 ± 10%
perf-profile.children.cycles-pp.add_to_page_cache_lru
1.04 ± 5% -21.6% 0.81 ± 7%
perf-profile.children.cycles-pp.ast_imageblit
0.97 ± 5% -22.3% 0.75 ± 8% perf-profile.children.cycles-pp.bit_cursor
0.00 ± -1% +Inf% 5.12 ± 1%
perf-profile.children.cycles-pp.clean_bdev_aliases
0.97 ± 5% -22.3% 0.75 ± 8%
perf-profile.children.cycles-pp.fb_flashcursor
0.00 ± -1% +Inf% 3.36 ± 3%
perf-profile.children.cycles-pp.find_get_pages
0.95 ± 12% -21.1% 0.75 ± 10%
perf-profile.children.cycles-pp.io_serial_in
7.20 ± 13% +52.9% 11.01 ± 3%
perf-profile.children.cycles-pp.iomap_write_begin
1.90 ± 9% -9.8% 1.71 ± 8%
perf-profile.children.cycles-pp.local_apic_timer_interrupt
0.00 ± -1% +Inf% 1.37 ± 9%
perf-profile.children.cycles-pp.lru_add_drain_cpu
0.96 ± 14% -82.9% 0.16 ± 16%
perf-profile.children.cycles-pp.lru_cache_add
0.95 ± 5% -20.1% 0.76 ± 8% perf-profile.children.cycles-pp.memcpy_erms
5.11 ± 11% -21.4% 4.02 ± 6%
perf-profile.children.cycles-pp.pagecache_get_page
0.00 ± -1% +Inf% 3.37 ± 3%
perf-profile.children.cycles-pp.pagevec_lookup
0.79 ± 16% +65.6% 1.30 ± 7%
perf-profile.children.cycles-pp.pagevec_lru_move_fn
6.71 ± 20% +80.4% 12.10 ± 39% perf-profile.children.cycles-pp.poll_idle
0.06 ± 58% +3083.3% 1.91 ± 8%
perf-profile.children.cycles-pp.radix_tree_next_chunk
1.72 ± 17% -27.8% 1.24 ± 8%
perf-profile.children.cycles-pp.serial8250_console_putchar
1.76 ± 17% -27.4% 1.28 ± 8%
perf-profile.children.cycles-pp.serial8250_console_write
4.74 ± 4% -10.6% 4.23 ± 6%
perf-profile.children.cycles-pp.smp_apic_timer_interrupt
0.97 ± 5% -22.3% 0.75 ± 8% perf-profile.children.cycles-pp.soft_cursor
1.15 ± 5% -13.6% 0.99 ± 9%
perf-profile.children.cycles-pp.tick_nohz_stop_sched_tick
1.72 ± 17% -27.8% 1.24 ± 8%
perf-profile.children.cycles-pp.uart_console_write
1.76 ± 17% -27.4% 1.28 ± 8%
perf-profile.children.cycles-pp.univ8250_console_write
1.72 ± 17% -28.6% 1.23 ± 9%
perf-profile.children.cycles-pp.wait_for_xmitr
0.00 ± -1% +Inf% 1.55 ± 6% perf-profile.self.cycles-pp.find_get_pages
0.95 ± 12% -21.1% 0.75 ± 10% perf-profile.self.cycles-pp.io_serial_in
0.95 ± 5% -20.1% 0.76 ± 8% perf-profile.self.cycles-pp.memcpy_erms
6.71 ± 20% +80.4% 12.10 ± 39% perf-profile.self.cycles-pp.poll_idle
0.06 ± 58% +3083.3% 1.91 ± 8%
perf-profile.self.cycles-pp.radix_tree_next_chunk
fio.time.system_time
80 O+-O--O--O--O-------------------------------------------O-----O--------O
| O O O O O O O O O O O O O O O O |
70 ++ |
60 *+.*..*..*..*..*..*..*..*..*..*..*..*.*..*..*..*..*..*..*..*..*..*..*..*
| |
50 ++ |
| |
40 ++ |
| |
30 ++ |
20 ++ |
| |
10 ++ |
| |
0 ++------------------------------------------------------------------O--+
fio.time.percent_of_cpu_this_job_got
30 ++---------------------------------------------------------------------+
O O O O O O O O O O O O O O O O O O O O O O O O
25 ++ |
| .*..*..*..*..*..*..*..*..*..*..*.*..*..*..*..*..*..*..*..*..*..*..*
*..*. |
20 ++ |
| |
15 ++ |
| |
10 ++ |
| |
| |
5 ++ |
| |
0 ++------------------------------------------------------------------O--+
[*] bisect-good sample
[O] bisect-bad sample
Disclaimer:
Results have been estimated based on internal Intel analysis and are provided
for informational purposes only. Any difference in system hardware or software
design or configuration may affect actual performance.
Thanks,
Ying Huang