* [vbabka:b4/sheaves-for-all-rebased] [slab] aa8fdb9e25: will-it-scale.per_process_ops 46.5% regression
@ 2026-01-13 13:57 kernel test robot
2026-01-28 10:31 ` Vlastimil Babka
0 siblings, 1 reply; 6+ messages in thread
From: kernel test robot @ 2026-01-13 13:57 UTC (permalink / raw)
To: Vlastimil Babka; +Cc: oe-lkp, lkp, linux-mm, oliver.sang
Hello,
kernel test robot noticed a 46.5% regression of will-it-scale.per_process_ops on:
commit: aa8fdb9e2516055552de11cabaacde4d77ad7d72 ("slab: refill sheaves from all nodes")
https://git.kernel.org/cgit/linux/kernel/git/vbabka/linux.git b4/sheaves-for-all-rebased
testcase: will-it-scale
config: x86_64-rhel-9.4
compiler: gcc-14
test machine: 192 threads 2 sockets Intel(R) Xeon(R) 6740E CPU @ 2.4GHz (Sierra Forest) with 256G memory
parameters:
nr_task: 100%
mode: process
test: mmap2
cpufreq_governor: performance
In addition to that, the commit also has significant impact on the following tests:
+------------------+----------------------------------------------------------------------------------------------------+
| testcase: change | stress-ng: stress-ng.pkey.ops_per_sec 28.4% regression |
| test machine | 192 threads 2 sockets Intel(R) Xeon(R) 6740E CPU @ 2.4GHz (Sierra Forest) with 256G memory |
| test parameters | cpufreq_governor=performance |
| | nr_threads=100% |
| | test=pkey |
| | testtime=60s |
+------------------+----------------------------------------------------------------------------------------------------+
| testcase: change | will-it-scale: will-it-scale.per_process_ops 32.8% regression |
| test machine | 224 threads 4 sockets Intel(R) Xeon(R) Platinum 8380H CPU @ 2.90GHz (Cooper Lake) with 192G memory |
| test parameters | cpufreq_governor=performance |
| | mode=process |
| | nr_task=100% |
| | test=brk2 |
+------------------+----------------------------------------------------------------------------------------------------+
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <oliver.sang@intel.com>
| Closes: https://lore.kernel.org/oe-lkp/202601132136.77efd6d7-lkp@intel.com
Details are as below:
-------------------------------------------------------------------------------------------------->
The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20260113/202601132136.77efd6d7-lkp@intel.com
=========================================================================================
compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase:
gcc-14/performance/x86_64-rhel-9.4/process/100%/debian-13-x86_64-20250902.cgz/lkp-srf-2sp2/mmap2/will-it-scale
commit:
6a67958ab0 ("slab: remove unused PREEMPT_RT specific macros")
aa8fdb9e25 ("slab: refill sheaves from all nodes")
6a67958ab000c3a7 aa8fdb9e2516055552de11cabaa
---------------- ---------------------------
%stddev %change %stddev
\ | \
4342 +310.0% 17807 vmstat.system.cs
254632 -2.1% 249196 vmstat.system.in
37383532 -46.5% 20003486 will-it-scale.192.processes
194705 -46.5% 104184 will-it-scale.per_process_ops
37383532 -46.5% 20003486 will-it-scale.workload
13704685 ± 2% -10.7% 12244063 meminfo.Memused
3930129 ± 4% -46.4% 2107850 meminfo.SUnreclaim
4077598 ± 4% -44.7% 2256779 meminfo.Slab
28750536 ± 35% -51.2% 14022086 meminfo.max_used_kB
0.20 -0.1 0.13 mpstat.cpu.all.irq%
37.67 -18.1 19.55 mpstat.cpu.all.soft%
59.17 +18.7 77.84 mpstat.cpu.all.sys%
2.37 -0.9 1.43 mpstat.cpu.all.usr%
1.544e+08 -65.4% 53459373 numa-numastat.node0.local_node
1.545e+08 -65.3% 53600318 numa-numastat.node0.numa_hit
1.614e+08 -65.0% 56529387 numa-numastat.node1.local_node
1.614e+08 -64.9% 56588026 numa-numastat.node1.numa_hit
7274 ± 13% -27.0% 5310 ± 16% perf-c2c.DRAM.local
1458 ± 14% +272.3% 5431 ± 10% perf-c2c.DRAM.remote
77502 ± 9% -58.6% 32066 ± 11% perf-c2c.HITM.local
150.83 ± 12% +2150.3% 3394 ± 12% perf-c2c.HITM.remote
77653 ± 9% -54.3% 35460 ± 10% perf-c2c.HITM.total
0.92 -43.6% 0.52 turbostat.IPC
1027292 ± 7% -1e+06 0.00 turbostat.PKG_%
64.83 -6.7% 60.50 ± 2% turbostat.PkgTmp
464.74 -11.6% 411.01 turbostat.PkgWatt
25.39 -11.8% 22.38 turbostat.RAMWatt
59906794 +2.1% 61151854 proc-vmstat.nr_free_pages_blocks
34605 +10.0% 38075 proc-vmstat.nr_kernel_stack
971757 ± 3% -46.2% 523132 ± 2% proc-vmstat.nr_slab_unreclaimable
3.16e+08 -65.1% 1.102e+08 proc-vmstat.numa_hit
3.158e+08 -65.2% 1.1e+08 proc-vmstat.numa_local
1.277e+09 -66.3% 4.298e+08 proc-vmstat.pgalloc_normal
1.275e+09 -66.4% 4.285e+08 proc-vmstat.pgfree
18.04 ± 3% -78.7% 3.85 ± 5% perf-sched.sch_delay.avg.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
18.04 ± 3% -78.7% 3.85 ± 5% perf-sched.total_sch_delay.average.ms
150.38 ± 2% -62.1% 57.06 perf-sched.total_wait_and_delay.average.ms
20506 ± 3% +307.0% 83458 perf-sched.total_wait_and_delay.count.ms
132.34 ± 2% -59.8% 53.21 perf-sched.total_wait_time.average.ms
150.38 ± 2% -62.1% 57.06 perf-sched.wait_and_delay.avg.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
20506 ± 3% +307.0% 83458 perf-sched.wait_and_delay.count.[unknown].[unknown].[unknown].[unknown].[unknown]
132.34 ± 2% -59.8% 53.21 perf-sched.wait_time.avg.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
615321 ± 36% -70.1% 183936 ± 54% numa-meminfo.node0.Active
614931 ± 36% -70.2% 183545 ± 54% numa-meminfo.node0.Active(anon)
368379 ± 51% -99.7% 976.97 ± 82% numa-meminfo.node0.AnonHugePages
503778 ± 42% -81.2% 94740 ± 8% numa-meminfo.node0.AnonPages
515242 ± 41% -79.4% 105999 ± 6% numa-meminfo.node0.AnonPages.max
17646 ± 2% +10.0% 19408 ± 2% numa-meminfo.node0.KernelStack
1999992 ± 4% -45.3% 1093661 numa-meminfo.node0.SUnreclaim
2064532 ± 5% -42.4% 1189588 ± 2% numa-meminfo.node0.Slab
3575860 ± 10% +22.6% 4384082 ± 6% numa-meminfo.node1.Active
3575470 ± 10% +22.6% 4383698 ± 6% numa-meminfo.node1.Active(anon)
1908870 ± 4% -47.3% 1006765 ± 2% numa-meminfo.node1.SUnreclaim
1991799 ± 4% -46.8% 1059762 ± 2% numa-meminfo.node1.Slab
153718 ± 36% -70.2% 45877 ± 54% numa-vmstat.node0.nr_active_anon
125968 ± 42% -81.2% 23674 ± 8% numa-vmstat.node0.nr_anon_pages
179.85 ± 51% -99.7% 0.48 ± 82% numa-vmstat.node0.nr_anon_transparent_hugepages
17648 ± 2% +10.0% 19407 ± 2% numa-vmstat.node0.nr_kernel_stack
505861 ± 5% -46.5% 270818 numa-vmstat.node0.nr_slab_unreclaimable
153718 ± 36% -70.2% 45877 ± 54% numa-vmstat.node0.nr_zone_active_anon
1.545e+08 -65.3% 53601656 numa-vmstat.node0.numa_hit
1.544e+08 -65.4% 53460711 numa-vmstat.node0.numa_local
893990 ± 10% +22.6% 1096173 ± 6% numa-vmstat.node1.nr_active_anon
479194 ± 3% -48.2% 248075 ± 3% numa-vmstat.node1.nr_slab_unreclaimable
893989 ± 10% +22.6% 1096172 ± 6% numa-vmstat.node1.nr_zone_active_anon
1.614e+08 -64.9% 56589247 numa-vmstat.node1.numa_hit
1.614e+08 -65.0% 56530608 numa-vmstat.node1.numa_local
0.81 +19.0% 0.96 perf-stat.i.MPKI
1.233e+11 -43.5% 6.963e+10 perf-stat.i.branch-instructions
0.09 +0.0 0.10 ± 20% perf-stat.i.branch-miss-rate%
1.032e+08 -40.7% 61185481 perf-stat.i.branch-misses
49.31 +12.7 62.02 perf-stat.i.cache-miss-rate%
4.48e+08 -33.1% 2.995e+08 perf-stat.i.cache-misses
9.107e+08 -47.1% 4.819e+08 perf-stat.i.cache-references
4173 +325.7% 17767 perf-stat.i.context-switches
1.09 +76.8% 1.93 perf-stat.i.cpi
223.77 +92.5% 430.85 ± 2% perf-stat.i.cpu-migrations
1364 +49.0% 2033 perf-stat.i.cycles-between-cache-misses
5.642e+11 -43.9% 3.166e+11 perf-stat.i.instructions
0.92 -43.4% 0.52 perf-stat.i.ipc
0.40 ± 15% -24.0% 0.30 ± 10% perf-stat.i.major-faults
0.79 +19.2% 0.95 perf-stat.overall.MPKI
0.08 +0.0 0.09 perf-stat.overall.branch-miss-rate%
49.22 +12.9 62.11 perf-stat.overall.cache-miss-rate%
1.09 +77.1% 1.92 perf-stat.overall.cpi
1366 +48.6% 2031 perf-stat.overall.cycles-between-cache-misses
0.92 -43.5% 0.52 perf-stat.overall.ipc
4624244 +4.3% 4822893 perf-stat.overall.path-length
1.224e+11 -43.3% 6.934e+10 perf-stat.ps.branch-instructions
1.024e+08 -40.6% 60870902 perf-stat.ps.branch-misses
4.45e+08 -32.9% 2.984e+08 perf-stat.ps.cache-misses
9.044e+08 -46.9% 4.805e+08 perf-stat.ps.cache-references
4123 +328.3% 17663 perf-stat.ps.context-switches
217.00 +95.7% 424.68 ± 2% perf-stat.ps.cpu-migrations
5.601e+11 -43.7% 3.152e+11 perf-stat.ps.instructions
0.37 ± 14% -26.4% 0.27 ± 10% perf-stat.ps.major-faults
1.729e+14 -44.2% 9.647e+13 perf-stat.total.instructions
18753172 +22.2% 22922183 sched_debug.cfs_rq:/.avg_vruntime.avg
18959408 +26.5% 23989376 sched_debug.cfs_rq:/.avg_vruntime.max
14142111 ± 5% +28.4% 18152087 ± 3% sched_debug.cfs_rq:/.avg_vruntime.min
388239 ± 11% +148.2% 963709 ± 5% sched_debug.cfs_rq:/.avg_vruntime.stddev
2.69 ± 4% +40.2% 3.78 ± 5% sched_debug.cfs_rq:/.h_nr_queued.max
0.21 ± 13% +98.9% 0.42 ± 9% sched_debug.cfs_rq:/.h_nr_queued.stddev
2.69 ± 4% +40.2% 3.78 ± 5% sched_debug.cfs_rq:/.h_nr_runnable.max
0.21 ± 13% +93.5% 0.41 ± 11% sched_debug.cfs_rq:/.h_nr_runnable.stddev
494388 ± 81% +273.6% 1846967 ± 26% sched_debug.cfs_rq:/.left_deadline.avg
6717106 ± 48% +253.4% 23740414 sched_debug.cfs_rq:/.left_deadline.max
1440488 ± 63% +294.1% 5677019 ± 11% sched_debug.cfs_rq:/.left_deadline.stddev
494379 ± 81% +273.6% 1846957 ± 26% sched_debug.cfs_rq:/.left_vruntime.avg
6717025 ± 48% +253.4% 23740319 sched_debug.cfs_rq:/.left_vruntime.max
1440465 ± 63% +294.1% 5676990 ± 11% sched_debug.cfs_rq:/.left_vruntime.stddev
33226 ± 65% +2177.9% 756855 ± 38% sched_debug.cfs_rq:/.load.avg
479692 ± 26% +10683.4% 51727275 ± 38% sched_debug.cfs_rq:/.load.max
85011 ± 43% +6510.2% 5619465 ± 38% sched_debug.cfs_rq:/.load.stddev
551.61 ± 21% +53.8% 848.36 ± 15% sched_debug.cfs_rq:/.load_avg.avg
1.44 ± 5% +150.0% 3.61 ± 6% sched_debug.cfs_rq:/.nr_queued.max
0.11 ± 35% +253.3% 0.39 ± 10% sched_debug.cfs_rq:/.nr_queued.stddev
494379 ± 81% +273.6% 1846964 ± 26% sched_debug.cfs_rq:/.right_vruntime.avg
6717047 ± 48% +253.4% 23740386 sched_debug.cfs_rq:/.right_vruntime.max
1440467 ± 63% +294.1% 5677009 ± 11% sched_debug.cfs_rq:/.right_vruntime.stddev
2097 ± 4% +41.0% 2957 ± 6% sched_debug.cfs_rq:/.runnable_avg.max
200.20 ± 10% +66.4% 333.16 ± 7% sched_debug.cfs_rq:/.runnable_avg.stddev
0.23 ± 89% +2760.3% 6.49 ± 45% sched_debug.cfs_rq:/.spread.avg
43.58 ± 89% +626.1% 316.47 ± 29% sched_debug.cfs_rq:/.spread.max
3.14 ± 89% +1031.7% 35.50 ± 32% sched_debug.cfs_rq:/.spread.stddev
701.97 ± 7% -15.1% 596.03 ± 5% sched_debug.cfs_rq:/.util_avg.min
54.04 ± 6% +16.7% 63.09 ± 8% sched_debug.cfs_rq:/.util_avg.stddev
1237 ± 4% +16.0% 1436 ± 8% sched_debug.cfs_rq:/.util_est.max
18741310 +22.3% 22916446 sched_debug.cfs_rq:/.zero_vruntime.avg
18951538 +26.6% 23983182 sched_debug.cfs_rq:/.zero_vruntime.max
14133757 ± 5% +28.4% 18146576 ± 3% sched_debug.cfs_rq:/.zero_vruntime.min
388690 ± 11% +147.8% 963119 ± 5% sched_debug.cfs_rq:/.zero_vruntime.stddev
525.43 ± 2% +275.4% 1972 ± 3% sched_debug.cpu.clock_task.stddev
4548 ± 14% -71.3% 1304 ± 47% sched_debug.cpu.curr->pid.min
962.66 ± 30% +53.3% 1475 ± 13% sched_debug.cpu.curr->pid.stddev
2.75 ± 5% +38.4% 3.81 ± 3% sched_debug.cpu.nr_running.max
0.22 ± 12% +95.7% 0.43 ± 8% sched_debug.cpu.nr_running.stddev
5122 ± 2% +208.6% 15807 sched_debug.cpu.nr_switches.avg
2701 ± 2% +167.1% 7215 ± 3% sched_debug.cpu.nr_switches.min
5212 ± 10% +55.8% 8122 ± 9% sched_debug.cpu.nr_switches.stddev
21.86 -21.9 0.00 perf-profile.calltrace.cycles-pp.__refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes.mas_preallocate
37.63 -20.4 17.26 ± 3% perf-profile.calltrace.cycles-pp.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm
37.62 -20.4 17.25 ± 3% perf-profile.calltrace.cycles-pp.rcu_core.handle_softirqs.run_ksoftirqd.smpboot_thread_fn.kthread
37.61 -20.4 17.25 ± 3% perf-profile.calltrace.cycles-pp.rcu_do_batch.rcu_core.handle_softirqs.run_ksoftirqd.smpboot_thread_fn
37.62 -20.4 17.25 ± 3% perf-profile.calltrace.cycles-pp.handle_softirqs.run_ksoftirqd.smpboot_thread_fn.kthread.ret_from_fork
37.62 -20.4 17.25 ± 3% perf-profile.calltrace.cycles-pp.run_ksoftirqd.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm
37.17 -20.3 16.90 ± 3% perf-profile.calltrace.cycles-pp.rcu_free_sheaf.rcu_do_batch.rcu_core.handle_softirqs.run_ksoftirqd
36.75 -19.3 17.47 ± 3% perf-profile.calltrace.cycles-pp.__kmem_cache_free_bulk.rcu_free_sheaf.rcu_do_batch.rcu_core.handle_softirqs
36.11 -18.9 17.22 ± 3% perf-profile.calltrace.cycles-pp.__slab_free.__kmem_cache_free_bulk.rcu_free_sheaf.rcu_do_batch.rcu_core
35.16 -18.4 16.78 ± 3% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__slab_free.__kmem_cache_free_bulk.rcu_free_sheaf.rcu_do_batch
34.64 -18.1 16.56 ± 3% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__slab_free.__kmem_cache_free_bulk.rcu_free_sheaf
37.63 -14.4 23.18 ± 2% perf-profile.calltrace.cycles-pp.kthread.ret_from_fork.ret_from_fork_asm
37.63 -14.4 23.18 ± 2% perf-profile.calltrace.cycles-pp.ret_from_fork.ret_from_fork_asm
37.63 -14.4 23.18 ± 2% perf-profile.calltrace.cycles-pp.ret_from_fork_asm
10.17 ± 2% -10.2 0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes
10.06 ± 2% -10.1 0.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof
20.47 -8.6 11.91 ± 2% perf-profile.calltrace.cycles-pp.__munmap
19.09 -7.9 11.16 ± 2% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__munmap
19.01 -7.9 11.12 ± 2% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__munmap
18.72 -7.8 10.97 ± 2% perf-profile.calltrace.cycles-pp.__x64_sys_munmap.do_syscall_64.entry_SYSCALL_64_after_hwframe.__munmap
18.71 -7.7 10.96 ± 2% perf-profile.calltrace.cycles-pp.__vm_munmap.__x64_sys_munmap.do_syscall_64.entry_SYSCALL_64_after_hwframe.__munmap
18.14 -7.5 10.65 ± 2% perf-profile.calltrace.cycles-pp.do_vmi_munmap.__vm_munmap.__x64_sys_munmap.do_syscall_64.entry_SYSCALL_64_after_hwframe
17.77 -7.3 10.45 ± 2% perf-profile.calltrace.cycles-pp.do_vmi_align_munmap.do_vmi_munmap.__vm_munmap.__x64_sys_munmap.do_syscall_64
6.70 ± 2% -6.7 0.00 perf-profile.calltrace.cycles-pp.alloc_from_new_slab.__refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes
6.52 ± 2% -6.5 0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.alloc_from_new_slab.__refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof
6.32 ± 2% -6.3 0.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.alloc_from_new_slab.__refill_objects.__pcs_replace_empty_main
10.74 -4.8 5.96 ± 2% perf-profile.calltrace.cycles-pp.vms_complete_munmap_vmas.do_vmi_align_munmap.do_vmi_munmap.__vm_munmap.__x64_sys_munmap
9.30 -4.2 5.14 ± 2% perf-profile.calltrace.cycles-pp.vms_clear_ptes.vms_complete_munmap_vmas.do_vmi_align_munmap.do_vmi_munmap.__vm_munmap
5.88 -2.7 3.19 ± 2% perf-profile.calltrace.cycles-pp.unmap_vmas.vms_clear_ptes.vms_complete_munmap_vmas.do_vmi_align_munmap.do_vmi_munmap
5.70 -2.6 3.09 ± 2% perf-profile.calltrace.cycles-pp.unmap_page_range.unmap_vmas.vms_clear_ptes.vms_complete_munmap_vmas.do_vmi_align_munmap
5.14 -2.4 2.78 ± 2% perf-profile.calltrace.cycles-pp.zap_pmd_range.unmap_page_range.unmap_vmas.vms_clear_ptes.vms_complete_munmap_vmas
2.99 -1.4 1.63 ± 2% perf-profile.calltrace.cycles-pp.perf_event_mmap.__mmap_region.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff
4.39 -1.4 3.04 ± 2% perf-profile.calltrace.cycles-pp.mas_store_gfp.do_vmi_align_munmap.do_vmi_munmap.__vm_munmap.__x64_sys_munmap
2.62 -1.2 1.42 ± 2% perf-profile.calltrace.cycles-pp.perf_event_mmap_event.perf_event_mmap.__mmap_region.do_mmap.vm_mmap_pgoff
2.83 -1.2 1.66 ± 2% perf-profile.calltrace.cycles-pp.free_pgtables.vms_clear_ptes.vms_complete_munmap_vmas.do_vmi_align_munmap.do_vmi_munmap
3.99 -1.1 2.85 ± 2% perf-profile.calltrace.cycles-pp.mas_store_prealloc.__mmap_new_vma.__mmap_region.do_mmap.vm_mmap_pgoff
2.21 -1.0 1.22 ± 2% perf-profile.calltrace.cycles-pp.__cond_resched.zap_pmd_range.unmap_page_range.unmap_vmas.vms_clear_ptes
2.15 -1.0 1.16 ± 2% perf-profile.calltrace.cycles-pp.vms_gather_munmap_vmas.do_vmi_align_munmap.do_vmi_munmap.__vm_munmap.__x64_sys_munmap
3.53 -0.9 2.58 ± 2% perf-profile.calltrace.cycles-pp.mas_wr_node_store.mas_store_prealloc.__mmap_new_vma.__mmap_region.do_mmap
3.48 -0.9 2.54 ± 2% perf-profile.calltrace.cycles-pp.mas_wr_node_store.mas_store_gfp.do_vmi_align_munmap.do_vmi_munmap.__vm_munmap
2.02 -0.9 1.16 ± 2% perf-profile.calltrace.cycles-pp.free_pgd_range.free_pgtables.vms_clear_ptes.vms_complete_munmap_vmas.do_vmi_align_munmap
1.94 -0.8 1.10 ± 2% perf-profile.calltrace.cycles-pp.free_p4d_range.free_pgd_range.free_pgtables.vms_clear_ptes.vms_complete_munmap_vmas
1.96 -0.8 1.18 ± 3% perf-profile.calltrace.cycles-pp.__get_unmapped_area.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64
1.90 -0.8 1.15 ± 3% perf-profile.calltrace.cycles-pp.shmem_get_unmapped_area.__get_unmapped_area.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff
1.77 -0.8 1.02 ± 2% perf-profile.calltrace.cycles-pp.free_pud_range.free_p4d_range.free_pgd_range.free_pgtables.vms_clear_ptes
1.74 -0.7 1.04 ± 3% perf-profile.calltrace.cycles-pp.arch_get_unmapped_area_topdown.shmem_get_unmapped_area.__get_unmapped_area.do_mmap.vm_mmap_pgoff
1.43 -0.6 0.88 ± 3% perf-profile.calltrace.cycles-pp.vm_unmapped_area.arch_get_unmapped_area_topdown.shmem_get_unmapped_area.__get_unmapped_area.do_mmap
