* [tip:sched/core] [sched/fair] 61b82dfb6b: aim7.jobs-per-min 2.1% improvement
@ 2025-01-07 13:01 kernel test robot
0 siblings, 0 replies; only message in thread
From: kernel test robot @ 2025-01-07 13:01 UTC (permalink / raw)
To: Vincent Guittot
Cc: oe-lkp, lkp, linux-kernel, x86, Peter Zijlstra, Dietmar Eggemann,
aubrey.li, yu.c.chen, oliver.sang
Hello,
kernel test robot noticed a 2.1% improvement of aim7.jobs-per-min on:
commit: 61b82dfb6b7e1f951fd1e95198a2aee2ccf6a167 ("sched/fair: Do not try to migrate delayed dequeue task")
https://git.kernel.org/cgit/linux/kernel/git/tip/tip.git sched/core
testcase: aim7
config: x86_64-rhel-9.4
compiler: gcc-12
test machine: 128 threads 2 sockets Intel(R) Xeon(R) Gold 6338 CPU @ 2.00GHz (Ice Lake) with 256G memory
parameters:
disk: 1BRD_48G
fs: xfs
test: sync_disk_rw
load: 600
cpufreq_governor: performance
Details are as below:
-------------------------------------------------------------------------------------------------->
The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20250107/202501072048.6f381b3a-lkp@intel.com
=========================================================================================
compiler/cpufreq_governor/disk/fs/kconfig/load/rootfs/tbox_group/test/testcase:
gcc-12/performance/1BRD_48G/xfs/x86_64-rhel-9.4/600/debian-12-x86_64-20240206.cgz/lkp-icl-2sp2/sync_disk_rw/aim7
commit:
736c55a02c ("sched/fair: Rename cfs_rq.nr_running into nr_queued")
61b82dfb6b ("sched/fair: Do not try to migrate delayed dequeue task")
736c55a02c477ad3 61b82dfb6b7e1f951fd1e95198a
---------------- ---------------------------
%stddev %change %stddev
\ | \
0.05 ± 31% +0.2 0.20 ±140% mpstat.cpu.all.iowait%
21.83 ± 18% -51.5% 10.58 ± 79% sched_debug.cfs_rq:/.load_avg.min
38.33 +1.9% 39.07 iostat.cpu.idle
60.71 -1.5% 59.80 iostat.cpu.system
47551 +2.1% 48556 aim7.jobs-per-min
7398 -2.0% 7250 aim7.time.percent_of_cpu_this_job_got
5583 -4.1% 5357 aim7.time.system_time
34860959 -1.5% 34349280 aim7.time.voluntary_context_switches
1.91 +2.6% 1.96 perf-stat.i.MPKI
1.319e+08 +2.1% 1.347e+08 perf-stat.i.cache-misses
837578 +1.7% 851427 perf-stat.i.context-switches
2.132e+11 -1.6% 2.099e+11 perf-stat.i.cpu-cycles
177061 -3.6% 170730 perf-stat.i.cpu-migrations
0.45 ± 2% +2.4% 0.46 perf-stat.i.ipc
2.13 +2.8% 2.19 perf-stat.overall.MPKI
1618 -3.6% 1560 perf-stat.overall.cycles-between-cache-misses
1.301e+08 +2.2% 1.331e+08 perf-stat.ps.cache-misses
826542 +1.7% 840923 perf-stat.ps.context-switches
2.106e+11 -1.4% 2.077e+11 perf-stat.ps.cpu-cycles
174640 -3.5% 168471 perf-stat.ps.cpu-migrations
4.691e+12 -2.2% 4.587e+12 perf-stat.total.instructions
0.27 ± 6% -15.9% 0.22 ± 7% perf-sched.sch_delay.avg.ms.__cond_resched.down_write.xfs_ilock_for_iomap.xfs_buffered_write_iomap_begin.iomap_iter