1.40 -0.5 0.86 ± 2% perf-profile.calltrace.cycles-pp.unmapped_area_topdown.vm_unmapped_area.arch_get_unmapped_area_topdown.shmem_get_unmapped_area.__get_unmapped_area
0.68 ± 6% -0.4 0.25 ±100% perf-profile.calltrace.cycles-pp.kvfree_call_rcu.mas_wr_node_store.mas_store_gfp.do_vmi_align_munmap.do_vmi_munmap
1.72 -0.4 1.31 ± 2% perf-profile.calltrace.cycles-pp.__pi_memcpy.mas_wr_node_store.mas_store_gfp.do_vmi_align_munmap.do_vmi_munmap
1.73 -0.4 1.32 ± 2% perf-profile.calltrace.cycles-pp.__pi_memcpy.mas_wr_node_store.mas_store_prealloc.__mmap_new_vma.__mmap_region
0.00 +0.8 0.79 ± 12% perf-profile.calltrace.cycles-pp.rcu_free_sheaf.rcu_do_batch.rcu_core.handle_softirqs.__irq_exit_rcu
0.00 +0.8 0.81 ± 12% perf-profile.calltrace.cycles-pp.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.kmem_cache_free_bulk.kvfree_rcu_bulk
0.00 +0.8 0.81 ± 12% perf-profile.calltrace.cycles-pp.handle_softirqs.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.kmem_cache_free_bulk
0.00 +0.8 0.81 ± 12% perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_monitor.process_one_work
0.00 +0.8 0.81 ± 12% perf-profile.calltrace.cycles-pp.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_monitor
0.00 +0.9 0.89 ± 30% perf-profile.calltrace.cycles-pp.rcu_do_batch.rcu_core.handle_softirqs.__irq_exit_rcu.sysvec_apic_timer_interrupt
0.00 +0.9 0.89 ± 30% perf-profile.calltrace.cycles-pp.rcu_core.handle_softirqs.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
0.00 +1.4 1.43 ± 6% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__slab_free.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_work
0.00 +1.5 1.46 ± 6% perf-profile.calltrace.cycles-pp.__slab_free.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_work.process_one_work
0.00 +1.8 1.83 ± 6% perf-profile.calltrace.cycles-pp.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_work.process_one_work.worker_thread
0.00 +1.8 1.83 ± 6% perf-profile.calltrace.cycles-pp.kvfree_rcu_bulk.kfree_rcu_work.process_one_work.worker_thread.kthread
0.00 +1.8 1.84 ± 6% perf-profile.calltrace.cycles-pp.kfree_rcu_work.process_one_work.worker_thread.kthread.ret_from_fork
0.00 +2.6 2.57 ± 17% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__slab_free.__refill_objects_node.__refill_objects_any
0.00 +2.6 2.58 ± 16% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__slab_free.__refill_objects_node.__refill_objects_any.refill_objects
0.00 +2.6 2.58 ± 17% perf-profile.calltrace.cycles-pp.__slab_free.__refill_objects_node.__refill_objects_any.refill_objects.__pcs_replace_empty_main
0.00 +2.6 2.59 ± 10% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__slab_free.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_monitor
0.00 +2.7 2.66 ± 9% perf-profile.calltrace.cycles-pp.__slab_free.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_monitor.process_one_work
0.00 +4.0 4.00 ± 5% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__slab_free.kmem_cache_free_bulk.kvfree_rcu_bulk
0.00 +4.0 4.04 ± 6% perf-profile.calltrace.cycles-pp.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_monitor.process_one_work.worker_thread
0.00 +4.0 4.04 ± 6% perf-profile.calltrace.cycles-pp.kvfree_rcu_bulk.kfree_rcu_monitor.process_one_work.worker_thread.kthread
0.00 +4.1 4.06 ± 6% perf-profile.calltrace.cycles-pp.kfree_rcu_monitor.process_one_work.worker_thread.kthread.ret_from_fork
0.00 +5.5 5.50 ± 7% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__slab_free.__refill_objects_node.refill_objects
0.00 +5.5 5.52 ± 7% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__slab_free.__refill_objects_node.refill_objects.__pcs_replace_empty_main
0.00 +5.6 5.56 ± 7% perf-profile.calltrace.cycles-pp.__slab_free.__refill_objects_node.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof
0.00 +5.9 5.92 ± 3% perf-profile.calltrace.cycles-pp.process_one_work.worker_thread.kthread.ret_from_fork.ret_from_fork_asm
0.00 +5.9 5.92 ± 3% perf-profile.calltrace.cycles-pp.worker_thread.kthread.ret_from_fork.ret_from_fork_asm
0.00 +6.6 6.58 ± 5% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.alloc_from_new_slab.refill_objects.__pcs_replace_empty_main
0.00 +6.7 6.68 ± 5% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.alloc_from_new_slab.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof
0.00 +6.8 6.84 ± 5% perf-profile.calltrace.cycles-pp.alloc_from_new_slab.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes
0.00 +13.3 13.30 ± 3% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__refill_objects_node.__refill_objects_any.refill_objects
0.00 +13.4 13.36 ± 3% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__refill_objects_node.__refill_objects_any.refill_objects.__pcs_replace_empty_main
0.00 +16.9 16.87 ± 5% perf-profile.calltrace.cycles-pp.__refill_objects_node.__refill_objects_any.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof
0.00 +17.1 17.06 ± 5% perf-profile.calltrace.cycles-pp.__refill_objects_any.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes
0.00 +20.8 20.84 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__refill_objects_node.refill_objects.__pcs_replace_empty_main
0.00 +21.0 20.96 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__refill_objects_node.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof
41.48 +23.1 64.56 perf-profile.calltrace.cycles-pp.__mmap
40.03 +23.7 63.74 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__mmap
39.95 +23.7 63.69 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__mmap
39.57 +23.9 63.49 perf-profile.calltrace.cycles-pp.ksys_mmap_pgoff.do_syscall_64.entry_SYSCALL_64_after_hwframe.__mmap
39.12 +24.1 63.24 perf-profile.calltrace.cycles-pp.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64.entry_SYSCALL_64_after_hwframe.__mmap
38.64 +24.3 62.97 perf-profile.calltrace.cycles-pp.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64.entry_SYSCALL_64_after_hwframe
36.37 +25.2 61.60 perf-profile.calltrace.cycles-pp.__mmap_region.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64
29.49 +28.3 57.78 perf-profile.calltrace.cycles-pp.__mmap_new_vma.__mmap_region.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff
0.00 +28.5 28.47 ± 2% perf-profile.calltrace.cycles-pp.__refill_objects_node.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes
23.95 +30.1 54.08 perf-profile.calltrace.cycles-pp.mas_preallocate.__mmap_new_vma.__mmap_region.do_mmap.vm_mmap_pgoff
23.34 ± 2% +30.4 53.74 ± 2% perf-profile.calltrace.cycles-pp.mas_alloc_nodes.mas_preallocate.__mmap_new_vma.__mmap_region.do_mmap
23.31 ± 2% +30.4 53.72 ± 2% perf-profile.calltrace.cycles-pp.kmem_cache_alloc_noprof.mas_alloc_nodes.mas_preallocate.__mmap_new_vma.__mmap_region
22.88 ± 2% +30.6 53.51 ± 2% perf-profile.calltrace.cycles-pp.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes.mas_preallocate.__mmap_new_vma
0.00 +52.8 52.82 ± 2% perf-profile.calltrace.cycles-pp.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes.mas_preallocate
22.07 -22.1 0.00 perf-profile.children.cycles-pp.__refill_objects
37.63 -20.4 17.26 ± 3% perf-profile.children.cycles-pp.smpboot_thread_fn
37.62 -20.4 17.25 ± 3% perf-profile.children.cycles-pp.run_ksoftirqd
38.56 -18.4 20.16 ± 2% perf-profile.children.cycles-pp.rcu_core
38.56 -18.4 20.16 ± 2% perf-profile.children.cycles-pp.rcu_do_batch
38.57 -18.4 20.17 ± 2% perf-profile.children.cycles-pp.handle_softirqs
38.04 -18.3 19.70 ± 2% perf-profile.children.cycles-pp.rcu_free_sheaf
37.64 -18.2 19.46 ± 2% perf-profile.children.cycles-pp.__kmem_cache_free_bulk
37.63 -14.4 23.18 ± 2% perf-profile.children.cycles-pp.ret_from_fork
37.63 -14.4 23.18 ± 2% perf-profile.children.cycles-pp.ret_from_fork_asm
37.63 -14.4 23.18 ± 2% perf-profile.children.cycles-pp.kthread
20.82 -8.7 12.10 ± 2% perf-profile.children.cycles-pp.__munmap
18.72 -7.8 10.97 ± 2% perf-profile.children.cycles-pp.__x64_sys_munmap
18.71 -7.7 10.96 ± 2% perf-profile.children.cycles-pp.__vm_munmap
18.14 -7.5 10.65 ± 2% perf-profile.children.cycles-pp.do_vmi_munmap
17.78 -7.3 10.46 ± 2% perf-profile.children.cycles-pp.do_vmi_align_munmap
38.64 -7.2 31.46 perf-profile.children.cycles-pp.__slab_free
10.77 -4.8 5.98 ± 2% perf-profile.children.cycles-pp.vms_complete_munmap_vmas
9.33 -4.2 5.14 ± 2% perf-profile.children.cycles-pp.vms_clear_ptes
5.89 -2.7 3.19 ± 2% perf-profile.children.cycles-pp.unmap_vmas
5.70 -2.6 3.09 ± 2% perf-profile.children.cycles-pp.unmap_page_range
5.53 -2.5 3.00 ± 2% perf-profile.children.cycles-pp.zap_pmd_range
7.03 -1.9 5.14 ± 2% perf-profile.children.cycles-pp.mas_wr_node_store
4.93 -1.6 3.33 ± 2% perf-profile.children.cycles-pp.mas_store_gfp
3.00 -1.4 1.64 ± 2% perf-profile.children.cycles-pp.perf_event_mmap
2.64 -1.2 1.43 ± 2% perf-profile.children.cycles-pp.perf_event_mmap_event
2.90 -1.2 1.70 ± 2% perf-profile.children.cycles-pp.free_pgtables
2.62 -1.2 1.44 ± 2% perf-profile.children.cycles-pp.__cond_resched
3.99 -1.1 2.85 ± 2% perf-profile.children.cycles-pp.mas_store_prealloc
2.16 -1.0 1.19 ± 2% perf-profile.children.cycles-pp.vms_gather_munmap_vmas
2.02 -0.8 1.17 ± 2% perf-profile.children.cycles-pp.free_pgd_range
3.48 -0.8 2.64 ± 2% perf-profile.children.cycles-pp.__pi_memcpy
1.95 -0.8 1.11 ± 2% perf-profile.children.cycles-pp.free_p4d_range
1.96 -0.8 1.18 ± 3% perf-profile.children.cycles-pp.__get_unmapped_area
1.78 -0.8 1.02 ± 2% perf-profile.children.cycles-pp.free_pud_range
1.90 -0.8 1.15 ± 3% perf-profile.children.cycles-pp.shmem_get_unmapped_area
1.75 -0.7 1.05 ± 2% perf-profile.children.cycles-pp.arch_get_unmapped_area_topdown
1.47 -0.6 0.87 ± 2% perf-profile.children.cycles-pp.mas_find
0.88 ± 3% -0.6 0.29 ± 5% perf-profile.children.cycles-pp.allocate_slab
1.26 -0.6 0.70 ± 3% perf-profile.children.cycles-pp.entry_SYSCALL_64
1.43 -0.6 0.88 ± 3% perf-profile.children.cycles-pp.vm_unmapped_area
1.41 -0.5 0.87 ± 3% perf-profile.children.cycles-pp.unmapped_area_topdown
1.03 -0.5 0.56 ± 2% perf-profile.children.cycles-pp.entry_SYSRETQ_unsafe_stack
0.90 -0.4 0.48 ± 3% perf-profile.children.cycles-pp.vm_area_alloc
0.62 ± 4% -0.4 0.20 ± 6% perf-profile.children.cycles-pp.shuffle_freelist
0.78 -0.4 0.38 ± 3% perf-profile.children.cycles-pp.zap_pte_range
0.84 -0.4 0.46 ± 3% perf-profile.children.cycles-pp.rcu_all_qs
0.71 -0.3 0.36 ± 2% perf-profile.children.cycles-pp.mas_prev_slot
0.48 ± 3% -0.3 0.16 ± 6% perf-profile.children.cycles-pp.setup_object
0.84 ± 5% -0.3 0.53 ± 5% perf-profile.children.cycles-pp.mas_update_gap
0.77 ± 5% -0.3 0.49 ± 4% perf-profile.children.cycles-pp.mas_leaf_max_gap
0.60 -0.3 0.33 ± 3% perf-profile.children.cycles-pp.d_path
0.62 -0.3 0.34 ± 2% perf-profile.children.cycles-pp.__build_id_parse
0.66 -0.3 0.40 ± 2% perf-profile.children.cycles-pp.mas_walk
0.51 -0.2 0.27 ± 2% perf-profile.children.cycles-pp.perf_iterate_sb
0.52 -0.2 0.30 ± 3% perf-profile.children.cycles-pp.mas_next_slot
0.55 ± 2% -0.2 0.34 ± 2% perf-profile.children.cycles-pp.kmem_cache_free
0.50 -0.2 0.29 ± 3% perf-profile.children.cycles-pp.mas_wr_store_type
0.65 -0.2 0.45 ± 3% perf-profile.children.cycles-pp.mas_empty_area_rev
0.79 ± 10% -0.2 0.60 ± 3% perf-profile.children.cycles-pp.__kfree_rcu_sheaf
0.43 -0.2 0.24 ± 3% perf-profile.children.cycles-pp.prepend_path
0.49 -0.2 0.30 ± 3% perf-profile.children.cycles-pp.__vma_start_write
0.43 -0.2 0.24 ± 2% perf-profile.children.cycles-pp.up_write
0.43 -0.2 0.24 ± 3% perf-profile.children.cycles-pp.unlink_file_vma_batch_process
0.41 -0.2 0.23 perf-profile.children.cycles-pp.shmem_mmap_prepare
0.40 -0.2 0.22 ± 3% perf-profile.children.cycles-pp.__rcu_free_sheaf_prepare
0.37 ± 3% -0.2 0.19 ± 3% perf-profile.children.cycles-pp.__memcg_slab_post_alloc_hook
0.38 -0.2 0.21 ± 3% perf-profile.children.cycles-pp.arch_exit_to_user_mode_prepare
0.36 ± 6% -0.2 0.19 ± 3% perf-profile.children.cycles-pp.__memcg_slab_free_hook
0.38 -0.2 0.21 ± 3% perf-profile.children.cycles-pp.touch_atime
0.35 -0.2 0.19 ± 3% perf-profile.children.cycles-pp.__pte_offset_map_lock
0.27 -0.2 0.11 ± 6% perf-profile.children.cycles-pp.__alloc_empty_sheaf
0.25 ± 2% -0.2 0.09 ± 4% perf-profile.children.cycles-pp.__alloc_frozen_pages_noprof
0.26 -0.2 0.10 ± 4% perf-profile.children.cycles-pp.__kmalloc_noprof
0.35 -0.2 0.20 ± 2% perf-profile.children.cycles-pp.down_write
0.33 -0.1 0.18 ± 3% perf-profile.children.cycles-pp.down_write_killable
0.32 -0.1 0.18 ± 2% perf-profile.children.cycles-pp.atime_needs_update
0.34 ± 2% -0.1 0.20 ± 3% perf-profile.children.cycles-pp.__vma_enter_locked
0.26 -0.1 0.12 ± 4% perf-profile.children.cycles-pp.security_vm_enough_memory_mm
0.22 ± 3% -0.1 0.08 ± 8% perf-profile.children.cycles-pp.get_page_from_freelist
0.28 -0.1 0.15 ± 2% perf-profile.children.cycles-pp.fget
0.46 -0.1 0.34 ± 3% perf-profile.children.cycles-pp.mas_rev_awalk
0.28 -0.1 0.16 ± 3% perf-profile.children.cycles-pp.fput
0.29 -0.1 0.17 ± 3% perf-profile.children.cycles-pp.freader_fetch
0.26 -0.1 0.14 ± 2% perf-profile.children.cycles-pp.up_read
0.32 -0.1 0.21 ± 4% perf-profile.children.cycles-pp.kfree
0.22 -0.1 0.11 ± 3% perf-profile.children.cycles-pp.remove_vma
0.14 ± 3% -0.1 0.04 ± 44% perf-profile.children.cycles-pp.rmqueue
0.22 -0.1 0.12 ± 4% perf-profile.children.cycles-pp.prepend_copy
0.22 ± 2% -0.1 0.13 ± 2% perf-profile.children.cycles-pp.tlb_finish_mmu
0.20 -0.1 0.11 ± 3% perf-profile.children.cycles-pp.freader_init_from_file
0.21 -0.1 0.12 ± 3% perf-profile.children.cycles-pp.__kmalloc_cache_noprof
0.17 ± 13% -0.1 0.08 ± 19% perf-profile.children.cycles-pp.strlen
0.18 ± 2% -0.1 0.10 ± 3% perf-profile.children.cycles-pp.tlb_gather_mmu
0.18 ± 2% -0.1 0.10 ± 3% perf-profile.children.cycles-pp.copy_from_kernel_nofault
0.14 ± 5% -0.1 0.06 ± 6% perf-profile.children.cycles-pp.syscall_return_via_sysret
0.18 ± 2% -0.1 0.11 ± 3% perf-profile.children.cycles-pp.current_time
0.17 ± 2% -0.1 0.10 ± 3% perf-profile.children.cycles-pp.entry_SYSCALL_64_safe_stack
0.37 ± 2% -0.1 0.30 ± 4% perf-profile.children.cycles-pp.build_detached_freelist
0.12 ± 4% -0.1 0.04 ± 44% perf-profile.children.cycles-pp.__rmqueue_pcplist
0.18 ± 2% -0.1 0.11 ± 3% perf-profile.children.cycles-pp.downgrade_write
0.15 -0.1 0.08 ± 4% perf-profile.children.cycles-pp.__filemap_get_folio_mpol
0.15 -0.1 0.08 ± 4% perf-profile.children.cycles-pp.__vm_enough_memory
0.15 ± 3% -0.1 0.09 ± 4% perf-profile.children.cycles-pp.mas_prev
0.14 ± 3% -0.1 0.07 ± 6% perf-profile.children.cycles-pp.vma_mark_detached
0.14 -0.1 0.08 ± 4% perf-profile.children.cycles-pp.vma_merge_new_range
0.18 ± 2% -0.1 0.12 perf-profile.children.cycles-pp.hrtimer_interrupt
0.19 ± 2% -0.1 0.13 ± 2% perf-profile.children.cycles-pp.__sysvec_apic_timer_interrupt
0.13 ± 2% -0.1 0.08 ± 6% perf-profile.children.cycles-pp.mas_wr_store_entry
0.13 ± 2% -0.1 0.08 ± 6% perf-profile.children.cycles-pp.testcase
0.11 -0.1 0.06 ± 6% perf-profile.children.cycles-pp.percpu_counter_add_batch
0.11 -0.1 0.06 perf-profile.children.cycles-pp.may_expand_vm
0.18 ± 4% -0.0 0.14 ± 7% perf-profile.children.cycles-pp.__free_frozen_pages
0.18 ± 2% -0.0 0.14 ± 4% perf-profile.children.cycles-pp.freader_get_folio
0.10 -0.0 0.06 ± 8% perf-profile.children.cycles-pp.mas_data_end
0.10 ± 3% -0.0 0.06 ± 6% perf-profile.children.cycles-pp.perf_event_mmap_output
0.09 ± 4% -0.0 0.05 perf-profile.children.cycles-pp.filemap_get_entry
0.07 ± 6% -0.0 0.03 ± 70% perf-profile.children.cycles-pp.free_pcppages_bulk
0.08 ± 5% -0.0 0.05 perf-profile.children.cycles-pp.mas_prev_setup
0.11 ± 3% -0.0 0.08 ± 4% perf-profile.children.cycles-pp.ktime_get_coarse_real_ts64_mg
0.12 ± 3% -0.0 0.09 ± 4% perf-profile.children.cycles-pp.__hrtimer_run_queues
0.11 -0.0 0.08 perf-profile.children.cycles-pp.tick_nohz_handler
0.10 ± 3% -0.0 0.07 perf-profile.children.cycles-pp.update_process_times
0.08 -0.0 0.06 ± 9% perf-profile.children.cycles-pp.free_frozen_page_commit
0.10 ± 5% -0.0 0.08 ± 10% perf-profile.children.cycles-pp._raw_spin_trylock
0.19 ± 3% +0.0 0.21 ± 3% perf-profile.children.cycles-pp.perf_session__process_events
0.19 ± 3% +0.0 0.21 ± 3% perf-profile.children.cycles-pp.reader__read_event
0.19 ± 3% +0.0 0.21 ± 3% perf-profile.children.cycles-pp.record__finish_output
0.00 +0.1 0.05 ± 7% perf-profile.children.cycles-pp.get_state_synchronize_rcu_full
0.14 ± 2% +0.1 0.23 ± 3% perf-profile.children.cycles-pp._raw_spin_lock
0.00 +1.8 1.84 ± 6% perf-profile.children.cycles-pp.kfree_rcu_work
1.08 ± 2% +2.0 3.04 ± 3% perf-profile.children.cycles-pp.sysvec_apic_timer_interrupt
1.10 ± 2% +2.0 3.06 ± 3% perf-profile.children.cycles-pp.asm_sysvec_apic_timer_interrupt
0.89 ± 3% +2.0 2.92 ± 4% perf-profile.children.cycles-pp.__irq_exit_rcu
0.00 +4.1 4.06 ± 6% perf-profile.children.cycles-pp.kfree_rcu_monitor
0.00 +5.9 5.86 ± 4% perf-profile.children.cycles-pp.kmem_cache_free_bulk
0.00 +5.9 5.86 ± 4% perf-profile.children.cycles-pp.kvfree_rcu_bulk
0.00 +5.9 5.92 ± 3% perf-profile.children.cycles-pp.process_one_work
0.00 +5.9 5.92 ± 3% perf-profile.children.cycles-pp.worker_thread
59.20 +15.8 74.96 perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
59.04 +15.8 74.88 perf-profile.children.cycles-pp.do_syscall_64
0.00 +17.1 17.06 ± 5% perf-profile.children.cycles-pp.__refill_objects_any
55.81 +17.2 73.06 perf-profile.children.cycles-pp._raw_spin_lock_irqsave
54.93 +17.5 72.46 perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath
41.80 +22.9 64.73 perf-profile.children.cycles-pp.__mmap
39.57 +23.9 63.50 perf-profile.children.cycles-pp.ksys_mmap_pgoff
39.12 +24.1 63.25 perf-profile.children.cycles-pp.vm_mmap_pgoff
38.65 +24.3 62.98 perf-profile.children.cycles-pp.do_mmap
36.41 +25.2 61.62 perf-profile.children.cycles-pp.__mmap_region
29.51 +28.3 57.79 perf-profile.children.cycles-pp.__mmap_new_vma
24.05 +30.1 54.12 perf-profile.children.cycles-pp.kmem_cache_alloc_noprof
23.96 +30.1 54.08 perf-profile.children.cycles-pp.mas_preallocate
23.34 ± 2% +30.4 53.74 ± 2% perf-profile.children.cycles-pp.mas_alloc_nodes
23.09 ± 2% +30.5 53.60 ± 2% perf-profile.children.cycles-pp.__pcs_replace_empty_main
0.00 +45.4 45.43 ± 3% perf-profile.children.cycles-pp.__refill_objects_node
0.00 +52.9 52.90 ± 2% perf-profile.children.cycles-pp.refill_objects
2.70 -1.2 1.48 ± 2% perf-profile.self.cycles-pp.zap_pmd_range
3.40 -0.8 2.57 ± 2% perf-profile.self.cycles-pp.__pi_memcpy
1.75 -0.8 0.96 ± 2% perf-profile.self.cycles-pp.__mmap_region
1.75 -0.7 1.00 ± 2% perf-profile.self.cycles-pp.free_pud_range
1.48 -0.7 0.81 ± 2% perf-profile.self.cycles-pp.__cond_resched
1.41 -0.6 0.78 ± 3% perf-profile.self.cycles-pp.mas_wr_node_store
1.02 -0.5 0.56 ± 2% perf-profile.self.cycles-pp.entry_SYSRETQ_unsafe_stack
1.02 -0.4 0.58 ± 3% perf-profile.self.cycles-pp.__slab_free
0.88 -0.4 0.47 ± 3% perf-profile.self.cycles-pp.mas_store_gfp
0.56 ± 4% -0.4 0.18 ± 5% perf-profile.self.cycles-pp.shuffle_freelist
0.68 -0.3 0.35 ± 2% perf-profile.self.cycles-pp.mas_prev_slot
0.72 -0.3 0.40 ± 3% perf-profile.self.cycles-pp.rcu_all_qs
0.65 -0.3 0.36 ± 3% perf-profile.self.cycles-pp.entry_SYSCALL_64
0.75 ± 6% -0.3 0.47 ± 5% perf-profile.self.cycles-pp.mas_leaf_max_gap
0.64 -0.3 0.38 ± 3% perf-profile.self.cycles-pp.mas_walk
0.85 ± 2% -0.3 0.60 ± 4% perf-profile.self.cycles-pp._raw_spin_lock_irqsave
0.54 -0.2 0.30 ± 3% perf-profile.self.cycles-pp.kmem_cache_alloc_noprof
0.51 -0.2 0.28 ± 2% perf-profile.self.cycles-pp.mas_next_slot
0.48 -0.2 0.26 ± 3% perf-profile.self.cycles-pp.do_vmi_align_munmap
0.49 -0.2 0.28 ± 3% perf-profile.self.cycles-pp.mas_wr_store_type
0.44 -0.2 0.24 ± 3% perf-profile.self.cycles-pp.__munmap
0.48 -0.2 0.28 ± 3% perf-profile.self.cycles-pp.__mmap
0.42 -0.2 0.23 ± 2% perf-profile.self.cycles-pp.up_write
0.40 -0.2 0.22 ± 3% perf-profile.self.cycles-pp.unmapped_area_topdown
0.39 -0.2 0.21 ± 2% perf-profile.self.cycles-pp.perf_iterate_sb
0.40 -0.2 0.22 ± 3% perf-profile.self.cycles-pp.__rcu_free_sheaf_prepare
0.41 -0.2 0.23 ± 2% perf-profile.self.cycles-pp.mas_store_prealloc
0.38 -0.2 0.21 ± 4% perf-profile.self.cycles-pp.arch_exit_to_user_mode_prepare
0.38 -0.2 0.21 ± 2% perf-profile.self.cycles-pp.mas_preallocate
0.38 -0.2 0.21 ± 4% perf-profile.self.cycles-pp.__vm_munmap
0.37 -0.2 0.20 ± 3% perf-profile.self.cycles-pp.perf_event_mmap
0.38 -0.2 0.22 perf-profile.self.cycles-pp.mas_find
0.34 -0.2 0.19 ± 3% perf-profile.self.cycles-pp.vm_area_alloc
0.30 -0.2 0.15 ± 3% perf-profile.self.cycles-pp.zap_pte_range
0.28 -0.1 0.14 ± 4% perf-profile.self.cycles-pp.__kfree_rcu_sheaf
0.31 -0.1 0.18 ± 4% perf-profile.self.cycles-pp.__vma_enter_locked
0.30 -0.1 0.17 ± 3% perf-profile.self.cycles-pp.down_write
0.28 -0.1 0.15 ± 3% perf-profile.self.cycles-pp.down_write_killable
0.27 -0.1 0.14 ± 5% perf-profile.self.cycles-pp.__memcg_slab_post_alloc_hook
0.28 -0.1 0.16 ± 3% perf-profile.self.cycles-pp.__mmap_new_vma
0.27 -0.1 0.14 ± 3% perf-profile.self.cycles-pp.fput
0.27 -0.1 0.15 ± 2% perf-profile.self.cycles-pp.prepend_path
0.27 -0.1 0.15 ± 2% perf-profile.self.cycles-pp.arch_get_unmapped_area_topdown
0.26 ± 9% -0.1 0.14 ± 2% perf-profile.self.cycles-pp.__memcg_slab_free_hook
0.29 -0.1 0.17 ± 5% perf-profile.self.cycles-pp.kfree
0.27 -0.1 0.15 ± 3% perf-profile.self.cycles-pp.fget
0.37 ± 2% -0.1 0.25 ± 4% perf-profile.self.cycles-pp.build_detached_freelist
0.25 -0.1 0.14 ± 3% perf-profile.self.cycles-pp.up_read
0.26 ± 8% -0.1 0.15 ± 10% perf-profile.self.cycles-pp.perf_event_mmap_event
0.23 -0.1 0.13 ± 4% perf-profile.self.cycles-pp.vms_gather_munmap_vmas
0.49 -0.1 0.39 ± 2% perf-profile.self.cycles-pp.kvfree_call_rcu
0.22 -0.1 0.12 ± 3% perf-profile.self.cycles-pp.tlb_finish_mmu
0.17 ± 2% -0.1 0.08 perf-profile.self.cycles-pp.free_p4d_range
0.19 ± 2% -0.1 0.11 ± 3% perf-profile.self.cycles-pp.freader_init_from_file
0.36 -0.1 0.28 ± 4% perf-profile.self.cycles-pp.mas_rev_awalk
0.18 ± 2% -0.1 0.10 ± 5% perf-profile.self.cycles-pp.do_syscall_64
0.19 ± 2% -0.1 0.10 ± 4% perf-profile.self.cycles-pp.free_pgtables
0.16 ± 11% -0.1 0.08 ± 17% perf-profile.self.cycles-pp.strlen
0.18 -0.1 0.10 ± 3% perf-profile.self.cycles-pp.tlb_gather_mmu
0.14 ± 5% -0.1 0.06 ± 6% perf-profile.self.cycles-pp.syscall_return_via_sysret
0.19 -0.1 0.11 ± 4% perf-profile.self.cycles-pp.__kmalloc_cache_noprof
0.17 ± 2% -0.1 0.10 ± 7% perf-profile.self.cycles-pp.do_mmap
0.18 -0.1 0.11 ± 4% perf-profile.self.cycles-pp.downgrade_write
0.13 -0.1 0.06 ± 6% perf-profile.self.cycles-pp.atime_needs_update
0.11 ± 3% -0.1 0.04 ± 44% perf-profile.self.cycles-pp.alloc_from_new_slab
0.19 -0.1 0.12 ± 3% perf-profile.self.cycles-pp.mas_empty_area_rev
0.16 ± 3% -0.1 0.10 ± 3% perf-profile.self.cycles-pp.__pte_offset_map_lock
0.14 ± 3% -0.1 0.08 ± 4% perf-profile.self.cycles-pp.unmap_page_range
0.12 ± 4% -0.1 0.06 ± 7% perf-profile.self.cycles-pp.__build_id_parse
0.15 -0.1 0.09 ± 4% perf-profile.self.cycles-pp.__vma_start_write
0.13 ± 3% -0.1 0.07 ± 6% perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe
0.13 ± 3% -0.1 0.07 ± 6% perf-profile.self.cycles-pp.entry_SYSCALL_64_safe_stack
0.13 -0.1 0.07 ± 5% perf-profile.self.cycles-pp.vma_mark_detached
0.13 ± 2% -0.1 0.07 ± 5% perf-profile.self.cycles-pp.__pcs_replace_empty_main
0.08 -0.1 0.02 ± 99% perf-profile.self.cycles-pp.testcase
0.12 -0.1 0.07 ± 5% perf-profile.self.cycles-pp.copy_from_kernel_nofault
0.14 ± 4% -0.1 0.09 ± 12% perf-profile.self.cycles-pp.shmem_get_unmapped_area
0.10 -0.0 0.05 ± 7% perf-profile.self.cycles-pp.security_vm_enough_memory_mm
0.09 -0.0 0.04 ± 44% perf-profile.self.cycles-pp.unmap_vmas
0.19 ± 5% -0.0 0.14 ± 4% perf-profile.self.cycles-pp.kmem_cache_free
0.10 -0.0 0.05 ± 8% perf-profile.self.cycles-pp.vma_merge_new_range
0.10 ± 4% -0.0 0.06 ± 6% perf-profile.self.cycles-pp.percpu_counter_add_batch
0.11 -0.0 0.06 ± 7% perf-profile.self.cycles-pp.mas_wr_store_entry
0.10 ± 5% -0.0 0.05 ± 7% perf-profile.self.cycles-pp.mas_data_end
0.08 ± 5% -0.0 0.04 ± 44% perf-profile.self.cycles-pp.mas_prev
0.10 ± 4% -0.0 0.06 perf-profile.self.cycles-pp.may_expand_vm
0.10 ± 3% -0.0 0.06 ± 6% perf-profile.self.cycles-pp.perf_event_mmap_output
0.09 -0.0 0.05 perf-profile.self.cycles-pp.vms_complete_munmap_vmas
0.10 ± 4% -0.0 0.07 ± 7% perf-profile.self.cycles-pp.vm_mmap_pgoff
0.11 ± 3% -0.0 0.08 ± 4% perf-profile.self.cycles-pp.ktime_get_coarse_real_ts64_mg
0.09 ± 5% -0.0 0.06 ± 11% perf-profile.self.cycles-pp._raw_spin_trylock
0.00 +0.1 0.05 ± 7% perf-profile.self.cycles-pp.get_state_synchronize_rcu_full
0.00 +0.1 0.06 ± 6% perf-profile.self.cycles-pp.freader_get_folio
0.14 ± 2% +0.1 0.22 ± 3% perf-profile.self.cycles-pp._raw_spin_lock
0.00 +0.2 0.16 ± 6% perf-profile.self.cycles-pp.refill_objects
0.00 +0.2 0.18 ± 5% perf-profile.self.cycles-pp.__refill_objects_any
0.00 +2.3 2.29 perf-profile.self.cycles-pp.__refill_objects_node
54.93 +17.5 72.46 perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath
***************************************************************************************************
lkp-srf-2sp3: 192 threads 2 sockets Intel(R) Xeon(R) 6740E CPU @ 2.4GHz (Sierra Forest) with 256G memory
=========================================================================================
compiler/cpufreq_governor/kconfig/nr_threads/rootfs/tbox_group/test/testcase/testtime:
gcc-14/performance/x86_64-rhel-9.4/100%/debian-13-x86_64-20250902.cgz/lkp-srf-2sp3/pkey/stress-ng/60s
commit:
6a67958ab0 ("slab: remove unused PREEMPT_RT specific macros")
aa8fdb9e25 ("slab: refill sheaves from all nodes")
6a67958ab000c3a7 aa8fdb9e2516055552de11cabaa
---------------- ---------------------------
%stddev %change %stddev
\ | \
4.001e+08 ± 6% +54.0% 6.163e+08 ± 11% cpuidle..time
156947 ± 16% +118.4% 342823 ± 9% cpuidle..usage
4331943 ± 61% +160.2% 11272469 ± 34% numa-meminfo.node0.FilePages
870538 ± 7% -18.9% 705628 ± 4% numa-meminfo.node1.SUnreclaim
979599 ± 7% -17.1% 812545 ± 4% numa-meminfo.node1.Slab
27817379 ± 5% +12.3% 31236755 vmstat.memory.cache
10123 ± 6% +130.8% 23360 vmstat.system.cs
475556 -2.9% 461553 vmstat.system.in
17747754 ± 11% -33.0% 11895491 ± 15% numa-numastat.node0.local_node
17831118 ± 10% -32.8% 11986853 ± 15% numa-numastat.node0.numa_hit
29067129 ± 2% -38.2% 17954639 ± 9% numa-numastat.node1.local_node
29182842 ± 2% -38.1% 18062464 ± 10% numa-numastat.node1.numa_hit
2.45 ± 9% +1.7 4.14 ± 15% mpstat.cpu.all.idle%
0.30 -0.1 0.23 ± 2% mpstat.cpu.all.irq%
19.77 ± 4% -5.9 13.83 ± 3% mpstat.cpu.all.soft%
66.39 +7.3 73.70 mpstat.cpu.all.sys%
11.09 -3.0 8.09 mpstat.cpu.all.usr%
1083627 ± 61% +160.2% 2820009 ± 34% numa-vmstat.node0.nr_file_pages
17839136 ± 10% -32.8% 11986742 ± 15% numa-vmstat.node0.numa_hit
17755627 ± 11% -33.0% 11895381 ± 15% numa-vmstat.node0.numa_local
209983 ± 7% -15.3% 177924 ± 4% numa-vmstat.node1.nr_slab_unreclaimable
29194095 ± 2% -38.1% 18061832 ± 10% numa-vmstat.node1.numa_hit
29078573 ± 2% -38.3% 17954008 ± 9% numa-vmstat.node1.numa_local
1.178e+09 -28.4% 8.427e+08 stress-ng.pkey.ops
19634814 -28.4% 14051726 stress-ng.pkey.ops_per_sec
265455 ± 8% +120.3% 584676 stress-ng.time.involuntary_context_switches
14980 +1.8% 15243 stress-ng.time.percent_of_cpu_this_job_got
7757 +6.7% 8279 stress-ng.time.system_time
1245 -29.2% 882.14 stress-ng.time.user_time
167683 ± 3% +37.8% 231015 ± 8% sched_debug.cfs_rq:/.avg_vruntime.stddev
112.17 ± 91% -85.7% 16.00 ±148% sched_debug.cfs_rq:/.runnable_avg.min
224.70 ± 6% +25.0% 280.94 ± 13% sched_debug.cfs_rq:/.runnable_avg.stddev
162.74 ± 8% +14.1% 185.74 ± 12% sched_debug.cfs_rq:/.util_avg.stddev
167731 ± 3% +37.6% 230810 ± 8% sched_debug.cfs_rq:/.zero_vruntime.stddev
3513 ± 10% +58.0% 5551 ± 3% sched_debug.cpu.nr_switches.avg
1611 ± 4% +51.4% 2440 ± 6% sched_debug.cpu.nr_switches.min
24416569 ± 6% +14.6% 27972054 meminfo.Active
24416553 ± 6% +14.6% 27972037 meminfo.Active(anon)
948184 ± 2% +10.2% 1044872 ± 3% meminfo.AnonPages
27174415 ± 5% +12.7% 30633367 meminfo.Cached
26095784 ± 5% +13.6% 29638241 meminfo.Committed_AS
726034 ± 8% +29.8% 942096 ± 8% meminfo.Mapped
32569469 ± 4% +10.6% 36020482 meminfo.Memused
23472099 ± 6% +14.7% 26931049 meminfo.Shmem
3.25 ± 6% +1.8 5.09 ± 12% turbostat.C1%
3.19 ± 6% +54.8% 4.94 ± 12% turbostat.CPU%c1
1.38 -25.7% 1.02 turbostat.IPC
469589 ± 11% -4.7e+05 0.00 turbostat.PKG_%
60.67 -2.7% 59.00 turbostat.PkgTmp
483.34 -9.9% 435.70 turbostat.PkgWatt
23.10 -3.8% 22.22 turbostat.RAMWatt
0.04 -25.0% 0.03 turbostat.SysWatt
6088135 ± 6% +14.7% 6981165 proc-vmstat.nr_active_anon
236856 ± 2% +10.2% 260993 ± 2% proc-vmstat.nr_anon_pages
5640903 -1.5% 5554633 proc-vmstat.nr_dirty_background_threshold
11295599 -1.5% 11122847 proc-vmstat.nr_dirty_threshold
6777790 ± 5% +12.8% 7646722 proc-vmstat.nr_file_pages
56774242 -1.5% 55910274 proc-vmstat.nr_free_pages
35659 +4.8% 37370 proc-vmstat.nr_kernel_stack
180794 ± 9% +29.8% 234610 ± 8% proc-vmstat.nr_mapped
5852212 ± 6% +14.8% 6721144 proc-vmstat.nr_shmem
47795 +4.0% 49717 proc-vmstat.nr_slab_reclaimable
6088134 ± 6% +14.7% 6981165 proc-vmstat.nr_zone_active_anon
47016156 ± 3% -36.1% 30053690 ± 3% proc-vmstat.numa_hit
46817079 ± 3% -36.2% 29854503 ± 4% proc-vmstat.numa_local
1.463e+08 ± 4% -50.3% 72679091 ± 6% proc-vmstat.pgalloc_normal
1.325e+08 ± 4% -56.7% 57326779 ± 7% proc-vmstat.pgfree
0.44 +18.3% 0.52 perf-stat.i.MPKI
1.653e+11 -26.6% 1.213e+11 perf-stat.i.branch-instructions
0.10 +0.0 0.11 perf-stat.i.branch-miss-rate%
1.702e+08 -24.0% 1.293e+08 perf-stat.i.branch-misses
53.44 +7.3 60.79 perf-stat.i.cache-miss-rate%
3.563e+08 -13.8% 3.073e+08 perf-stat.i.cache-misses
6.643e+08 -24.4% 5.019e+08 perf-stat.i.cache-references
9933 ± 7% +137.5% 23593 perf-stat.i.context-switches
0.73 +35.4% 0.99 perf-stat.i.cpi
5.999e+11 -1.7% 5.899e+11 perf-stat.i.cpu-cycles
532.61 ± 3% +115.9% 1150 ± 2% perf-stat.i.cpu-migrations
1714 +16.6% 1999 perf-stat.i.cycles-between-cache-misses
8.274e+11 -26.8% 6.054e+11 perf-stat.i.instructions
1.38 -25.6% 1.02 perf-stat.i.ipc
0.43 +17.7% 0.51 perf-stat.overall.MPKI
0.10 +0.0 0.11 perf-stat.overall.branch-miss-rate%
53.79 +7.5 61.32 perf-stat.overall.cache-miss-rate%
0.73 +34.5% 0.98 perf-stat.overall.cpi
1675 +14.3% 1915 perf-stat.overall.cycles-between-cache-misses
1.38 -25.6% 1.02 perf-stat.overall.ipc
1.623e+11 -26.7% 1.19e+11 perf-stat.ps.branch-instructions
1.671e+08 -24.2% 1.267e+08 perf-stat.ps.branch-misses
3.521e+08 -13.9% 3.03e+08 perf-stat.ps.cache-misses
6.547e+08 -24.5% 4.942e+08 perf-stat.ps.cache-references
9668 ± 7% +140.0% 23203 perf-stat.ps.context-switches
5.9e+11 -1.7% 5.803e+11 perf-stat.ps.cpu-cycles
519.80 ± 3% +117.3% 1129 ± 2% perf-stat.ps.cpu-migrations
8.124e+11 -26.9% 5.941e+11 perf-stat.ps.instructions
4.938e+13 -26.4% 3.633e+13 perf-stat.total.instructions
***************************************************************************************************
lkp-cpl-4sp2: 224 threads 4 sockets Intel(R) Xeon(R) Platinum 8380H CPU @ 2.90GHz (Cooper Lake) with 192G memory
=========================================================================================
compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase:
gcc-14/performance/x86_64-rhel-9.4/process/100%/debian-13-x86_64-20250902.cgz/lkp-cpl-4sp2/brk2/will-it-scale
commit:
6a67958ab0 ("slab: remove unused PREEMPT_RT specific macros")
aa8fdb9e25 ("slab: refill sheaves from all nodes")
6a67958ab000c3a7 aa8fdb9e2516055552de11cabaa
---------------- ---------------------------
%stddev %change %stddev
\ | \
482268 ± 5% +474.2% 2769004 ±173% cpuidle..usage
5608 +527.0% 35168 ± 5% vmstat.system.cs
4055429 ± 3% -41.6% 2367093 ± 5% meminfo.SUnreclaim
4204255 ± 3% -40.2% 2515845 ± 5% meminfo.Slab
20631562 -29.4% 14569336 ± 10% meminfo.max_used_kB
181.50 ± 17% +6369.9% 11742 ± 17% perf-c2c.DRAM.remote
36602 ± 12% -31.1% 25221 ± 11% perf-c2c.HITM.local
33.67 ± 24% +23228.2% 7853 ± 17% perf-c2c.HITM.remote
0.81 -0.2 0.62 ± 2% mpstat.cpu.all.irq%
25.48 -12.9 12.54 ± 6% mpstat.cpu.all.soft%
69.98 +11.3 81.25 ± 6% mpstat.cpu.all.sys%
3.17 -1.0 2.22 ± 5% mpstat.cpu.all.usr%
3454 +7.2% 3702 turbostat.Bzy_MHz
0.47 -36.2% 0.30 turbostat.IPC
298.04 -119.9 178.13 ± 5% turbostat.PKG_%
32.63 -6.7% 30.45 ± 2% turbostat.RAMWatt
1.124e+08 -32.8% 75518130 will-it-scale.224.processes
0.12 ± 3% +26.8% 0.15 will-it-scale.224.processes_idle
501772 -32.8% 337134 will-it-scale.per_process_ops
1.124e+08 -32.8% 75518130 will-it-scale.workload
1.224e+08 -63.6% 44526750 numa-numastat.node0.local_node
1.225e+08 -63.6% 44595833 numa-numastat.node0.numa_hit
1.247e+08 -64.8% 43916474 numa-numastat.node1.local_node
1.248e+08 -64.7% 44011639 numa-numastat.node1.numa_hit
1.224e+08 -63.7% 44470373 numa-numastat.node2.local_node
1.225e+08 -63.6% 44552932 numa-numastat.node2.numa_hit
1.198e+08 -64.2% 42880598 numa-numastat.node3.local_node
1.199e+08 -64.2% 42983939 numa-numastat.node3.numa_hit
9.79 ± 8% -85.1% 1.46 perf-sched.sch_delay.avg.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
9.79 ± 8% -85.1% 1.46 perf-sched.total_sch_delay.average.ms
116.90 ± 6% -70.8% 34.12 ± 2% perf-sched.total_wait_and_delay.average.ms
29376 ± 7% +457.0% 163637 perf-sched.total_wait_and_delay.count.ms
107.11 ± 6% -69.5% 32.66 ± 2% perf-sched.total_wait_time.average.ms
116.90 ± 6% -70.8% 34.12 ± 2% perf-sched.wait_and_delay.avg.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
29376 ± 7% +457.0% 163637 perf-sched.wait_and_delay.count.[unknown].[unknown].[unknown].[unknown].[unknown]
107.11 ± 6% -69.5% 32.66 ± 2% perf-sched.wait_time.avg.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
43891245 +3.5% 45412620 proc-vmstat.nr_free_pages_blocks
42571 +4.1% 44304 proc-vmstat.nr_kernel_stack
11026 -3.1% 10681 ± 4% proc-vmstat.nr_page_table_pages
1023111 ± 5% -42.6% 587142 ± 4% proc-vmstat.nr_slab_unreclaimable
4.897e+08 -64.0% 1.762e+08 proc-vmstat.numa_hit
4.893e+08 -64.1% 1.758e+08 proc-vmstat.numa_local
1.988e+09 -65.5% 6.856e+08 proc-vmstat.pgalloc_normal
1.987e+09 -65.6% 6.843e+08 proc-vmstat.pgfree
1035692 ± 6% -39.7% 624584 ± 5% numa-meminfo.node0.SUnreclaim
1058283 ± 6% -37.1% 665752 ± 7% numa-meminfo.node0.Slab
41586 ± 66% -73.5% 11030 ±103% numa-meminfo.node1.Mapped
3109692 ± 44% -54.5% 1413986 ± 23% numa-meminfo.node1.MemUsed
1010583 ± 8% -41.1% 595132 ± 7% numa-meminfo.node1.SUnreclaim
1070684 ± 7% -41.0% 631765 ± 7% numa-meminfo.node1.Slab
974105 ± 7% -41.5% 569763 ± 7% numa-meminfo.node2.SUnreclaim