0.14 ± 76% -99.8% 0.00 ±223% perf-sched.sch_delay.avg.ms.__cond_resched.kmem_cache_alloc_noprof.alloc_empty_file.path_openat.do_filp_open
0.08 ± 2% -14.7% 0.07 ± 2% perf-sched.sch_delay.avg.ms.io_schedule.folio_wait_bit_common.folio_wait_writeback.__filemap_fdatawait_range
0.08 ± 4% -14.3% 0.07 ± 3% perf-sched.sch_delay.avg.ms.schedule_timeout.__wait_for_common.__flush_workqueue.xlog_cil_push_now.isra
0.01 ± 52% -52.7% 0.01 ± 26% perf-sched.sch_delay.avg.ms.wait_for_partner.fifo_open.do_dentry_open.vfs_open
0.16 ± 85% -99.8% 0.00 ±223% perf-sched.sch_delay.max.ms.__cond_resched.kmem_cache_alloc_noprof.alloc_empty_file.path_openat.do_filp_open
1.62 ± 45% -53.6% 0.75 ± 14% perf-sched.sch_delay.max.ms.__cond_resched.writeback_get_folio.writeback_iter.iomap_writepages.xfs_vm_writepages
0.02 ± 15% +130.5% 0.04 ± 58% perf-sched.sch_delay.max.ms.schedule_timeout.kcompactd.kthread.ret_from_fork
0.21 ±105% -88.3% 0.02 ± 64% perf-sched.sch_delay.max.ms.wait_for_partner.fifo_open.do_dentry_open.vfs_open
0.07 ± 4% -12.0% 0.06 ± 2% perf-sched.total_sch_delay.average.ms
750.06 ± 6% -11.1% 666.52 ± 18% perf-sched.wait_and_delay.avg.ms.__x64_sys_pause.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown]
168.85 ±141% +205.4% 515.61 perf-sched.wait_and_delay.avg.ms.do_task_dead.do_exit.__x64_sys_exit.x64_sys_call.do_syscall_64
31.35 ± 7% -10.6% 28.01 perf-sched.wait_and_delay.avg.ms.do_task_dead.do_exit.do_group_exit.__x64_sys_exit_group.x64_sys_call
832.00 ± 17% +20.6% 1003 ± 4% perf-sched.wait_and_delay.count.__cond_resched.__wait_for_common.affine_move_task.__set_cpus_allowed_ptr.__sched_setaffinity
2113 ± 10% +15.7% 2445 ± 2% perf-sched.wait_and_delay.count.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm
75.84 ± 18% -14.9% 64.53 ± 4% perf-sched.wait_and_delay.max.ms.__cond_resched.__wait_for_common.__flush_workqueue.xlog_cil_push_now.isra
168.85 ±141% +205.4% 515.61 perf-sched.wait_and_delay.max.ms.do_task_dead.do_exit.__x64_sys_exit.x64_sys_call.do_syscall_64
1.15 ± 9% -33.4% 0.77 ± 24% perf-sched.wait_time.avg.ms.__cond_resched.__wait_for_common.wait_for_completion_state.kernel_clone.__x64_sys_vfork
0.96 ± 17% -25.9% 0.71 ± 7% perf-sched.wait_time.avg.ms.__cond_resched.writeback_get_folio.writeback_iter.iomap_writepages.xfs_vm_writepages
168.85 ±141% +205.4% 515.61 perf-sched.wait_time.avg.ms.do_task_dead.do_exit.__x64_sys_exit.x64_sys_call.do_syscall_64
31.35 ± 7% -10.6% 28.01 perf-sched.wait_time.avg.ms.do_task_dead.do_exit.do_group_exit.__x64_sys_exit_group.x64_sys_call
9.88 ±118% -76.9% 2.28 ± 55% perf-sched.wait_time.max.ms.__cond_resched.writeback_get_folio.writeback_iter.iomap_writepages.xfs_vm_writepages
168.85 ±141% +205.4% 515.61 perf-sched.wait_time.max.ms.do_task_dead.do_exit.__x64_sys_exit.x64_sys_call.do_syscall_64
226.05 ±153% -70.8% 65.98 ± 4% perf-sched.wait_time.max.ms.schedule_preempt_disabled.__mutex_lock.constprop.0.__flush_workqueue
333.54 ±141% -100.0% 0.02 ± 64% perf-sched.wait_time.max.ms.wait_for_partner.fifo_open.do_dentry_open.vfs_open
70.43 -1.1 69.31 perf-profile.calltrace.cycles-pp.__mutex_lock.__flush_workqueue.xlog_cil_push_now.xlog_cil_force_seq.xfs_log_force_seq
68.32 -1.1 67.21 perf-profile.calltrace.cycles-pp.osq_lock.__mutex_lock.__flush_workqueue.xlog_cil_push_now.xlog_cil_force_seq
74.65 -1.0 73.63 perf-profile.calltrace.cycles-pp.__flush_workqueue.xlog_cil_push_now.xlog_cil_force_seq.xfs_log_force_seq.xfs_file_fsync
81.36 -1.0 80.34 perf-profile.calltrace.cycles-pp.xfs_log_force_seq.xfs_file_fsync.xfs_file_buffered_write.vfs_write.ksys_write
75.60 -1.0 74.61 perf-profile.calltrace.cycles-pp.xlog_cil_push_now.xlog_cil_force_seq.xfs_log_force_seq.xfs_file_fsync.xfs_file_buffered_write
77.63 -1.0 76.66 perf-profile.calltrace.cycles-pp.xlog_cil_force_seq.xfs_log_force_seq.xfs_file_fsync.xfs_file_buffered_write.vfs_write
90.80 -0.4 90.38 perf-profile.calltrace.cycles-pp.xfs_file_fsync.xfs_file_buffered_write.vfs_write.ksys_write.do_syscall_64
93.24 -0.3 92.92 perf-profile.calltrace.cycles-pp.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.write
93.52 -0.3 93.20 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.write
93.27 -0.3 92.95 perf-profile.calltrace.cycles-pp.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.write
93.52 -0.3 93.21 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.write
93.12 -0.3 92.81 perf-profile.calltrace.cycles-pp.xfs_file_buffered_write.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe
93.75 -0.3 93.44 perf-profile.calltrace.cycles-pp.write
0.52 +0.0 0.56 perf-profile.calltrace.cycles-pp.__sysvec_call_function_single.sysvec_call_function_single.asm_sysvec_call_function_single.acpi_safe_halt.acpi_idle_do_entry
0.64 +0.0 0.68 perf-profile.calltrace.cycles-pp.sysvec_call_function_single.asm_sysvec_call_function_single.acpi_safe_halt.acpi_idle_do_entry.acpi_idle_enter
0.96 +0.0 1.00 perf-profile.calltrace.cycles-pp.iomap_write_iter.iomap_file_buffered_write.xfs_file_buffered_write.vfs_write.ksys_write
1.24 +0.0 1.29 perf-profile.calltrace.cycles-pp.iomap_file_buffered_write.xfs_file_buffered_write.vfs_write.ksys_write.do_syscall_64
0.88 +0.1 0.93 perf-profile.calltrace.cycles-pp.acpi_safe_halt.acpi_idle_do_entry.acpi_idle_enter.cpuidle_enter_state.cpuidle_enter
0.85 +0.1 0.92 ± 4% perf-profile.calltrace.cycles-pp.xfs_iomap_write_unwritten.xfs_end_ioend.xfs_end_io.process_one_work.worker_thread
1.10 ± 5% +0.1 1.17 ± 3% perf-profile.calltrace.cycles-pp.__folio_end_writeback.folio_end_writeback.iomap_finish_ioend.iomap_finish_ioends.xfs_end_ioend
1.31 ± 3% +0.1 1.40 ± 2% perf-profile.calltrace.cycles-pp.iomap_finish_ioends.xfs_end_ioend.xfs_end_io.process_one_work.worker_thread