2867 ± 17% +4632.2% 135676 ±107% numa-meminfo.node2.Shmem
995610 ± 8% -39.4% 603137 ± 7% numa-meminfo.node2.Slab
986763 ± 9% -42.8% 564018 ± 5% numa-meminfo.node3.SUnreclaim
1031401 ± 9% -41.7% 601587 ± 4% numa-meminfo.node3.Slab
246455 ± 6% -36.4% 156760 ± 4% numa-vmstat.node0.nr_slab_unreclaimable
1.225e+08 -63.6% 44594040 numa-vmstat.node0.numa_hit
1.224e+08 -63.6% 44524957 numa-vmstat.node0.numa_local
10428 ± 66% -73.6% 2748 ±103% numa-vmstat.node1.nr_mapped
234207 ± 9% -35.7% 150515 ± 8% numa-vmstat.node1.nr_slab_unreclaimable
1.248e+08 -64.7% 44009980 numa-vmstat.node1.numa_hit
1.247e+08 -64.8% 43914814 numa-vmstat.node1.numa_local
716.36 ± 17% +4634.4% 33915 ±107% numa-vmstat.node2.nr_shmem
224298 ± 7% -36.2% 143071 ± 6% numa-vmstat.node2.nr_slab_unreclaimable
1.225e+08 -63.6% 44550930 numa-vmstat.node2.numa_hit
1.224e+08 -63.7% 44468383 numa-vmstat.node2.numa_local
222072 ± 10% -36.2% 141724 ± 5% numa-vmstat.node3.nr_slab_unreclaimable
1.199e+08 -64.2% 42982546 numa-vmstat.node3.numa_hit
1.198e+08 -64.2% 42879206 numa-vmstat.node3.numa_local
1.29 +51.7% 1.96 perf-stat.i.MPKI
1.196e+11 -33.0% 8.019e+10 ± 6% perf-stat.i.branch-instructions
0.14 +0.0 0.17 ± 25% perf-stat.i.branch-miss-rate%
1.498e+08 -28.6% 1.069e+08 ± 6% perf-stat.i.branch-misses
33.64 +26.5 60.19 ± 5% perf-stat.i.cache-miss-rate%
2.079e+09 -44.3% 1.158e+09 ± 5% perf-stat.i.cache-references
5465 +543.8% 35186 ± 6% perf-stat.i.context-switches
1.41 +61.7% 2.28 ± 7% perf-stat.i.cpi
430.92 +175.4% 1186 ± 5% perf-stat.i.cpu-migrations
1090 +8.7% 1185 ± 12% perf-stat.i.cycles-between-cache-misses
5.413e+11 -33.3% 3.612e+11 ± 6% perf-stat.i.instructions
0.71 -36.7% 0.45 ± 2% perf-stat.i.ipc
1.29 +53.0% 1.98 perf-stat.overall.MPKI
0.13 +0.0 0.13 perf-stat.overall.branch-miss-rate%
33.65 +28.0 61.66 perf-stat.overall.cache-miss-rate%
1.41 +56.4% 2.20 perf-stat.overall.cpi
1090 +2.3% 1115 perf-stat.overall.cycles-between-cache-misses
0.71 -36.1% 0.45 perf-stat.overall.ipc
1465595 +1.5% 1487296 perf-stat.overall.path-length
1.191e+11 -32.8% 8e+10 ± 6% perf-stat.ps.branch-instructions
1.491e+08 -28.5% 1.066e+08 ± 6% perf-stat.ps.branch-misses
2.07e+09 -44.2% 1.155e+09 ± 5% perf-stat.ps.cache-references
5439 +545.3% 35099 ± 5% perf-stat.ps.context-switches
428.68 +176.1% 1183 ± 5% perf-stat.ps.cpu-migrations
5.39e+11 -33.1% 3.603e+11 ± 6% perf-stat.ps.instructions
1.647e+14 -31.8% 1.123e+14 perf-stat.total.instructions
27898253 +15.5% 32215787 ± 5% sched_debug.cfs_rq:/.avg_vruntime.avg
28320859 +17.8% 33369238 ± 5% sched_debug.cfs_rq:/.avg_vruntime.max
20058481 ± 4% +23.3% 24736033 ± 6% sched_debug.cfs_rq:/.avg_vruntime.min
655489 ± 6% +49.4% 979168 ± 5% sched_debug.cfs_rq:/.avg_vruntime.stddev
12547618 ± 35% +151.9% 31606527 ± 5% sched_debug.cfs_rq:/.left_deadline.max
3553201 ± 37% +81.6% 6453093 ± 16% sched_debug.cfs_rq:/.left_deadline.stddev
12547461 ± 35% +151.9% 31606425 ± 5% sched_debug.cfs_rq:/.left_vruntime.max
3553163 ± 37% +81.6% 6453059 ± 16% sched_debug.cfs_rq:/.left_vruntime.stddev
38375 ± 43% +2505.3% 999785 ± 45% sched_debug.cfs_rq:/.load.avg
392259 ± 39% +16269.0% 64208761 ± 24% sched_debug.cfs_rq:/.load.max
2140 ± 12% +44.3% 3088 ± 17% sched_debug.cfs_rq:/.load.min
101800 ± 34% +7070.3% 7299363 ± 31% sched_debug.cfs_rq:/.load.stddev
580.87 ± 33% +102.7% 1177 ± 14% sched_debug.cfs_rq:/.load_avg.avg
13113 ± 6% +37.3% 18008 ± 6% sched_debug.cfs_rq:/.load_avg.max
1.64 ± 14% +434.4% 8.76 ± 48% sched_debug.cfs_rq:/.load_avg.min
2024 ± 20% +36.0% 2752 ± 16% sched_debug.cfs_rq:/.load_avg.stddev
1.36 ± 10% +129.4% 3.12 ± 20% sched_debug.cfs_rq:/.nr_queued.max
0.14 ± 20% +110.4% 0.30 ± 29% sched_debug.cfs_rq:/.nr_queued.stddev
12547461 ± 35% +151.9% 31606429 ± 5% sched_debug.cfs_rq:/.right_vruntime.max
3553163 ± 37% +81.6% 6453072 ± 16% sched_debug.cfs_rq:/.right_vruntime.stddev
242.06 ± 21% +56.6% 379.09 ± 15% sched_debug.cfs_rq:/.util_avg.min
464.83 ± 5% -9.4% 421.15 ± 7% sched_debug.cfs_rq:/.util_est.avg
27882255 +15.5% 32212264 ± 5% sched_debug.cfs_rq:/.zero_vruntime.avg
28305796 +17.9% 33365700 ± 5% sched_debug.cfs_rq:/.zero_vruntime.max
20048781 ± 4% +23.4% 24732987 ± 6% sched_debug.cfs_rq:/.zero_vruntime.min
655026 ± 6% +49.4% 978931 ± 5% sched_debug.cfs_rq:/.zero_vruntime.stddev
42294 ± 7% +29.0% 54565 ± 4% sched_debug.cpu.avg_idle.stddev
939.51 +45.0% 1362 ± 2% sched_debug.cpu.clock_task.stddev
0.22 ± 11% +48.5% 0.33 ± 25% sched_debug.cpu.nr_running.stddev
4751 +443.2% 25812 ± 5% sched_debug.cpu.nr_switches.avg
21779 ± 17% +130.7% 50250 ± 3% sched_debug.cpu.nr_switches.max
2615 +383.8% 12654 ± 5% sched_debug.cpu.nr_switches.min
2351 ± 7% +304.5% 9510 ± 3% sched_debug.cpu.nr_switches.stddev
0.00 -51.2% 0.00 ± 5% sched_debug.rt_rq:.rt_nr_running.avg
0.33 -51.2% 0.16 ± 5% sched_debug.rt_rq:.rt_nr_running.max
0.02 -51.2% 0.01 ± 5% sched_debug.rt_rq:.rt_nr_running.stddev
28.25 -17.4 10.82 ± 2% perf-profile.calltrace.cycles-pp.handle_softirqs.run_ksoftirqd.smpboot_thread_fn.kthread.ret_from_fork
28.25 -17.4 10.82 ± 2% perf-profile.calltrace.cycles-pp.run_ksoftirqd.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm
28.26 -17.4 10.84 ± 2% perf-profile.calltrace.cycles-pp.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm
28.24 -17.4 10.82 ± 2% perf-profile.calltrace.cycles-pp.rcu_core.handle_softirqs.run_ksoftirqd.smpboot_thread_fn.kthread
28.24 -17.4 10.82 ± 2% perf-profile.calltrace.cycles-pp.rcu_do_batch.rcu_core.handle_softirqs.run_ksoftirqd.smpboot_thread_fn
27.84 -17.3 10.56 ± 2% perf-profile.calltrace.cycles-pp.rcu_free_sheaf.rcu_do_batch.rcu_core.handle_softirqs.run_ksoftirqd
27.21 -16.7 10.53 ± 2% perf-profile.calltrace.cycles-pp.__kmem_cache_free_bulk.rcu_free_sheaf.rcu_do_batch.rcu_core.handle_softirqs
25.64 -16.3 9.29 ± 3% perf-profile.calltrace.cycles-pp.__slab_free.__kmem_cache_free_bulk.rcu_free_sheaf.rcu_do_batch.rcu_core
24.89 -16.0 8.88 ± 2% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__slab_free.__kmem_cache_free_bulk.rcu_free_sheaf.rcu_do_batch
24.62 -15.9 8.76 ± 2% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__slab_free.__kmem_cache_free_bulk.rcu_free_sheaf
28.26 -13.1 15.19 perf-profile.calltrace.cycles-pp.kthread.ret_from_fork.ret_from_fork_asm
28.26 -13.1 15.19 perf-profile.calltrace.cycles-pp.ret_from_fork.ret_from_fork_asm
28.26 -13.1 15.19 perf-profile.calltrace.cycles-pp.ret_from_fork_asm
5.98 -6.0 0.00 perf-profile.calltrace.cycles-pp.__refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes.mas_preallocate
13.29 -5.9 7.35 perf-profile.calltrace.cycles-pp.do_brk_flags.__do_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe.brk
9.69 -4.2 5.46 perf-profile.calltrace.cycles-pp.vms_complete_munmap_vmas.do_vmi_align_munmap.__do_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe
7.32 -3.2 4.12 perf-profile.calltrace.cycles-pp.vma_merge_new_range.do_brk_flags.__do_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe
6.59 -2.9 3.70 perf-profile.calltrace.cycles-pp.vma_expand.vma_merge_new_range.do_brk_flags.__do_sys_brk.do_syscall_64
5.85 -2.7 3.19 perf-profile.calltrace.cycles-pp.vms_clear_ptes.vms_complete_munmap_vmas.do_vmi_align_munmap.__do_sys_brk.do_syscall_64
5.21 -2.4 2.82 perf-profile.calltrace.cycles-pp.commit_merge.vma_expand.vma_merge_new_range.do_brk_flags.__do_sys_brk
6.85 -2.1 4.70 perf-profile.calltrace.cycles-pp.mas_store_gfp.do_vmi_align_munmap.__do_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe
3.48 -1.6 1.84 perf-profile.calltrace.cycles-pp.unmap_vmas.vms_clear_ptes.vms_complete_munmap_vmas.do_vmi_align_munmap.__do_sys_brk
3.57 -1.6 1.93 perf-profile.calltrace.cycles-pp.perf_event_mmap.do_brk_flags.__do_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe
6.15 -1.5 4.64 perf-profile.calltrace.cycles-pp.vma_complete.__split_vma.vms_gather_munmap_vmas.do_vmi_align_munmap.__do_sys_brk
5.82 -1.4 4.46 perf-profile.calltrace.cycles-pp.mas_store_prealloc.vma_complete.__split_vma.vms_gather_munmap_vmas.do_vmi_align_munmap
2.91 -1.3 1.58 perf-profile.calltrace.cycles-pp.perf_event_mmap_event.perf_event_mmap.do_brk_flags.__do_sys_brk.do_syscall_64
3.68 -1.3 2.41 perf-profile.calltrace.cycles-pp.mas_wr_node_store.mas_store_gfp.do_vmi_align_munmap.__do_sys_brk.do_syscall_64
2.66 -1.3 1.41 perf-profile.calltrace.cycles-pp.unmap_page_range.unmap_vmas.vms_clear_ptes.vms_complete_munmap_vmas.do_vmi_align_munmap
1.94 -1.2 0.75 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64.brk
2.49 -1.2 1.30 perf-profile.calltrace.cycles-pp.mas_find.__do_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe.brk
2.42 -1.1 1.34 perf-profile.calltrace.cycles-pp.mas_store_prealloc.commit_merge.vma_expand.vma_merge_new_range.do_brk_flags
2.11 -1.0 1.14 perf-profile.calltrace.cycles-pp.clear_bhb_loop.brk
4.84 -0.9 3.92 perf-profile.calltrace.cycles-pp.mas_wr_node_store.mas_store_prealloc.vma_complete.__split_vma.vms_gather_munmap_vmas
1.91 -0.9 1.00 perf-profile.calltrace.cycles-pp.mas_walk.mas_find.__do_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe
2.05 -0.8 1.20 perf-profile.calltrace.cycles-pp.vm_area_dup.__split_vma.vms_gather_munmap_vmas.do_vmi_align_munmap.__do_sys_brk
1.74 -0.8 0.93 perf-profile.calltrace.cycles-pp.zap_pmd_range.unmap_page_range.unmap_vmas.vms_clear_ptes.vms_complete_munmap_vmas
1.72 -0.8 0.96 perf-profile.calltrace.cycles-pp.kmem_cache_alloc_noprof.vm_area_dup.__split_vma.vms_gather_munmap_vmas.do_vmi_align_munmap
1.43 -0.7 0.75 perf-profile.calltrace.cycles-pp.mas_preallocate.commit_merge.vma_expand.vma_merge_new_range.do_brk_flags
1.36 ± 3% -0.6 0.72 perf-profile.calltrace.cycles-pp.perf_iterate_sb.perf_event_mmap_event.perf_event_mmap.do_brk_flags.__do_sys_brk
1.59 -0.6 0.98 perf-profile.calltrace.cycles-pp.kmem_cache_free.vms_complete_munmap_vmas.do_vmi_align_munmap.__do_sys_brk.do_syscall_64
1.29 -0.6 0.69 perf-profile.calltrace.cycles-pp.mas_find.vms_gather_munmap_vmas.do_vmi_align_munmap.__do_sys_brk.do_syscall_64
1.26 -0.6 0.68 perf-profile.calltrace.cycles-pp.zap_pte_range.zap_pmd_range.unmap_page_range.unmap_vmas.vms_clear_ptes
1.18 -0.6 0.62 perf-profile.calltrace.cycles-pp.mas_prev_slot.vms_gather_munmap_vmas.do_vmi_align_munmap.__do_sys_brk.do_syscall_64
1.15 -0.6 0.59 perf-profile.calltrace.cycles-pp.check_brk_limits.__do_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe.brk
1.26 -0.5 0.72 perf-profile.calltrace.cycles-pp.free_pgtables.vms_clear_ptes.vms_complete_munmap_vmas.do_vmi_align_munmap.__do_sys_brk
1.12 -0.5 0.60 perf-profile.calltrace.cycles-pp.__memcg_slab_post_alloc_hook.kmem_cache_alloc_noprof.vm_area_dup.__split_vma.vms_gather_munmap_vmas
1.21 -0.5 0.70 perf-profile.calltrace.cycles-pp.mas_update_gap.mas_store_prealloc.commit_merge.vma_expand.vma_merge_new_range
1.19 -0.4 0.76 perf-profile.calltrace.cycles-pp.mas_store_gfp.vms_gather_munmap_vmas.do_vmi_align_munmap.__do_sys_brk.do_syscall_64
0.88 -0.4 0.50 perf-profile.calltrace.cycles-pp.mas_leaf_max_gap.mas_update_gap.mas_store_prealloc.commit_merge.vma_expand
0.98 -0.4 0.62 perf-profile.calltrace.cycles-pp.__vma_start_write.__split_vma.vms_gather_munmap_vmas.do_vmi_align_munmap.__do_sys_brk
0.88 ± 3% -0.3 0.56 ± 2% perf-profile.calltrace.cycles-pp.__memcg_slab_free_hook.kmem_cache_free.vms_complete_munmap_vmas.do_vmi_align_munmap.__do_sys_brk
0.92 -0.3 0.61 perf-profile.calltrace.cycles-pp.__vma_start_write.vma_expand.vma_merge_new_range.do_brk_flags.__do_sys_brk
0.82 -0.2 0.60 ± 2% perf-profile.calltrace.cycles-pp.kvfree_call_rcu.mas_wr_node_store.mas_store_gfp.do_vmi_align_munmap.__do_sys_brk
0.89 -0.2 0.69 perf-profile.calltrace.cycles-pp.kvfree_call_rcu.mas_wr_node_store.mas_store_prealloc.vma_complete.__split_vma
1.46 ± 2% +0.1 1.58 perf-profile.calltrace.cycles-pp.memcpy_orig.mas_wr_node_store.mas_store_prealloc.vma_complete.__split_vma
0.39 ± 71% +0.3 0.66 ± 9% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.free_pcppages_bulk.free_frozen_page_commit.__free_frozen_pages.__kmem_cache_free_bulk
1.79 ± 3% +0.4 2.22 ± 2% perf-profile.calltrace.cycles-pp.__pi_memcpy.mas_wr_node_store.mas_store_gfp.do_vmi_align_munmap.__do_sys_brk
0.19 ±141% +0.5 0.65 ± 9% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.free_pcppages_bulk.free_frozen_page_commit.__free_frozen_pages
0.00 +0.5 0.53 ± 4% perf-profile.calltrace.cycles-pp.rcu_free_sheaf.rcu_do_batch.rcu_core.handle_softirqs.__irq_exit_rcu
0.00 +0.5 0.54 ± 3% perf-profile.calltrace.cycles-pp.rcu_do_batch.rcu_core.handle_softirqs.__irq_exit_rcu.sysvec_apic_timer_interrupt
0.00 +0.5 0.54 ± 4% perf-profile.calltrace.cycles-pp.rcu_core.handle_softirqs.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
0.00 +0.5 0.54 ± 4% perf-profile.calltrace.cycles-pp.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.kmem_cache_free_bulk.kvfree_rcu_bulk
0.00 +0.5 0.54 ± 4% perf-profile.calltrace.cycles-pp.handle_softirqs.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.kmem_cache_free_bulk
0.00 +0.5 0.55 ± 4% perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_monitor.process_one_work
0.00 +0.5 0.55 ± 4% perf-profile.calltrace.cycles-pp.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_monitor
0.00 +1.0 1.05 ± 6% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__slab_free.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_work
0.00 +1.1 1.11 ± 6% perf-profile.calltrace.cycles-pp.__slab_free.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_work.process_one_work
0.00 +1.5 1.46 ± 7% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__slab_free.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_monitor
0.00 +1.5 1.52 ± 7% perf-profile.calltrace.cycles-pp.__slab_free.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_monitor.process_one_work
0.00 +1.8 1.78 ± 4% perf-profile.calltrace.cycles-pp.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_work.process_one_work.worker_thread
0.00 +1.8 1.78 ± 4% perf-profile.calltrace.cycles-pp.kvfree_rcu_bulk.kfree_rcu_work.process_one_work.worker_thread.kthread
0.00 +1.8 1.79 ± 4% perf-profile.calltrace.cycles-pp.kfree_rcu_work.process_one_work.worker_thread.kthread.ret_from_fork
0.00 +2.2 2.24 ± 3% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__slab_free.__refill_objects_node.__refill_objects_any
0.00 +2.2 2.25 ± 3% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__slab_free.__refill_objects_node.__refill_objects_any.refill_objects
0.00 +2.3 2.25 ± 3% perf-profile.calltrace.cycles-pp.__slab_free.__refill_objects_node.__refill_objects_any.refill_objects.__pcs_replace_empty_main
0.00 +2.3 2.34 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__slab_free.__refill_objects_node.refill_objects
0.00 +2.4 2.36 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__slab_free.__refill_objects_node.refill_objects.__pcs_replace_empty_main
0.00 +2.4 2.41 perf-profile.calltrace.cycles-pp.__slab_free.__refill_objects_node.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof
0.00 +2.5 2.47 ± 3% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__slab_free.kmem_cache_free_bulk.kvfree_rcu_bulk
0.00 +2.5 2.50 ± 5% perf-profile.calltrace.cycles-pp.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_monitor.process_one_work.worker_thread
0.00 +2.5 2.50 ± 6% perf-profile.calltrace.cycles-pp.kvfree_rcu_bulk.kfree_rcu_monitor.process_one_work.worker_thread.kthread
0.00 +2.5 2.52 ± 5% perf-profile.calltrace.cycles-pp.kfree_rcu_monitor.process_one_work.worker_thread.kthread.ret_from_fork
0.00 +4.1 4.08 ± 2% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.alloc_from_new_slab.refill_objects.__pcs_replace_empty_main
0.00 +4.2 4.18 ± 2% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.alloc_from_new_slab.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof
0.00 +4.3 4.34 ± 2% perf-profile.calltrace.cycles-pp.process_one_work.worker_thread.kthread.ret_from_fork.ret_from_fork_asm
0.00 +4.4 4.35 ± 2% perf-profile.calltrace.cycles-pp.worker_thread.kthread.ret_from_fork.ret_from_fork_asm
0.00 +4.4 4.39 ± 2% perf-profile.calltrace.cycles-pp.alloc_from_new_slab.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes
0.00 +11.5 11.46 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__refill_objects_node.refill_objects.__pcs_replace_empty_main
0.00 +11.7 11.66 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__refill_objects_node.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof
71.31 +12.7 83.99 perf-profile.calltrace.cycles-pp.brk
65.80 +15.1 80.88 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.brk
65.43 +15.2 80.67 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.brk
63.58 +16.0 79.57 perf-profile.calltrace.cycles-pp.__do_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe.brk
0.00 +16.5 16.54 perf-profile.calltrace.cycles-pp.__refill_objects_node.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes
0.00 +19.4 19.42 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__refill_objects_node.__refill_objects_any.refill_objects
0.00 +19.6 19.60 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__refill_objects_node.__refill_objects_any.refill_objects.__pcs_replace_empty_main