1.29 ± 3% +0.1 1.38 ± 2% perf-profile.calltrace.cycles-pp.folio_end_writeback.iomap_finish_ioend.iomap_finish_ioends.xfs_end_ioend.xfs_end_io
1.31 ± 3% +0.1 1.40 ± 2% perf-profile.calltrace.cycles-pp.iomap_finish_ioend.iomap_finish_ioends.xfs_end_ioend.xfs_end_io.process_one_work
1.54 +0.1 1.63 perf-profile.calltrace.cycles-pp.acpi_idle_do_entry.acpi_idle_enter.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call
1.55 +0.1 1.64 perf-profile.calltrace.cycles-pp.acpi_idle_enter.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle
1.60 +0.1 1.70 perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry
1.62 +0.1 1.71 perf-profile.calltrace.cycles-pp.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary
2.06 +0.1 2.16 perf-profile.calltrace.cycles-pp.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary.common_startup_64
1.94 +0.1 2.05 perf-profile.calltrace.cycles-pp.asm_sysvec_call_function_single.acpi_safe_halt.acpi_idle_do_entry.acpi_idle_enter.cpuidle_enter_state
0.42 ± 44% +0.1 0.54 perf-profile.calltrace.cycles-pp.__flush_smp_call_function_queue.__sysvec_call_function_single.sysvec_call_function_single.asm_sysvec_call_function_single.acpi_safe_halt
2.46 +0.1 2.58 perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.common_startup_64
2.46 +0.1 2.59 perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.common_startup_64
2.46 +0.1 2.59 perf-profile.calltrace.cycles-pp.start_secondary.common_startup_64
2.49 +0.1 2.62 perf-profile.calltrace.cycles-pp.common_startup_64
2.18 ± 2% +0.2 2.33 ± 2% perf-profile.calltrace.cycles-pp.xfs_end_io.process_one_work.worker_thread.kthread.ret_from_fork
2.17 ± 2% +0.2 2.32 ± 2% perf-profile.calltrace.cycles-pp.xfs_end_ioend.xfs_end_io.process_one_work.worker_thread.kthread
2.92 +0.2 3.08 ± 2% perf-profile.calltrace.cycles-pp.process_one_work.worker_thread.kthread.ret_from_fork.ret_from_fork_asm
3.22 +0.2 3.38 perf-profile.calltrace.cycles-pp.worker_thread.kthread.ret_from_fork.ret_from_fork_asm
3.23 +0.2 3.40 perf-profile.calltrace.cycles-pp.kthread.ret_from_fork.ret_from_fork_asm
3.23 +0.2 3.40 perf-profile.calltrace.cycles-pp.ret_from_fork.ret_from_fork_asm
3.23 +0.2 3.40 perf-profile.calltrace.cycles-pp.ret_from_fork_asm
3.58 ± 5% +0.2 3.83 ± 3% perf-profile.calltrace.cycles-pp.__folio_end_writeback.folio_end_writeback.iomap_finish_ioend.__submit_bio.__submit_bio_noacct
3.71 ± 5% +0.3 3.97 ± 3% perf-profile.calltrace.cycles-pp.folio_end_writeback.iomap_finish_ioend.__submit_bio.__submit_bio_noacct.iomap_submit_ioend
3.74 ± 5% +0.3 4.00 ± 3% perf-profile.calltrace.cycles-pp.iomap_finish_ioend.__submit_bio.__submit_bio_noacct.iomap_submit_ioend.iomap_writepages
4.46 ± 4% +0.3 4.75 ± 2% perf-profile.calltrace.cycles-pp.iomap_submit_ioend.iomap_writepages.xfs_vm_writepages.do_writepages.filemap_fdatawrite_wbc
4.41 ± 4% +0.3 4.70 ± 2% perf-profile.calltrace.cycles-pp.__submit_bio_noacct.iomap_submit_ioend.iomap_writepages.xfs_vm_writepages.do_writepages
4.40 ± 4% +0.3 4.69 ± 2% perf-profile.calltrace.cycles-pp.__submit_bio.__submit_bio_noacct.iomap_submit_ioend.iomap_writepages.xfs_vm_writepages
8.96 ± 5% +0.6 9.55 ± 3% perf-profile.calltrace.cycles-pp.do_writepages.filemap_fdatawrite_wbc.__filemap_fdatawrite_range.file_write_and_wait_range.xfs_file_fsync
9.01 ± 5% +0.6 9.60 ± 3% perf-profile.calltrace.cycles-pp.filemap_fdatawrite_wbc.__filemap_fdatawrite_range.file_write_and_wait_range.xfs_file_fsync.xfs_file_buffered_write
8.92 ± 5% +0.6 9.51 ± 3% perf-profile.calltrace.cycles-pp.iomap_writepages.xfs_vm_writepages.do_writepages.filemap_fdatawrite_wbc.__filemap_fdatawrite_range
9.02 ± 5% +0.6 9.61 ± 3% perf-profile.calltrace.cycles-pp.__filemap_fdatawrite_range.file_write_and_wait_range.xfs_file_fsync.xfs_file_buffered_write.vfs_write
8.94 ± 5% +0.6 9.53 ± 3% perf-profile.calltrace.cycles-pp.xfs_vm_writepages.do_writepages.filemap_fdatawrite_wbc.__filemap_fdatawrite_range.file_write_and_wait_range
9.28 ± 5% +0.6 9.88 ± 3% perf-profile.calltrace.cycles-pp.file_write_and_wait_range.xfs_file_fsync.xfs_file_buffered_write.vfs_write.ksys_write
70.43 -1.1 69.31 perf-profile.children.cycles-pp.__mutex_lock
68.34 -1.1 67.23 perf-profile.children.cycles-pp.osq_lock
74.66 -1.0 73.63 perf-profile.children.cycles-pp.__flush_workqueue
81.36 -1.0 80.34 perf-profile.children.cycles-pp.xfs_log_force_seq
75.60 -1.0 74.61 perf-profile.children.cycles-pp.xlog_cil_push_now
77.64 -1.0 76.66 perf-profile.children.cycles-pp.xlog_cil_force_seq
90.80 -0.4 90.38 perf-profile.children.cycles-pp.xfs_file_fsync
93.30 -0.3 92.98 perf-profile.children.cycles-pp.ksys_write
93.27 -0.3 92.95 perf-profile.children.cycles-pp.vfs_write
93.13 -0.3 92.81 perf-profile.children.cycles-pp.xfs_file_buffered_write
93.82 -0.3 93.51 perf-profile.children.cycles-pp.write
93.81 -0.3 93.51 perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
93.78 -0.3 93.49 perf-profile.children.cycles-pp.do_syscall_64
0.20 +0.0 0.21 perf-profile.children.cycles-pp.writeback_iter
0.31 +0.0 0.33 ± 2% perf-profile.children.cycles-pp.iomap_write_begin
0.69 +0.0 0.71 perf-profile.children.cycles-pp.ttwu_do_activate
0.35 +0.0 0.37 perf-profile.children.cycles-pp.iomap_write_end
0.55 +0.0 0.58 perf-profile.children.cycles-pp.sched_ttwu_pending
0.64 +0.0 0.66 perf-profile.children.cycles-pp.__flush_smp_call_function_queue
0.65 +0.0 0.68 perf-profile.children.cycles-pp.__sysvec_call_function_single
0.79 +0.0 0.82 perf-profile.children.cycles-pp.sysvec_call_function_single
0.96 +0.0 1.00 perf-profile.children.cycles-pp.iomap_write_iter
1.24 +0.0 1.29 perf-profile.children.cycles-pp.iomap_file_buffered_write
1.18 ± 2% +0.1 1.23 ± 2% perf-profile.children.cycles-pp.__xfs_trans_commit