0.00 +23.2 23.16 perf-profile.calltrace.cycles-pp.__refill_objects_node.__refill_objects_any.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof
0.00 +23.6 23.60 perf-profile.calltrace.cycles-pp.__refill_objects_any.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes
43.10 +25.3 68.41 perf-profile.calltrace.cycles-pp.do_vmi_align_munmap.__do_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe.brk
25.61 +32.1 57.73 perf-profile.calltrace.cycles-pp.vms_gather_munmap_vmas.do_vmi_align_munmap.__do_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe
20.04 +34.5 54.56 perf-profile.calltrace.cycles-pp.__split_vma.vms_gather_munmap_vmas.do_vmi_align_munmap.__do_sys_brk.do_syscall_64
8.83 +38.1 46.98 perf-profile.calltrace.cycles-pp.mas_preallocate.__split_vma.vms_gather_munmap_vmas.do_vmi_align_munmap.__do_sys_brk
7.38 +38.8 46.20 perf-profile.calltrace.cycles-pp.mas_alloc_nodes.mas_preallocate.__split_vma.vms_gather_munmap_vmas.do_vmi_align_munmap
7.31 +38.8 46.16 perf-profile.calltrace.cycles-pp.kmem_cache_alloc_noprof.mas_alloc_nodes.mas_preallocate.__split_vma.vms_gather_munmap_vmas
6.47 +39.2 45.68 perf-profile.calltrace.cycles-pp.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes.mas_preallocate.__split_vma
0.00 +45.2 45.23 perf-profile.calltrace.cycles-pp.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes.mas_preallocate
28.25 -17.4 10.82 ± 2% perf-profile.children.cycles-pp.run_ksoftirqd
28.26 -17.4 10.84 ± 2% perf-profile.children.cycles-pp.smpboot_thread_fn
29.03 -15.0 14.06 perf-profile.children.cycles-pp.rcu_core
29.02 -15.0 14.05 perf-profile.children.cycles-pp.rcu_do_batch
29.04 -14.9 14.09 perf-profile.children.cycles-pp.handle_softirqs
28.56 -14.9 13.69 perf-profile.children.cycles-pp.rcu_free_sheaf
27.97 -14.7 13.31 perf-profile.children.cycles-pp.__kmem_cache_free_bulk
28.26 -13.1 15.19 perf-profile.children.cycles-pp.kthread
28.26 -13.1 15.19 perf-profile.children.cycles-pp.ret_from_fork
28.26 -13.1 15.19 perf-profile.children.cycles-pp.ret_from_fork_asm
26.68 -7.4 19.28 perf-profile.children.cycles-pp.__slab_free
6.18 -6.2 0.00 perf-profile.children.cycles-pp.__refill_objects
13.34 -6.0 7.37 perf-profile.children.cycles-pp.do_brk_flags
9.76 -4.3 5.49 perf-profile.children.cycles-pp.vms_complete_munmap_vmas
7.37 -3.2 4.14 perf-profile.children.cycles-pp.vma_merge_new_range
6.64 -2.9 3.72 perf-profile.children.cycles-pp.vma_expand
5.90 -2.7 3.22 perf-profile.children.cycles-pp.vms_clear_ptes
8.15 -2.6 5.52 perf-profile.children.cycles-pp.mas_store_gfp
5.56 -2.6 2.97 perf-profile.children.cycles-pp.mas_find
8.26 -2.5 5.81 perf-profile.children.cycles-pp.mas_store_prealloc
5.32 -2.4 2.88 perf-profile.children.cycles-pp.commit_merge
9.46 -2.0 7.45 perf-profile.children.cycles-pp.mas_wr_node_store
3.51 -1.7 1.85 perf-profile.children.cycles-pp.unmap_vmas
3.58 -1.6 1.94 perf-profile.children.cycles-pp.perf_event_mmap
6.33 -1.6 4.75 perf-profile.children.cycles-pp.vma_complete
3.06 -1.4 1.66 perf-profile.children.cycles-pp.perf_event_mmap_event
2.70 -1.3 1.43 perf-profile.children.cycles-pp.unmap_page_range
2.67 -1.2 1.45 perf-profile.children.cycles-pp.mas_update_gap
2.40 -1.1 1.27 perf-profile.children.cycles-pp.mas_walk
2.17 -1.0 1.16 perf-profile.children.cycles-pp.mas_wr_store_type
2.13 -1.0 1.15 perf-profile.children.cycles-pp.clear_bhb_loop
1.41 -0.9 0.52 ± 2% perf-profile.children.cycles-pp.allocate_slab
2.08 -0.9 1.22 perf-profile.children.cycles-pp.vm_area_dup
1.79 -0.8 0.96 perf-profile.children.cycles-pp.mas_leaf_max_gap
1.79 -0.8 0.96 perf-profile.children.cycles-pp.zap_pmd_range
1.62 -0.8 0.86 perf-profile.children.cycles-pp.mas_prev_slot
1.58 -0.7 0.86 perf-profile.children.cycles-pp.mas_next_slot
1.67 -0.7 0.97 perf-profile.children.cycles-pp.its_return_thunk
1.09 -0.7 0.39 ± 2% perf-profile.children.cycles-pp.shuffle_freelist
1.94 -0.7 1.25 perf-profile.children.cycles-pp.__vma_start_write
1.39 ± 3% -0.7 0.74 perf-profile.children.cycles-pp.perf_iterate_sb
1.42 -0.6 0.80 perf-profile.children.cycles-pp.free_pgtables
1.60 -0.6 0.98 perf-profile.children.cycles-pp.kmem_cache_free
1.30 -0.6 0.70 perf-profile.children.cycles-pp.zap_pte_range
1.16 -0.6 0.60 perf-profile.children.cycles-pp.check_brk_limits
1.42 -0.6 0.87 perf-profile.children.cycles-pp.entry_SYSCALL_64
1.12 -0.5 0.59 perf-profile.children.cycles-pp.__cond_resched
1.13 -0.5 0.60 perf-profile.children.cycles-pp.__memcg_slab_post_alloc_hook
0.99 -0.5 0.51 perf-profile.children.cycles-pp.__get_unmapped_area
1.34 -0.5 0.87 perf-profile.children.cycles-pp.__vma_enter_locked
0.95 -0.4 0.51 perf-profile.children.cycles-pp.security_vm_enough_memory_mm
0.81 -0.4 0.40 perf-profile.children.cycles-pp.init_multi_vma_prep
0.88 ± 3% -0.3 0.56 ± 2% perf-profile.children.cycles-pp.__memcg_slab_free_hook
0.48 -0.3 0.17 ± 2% perf-profile.children.cycles-pp.setup_object
0.75 -0.3 0.44 perf-profile.children.cycles-pp.entry_SYSRETQ_unsafe_stack
0.68 -0.3 0.40 perf-profile.children.cycles-pp.down_write_killable
0.57 -0.3 0.30 perf-profile.children.cycles-pp.mas_wr_store_entry
0.60 -0.3 0.35 perf-profile.children.cycles-pp.__pte_offset_map_lock
0.50 -0.2 0.25 perf-profile.children.cycles-pp.syscall_return_via_sysret
0.76 -0.2 0.53 ± 2% perf-profile.children.cycles-pp.__kfree_rcu_sheaf
0.51 ± 2% -0.2 0.29 ± 2% perf-profile.children.cycles-pp.perf_event_mmap_output
0.56 -0.2 0.34 ± 2% perf-profile.children.cycles-pp.__rcu_free_sheaf_prepare
0.46 -0.2 0.25 perf-profile.children.cycles-pp.arch_get_unmapped_area_topdown
0.41 -0.2 0.20 perf-profile.children.cycles-pp.vma_adjust_trans_huge
0.42 -0.2 0.21 ± 2% perf-profile.children.cycles-pp.mas_prev_range
0.44 -0.2 0.26 perf-profile.children.cycles-pp.mas_prev
0.41 -0.2 0.23 perf-profile.children.cycles-pp.can_vma_merge_left
0.53 -0.2 0.35 perf-profile.children.cycles-pp.arch_exit_to_user_mode_prepare
0.38 -0.2 0.20 ± 3% perf-profile.children.cycles-pp.static_key_count
2.23 -0.2 2.06 perf-profile.children.cycles-pp.memcpy_orig
0.35 -0.2 0.18 ± 2% perf-profile.children.cycles-pp.x64_sys_call
0.30 -0.2 0.13 perf-profile.children.cycles-pp.__alloc_frozen_pages_noprof
0.45 -0.2 0.28 perf-profile.children.cycles-pp.up_read
0.35 -0.2 0.19 perf-profile.children.cycles-pp.sized_strscpy
1.75 -0.2 1.59 perf-profile.children.cycles-pp.kvfree_call_rcu
0.75 -0.2 0.60 perf-profile.children.cycles-pp.build_detached_freelist
0.33 -0.2 0.18 ± 2% perf-profile.children.cycles-pp.rcu_all_qs
0.34 ± 7% -0.2 0.19 perf-profile.children.cycles-pp.obj_cgroup_charge_account
0.31 -0.1 0.16 ± 2% perf-profile.children.cycles-pp.percpu_counter_add_batch
0.33 ± 2% -0.1 0.18 ± 2% perf-profile.children.cycles-pp.__vm_enough_memory
0.38 -0.1 0.24 perf-profile.children.cycles-pp.entry_SYSCALL_64_safe_stack
0.25 -0.1 0.10 ± 3% perf-profile.children.cycles-pp.get_page_from_freelist
0.37 -0.1 0.22 ± 2% perf-profile.children.cycles-pp.tlb_gather_mmu
0.36 ± 3% -0.1 0.21 perf-profile.children.cycles-pp.refill_obj_stock
0.35 -0.1 0.21 perf-profile.children.cycles-pp.tlb_finish_mmu
0.30 -0.1 0.18 ± 2% perf-profile.children.cycles-pp.__kmalloc_noprof
0.30 -0.1 0.18 ± 2% perf-profile.children.cycles-pp.__alloc_empty_sheaf
0.35 ± 2% -0.1 0.22 ± 2% perf-profile.children.cycles-pp.kfree
0.27 -0.1 0.15 perf-profile.children.cycles-pp.mas_prev_setup
0.36 ± 3% -0.1 0.24 ± 6% perf-profile.children.cycles-pp.__sysvec_apic_timer_interrupt
0.25 -0.1 0.13 perf-profile.children.cycles-pp.mas_wr_slot_store
0.36 ± 3% -0.1 0.23 ± 5% perf-profile.children.cycles-pp.hrtimer_interrupt
0.23 -0.1 0.11 ± 4% perf-profile.children.cycles-pp.userfaultfd_unmap_complete
0.25 -0.1 0.14 ± 2% perf-profile.children.cycles-pp.remove_vma
0.30 ± 3% -0.1 0.20 ± 7% perf-profile.children.cycles-pp.__hrtimer_run_queues
0.25 -0.1 0.14 ± 3% perf-profile.children.cycles-pp.vm_area_free
0.24 ± 3% -0.1 0.14 ± 11% perf-profile.children.cycles-pp.tick_nohz_handler
0.26 -0.1 0.16 ± 2% perf-profile.children.cycles-pp.vma_mark_detached
0.20 -0.1 0.11 ± 4% perf-profile.children.cycles-pp.strnlen
0.22 ± 4% -0.1 0.13 ± 2% perf-profile.children.cycles-pp.__account_obj_stock
0.22 ± 3% -0.1 0.13 ± 9% perf-profile.children.cycles-pp.update_process_times
0.20 -0.1 0.11 perf-profile.children.cycles-pp.free_pgd_range
0.28 -0.1 0.19 perf-profile.children.cycles-pp.downgrade_write
0.16 -0.1 0.07 perf-profile.children.cycles-pp.rmqueue
0.24 -0.1 0.15 perf-profile.children.cycles-pp.unlink_anon_vmas
0.17 ± 2% -0.1 0.09 perf-profile.children.cycles-pp.anon_vma_clone
0.16 ± 2% -0.1 0.08 ± 6% perf-profile.children.cycles-pp.security_mmap_addr
0.18 ± 2% -0.1 0.10 ± 3% perf-profile.children.cycles-pp.vma_prepare
0.14 -0.1 0.06 ± 6% perf-profile.children.cycles-pp.__rmqueue_pcplist
0.15 ± 2% -0.1 0.08 ± 6% perf-profile.children.cycles-pp.mm_get_unmapped_area_vmflags
0.15 -0.1 0.08 perf-profile.children.cycles-pp.unmap_single_vma
0.20 -0.1 0.13 ± 3% perf-profile.children.cycles-pp.up_write
0.16 ± 2% -0.1 0.09 perf-profile.children.cycles-pp.__build_id_parse
0.14 ± 3% -0.1 0.07 perf-profile.children.cycles-pp.rcu_cblist_dequeue
0.14 ± 3% -0.1 0.07 perf-profile.children.cycles-pp.strlen
0.22 ± 4% -0.1 0.15 ± 3% perf-profile.children.cycles-pp.vm_area_init_from
0.13 -0.1 0.07 ± 7% perf-profile.children.cycles-pp.__free_one_page
0.13 -0.1 0.07 perf-profile.children.cycles-pp.is_mergeable_anon_vma
0.12 ± 4% -0.1 0.06 perf-profile.children.cycles-pp.___pte_offset_map
0.12 ± 6% -0.1 0.07 perf-profile.children.cycles-pp.cap_capable
0.12 ± 3% -0.1 0.06 ± 7% perf-profile.children.cycles-pp.may_expand_vm
0.11 ± 4% -0.1 0.06 perf-profile.children.cycles-pp.brk@plt
0.11 -0.1 0.06 perf-profile.children.cycles-pp.__x64_sys_brk
0.10 ± 4% -0.0 0.05 perf-profile.children.cycles-pp.userfaultfd_unmap_prep
0.12 ± 3% -0.0 0.08 ± 6% perf-profile.children.cycles-pp.finish_rcuwait
0.24 ± 2% -0.0 0.20 ± 2% perf-profile.children.cycles-pp._raw_spin_lock
0.07 ± 15% -0.0 0.04 ± 71% perf-profile.children.cycles-pp.cap_vm_enough_memory
0.10 ± 8% -0.0 0.06 ± 6% perf-profile.children.cycles-pp.is_vmalloc_addr
0.10 ± 3% -0.0 0.07 ± 5% perf-profile.children.cycles-pp.khugepaged_enter_vma
0.08 ± 12% -0.0 0.05 perf-profile.children.cycles-pp.testcase
0.10 ± 4% -0.0 0.08 ± 5% perf-profile.children.cycles-pp.sched_tick
0.28 ± 8% +0.0 0.31 ± 7% perf-profile.children.cycles-pp.cmd_record
0.20 ± 6% +0.0 0.25 ± 10% perf-profile.children.cycles-pp.reader__read_event
0.16 ± 16% +0.1 0.21 ± 11% perf-profile.children.cycles-pp.ordered_events__queue
0.16 ± 14% +0.1 0.22 ± 9% perf-profile.children.cycles-pp.process_simple
0.15 ± 17% +0.1 0.21 ± 11% perf-profile.children.cycles-pp.queue_event
0.00 +0.1 0.12 ± 3% perf-profile.children.cycles-pp.get_state_synchronize_rcu_full
1.01 ± 2% +0.2 1.19 ± 2% perf-profile.children.cycles-pp.__pi_memcpy
0.70 ± 9% +0.6 1.28 ± 6% perf-profile.children.cycles-pp.free_pcppages_bulk
0.78 ± 8% +0.6 1.39 ± 6% perf-profile.children.cycles-pp.__free_frozen_pages
0.72 ± 9% +0.6 1.34 ± 6% perf-profile.children.cycles-pp.free_frozen_page_commit
0.00 +1.8 1.79 ± 4% perf-profile.children.cycles-pp.kfree_rcu_work
1.14 ± 3% +2.4 3.50 ± 4% perf-profile.children.cycles-pp.sysvec_apic_timer_interrupt
1.17 ± 3% +2.4 3.54 ± 4% perf-profile.children.cycles-pp.asm_sysvec_apic_timer_interrupt
0.77 ± 3% +2.5 3.27 ± 4% perf-profile.children.cycles-pp.__irq_exit_rcu
0.00 +2.5 2.52 ± 5% perf-profile.children.cycles-pp.kfree_rcu_monitor
0.65 +3.7 4.39 ± 2% perf-profile.children.cycles-pp.alloc_from_new_slab
0.00 +4.3 4.28 ± 2% perf-profile.children.cycles-pp.kmem_cache_free_bulk
0.00 +4.3 4.28 ± 2% perf-profile.children.cycles-pp.kvfree_rcu_bulk
0.00 +4.3 4.34 ± 2% perf-profile.children.cycles-pp.process_one_work
0.00 +4.4 4.35 ± 2% perf-profile.children.cycles-pp.worker_thread
71.34 +13.1 84.41 perf-profile.children.cycles-pp.brk
65.92 +15.0 80.97 perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
65.55 +15.2 80.75 perf-profile.children.cycles-pp.do_syscall_64
63.69 +15.9 79.63 perf-profile.children.cycles-pp.__do_sys_brk
0.00 +23.6 23.61 perf-profile.children.cycles-pp.__refill_objects_any
43.14 +25.3 68.42 perf-profile.children.cycles-pp.do_vmi_align_munmap
28.16 +27.0 55.18 perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath
28.78 +27.2 55.94 perf-profile.children.cycles-pp._raw_spin_lock_irqsave
25.72 +32.1 57.79 perf-profile.children.cycles-pp.vms_gather_munmap_vmas
20.22 +34.4 54.65 perf-profile.children.cycles-pp.__split_vma
10.29 +37.5 47.74 perf-profile.children.cycles-pp.mas_preallocate
9.46 +37.9 47.36 perf-profile.children.cycles-pp.kmem_cache_alloc_noprof
7.40 +38.8 46.21 perf-profile.children.cycles-pp.mas_alloc_nodes
6.68 +39.1 45.80 perf-profile.children.cycles-pp.__pcs_replace_empty_main
0.00 +39.8 39.83 perf-profile.children.cycles-pp.__refill_objects_node
0.00 +45.4 45.35 perf-profile.children.cycles-pp.refill_objects
2.24 -1.1 1.16 perf-profile.self.cycles-pp.mas_walk
3.85 -1.0 2.87 perf-profile.self.cycles-pp.mas_wr_node_store
2.10 -1.0 1.11 perf-profile.self.cycles-pp.mas_wr_store_type
2.11 -1.0 1.14 perf-profile.self.cycles-pp.clear_bhb_loop
1.74 -0.9 0.87 perf-profile.self.cycles-pp.__do_sys_brk
1.93 -0.8 1.10 perf-profile.self.cycles-pp.mas_store_gfp
1.72 -0.8 0.90 perf-profile.self.cycles-pp.mas_leaf_max_gap
1.65 -0.8 0.87 perf-profile.self.cycles-pp.mas_store_prealloc
1.59 -0.8 0.83 perf-profile.self.cycles-pp.mas_prev_slot
1.58 -0.8 0.82 perf-profile.self.cycles-pp.mas_find
1.52 -0.7 0.78 perf-profile.self.cycles-pp.mas_preallocate
1.46 -0.7 0.79 perf-profile.self.cycles-pp.mas_next_slot
1.34 -0.6 0.73 perf-profile.self.cycles-pp.its_return_thunk
0.94 ± 2% -0.6 0.34 ± 3% perf-profile.self.cycles-pp.shuffle_freelist
1.38 -0.6 0.81 perf-profile.self.cycles-pp.kmem_cache_alloc_noprof
0.98 -0.5 0.52 perf-profile.self.cycles-pp.do_brk_flags
0.85 ± 4% -0.4 0.43 ± 3% perf-profile.self.cycles-pp.perf_iterate_sb
1.17 -0.4 0.76 perf-profile.self.cycles-pp.__vma_enter_locked
0.77 -0.4 0.37 perf-profile.self.cycles-pp.init_multi_vma_prep
0.77 -0.4 0.39 perf-profile.self.cycles-pp.unmap_page_range
0.76 ± 3% -0.4 0.40 ± 2% perf-profile.self.cycles-pp.__memcg_slab_post_alloc_hook
0.77 -0.4 0.41 perf-profile.self.cycles-pp.__split_vma
0.78 -0.4 0.42 perf-profile.self.cycles-pp.mas_update_gap
0.70 -0.4 0.35 perf-profile.self.cycles-pp.__cond_resched
0.73 -0.4 0.38 perf-profile.self.cycles-pp.do_syscall_64
0.73 -0.3 0.39 perf-profile.self.cycles-pp.do_vmi_align_munmap
0.74 -0.3 0.41 perf-profile.self.cycles-pp.perf_event_mmap_event
0.62 -0.3 0.30 perf-profile.self.cycles-pp.zap_pte_range
0.77 -0.3 0.46 perf-profile.self.cycles-pp.vms_gather_munmap_vmas
0.72 -0.3 0.42 perf-profile.self.cycles-pp.entry_SYSRETQ_unsafe_stack
0.70 ± 3% -0.3 0.41 ± 2% perf-profile.self.cycles-pp.kmem_cache_free
0.56 -0.3 0.30 perf-profile.self.cycles-pp.commit_merge
0.71 -0.3 0.46 ± 2% perf-profile.self.cycles-pp.entry_SYSCALL_64
0.51 ± 2% -0.2 0.26 perf-profile.self.cycles-pp.perf_event_mmap
0.49 -0.2 0.25 perf-profile.self.cycles-pp.syscall_return_via_sysret
0.49 -0.2 0.26 perf-profile.self.cycles-pp.mas_wr_store_entry
0.46 ± 2% -0.2 0.23 ± 2% perf-profile.self.cycles-pp.__kfree_rcu_sheaf
0.50 ± 2% -0.2 0.28 perf-profile.self.cycles-pp.perf_event_mmap_output
0.75 -0.2 0.52 perf-profile.self.cycles-pp.__slab_free
0.56 -0.2 0.34 ± 2% perf-profile.self.cycles-pp.__rcu_free_sheaf_prepare
0.43 ± 2% -0.2 0.21 ± 2% perf-profile.self.cycles-pp.security_vm_enough_memory_mm
0.44 -0.2 0.24 perf-profile.self.cycles-pp.arch_get_unmapped_area_topdown
0.47 ± 2% -0.2 0.27 ± 2% perf-profile.self.cycles-pp.vma_complete
0.55 -0.2 0.35 ± 2% perf-profile.self.cycles-pp.__vma_start_write
0.45 -0.2 0.25 perf-profile.self.cycles-pp.free_pgtables
0.42 -0.2 0.22 ± 3% perf-profile.self.cycles-pp.vms_complete_munmap_vmas
0.75 ± 2% -0.2 0.55 perf-profile.self.cycles-pp.build_detached_freelist
0.38 -0.2 0.18 ± 2% perf-profile.self.cycles-pp.vma_adjust_trans_huge
2.14 -0.2 1.95 perf-profile.self.cycles-pp.memcpy_orig
0.50 ± 8% -0.2 0.32 ± 3% perf-profile.self.cycles-pp.__memcg_slab_free_hook
0.38 -0.2 0.20 perf-profile.self.cycles-pp.unmap_vmas
0.50 -0.2 0.33 ± 2% perf-profile.self.cycles-pp.arch_exit_to_user_mode_prepare
0.33 -0.2 0.16 ± 3% perf-profile.self.cycles-pp.x64_sys_call
0.34 -0.2 0.18 ± 2% perf-profile.self.cycles-pp.zap_pmd_range
0.40 -0.2 0.24 perf-profile.self.cycles-pp.vma_expand
0.43 -0.2 0.27 perf-profile.self.cycles-pp.up_read
0.36 -0.1 0.21 perf-profile.self.cycles-pp.tlb_gather_mmu
0.34 -0.1 0.20 perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe
0.40 -0.1 0.26 perf-profile.self.cycles-pp.down_write_killable
0.26 -0.1 0.12 perf-profile.self.cycles-pp.__get_unmapped_area
0.29 -0.1 0.15 perf-profile.self.cycles-pp.percpu_counter_add_batch
0.30 -0.1 0.16 perf-profile.self.cycles-pp.sized_strscpy
0.28 -0.1 0.14 perf-profile.self.cycles-pp.mas_prev_range
0.29 -0.1 0.16 ± 4% perf-profile.self.cycles-pp.static_key_count
0.30 -0.1 0.17 perf-profile.self.cycles-pp.mas_prev
0.32 -0.1 0.19 perf-profile.self.cycles-pp.tlb_finish_mmu
0.26 -0.1 0.14 ± 2% perf-profile.self.cycles-pp.vms_clear_ptes
0.24 -0.1 0.12 perf-profile.self.cycles-pp.mas_wr_slot_store
0.24 -0.1 0.12 ± 3% perf-profile.self.cycles-pp.vma_merge_new_range
0.24 -0.1 0.13 perf-profile.self.cycles-pp.mas_prev_setup
1.12 ± 2% -0.1 1.01 perf-profile.self.cycles-pp.brk
0.23 -0.1 0.12 ± 3% perf-profile.self.cycles-pp.rcu_all_qs
0.27 -0.1 0.16 ± 2% perf-profile.self.cycles-pp.kfree
0.24 -0.1 0.13 ± 2% perf-profile.self.cycles-pp.can_vma_merge_left
0.22 ± 4% -0.1 0.12 ± 3% perf-profile.self.cycles-pp.refill_obj_stock
0.24 -0.1 0.14 ± 3% perf-profile.self.cycles-pp.__pte_offset_map_lock
0.20 -0.1 0.10 perf-profile.self.cycles-pp.userfaultfd_unmap_complete
0.19 -0.1 0.10 perf-profile.self.cycles-pp.strnlen
0.21 ± 4% -0.1 0.12 perf-profile.self.cycles-pp.__account_obj_stock
0.27 -0.1 0.18 perf-profile.self.cycles-pp.downgrade_write
0.22 ± 2% -0.1 0.14 perf-profile.self.cycles-pp.vma_mark_detached
0.17 ± 4% -0.1 0.08 ± 5% perf-profile.self.cycles-pp.__vm_enough_memory
0.17 ± 4% -0.1 0.09 perf-profile.self.cycles-pp.obj_cgroup_charge_account
0.18 ± 2% -0.1 0.10 perf-profile.self.cycles-pp.free_pgd_range
0.20 -0.1 0.12 perf-profile.self.cycles-pp.entry_SYSCALL_64_safe_stack
0.17 ± 2% -0.1 0.10 ± 5% perf-profile.self.cycles-pp.check_brk_limits
0.18 -0.1 0.11 perf-profile.self.cycles-pp.unlink_anon_vmas
0.14 -0.1 0.07 perf-profile.self.cycles-pp.unmap_single_vma
0.15 ± 2% -0.1 0.08 perf-profile.self.cycles-pp.anon_vma_clone
0.20 -0.1 0.13 perf-profile.self.cycles-pp.up_write
0.17 ± 2% -0.1 0.10 ± 4% perf-profile.self.cycles-pp.alloc_from_new_slab
0.14 ± 3% -0.1 0.07 perf-profile.self.cycles-pp.rcu_cblist_dequeue
0.16 ± 3% -0.1 0.09 perf-profile.self.cycles-pp.__build_id_parse
0.83 -0.1 0.76 perf-profile.self.cycles-pp.kvfree_call_rcu
0.12 -0.1 0.06 perf-profile.self.cycles-pp.__free_one_page
0.13 -0.1 0.07 perf-profile.self.cycles-pp.vma_prepare
0.12 ± 3% -0.1 0.06 perf-profile.self.cycles-pp.strlen
0.20 ± 4% -0.1 0.14 perf-profile.self.cycles-pp.vm_area_init_from
0.12 ± 6% -0.1 0.07 perf-profile.self.cycles-pp.cap_capable
0.11 ± 3% -0.1 0.06 perf-profile.self.cycles-pp.may_expand_vm
0.11 ± 3% -0.1 0.06 perf-profile.self.cycles-pp.mm_get_unmapped_area_vmflags
0.10 -0.1 0.05 perf-profile.self.cycles-pp.mas_alloc_nodes
0.11 -0.1 0.06 perf-profile.self.cycles-pp.___pte_offset_map
0.11 -0.1 0.06 perf-profile.self.cycles-pp.is_mergeable_anon_vma
0.07 ± 14% -0.0 0.03 ±100% perf-profile.self.cycles-pp.cap_vm_enough_memory
0.23 -0.0 0.19 perf-profile.self.cycles-pp._raw_spin_lock
0.10 ± 4% -0.0 0.06 ± 7% perf-profile.self.cycles-pp.finish_rcuwait
0.13 ± 3% -0.0 0.09 perf-profile.self.cycles-pp.vm_area_dup
0.07 ± 10% -0.0 0.04 ± 44% perf-profile.self.cycles-pp.testcase
0.07 ± 5% -0.0 0.04 ± 44% perf-profile.self.cycles-pp.__pi_memcpy
0.07 ± 11% -0.0 0.04 ± 44% perf-profile.self.cycles-pp.is_vmalloc_addr
0.08 -0.0 0.06 ± 8% perf-profile.self.cycles-pp.khugepaged_enter_vma
0.09 ± 4% -0.0 0.07 perf-profile.self.cycles-pp.rcu_free_sheaf
0.11 -0.0 0.10 perf-profile.self.cycles-pp.__pcs_replace_empty_main
0.15 ± 19% +0.1 0.20 ± 10% perf-profile.self.cycles-pp.queue_event
0.00 +0.1 0.12 ± 3% perf-profile.self.cycles-pp.get_state_synchronize_rcu_full
0.62 +0.2 0.78 perf-profile.self.cycles-pp._raw_spin_lock_irqsave
0.00 +0.2 0.18 ± 2% perf-profile.self.cycles-pp.refill_objects
0.00 +0.4 0.43 ± 2% perf-profile.self.cycles-pp.__refill_objects_any
0.00 +3.0 3.02 perf-profile.self.cycles-pp.__refill_objects_node
28.15 +27.0 55.18 perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath
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] 6+ messages in thread
* Re: [vbabka:b4/sheaves-for-all-rebased] [slab] aa8fdb9e25: will-it-scale.per_process_ops 46.5% regression
2026-01-13 13:57 [vbabka:b4/sheaves-for-all-rebased] [slab] aa8fdb9e25: will-it-scale.per_process_ops 46.5% regression kernel test robot
@ 2026-01-28 10:31 ` Vlastimil Babka
2026-01-29 7:05 ` Hao Li
2026-01-30 1:24 ` Oliver Sang
0 siblings, 2 replies; 6+ messages in thread
From: Vlastimil Babka @ 2026-01-28 10:31 UTC (permalink / raw)
To: kernel test robot; +Cc: oe-lkp, lkp, linux-mm, Harry Yoo, Hao Li, Mateusz Guzik
On 1/13/26 14:57, kernel test robot wrote:
>
>
> Hello,
>
> kernel test robot noticed a 46.5% regression of will-it-scale.per_process_ops on:
>
>
> commit: aa8fdb9e2516055552de11cabaacde4d77ad7d72 ("slab: refill sheaves from all nodes")
> https://git.kernel.org/cgit/linux/kernel/git/vbabka/linux.git b4/sheaves-for-all-rebased
>
> testcase: will-it-scale
> config: x86_64-rhel-9.4
> compiler: gcc-14
> test machine: 192 threads 2 sockets Intel(R) Xeon(R) 6740E CPU @ 2.4GHz (Sierra Forest) with 256G memory
> parameters:
>
> nr_task: 100%
> mode: process
> test: mmap2
> cpufreq_governor: performance
>
>
> In addition to that, the commit also has significant impact on the following tests:
>
> +------------------+----------------------------------------------------------------------------------------------------+
> | testcase: change | stress-ng: stress-ng.pkey.ops_per_sec 28.4% regression |
> | test machine | 192 threads 2 sockets Intel(R) Xeon(R) 6740E CPU @ 2.4GHz (Sierra Forest) with 256G memory |
> | test parameters | cpufreq_governor=performance |
> | | nr_threads=100% |
> | | test=pkey |
> | | testtime=60s |
> +------------------+----------------------------------------------------------------------------------------------------+
> | testcase: change | will-it-scale: will-it-scale.per_process_ops 32.8% regression |
> | test machine | 224 threads 4 sockets Intel(R) Xeon(R) Platinum 8380H CPU @ 2.90GHz (Cooper Lake) with 192G memory |
> | test parameters | cpufreq_governor=performance |
> | | mode=process |
> | | nr_task=100% |
> | | test=brk2 |
> +------------------+----------------------------------------------------------------------------------------------------+
>
>
> If you fix the issue in a separate patch/commit (i.e. not just a new version of
> the same patch/commit), kindly add following tags
> | Reported-by: kernel test robot <oliver.sang@intel.com>
> | Closes: https://lore.kernel.org/oe-lkp/202601132136.77efd6d7-lkp@intel.com
>
>
> Details are as below:
> -------------------------------------------------------------------------------------------------->
>
>
> The kernel config and materials to reproduce are available at:
> https://download.01.org/0day-ci/archive/20260113/202601132136.77efd6d7-lkp@intel.com
>
> =========================================================================================
> compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase:
> gcc-14/performance/x86_64-rhel-9.4/process/100%/debian-13-x86_64-20250902.cgz/lkp-srf-2sp2/mmap2/will-it-scale
>
> commit:
> 6a67958ab0 ("slab: remove unused PREEMPT_RT specific macros")
> aa8fdb9e25 ("slab: refill sheaves from all nodes")
Hi,
as discussed at [1] this particular commit restores a behavior analogical to
one that existed before sheaves, so while it may show a regression in
isolation, there should hopefully be also corresponding improvement in an
earlier commit, and those two more or less cancelled out.
What would be more useful is to know the whole series effect (excluding some
preparatory patches). Could you please compare that if anything stands out?
In next-20260127 that would be:
before: d86c9915f4b5 ("mm/slab: make caches with sheaves mergeable")
after: ca43eb67282a ("mm/slub: cleanup and repurpose some stat items")
Additionally, does the patch below improve anything? (on top of
ca43eb67282a). Thanks!
[1] https://lore.kernel.org/all/85d872a3-8192-4668-b5c4-c81ffadc74da@suse.cz/
----8<----
From 5ac96a0bde0c3ea5cecfb4e478e49c9f6deb9c19 Mon Sep 17 00:00:00 2001
From: Vlastimil Babka <vbabka@suse.cz>
Date: Tue, 27 Jan 2026 22:40:26 +0100
Subject: [PATCH] slub: avoid list_lock contention from __refill_objects_any()
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
---
mm/slub.c | 19 +++++++++++++------
1 file changed, 13 insertions(+), 6 deletions(-)
diff --git a/mm/slub.c b/mm/slub.c
index 7d7e1ae1922f..3458dfbab85d 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -3378,7 +3378,8 @@ static inline bool pfmemalloc_match(struct slab *slab, gfp_t gfpflags);
static bool get_partial_node_bulk(struct kmem_cache *s,
struct kmem_cache_node *n,
- struct partial_bulk_context *pc)
+ struct partial_bulk_context *pc,
+ bool allow_spin)
{
struct slab *slab, *slab2;
unsigned int total_free = 0;
@@ -3390,7 +3391,10 @@ static bool get_partial_node_bulk(struct kmem_cache *s,
INIT_LIST_HEAD(&pc->slabs);
- spin_lock_irqsave(&n->list_lock, flags);
+ if (allow_spin)
+ spin_lock_irqsave(&n->list_lock, flags);
+ else if (!spin_trylock_irqsave(&n->list_lock, flags))
+ return false;
list_for_each_entry_safe(slab, slab2, &n->partial, slab_list) {
struct freelist_counters flc;
@@ -6544,7 +6548,8 @@ EXPORT_SYMBOL(kmem_cache_free_bulk);
static unsigned int
__refill_objects_node(struct kmem_cache *s, void **p, gfp_t gfp, unsigned int min,
- unsigned int max, struct kmem_cache_node *n)
+ unsigned int max, struct kmem_cache_node *n,
+ bool allow_spin)
{
struct partial_bulk_context pc;
struct slab *slab, *slab2;
@@ -6556,7 +6561,7 @@ __refill_objects_node(struct kmem_cache *s, void **p, gfp_t gfp, unsigned int mi
pc.min_objects = min;
pc.max_objects = max;
- if (!get_partial_node_bulk(s, n, &pc))
+ if (!get_partial_node_bulk(s, n, &pc, allow_spin))
return 0;
list_for_each_entry_safe(slab, slab2, &pc.slabs, slab_list) {
@@ -6650,7 +6655,8 @@ __refill_objects_any(struct kmem_cache *s, void **p, gfp_t gfp, unsigned int min
n->nr_partial <= s->min_partial)
continue;
- r = __refill_objects_node(s, p, gfp, min, max, n);
+ r = __refill_objects_node(s, p, gfp, min, max, n,
+ /* allow_spin = */ false);
refilled += r;
if (r >= min) {
@@ -6691,7 +6697,8 @@ refill_objects(struct kmem_cache *s, void **p, gfp_t gfp, unsigned int min,
return 0;
refilled = __refill_objects_node(s, p, gfp, min, max,
- get_node(s, local_node));
+ get_node(s, local_node),
+ /* allow_spin = */ true);
if (refilled >= min)
return refilled;
--
2.52.0
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [vbabka:b4/sheaves-for-all-rebased] [slab] aa8fdb9e25: will-it-scale.per_process_ops 46.5% regression
2026-01-28 10:31 ` Vlastimil Babka
@ 2026-01-29 7:05 ` Hao Li
2026-01-29 8:47 ` Vlastimil Babka
2026-01-30 1:24 ` Oliver Sang
1 sibling, 1 reply; 6+ messages in thread
From: Hao Li @ 2026-01-29 7:05 UTC (permalink / raw)
To: Vlastimil Babka
Cc: kernel test robot, oe-lkp, lkp, linux-mm, Harry Yoo, Mateusz Guzik
On Wed, Jan 28, 2026 at 11:31:59AM +0100, Vlastimil Babka wrote:
> On 1/13/26 14:57, kernel test robot wrote:
> >
> >
> > Hello,
> >
> > kernel test robot noticed a 46.5% regression of will-it-scale.per_process_ops on:
> >
> >
> > commit: aa8fdb9e2516055552de11cabaacde4d77ad7d72 ("slab: refill sheaves from all nodes")
> > https://git.kernel.org/cgit/linux/kernel/git/vbabka/linux.git b4/sheaves-for-all-rebased
> >
> > testcase: will-it-scale
> > config: x86_64-rhel-9.4
> > compiler: gcc-14
> > test machine: 192 threads 2 sockets Intel(R) Xeon(R) 6740E CPU @ 2.4GHz (Sierra Forest) with 256G memory
> > parameters:
> >
> > nr_task: 100%
> > mode: process
> > test: mmap2
> > cpufreq_governor: performance
> >
> >
> > In addition to that, the commit also has significant impact on the following tests:
> >
> > +------------------+----------------------------------------------------------------------------------------------------+
> > | testcase: change | stress-ng: stress-ng.pkey.ops_per_sec 28.4% regression |
> > | test machine | 192 threads 2 sockets Intel(R) Xeon(R) 6740E CPU @ 2.4GHz (Sierra Forest) with 256G memory |
> > | test parameters | cpufreq_governor=performance |
> > | | nr_threads=100% |
> > | | test=pkey |
> > | | testtime=60s |
> > +------------------+----------------------------------------------------------------------------------------------------+
> > | testcase: change | will-it-scale: will-it-scale.per_process_ops 32.8% regression |
> > | test machine | 224 threads 4 sockets Intel(R) Xeon(R) Platinum 8380H CPU @ 2.90GHz (Cooper Lake) with 192G memory |
> > | test parameters | cpufreq_governor=performance |
> > | | mode=process |
> > | | nr_task=100% |
> > | | test=brk2 |
> > +------------------+----------------------------------------------------------------------------------------------------+
> >
> >
> > If you fix the issue in a separate patch/commit (i.e. not just a new version of
> > the same patch/commit), kindly add following tags
> > | Reported-by: kernel test robot <oliver.sang@intel.com>
> > | Closes: https://lore.kernel.org/oe-lkp/202601132136.77efd6d7-lkp@intel.com
> >
> >
> > Details are as below:
> > -------------------------------------------------------------------------------------------------->
> >
> >
> > The kernel config and materials to reproduce are available at:
> > https://download.01.org/0day-ci/archive/20260113/202601132136.77efd6d7-lkp@intel.com
> >
> > =========================================================================================
> > compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase:
> > gcc-14/performance/x86_64-rhel-9.4/process/100%/debian-13-x86_64-20250902.cgz/lkp-srf-2sp2/mmap2/will-it-scale
> >
> > commit:
> > 6a67958ab0 ("slab: remove unused PREEMPT_RT specific macros")
> > aa8fdb9e25 ("slab: refill sheaves from all nodes")
>
> Hi,
>
> as discussed at [1] this particular commit restores a behavior analogical to
> one that existed before sheaves, so while it may show a regression in
> isolation, there should hopefully be also corresponding improvement in an
> earlier commit, and those two more or less cancelled out.
>
> What would be more useful is to know the whole series effect (excluding some
> preparatory patches). Could you please compare that if anything stands out?
> In next-20260127 that would be:
>
> before: d86c9915f4b5 ("mm/slab: make caches with sheaves mergeable")
>
> after: ca43eb67282a ("mm/slub: cleanup and repurpose some stat items")
>
> Additionally, does the patch below improve anything? (on top of
> ca43eb67282a). Thanks!
>
> [1] https://lore.kernel.org/all/85d872a3-8192-4668-b5c4-c81ffadc74da@suse.cz/
>
> ----8<----
> From 5ac96a0bde0c3ea5cecfb4e478e49c9f6deb9c19 Mon Sep 17 00:00:00 2001
> From: Vlastimil Babka <vbabka@suse.cz>
> Date: Tue, 27 Jan 2026 22:40:26 +0100
> Subject: [PATCH] slub: avoid list_lock contention from __refill_objects_any()
>
> Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
> ---
> mm/slub.c | 19 +++++++++++++------
> 1 file changed, 13 insertions(+), 6 deletions(-)
Hi Vlastimil,
I conducted a few performance tests on my machine, and I'd like to share my
findings. While I'm not an expert in LKP-style performance testing, I hope these
results can still serve as a useful reference.
Machine Configuration:
- CPU: AMD, 2 sockets, 2 nodes per socket, total 192 CPUs
- SMT: Disabled
Kernel Version:
All tests were based on modifications to the 6.19-rc5 kernel.
Test Scenarios:
0. 6.19-rc5 + Completely disabled the sheaf mechanism
- This was done by set s->cpu_sheaves to NULL
1. Unmodified 6.19-rc5
2. 6.19-rc5 + sheaves-for-all patchset
3. 6.19-rc5 + sheaves-for-all patchset + list_lock contention patch
4. 6.19-rc5 + sheaves-for-all patchset + list_lock contention patch + increased
the maple node sheaf capacity to 128.
Results:
- Performance change of 1 relative to 0:
```
will-it-scale.64.processes -25.3%
will-it-scale.128.processes -22.7%
will-it-scale.192.processes -24.4%
will-it-scale.per_process_ops -24.2%
```
- Performance change of 2 relative to 1:
```
will-it-scale.64.processes -34.2%
will-it-scale.128.processes -32.9%
will-it-scale.192.processes -36.1%
will-it-scale.per_process_ops -34.4%
```
- Performance change of 3 relative to 1:
```
will-it-scale.64.processes -24.8%
will-it-scale.128.processes -26.5%
will-it-scale.192.processes -29.24%
will-it-scale.per_process_ops -26.7%
```
- Performance change of 4 relative to 1:
```
will-it-scale.64.processes +18.0%
will-it-scale.128.processes +22.4%
will-it-scale.192.processes +26.9%
will-it-scale.per_process_ops +22.2%
```
- Performance change of 4 relative to 0:
```
will-it-scale.64.processes -11.9%
will-it-scale.128.processes -5.3%
will-it-scale.192.processes -4.1%
will-it-scale.per_process_ops -7.3%
```
From these results, enabling sheaves and increasing the sheaf capacity to 128
seems to bring the behavior closer to the old percpu partial list mechanism.
However, I previously noticed differences[1] between my results on the AMD
platform and Zhao Liu's results on the Intel platform. This leads me to consider
the possibility of other influencing factors, such as CPU architecture
differences or platform-specific behaviors, that might be impacting the
performance results.
I hope these results are helpful. I'd be happy to hear any feedback or
suggestions for further testing.
[1]: https://lore.kernel.org/linux-mm/3ozekmmsscrarwoa7vcytwjn5rxsiyxjrcsirlu3bhmlwtdxzn@s7a6rcxnqadc/
---
The original testing data are shown below.