1.48 +0.1 1.54 perf-profile.children.cycles-pp.asm_sysvec_call_function_single
0.85 +0.1 0.92 ± 4% perf-profile.children.cycles-pp.xfs_iomap_write_unwritten
1.31 ± 3% +0.1 1.40 ± 2% perf-profile.children.cycles-pp.iomap_finish_ioends
1.57 +0.1 1.66 perf-profile.children.cycles-pp.acpi_idle_enter
1.57 +0.1 1.66 perf-profile.children.cycles-pp.acpi_safe_halt
1.63 +0.1 1.72 perf-profile.children.cycles-pp.cpuidle_enter_state
1.64 +0.1 1.73 perf-profile.children.cycles-pp.cpuidle_enter
1.57 +0.1 1.66 perf-profile.children.cycles-pp.acpi_idle_do_entry
2.08 +0.1 2.19 perf-profile.children.cycles-pp.cpuidle_idle_call
2.49 +0.1 2.62 perf-profile.children.cycles-pp.do_idle
2.46 +0.1 2.59 perf-profile.children.cycles-pp.start_secondary
2.49 +0.1 2.62 perf-profile.children.cycles-pp.common_startup_64
2.49 +0.1 2.62 perf-profile.children.cycles-pp.cpu_startup_entry
2.17 ± 2% +0.2 2.32 ± 2% perf-profile.children.cycles-pp.xfs_end_ioend
2.18 ± 2% +0.2 2.34 ± 2% perf-profile.children.cycles-pp.xfs_end_io
2.92 +0.2 3.08 ± 2% perf-profile.children.cycles-pp.process_one_work
3.24 +0.2 3.40 perf-profile.children.cycles-pp.ret_from_fork
3.24 +0.2 3.40 perf-profile.children.cycles-pp.ret_from_fork_asm
3.23 +0.2 3.40 perf-profile.children.cycles-pp.kthread
3.22 +0.2 3.39 perf-profile.children.cycles-pp.worker_thread
4.46 ± 4% +0.3 4.75 ± 2% perf-profile.children.cycles-pp.iomap_submit_ioend
4.46 ± 4% +0.3 4.75 ± 2% perf-profile.children.cycles-pp.__submit_bio_noacct
4.46 ± 4% +0.3 4.74 ± 2% perf-profile.children.cycles-pp.__submit_bio
4.68 ± 5% +0.3 5.01 ± 3% perf-profile.children.cycles-pp.__folio_end_writeback
5.00 ± 5% +0.3 5.34 ± 3% perf-profile.children.cycles-pp.folio_end_writeback
5.05 ± 5% +0.3 5.40 ± 3% perf-profile.children.cycles-pp.iomap_finish_ioend
8.96 ± 5% +0.6 9.55 ± 3% perf-profile.children.cycles-pp.do_writepages
9.01 ± 5% +0.6 9.60 ± 3% perf-profile.children.cycles-pp.filemap_fdatawrite_wbc
9.02 ± 5% +0.6 9.61 ± 3% perf-profile.children.cycles-pp.__filemap_fdatawrite_range
8.94 ± 5% +0.6 9.53 ± 3% perf-profile.children.cycles-pp.xfs_vm_writepages
8.92 ± 5% +0.6 9.52 ± 3% perf-profile.children.cycles-pp.iomap_writepages
9.28 ± 5% +0.6 9.88 ± 3% perf-profile.children.cycles-pp.file_write_and_wait_range
67.44 -1.1 66.35 perf-profile.self.cycles-pp.osq_lock
0.21 +0.0 0.22 perf-profile.self.cycles-pp.iomap_set_range_uptodate
0.64 +0.0 0.66 perf-profile.self.cycles-pp._raw_spin_lock_irqsave
0.75 +0.0 0.79 perf-profile.self.cycles-pp.acpi_safe_halt
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/wiki
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2025-01-07 13:01 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-01-07 13:01 [tip:sched/core] [sched/fair] 61b82dfb6b: aim7.jobs-per-min 2.1% improvement kernel test robot
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.