0. 6.19-rc5 + Completely disabled the sheaf mechanism
"time.elapsed_time": 93.85333333333334,
"time.elapsed_time.max": 93.85333333333334,
"time.file_system_inputs": 56,
"time.file_system_outputs": 128,
"time.involuntary_context_switches": 2698703.3333333335,
"time.major_page_faults": 50.333333333333336,
"time.maximum_resident_set_size": 90016,
"time.minor_page_faults": 80592,
"time.page_size": 4096,
"time.percent_of_cpu_this_job_got": 5772,
"time.system_time": 5265.683333333333,
"time.user_time": 152.25666666666666,
"time.voluntary_context_switches": 2453,
"will-it-scale.128.processes": 49465360,
"will-it-scale.128.processes_idle": 33.25,
"will-it-scale.192.processes": 71529124,
"will-it-scale.192.processes_idle": 1.2666666666666668,
"will-it-scale.64.processes": 27582414.666666668,
"will-it-scale.64.processes_idle": 66.57,
"will-it-scale.per_process_ops": 396656.3333333333,
"will-it-scale.time.elapsed_time": 93.85333333333334,
"will-it-scale.time.elapsed_time.max": 93.85333333333334,
"will-it-scale.time.file_system_inputs": 56,
"will-it-scale.time.file_system_outputs": 128,
"will-it-scale.time.involuntary_context_switches": 2698703.3333333335,
"will-it-scale.time.major_page_faults": 50.333333333333336,
"will-it-scale.time.maximum_resident_set_size": 90016,
"will-it-scale.time.minor_page_faults": 80592,
"will-it-scale.time.page_size": 4096,
"will-it-scale.time.percent_of_cpu_this_job_got": 5772,
"will-it-scale.time.system_time": 5265.683333333333,
"will-it-scale.time.user_time": 152.25666666666666,
"will-it-scale.time.voluntary_context_switches": 2453,
"will-it-scale.workload": 148576898.66666666
1. Unmodified 6.19-rc5
"time.elapsed_time": 93.86000000000001,
"time.elapsed_time.max": 93.86000000000001,
"time.file_system_inputs": 1952,
"time.file_system_outputs": 160,
"time.involuntary_context_switches": 766225,
"time.major_page_faults": 50.666666666666664,
"time.maximum_resident_set_size": 90012,
"time.minor_page_faults": 80635,
"time.page_size": 4096,
"time.percent_of_cpu_this_job_got": 5738,
"time.system_time": 5251.88,
"time.user_time": 134.57666666666665,
"time.voluntary_context_switches": 2539,
"will-it-scale.128.processes": 38223543.333333336,
"will-it-scale.128.processes_idle": 33.833333333333336,
"will-it-scale.192.processes": 54039039,
"will-it-scale.192.processes_idle": 1.26,
"will-it-scale.64.processes": 20579207.666666668,
"will-it-scale.64.processes_idle": 66.74333333333334,
"will-it-scale.per_process_ops": 300541,
"will-it-scale.time.elapsed_time": 93.86000000000001,
"will-it-scale.time.elapsed_time.max": 93.86000000000001,
"will-it-scale.time.file_system_inputs": 1952,
"will-it-scale.time.file_system_outputs": 160,
"will-it-scale.time.involuntary_context_switches": 766225,
"will-it-scale.time.major_page_faults": 50.666666666666664,
"will-it-scale.time.maximum_resident_set_size": 90012,
"will-it-scale.time.minor_page_faults": 80635,
"will-it-scale.time.page_size": 4096,
"will-it-scale.time.percent_of_cpu_this_job_got": 5738,
"will-it-scale.time.system_time": 5251.88,
"will-it-scale.time.user_time": 134.57666666666665,
"will-it-scale.time.voluntary_context_switches": 2539,
"will-it-scale.workload": 112841790
2. 6.19-rc5 + sheaves-for-all patchset
"time.elapsed_time": 93.88,
"time.elapsed_time.max": 93.88,
"time.file_system_outputs": 128,
"time.involuntary_context_switches": 450569.6666666667,
"time.major_page_faults": 49.333333333333336,
"time.maximum_resident_set_size": 90012,
"time.minor_page_faults": 80581,
"time.page_size": 4096,
"time.percent_of_cpu_this_job_got": 5580,
"time.system_time": 5162.076666666667,
"time.user_time": 76.91666666666667,
"time.voluntary_context_switches": 2467.6666666666665,
"will-it-scale.128.processes": 25617118,
"will-it-scale.128.processes_idle": 33.839999999999996,
"will-it-scale.192.processes": 34502474,
"will-it-scale.192.processes_idle": 1.3133333333333335,
"will-it-scale.64.processes": 13540542.333333334,
"will-it-scale.64.processes_idle": 66.74000000000001,
"will-it-scale.per_process_ops": 197134.33333333334,
"will-it-scale.time.elapsed_time": 93.88,
"will-it-scale.time.elapsed_time.max": 93.88,
"will-it-scale.time.file_system_outputs": 128,
"will-it-scale.time.involuntary_context_switches": 450569.6666666667,
"will-it-scale.time.major_page_faults": 49.333333333333336,
"will-it-scale.time.maximum_resident_set_size": 90012,
"will-it-scale.time.minor_page_faults": 80581,
"will-it-scale.time.page_size": 4096,
"will-it-scale.time.percent_of_cpu_this_job_got": 5580,
"will-it-scale.time.system_time": 5162.076666666667,
"will-it-scale.time.user_time": 76.91666666666667,
"will-it-scale.time.voluntary_context_switches": 2467.6666666666665,
"will-it-scale.workload": 73660134.33333333
3. 6.19-rc5 + sheaves-for-all patchset + list_lock contention patch
"time.elapsed_time": 93.86666666666667,
"time.elapsed_time.max": 93.86666666666667,
"time.file_system_inputs": 1800,
"time.file_system_outputs": 149.33333333333334,
"time.involuntary_context_switches": 421120,
"time.major_page_faults": 37,
"time.maximum_resident_set_size": 90016,
"time.minor_page_faults": 80645,
"time.page_size": 4096,
"time.percent_of_cpu_this_job_got": 5714.666666666667,
"time.system_time": 5256.176666666667,
"time.user_time": 108.88333333333333,
"time.voluntary_context_switches": 2513,
"will-it-scale.128.processes": 28067051.333333332,
"will-it-scale.128.processes_idle": 33.82,
"will-it-scale.192.processes": 38232965.666666664,
"will-it-scale.192.processes_idle": 1.2733333333333334,
"will-it-scale.64.processes": 15464041.333333334,
"will-it-scale.64.processes_idle": 66.76333333333334,
"will-it-scale.per_process_ops": 220009.33333333334,
"will-it-scale.time.elapsed_time": 93.86666666666667,
"will-it-scale.time.elapsed_time.max": 93.86666666666667,
"will-it-scale.time.file_system_inputs": 1800,
"will-it-scale.time.file_system_outputs": 149.33333333333334,
"will-it-scale.time.involuntary_context_switches": 421120,
"will-it-scale.time.major_page_faults": 37,
"will-it-scale.time.maximum_resident_set_size": 90016,
"will-it-scale.time.minor_page_faults": 80645,
"will-it-scale.time.page_size": 4096,
"will-it-scale.time.percent_of_cpu_this_job_got": 5714.666666666667,
"will-it-scale.time.system_time": 5256.176666666667,
"will-it-scale.time.user_time": 108.88333333333333,
"will-it-scale.time.voluntary_context_switches": 2513,
"will-it-scale.workload": 81764058.33333333
4. 6.19-rc5 + sheaves-for-all patchset + list_lock contention patch + increased
the maple node sheaf capacity to 128
"time.elapsed_time": 93.85000000000001,
"time.elapsed_time.max": 93.85000000000001,
"time.file_system_inputs": 1832,
"time.file_system_outputs": 149.33333333333334,
"time.involuntary_context_switches": 208686.33333333334,
"time.major_page_faults": 57.666666666666664,
"time.maximum_resident_set_size": 90016,
"time.minor_page_faults": 80622,
"time.page_size": 4096,
"time.percent_of_cpu_this_job_got": 5788.333333333333,
"time.system_time": 5295.993333333333,
"time.user_time": 136.89333333333332,
"time.voluntary_context_switches": 2521.3333333333335,
"will-it-scale.128.processes": 46820500.666666664,
"will-it-scale.128.processes_idle": 33.806666666666665,
"will-it-scale.192.processes": 68584324.33333333,
"will-it-scale.192.processes_idle": 1.2566666666666668,
"will-it-scale.64.processes": 24292108.666666668,
"will-it-scale.64.processes_idle": 66.74,
"will-it-scale.per_process_ops": 367519.3333333333,
"will-it-scale.time.elapsed_time": 93.85000000000001,
"will-it-scale.time.elapsed_time.max": 93.85000000000001,
"will-it-scale.time.file_system_inputs": 1832,
"will-it-scale.time.file_system_outputs": 149.33333333333334,
"will-it-scale.time.involuntary_context_switches": 208686.33333333334,
"will-it-scale.time.major_page_faults": 57.666666666666664,
"will-it-scale.time.maximum_resident_set_size": 90016,
"will-it-scale.time.minor_page_faults": 80622,
"will-it-scale.time.page_size": 4096,
"will-it-scale.time.percent_of_cpu_this_job_got": 5788.333333333333,
"will-it-scale.time.system_time": 5295.993333333333,
"will-it-scale.time.user_time": 136.89333333333332,
"will-it-scale.time.voluntary_context_switches": 2521.3333333333335,
"will-it-scale.workload": 139696933.66666666
>
> diff --git a/mm/slub.c b/mm/slub.c
> index 7d7e1ae1922f..3458dfbab85d 100644
> --- a/mm/slub.c
> +++ b/mm/slub.c
> @@ -3378,7 +3378,8 @@ static inline bool pfmemalloc_match(struct slab *slab, gfp_t gfpflags);
>
> static bool get_partial_node_bulk(struct kmem_cache *s,
> struct kmem_cache_node *n,
> - struct partial_bulk_context *pc)
> + struct partial_bulk_context *pc,
> + bool allow_spin)
> {
> struct slab *slab, *slab2;
> unsigned int total_free = 0;
> @@ -3390,7 +3391,10 @@ static bool get_partial_node_bulk(struct kmem_cache *s,
>
> INIT_LIST_HEAD(&pc->slabs);
>
> - spin_lock_irqsave(&n->list_lock, flags);
> + if (allow_spin)
> + spin_lock_irqsave(&n->list_lock, flags);
> + else if (!spin_trylock_irqsave(&n->list_lock, flags))
> + return false;
>
> list_for_each_entry_safe(slab, slab2, &n->partial, slab_list) {
> struct freelist_counters flc;
> @@ -6544,7 +6548,8 @@ EXPORT_SYMBOL(kmem_cache_free_bulk);
>
> static unsigned int
> __refill_objects_node(struct kmem_cache *s, void **p, gfp_t gfp, unsigned int min,
> - unsigned int max, struct kmem_cache_node *n)
> + unsigned int max, struct kmem_cache_node *n,
> + bool allow_spin)
> {
> struct partial_bulk_context pc;
> struct slab *slab, *slab2;
> @@ -6556,7 +6561,7 @@ __refill_objects_node(struct kmem_cache *s, void **p, gfp_t gfp, unsigned int mi
> pc.min_objects = min;
> pc.max_objects = max;
>
> - if (!get_partial_node_bulk(s, n, &pc))
> + if (!get_partial_node_bulk(s, n, &pc, allow_spin))
> return 0;
>
> list_for_each_entry_safe(slab, slab2, &pc.slabs, slab_list) {
> @@ -6650,7 +6655,8 @@ __refill_objects_any(struct kmem_cache *s, void **p, gfp_t gfp, unsigned int min
> n->nr_partial <= s->min_partial)
> continue;
>
> - r = __refill_objects_node(s, p, gfp, min, max, n);
> + r = __refill_objects_node(s, p, gfp, min, max, n,
> + /* allow_spin = */ false);
> refilled += r;
>
> if (r >= min) {
> @@ -6691,7 +6697,8 @@ refill_objects(struct kmem_cache *s, void **p, gfp_t gfp, unsigned int min,
> return 0;
>
> refilled = __refill_objects_node(s, p, gfp, min, max,
> - get_node(s, local_node));
> + get_node(s, local_node),
> + /* allow_spin = */ true);
> if (refilled >= min)
> return refilled;
>
> --
> 2.52.0
>
>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [vbabka:b4/sheaves-for-all-rebased] [slab] aa8fdb9e25: will-it-scale.per_process_ops 46.5% regression
2026-01-29 7:05 ` Hao Li
@ 2026-01-29 8:47 ` Vlastimil Babka
2026-01-29 14:49 ` Hao Li
0 siblings, 1 reply; 6+ messages in thread
From: Vlastimil Babka @ 2026-01-29 8:47 UTC (permalink / raw)
To: Hao Li
Cc: kernel test robot, oe-lkp, lkp, linux-mm, Harry Yoo,
Mateusz Guzik, Petr Tesarik
On 1/29/26 08:05, Hao Li wrote:
> On Wed, Jan 28, 2026 at 11:31:59AM +0100, Vlastimil Babka wrote:
> Hi Vlastimil,
>
> I conducted a few performance tests on my machine, and I'd like to share my
> findings. While I'm not an expert in LKP-style performance testing, I hope these
> results can still serve as a useful reference.
>
> Machine Configuration:
> - CPU: AMD, 2 sockets, 2 nodes per socket, total 192 CPUs
> - SMT: Disabled
>
> Kernel Version:
> All tests were based on modifications to the 6.19-rc5 kernel.
>
> Test Scenarios:
> 0. 6.19-rc5 + Completely disabled the sheaf mechanism
> - This was done by set s->cpu_sheaves to NULL
> 1. Unmodified 6.19-rc5
> 2. 6.19-rc5 + sheaves-for-all patchset
> 3. 6.19-rc5 + sheaves-for-all patchset + list_lock contention patch
> 4. 6.19-rc5 + sheaves-for-all patchset + list_lock contention patch + increased
> the maple node sheaf capacity to 128.
>
> Results:
>
> - Performance change of 1 relative to 0:
>
> ```
> will-it-scale.64.processes -25.3%
> will-it-scale.128.processes -22.7%
> will-it-scale.192.processes -24.4%
> will-it-scale.per_process_ops -24.2%
> ```
>
> - Performance change of 2 relative to 1:
>
> ```
> will-it-scale.64.processes -34.2%
> will-it-scale.128.processes -32.9%
> will-it-scale.192.processes -36.1%
> will-it-scale.per_process_ops -34.4%
> ```
>
> - Performance change of 3 relative to 1:
>
> ```
> will-it-scale.64.processes -24.8%
> will-it-scale.128.processes -26.5%
> will-it-scale.192.processes -29.24%
> will-it-scale.per_process_ops -26.7%
> ```
Oh cool, that shows the patch helps, so I'll proceed with it.
IIUC with that the sheaves-for-all doesn't regress this benchmark anymore,
the regression is from 6.18 initial sheaves introduction and related to
maple tree sheaf size.
> - Performance change of 4 relative to 1:
>
> ```
> will-it-scale.64.processes +18.0%
> will-it-scale.128.processes +22.4%
> will-it-scale.192.processes +26.9%
> will-it-scale.per_process_ops +22.2%
> ```
>
> - Performance change of 4 relative to 0:
>
> ```
> will-it-scale.64.processes -11.9%
> will-it-scale.128.processes -5.3%
> will-it-scale.192.processes -4.1%
> will-it-scale.per_process_ops -7.3%
> ```
>
> From these results, enabling sheaves and increasing the sheaf capacity to 128
> seems to bring the behavior closer to the old percpu partial list mechanism.
Yeah but it's a tradeoff so not something to do based on one microbenchmark.
> However, I previously noticed differences[1] between my results on the AMD
> platform and Zhao Liu's results on the Intel platform. This leads me to consider
> the possibility of other influencing factors, such as CPU architecture
> differences or platform-specific behaviors, that might be impacting the
> performance results.
Yeah, these will-it-scale benchmarks are quite sensitive to that.
> I hope these results are helpful. I'd be happy to hear any feedback or
Very helpful, thanks!
> suggestions for further testing.
I've had Petr Tesarik running various mmtests, but those results are now
invalidated due to the memory leak, and resuming them is pending some infra
move to finish. But it might be rather non-obvious how to configure them or
even what subset to take. I was interested in netperf and then a bit of
everything just to see there are no unpleasant surprises.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [vbabka:b4/sheaves-for-all-rebased] [slab] aa8fdb9e25: will-it-scale.per_process_ops 46.5% regression
2026-01-29 8:47 ` Vlastimil Babka
@ 2026-01-29 14:49 ` Hao Li
0 siblings, 0 replies; 6+ messages in thread
From: Hao Li @ 2026-01-29 14:49 UTC (permalink / raw)
To: Vlastimil Babka
Cc: kernel test robot, oe-lkp, lkp, linux-mm, Harry Yoo,
Mateusz Guzik, Petr Tesarik
On Thu, Jan 29, 2026 at 09:47:02AM +0100, Vlastimil Babka wrote:
> On 1/29/26 08:05, Hao Li wrote:
> > On Wed, Jan 28, 2026 at 11:31:59AM +0100, Vlastimil Babka wrote:
> > Hi Vlastimil,
> >
> > I conducted a few performance tests on my machine, and I'd like to share my
> > findings. While I'm not an expert in LKP-style performance testing, I hope these
> > results can still serve as a useful reference.
> >
> > Machine Configuration:
> > - CPU: AMD, 2 sockets, 2 nodes per socket, total 192 CPUs
> > - SMT: Disabled
> >
> > Kernel Version:
> > All tests were based on modifications to the 6.19-rc5 kernel.
> >
> > Test Scenarios:
> > 0. 6.19-rc5 + Completely disabled the sheaf mechanism
> > - This was done by set s->cpu_sheaves to NULL
> > 1. Unmodified 6.19-rc5
> > 2. 6.19-rc5 + sheaves-for-all patchset
> > 3. 6.19-rc5 + sheaves-for-all patchset + list_lock contention patch
> > 4. 6.19-rc5 + sheaves-for-all patchset + list_lock contention patch + increased
> > the maple node sheaf capacity to 128.
> >
> > Results:
> >
> > - Performance change of 1 relative to 0:
> >
> > ```
> > will-it-scale.64.processes -25.3%
> > will-it-scale.128.processes -22.7%
> > will-it-scale.192.processes -24.4%
> > will-it-scale.per_process_ops -24.2%
> > ```
> >
> > - Performance change of 2 relative to 1:
> >
> > ```
> > will-it-scale.64.processes -34.2%
> > will-it-scale.128.processes -32.9%
> > will-it-scale.192.processes -36.1%
> > will-it-scale.per_process_ops -34.4%
> > ```
> >
> > - Performance change of 3 relative to 1:
> >
> > ```
> > will-it-scale.64.processes -24.8%
> > will-it-scale.128.processes -26.5%
> > will-it-scale.192.processes -29.24%
> > will-it-scale.per_process_ops -26.7%
> > ```
>
> Oh cool, that shows the patch helps, so I'll proceed with it.
> IIUC with that the sheaves-for-all doesn't regress this benchmark anymore,
> the regression is from 6.18 initial sheaves introduction and related to
> maple tree sheaf size.
Yes, one of the factors contributing to the regression does seem to be the capacity
of the sheaf.
And I feel that this regression may be difficult to completely resolve with this
lock optimization patch. I'll share my latest test results in response to the v4
patchset a bit later, where we can continue the discussion in more detail.
However, I believe this regression doesn't need to block the progress of the v4
patchset.
>
> > - Performance change of 4 relative to 1:
> >
> > ```
> > will-it-scale.64.processes +18.0%
> > will-it-scale.128.processes +22.4%
> > will-it-scale.192.processes +26.9%
> > will-it-scale.per_process_ops +22.2%
> > ```
> >
> > - Performance change of 4 relative to 0:
> >
> > ```
> > will-it-scale.64.processes -11.9%
> > will-it-scale.128.processes -5.3%
> > will-it-scale.192.processes -4.1%
> > will-it-scale.per_process_ops -7.3%
> > ```
> >
> > From these results, enabling sheaves and increasing the sheaf capacity to 128
> > seems to bring the behavior closer to the old percpu partial list mechanism.
>
> Yeah but it's a tradeoff so not something to do based on one microbenchmark.
Sure, exactly.
>
> > However, I previously noticed differences[1] between my results on the AMD
> > platform and Zhao Liu's results on the Intel platform. This leads me to consider
> > the possibility of other influencing factors, such as CPU architecture
> > differences or platform-specific behaviors, that might be impacting the
> > performance results.
>
> Yeah, these will-it-scale benchmarks are quite sensitive to that.
>
> > I hope these results are helpful. I'd be happy to hear any feedback or
>
> Very helpful, thanks!
>
> > suggestions for further testing.
>
> I've had Petr Tesarik running various mmtests, but those results are now
> invalidated due to the memory leak, and resuming them is pending some infra
> move to finish. But it might be rather non-obvious how to configure them or
> even what subset to take. I was interested in netperf and then a bit of
> everything just to see there are no unpleasant surprises.
Thanks for the update. Looking forward to the test results whenever they're
ready.
--
Thanks,
Hao
>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [vbabka:b4/sheaves-for-all-rebased] [slab] aa8fdb9e25: will-it-scale.per_process_ops 46.5% regression
2026-01-28 10:31 ` Vlastimil Babka
2026-01-29 7:05 ` Hao Li
@ 2026-01-30 1:24 ` Oliver Sang
1 sibling, 0 replies; 6+ messages in thread
From: Oliver Sang @ 2026-01-30 1:24 UTC (permalink / raw)
To: Vlastimil Babka
Cc: oe-lkp, lkp, linux-mm, Harry Yoo, Hao Li, Mateusz Guzik, oliver.sang
hi, Vlastimil Babka,
On Wed, Jan 28, 2026 at 11:31:59AM +0100, Vlastimil Babka wrote:
[...]
> Hi,
>
> as discussed at [1] this particular commit restores a behavior analogical to
> one that existed before sheaves, so while it may show a regression in
> isolation, there should hopefully be also corresponding improvement in an
> earlier commit, and those two more or less cancelled out.
>
> What would be more useful is to know the whole series effect (excluding some
> preparatory patches). Could you please compare that if anything stands out?
> In next-20260127 that would be:
>
> before: d86c9915f4b5 ("mm/slab: make caches with sheaves mergeable")
>
> after: ca43eb67282a ("mm/slub: cleanup and repurpose some stat items")
>
> Additionally, does the patch below improve anything? (on top of
> ca43eb67282a). Thanks!
we see a 60.3% regression if comparing ca43eb67282a with d86c9915f4b5 by same
tests in our original report. a8ce496508 is the commit by applying
"[PATCH] slub: avoid list_lock contention from __refill_objects_any()"
on top of ca43eb67282a. it really recovers performance, in a way of half,
if regards d86c9915f4b5 as base.
(still 31.7% regression comparing to d86c9915f4b5)
more details are attached as [1]
Tested-by: kernel test robot <oliver.sang@intel.com>
=========================================================================================
compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase:
gcc-14/performance/x86_64-rhel-9.4/process/100%/debian-13-x86_64-20250902.cgz/lkp-srf-2sp2/mmap2/will-it-scale
commit:
d86c9915f4 ("mm/slab: make caches with sheaves mergeable")
ca43eb6728 ("mm/slub: cleanup and repurpose some stat items")
a8ce496508 ("slub: avoid list_lock contention from __refill_objects_any()")
d86c9915f4b57ff3 ca43eb67282a4a1b4be449b004a a8ce496508b9ac28b71ce797972
---------------- --------------------------- ---------------------------
%stddev %change %stddev %change %stddev
\ | \ | \
264131 -60.3% 104932 -31.7% 180457 will-it-scale.per_process_ops
for easy comparison, also put summary results in our origil report here.
=========================================================================================
compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase:
gcc-14/performance/x86_64-rhel-9.4/process/100%/debian-13-x86_64-20250902.cgz/lkp-srf-2sp2/mmap2/will-it-scale
commit:
6a67958ab0 ("slab: remove unused PREEMPT_RT specific macros")
aa8fdb9e25 ("slab: refill sheaves from all nodes")
6a67958ab000c3a7 aa8fdb9e2516055552de11cabaa
---------------- ---------------------------
%stddev %change %stddev
\ | \
194705 -46.5% 104184 will-it-scale.per_process_ops
[1]
=========================================================================================
compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase:
gcc-14/performance/x86_64-rhel-9.4/process/100%/debian-13-x86_64-20250902.cgz/lkp-srf-2sp2/mmap2/will-it-scale
commit:
d86c9915f4 ("mm/slab: make caches with sheaves mergeable")
ca43eb6728 ("mm/slub: cleanup and repurpose some stat items")
a8ce496508 ("slub: avoid list_lock contention from __refill_objects_any()")
d86c9915f4b57ff3 ca43eb67282a4a1b4be449b004a a8ce496508b9ac28b71ce797972
---------------- --------------------------- ---------------------------
%stddev %change %stddev %change %stddev
\ | \ | \
307.79 -1.5% 303.32 -0.4% 306.64 time.elapsed_time
307.79 -1.5% 303.32 -0.4% 306.64 time.elapsed_time.max
50713270 -60.3% 20147177 -31.7% 34647924 will-it-scale.192.processes
264131 -60.3% 104932 -31.7% 180457 will-it-scale.per_process_ops
50713270 -60.3% 20147177 -31.7% 34647924 will-it-scale.workload
0.27 ± 3% -0.1 0.13 -0.1 0.18 ± 2% mpstat.cpu.all.irq%
36.63 -17.2 19.47 -2.7 33.97 mpstat.cpu.all.soft%
59.40 +19.0 78.40 +3.6 63.05 mpstat.cpu.all.sys%
3.09 -1.7 1.40 -0.9 2.19 mpstat.cpu.all.usr%
2.163e+08 -75.8% 52417003 -39.0% 1.32e+08 numa-numastat.node0.local_node
2.164e+08 -75.7% 52529330 -39.0% 1.32e+08 numa-numastat.node0.numa_hit
2.174e+08 -74.6% 55136459 -32.2% 1.475e+08 numa-numastat.node1.local_node
2.175e+08 -74.6% 55223414 -32.1% 1.476e+08 numa-numastat.node1.numa_hit
6993580 ± 2% +8.4% 7579245 ± 2% +4.1% 7281381 ± 2% vmstat.memory.cache
253.22 ± 2% -3.9% 243.31 ± 2% -5.3% 239.92 ± 2% vmstat.procs.r
12499 ± 3% +44.3% 18039 -61.8% 4769 vmstat.system.cs
256261 -2.3% 250282 -0.9% 253890 vmstat.system.in
1441 ± 14% +159.4% 3740 ± 19% +293.1% 5668 ± 13% perf-c2c.DRAM.local
1060 ± 35% +410.9% 5420 ± 6% +40.3% 1488 ± 14% perf-c2c.DRAM.remote
39267 ± 13% -16.3% 32871 ± 12% +81.6% 71315 ± 10% perf-c2c.HITM.local
569.80 ± 48% +519.1% 3527 ± 8% -49.6% 287.00 ± 13% perf-c2c.HITM.remote
39837 ± 14% -8.6% 36398 ± 11% +79.7% 71602 ± 10% perf-c2c.HITM.total
2135346 ± 4% -48.8% 1092768 ± 6% -11.5% 1890717 ± 4% numa-meminfo.node0.SUnreclaim
2225620 ± 3% -47.3% 1173754 ± 7% -11.5% 1969261 ± 5% numa-meminfo.node0.Slab
18307 +1.4% 18567 ± 2% -6.8% 17054 ± 3% numa-meminfo.node1.KernelStack
1959677 ± 4% -47.0% 1037682 ± 4% -8.1% 1801834 ± 3% numa-meminfo.node1.SUnreclaim
3090819 ± 6% +19.0% 3677082 ± 6% +3.3% 3191592 ± 6% numa-meminfo.node1.Shmem
2028184 ± 5% -45.5% 1105393 ± 5% -7.8% 1870220 ± 4% numa-meminfo.node1.Slab
3166 +1.1% 3201 +1.1% 3201 turbostat.Bzy_MHz
65.50 ± 2% -6.7% 61.10 -2.3% 64.00 turbostat.CoreTmp
1.24 -57.8% 0.52 -31.0% 0.86 turbostat.IPC
1798555 ± 3% -1.8e+06 0.00 -1.7e+06 109073 ± 21% turbostat.PKG_%
65.30 ± 2% -6.0% 61.40 -2.1% 63.90 turbostat.PkgTmp
470.72 -12.5% 411.86 -2.6% 458.48 turbostat.PkgWatt
28.14 -20.4% 22.40 -11.3% 24.95 turbostat.RAMWatt
529155 ± 4% -48.9% 270394 ± 6% -10.1% 475663 ± 4% numa-vmstat.node0.nr_slab_unreclaimable
2.164e+08 -75.7% 52529459 -39.0% 1.32e+08 numa-vmstat.node0.numa_hit
2.163e+08 -75.8% 52417132 -39.0% 1.319e+08 numa-vmstat.node0.numa_local
18306 +1.4% 18566 ± 2% -6.8% 17057 ± 3% numa-vmstat.node1.nr_kernel_stack
773230 ± 6% +18.9% 919014 ± 6% +3.2% 797828 ± 6% numa-vmstat.node1.nr_shmem
482496 ± 4% -46.7% 257120 ± 4% -6.3% 452095 ± 3% numa-vmstat.node1.nr_slab_unreclaimable
2.175e+08 -74.6% 55223352 -32.1% 1.476e+08 numa-vmstat.node1.numa_hit
2.174e+08 -74.6% 55136398 -32.2% 1.475e+08 numa-vmstat.node1.numa_local
9.61 ± 26% -59.9% 3.85 ± 6% +63.2% 15.69 ± 4% perf-sched.sch_delay.avg.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
9.61 ± 26% -59.9% 3.85 ± 6% +63.2% 15.69 ± 4% perf-sched.total_sch_delay.average.ms
97.85 ± 24% -42.5% 56.29 +44.5% 141.41 ± 3% perf-sched.total_wait_and_delay.average.ms
44235 ± 30% +92.1% 84980 ± 2% -51.2% 21596 ± 3% perf-sched.total_wait_and_delay.count.ms
88.23 ± 23% -40.6% 52.44 +42.5% 125.73 ± 3% perf-sched.total_wait_time.average.ms
97.85 ± 24% -42.5% 56.29 +44.5% 141.41 ± 3% perf-sched.wait_and_delay.avg.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
44235 ± 30% +92.1% 84980 ± 2% -51.2% 21596 ± 3% perf-sched.wait_and_delay.count.[unknown].[unknown].[unknown].[unknown].[unknown]
88.23 ± 23% -40.6% 52.44 +42.5% 125.73 ± 3% perf-sched.wait_time.avg.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
4034071 ± 4% +14.8% 4629394 ± 3% +7.4% 4334119 ± 4% meminfo.Active
4033292 ± 4% +14.8% 4628614 ± 3% +7.4% 4333340 ± 4% meminfo.Active(anon)
6831441 ± 2% +8.7% 7426973 ± 2% +4.4% 7130163 ± 2% meminfo.Cached
35007579 +8.6% 38009265 +4.2% 36465336 meminfo.Committed_AS
158780 -6.4% 148696 -7.5% 146927 meminfo.KReclaimable
37172 +2.3% 38025 -5.8% 35014 meminfo.KernelStack
13747866 -10.4% 12317640 -0.8% 13631524 ± 2% meminfo.Memused
158780 -6.4% 148696 -7.5% 146927 meminfo.SReclaimable
4099705 ± 4% -48.4% 2115765 -9.5% 3711663 ± 4% meminfo.SUnreclaim
3154355 ± 5% +18.9% 3749886 ± 4% +9.5% 3453072 ± 5% meminfo.Shmem
4258485 ± 4% -46.8% 2264461 -9.4% 3858590 ± 4% meminfo.Slab
24903807 ± 24% -41.8% 14487867 ± 12% +8.5% 27008355 ± 29% meminfo.max_used_kB
1008212 ± 4% +14.8% 1157016 ± 3% +7.5% 1083507 ± 4% proc-vmstat.nr_active_anon
1707723 ± 2% +8.7% 1856620 ± 2% +4.4% 1782730 ± 2% proc-vmstat.nr_file_pages
59650535 +2.6% 61174899 +0.6% 60036028 proc-vmstat.nr_free_pages_blocks
37173 +2.3% 38023 -5.8% 35012 proc-vmstat.nr_kernel_stack
788450 ± 5% +18.9% 937347 ± 4% +9.5% 863456 ± 5% proc-vmstat.nr_shmem
39695 -6.4% 37173 -7.5% 36731 proc-vmstat.nr_slab_reclaimable
1013008 ± 4% -47.6% 530940 -8.4% 928206 ± 3% proc-vmstat.nr_slab_unreclaimable
1008212 ± 4% +14.8% 1157016 ± 3% +7.5% 1083507 ± 4% proc-vmstat.nr_zone_active_anon
4.339e+08 -75.2% 1.078e+08 -35.6% 2.796e+08 proc-vmstat.numa_hit
4.337e+08 -75.2% 1.076e+08 -35.6% 2.794e+08 proc-vmstat.numa_local
1.764e+09 -76.2% 4.196e+08 -36.1% 1.127e+09 proc-vmstat.pgalloc_normal
1.762e+09 -76.3% 4.182e+08 -36.1% 1.126e+09 proc-vmstat.pgfree
0.77 +23.5% 0.96 +5.6% 0.82 perf-stat.i.MPKI
1.644e+11 -57.1% 7.045e+10 -30.1% 1.149e+11 perf-stat.i.branch-instructions
0.08 +0.0 0.09 +0.0 0.09 perf-stat.i.branch-miss-rate%
1.264e+08 -50.8% 62178838 -23.6% 96551454 perf-stat.i.branch-misses
84.23 -21.3 62.92 -32.7 51.48 perf-stat.i.cache-miss-rate%
5.748e+08 -47.3% 3.032e+08 -26.3% 4.237e+08 perf-stat.i.cache-misses
6.82e+08 -29.3% 4.823e+08 +20.9% 8.245e+08 perf-stat.i.cache-references
12018 ± 3% +49.4% 17960 -61.6% 4609 perf-stat.i.context-switches
0.81 +137.3% 1.91 +44.9% 1.17 perf-stat.i.cpi
6.054e+11 +1.0% 6.114e+11 +1.1% 6.122e+11 perf-stat.i.cpu-cycles
496.60 ± 2% -15.4% 420.01 ± 2% -44.2% 277.21 perf-stat.i.cpu-migrations
1051 +92.3% 2021 +37.2% 1442 perf-stat.i.cycles-between-cache-misses
7.538e+11 -57.5% 3.204e+11 -30.3% 5.257e+11 perf-stat.i.instructions
1.24 -57.8% 0.53 -30.9% 0.86 perf-stat.i.ipc
0.76 +24.1% 0.95 +5.7% 0.81 perf-stat.overall.MPKI
0.08 +0.0 0.09 +0.0 0.08 perf-stat.overall.branch-miss-rate%
84.21 -21.4 62.81 -32.8 51.38 perf-stat.overall.cache-miss-rate%
0.80 +137.4% 1.91 +44.8% 1.16 perf-stat.overall.cpi
1053 +91.3% 2016 +37.1% 1444 perf-stat.overall.cycles-between-cache-misses
1.24 -57.9% 0.52 -30.9% 0.86 perf-stat.overall.ipc
4561894 +5.7% 4820717 +1.8% 4642538 perf-stat.overall.path-length
1.633e+11 -57.1% 7.014e+10 -30.0% 1.143e+11 perf-stat.ps.branch-instructions
1.256e+08 -50.8% 61848881 -23.6% 95977841 perf-stat.ps.branch-misses
5.714e+08 -47.2% 3.019e+08 -26.2% 4.215e+08 perf-stat.ps.cache-misses
6.786e+08 -29.1% 4.808e+08 +20.9% 8.204e+08 perf-stat.ps.cache-references
12045 ± 3% +48.2% 17857 -62.2% 4556 perf-stat.ps.context-switches
6.021e+11 +1.1% 6.088e+11 +1.1% 6.088e+11 perf-stat.ps.cpu-cycles
489.88 ± 2% -15.5% 413.86 ± 2% -44.8% 270.39 perf-stat.ps.cpu-migrations
7.49e+11 -57.4% 3.19e+11 -30.2% 5.23e+11 perf-stat.ps.instructions
2.313e+14 -58.0% 9.712e+13 -30.5% 1.609e+14 perf-stat.total.instructions
18830824 +21.8% 22927609 +5.5% 19865940 sched_debug.cfs_rq:/.avg_vruntime.avg
19651669 +22.3% 24034159 +3.2% 20278357 sched_debug.cfs_rq:/.avg_vruntime.max
14795521 ± 4% +26.1% 18656404 ± 5% +1.9% 15075489 ± 5% sched_debug.cfs_rq:/.avg_vruntime.min
519008 ± 5% +87.5% 973121 ± 4% -0.3% 517319 ± 8% sched_debug.cfs_rq:/.avg_vruntime.stddev
13541640 ± 37% +74.5% 23630973 ± 2% -33.6% 8995423 ± 34% sched_debug.cfs_rq:/.left_deadline.max
2891535 ± 42% +76.2% 5095880 ± 23% -45.9% 1565184 ± 81% sched_debug.cfs_rq:/.left_deadline.stddev
13541466 ± 37% +74.5% 23630825 ± 2% -33.6% 8995322 ± 34% sched_debug.cfs_rq:/.left_vruntime.max
2891496 ± 42% +76.2% 5095853 ± 23% -45.9% 1565162 ± 81% sched_debug.cfs_rq:/.left_vruntime.stddev
163721 ± 89% +480.9% 951119 ± 29% -82.5% 28651 ±115% sched_debug.cfs_rq:/.load.avg
13002869 ±111% +391.8% 63948635 ± 17% -95.6% 572064 ± 26% sched_debug.cfs_rq:/.load.max
1228727 ±114% +476.8% 7087338 ± 19% -93.5% 79390 ± 64% sched_debug.cfs_rq:/.load.stddev
409.08 ± 25% +173.7% 1119 ± 15% +16.3% 475.77 ± 32% sched_debug.cfs_rq:/.load_avg.avg
3076 ± 37% +460.2% 17235 ± 15% +278.2% 11637 ± 24% sched_debug.cfs_rq:/.load_avg.max
527.03 ± 40% +366.2% 2457 ± 16% +183.8% 1495 ± 31% sched_debug.cfs_rq:/.load_avg.stddev
13541496 ± 37% +74.5% 23630838 ± 2% -33.6% 8995348 ± 34% sched_debug.cfs_rq:/.right_vruntime.max
2891500 ± 42% +76.2% 5095868 ± 23% -45.9% 1565165 ± 81% sched_debug.cfs_rq:/.right_vruntime.stddev
246.98 ± 10% +13.1% 279.43 ± 15% -21.4% 194.19 ± 15% sched_debug.cfs_rq:/.runnable_avg.stddev
684.02 ± 4% -11.3% 607.05 ± 4% -0.4% 681.35 ± 7% sched_debug.cfs_rq:/.util_avg.min
52.86 ± 7% +17.6% 62.17 ± 5% +2.0% 53.91 ± 9% sched_debug.cfs_rq:/.util_avg.stddev
18820626 +21.8% 22922813 +5.5% 19855131 sched_debug.cfs_rq:/.zero_vruntime.avg
19640361 +22.3% 24029703 +3.2% 20270726 sched_debug.cfs_rq:/.zero_vruntime.max
14788107 ± 4% +26.1% 18652511 ± 5% +1.9% 15064048 ± 5% sched_debug.cfs_rq:/.zero_vruntime.min
518981 ± 5% +87.4% 972714 ± 4% -0.2% 518098 ± 8% sched_debug.cfs_rq:/.zero_vruntime.stddev
41.86 ± 17% -22.9% 32.26 ± 12% -30.6% 29.06 ± 24% sched_debug.cpu.clock.stddev
1305 ± 7% +44.8% 1890 ± 4% -60.1% 520.93 ± 2% sched_debug.cpu.clock_task.stddev
0.00 ± 4% -9.0% 0.00 ± 4% -12.3% 0.00 ± 6% sched_debug.cpu.next_balance.stddev
10990 ± 2% +44.7% 15903 -51.1% 5372 ± 2% sched_debug.cpu.nr_switches.avg
57218 ± 19% +33.7% 76496 ± 7% -5.9% 53846 ± 20% sched_debug.cpu.nr_switches.max
3823 ± 5% +91.7% 7331 ± 3% -23.0% 2944 sched_debug.cpu.nr_switches.min
7144 ± 8% +17.1% 8366 ± 2% -31.4% 4899 ± 18% sched_debug.cpu.nr_switches.stddev
36.02 ± 2% -18.9 17.16 ± 4% -2.5 33.54 ± 2% perf-profile.calltrace.cycles-pp.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm
36.01 ± 2% -18.9 17.16 ± 4% -2.5 33.53 ± 2% perf-profile.calltrace.cycles-pp.handle_softirqs.run_ksoftirqd.smpboot_thread_fn.kthread.ret_from_fork
36.01 ± 2% -18.9 17.16 ± 4% -2.5 33.53 ± 2% perf-profile.calltrace.cycles-pp.run_ksoftirqd.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm
36.01 ± 2% -18.9 17.16 ± 4% -2.5 33.53 ± 2% perf-profile.calltrace.cycles-pp.rcu_core.handle_softirqs.run_ksoftirqd.smpboot_thread_fn.kthread
36.00 ± 2% -18.9 17.15 ± 4% -2.5 33.53 ± 2% perf-profile.calltrace.cycles-pp.rcu_do_batch.rcu_core.handle_softirqs.run_ksoftirqd.smpboot_thread_fn
35.33 ± 2% -18.5 16.80 ± 4% -2.4 32.92 ± 2% perf-profile.calltrace.cycles-pp.rcu_free_sheaf.rcu_do_batch.rcu_core.handle_softirqs.run_ksoftirqd
34.83 ± 2% -17.4 17.43 ± 5% -2.3 32.52 ± 2% perf-profile.calltrace.cycles-pp.__kmem_cache_free_bulk.rcu_free_sheaf.rcu_do_batch.rcu_core.handle_softirqs
28.98 -17.2 11.74 ± 2% -10.3 18.65 perf-profile.calltrace.cycles-pp.__munmap
26.57 -15.6 11.00 ± 2% -9.2 17.41 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__munmap
26.46 -15.5 10.96 ± 2% -9.1 17.34 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__munmap
26.06 -15.3 10.81 ± 2% -9.0 17.08 perf-profile.calltrace.cycles-pp.__x64_sys_munmap.do_syscall_64.entry_SYSCALL_64_after_hwframe.__munmap
26.05 -15.2 10.80 ± 2% -9.0 17.07 perf-profile.calltrace.cycles-pp.__vm_munmap.__x64_sys_munmap.do_syscall_64.entry_SYSCALL_64_after_hwframe.__munmap
25.26 -14.8 10.49 ± 2% -8.7 16.54 perf-profile.calltrace.cycles-pp.do_vmi_munmap.__vm_munmap.__x64_sys_munmap.do_syscall_64.entry_SYSCALL_64_after_hwframe
37.77 -14.5 23.29 ± 3% -4.1 33.62 ± 2% perf-profile.calltrace.cycles-pp.kthread.ret_from_fork.ret_from_fork_asm
37.77 -14.5 23.29 ± 3% -4.1 33.62 ± 2% perf-profile.calltrace.cycles-pp.ret_from_fork.ret_from_fork_asm
37.77 -14.5 23.29 ± 3% -4.1 33.62 ± 2% perf-profile.calltrace.cycles-pp.ret_from_fork_asm
24.74 -14.5 10.29 ± 2% -8.5 16.22 perf-profile.calltrace.cycles-pp.do_vmi_align_munmap.do_vmi_munmap.__vm_munmap.__x64_sys_munmap.do_syscall_64
11.67 -11.7 0.00 -11.7 0.00 perf-profile.calltrace.cycles-pp.__put_partials.__kmem_cache_free_bulk.rcu_free_sheaf.rcu_do_batch.rcu_core
11.47 -11.5 0.00 -11.5 0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__put_partials.__kmem_cache_free_bulk.rcu_free_sheaf.rcu_do_batch
11.41 -11.4 0.00 -11.4 0.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__put_partials.__kmem_cache_free_bulk.rcu_free_sheaf
15.33 -9.5 5.79 ± 2% -5.6 9.69 perf-profile.calltrace.cycles-pp.vms_complete_munmap_vmas.do_vmi_align_munmap.do_vmi_munmap.__vm_munmap.__x64_sys_munmap
13.18 -8.2 4.99 ± 3% -4.8 8.41 perf-profile.calltrace.cycles-pp.vms_clear_ptes.vms_complete_munmap_vmas.do_vmi_align_munmap.do_vmi_munmap.__vm_munmap
22.58 ± 2% -5.4 17.20 ± 5% +9.4 32.00 ± 2% perf-profile.calltrace.cycles-pp.__slab_free.__kmem_cache_free_bulk.rcu_free_sheaf.rcu_do_batch.rcu_core
8.36 -5.2 3.19 ± 3% -3.0 5.38 perf-profile.calltrace.cycles-pp.unmap_vmas.vms_clear_ptes.vms_complete_munmap_vmas.do_vmi_align_munmap.do_vmi_munmap
8.03 -4.9 3.09 ± 3% -2.8 5.22 perf-profile.calltrace.cycles-pp.unmap_page_range.unmap_vmas.vms_clear_ptes.vms_complete_munmap_vmas.do_vmi_align_munmap
21.18 ± 2% -4.4 16.76 ± 5% +9.9 31.10 ± 2% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__slab_free.__kmem_cache_free_bulk.rcu_free_sheaf.rcu_do_batch
20.93 ± 2% -4.4 16.55 ± 5% +9.7 30.68 ± 2% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__slab_free.__kmem_cache_free_bulk.rcu_free_sheaf
7.13 -4.4 2.77 ± 3% -2.4 4.68 perf-profile.calltrace.cycles-pp.zap_pmd_range.unmap_page_range.unmap_vmas.vms_clear_ptes.vms_complete_munmap_vmas
4.39 -2.8 1.55 ± 2% -1.7 2.69 perf-profile.calltrace.cycles-pp.perf_event_mmap.__mmap_region.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff
5.89 -2.8 3.12 ± 2% -1.7 4.16 ± 2% perf-profile.calltrace.cycles-pp.mas_store_gfp.do_vmi_align_munmap.do_vmi_munmap.__vm_munmap.__x64_sys_munmap
4.17 -2.6 1.56 ± 2% -1.6 2.62 perf-profile.calltrace.cycles-pp.free_pgtables.vms_clear_ptes.vms_complete_munmap_vmas.do_vmi_align_munmap.do_vmi_munmap
5.44 -2.6 2.89 ± 2% -1.7 3.76 ± 2% perf-profile.calltrace.cycles-pp.mas_store_prealloc.__mmap_new_vma.__mmap_region.do_mmap.vm_mmap_pgoff
3.87 -2.5 1.34 ± 2% -1.6 2.31 perf-profile.calltrace.cycles-pp.perf_event_mmap_event.perf_event_mmap.__mmap_region.do_mmap.vm_mmap_pgoff
3.67 -2.5 1.22 ± 3% -1.6 2.06 perf-profile.calltrace.cycles-pp.__cond_resched.zap_pmd_range.unmap_page_range.unmap_vmas.vms_clear_ptes
4.81 -2.2 2.62 ± 2% -1.5 3.33 ± 2% perf-profile.calltrace.cycles-pp.mas_wr_node_store.mas_store_prealloc.__mmap_new_vma.__mmap_region.do_mmap
4.66 -2.1 2.60 ± 2% -1.4 3.30 ± 2% perf-profile.calltrace.cycles-pp.mas_wr_node_store.mas_store_gfp.do_vmi_align_munmap.do_vmi_munmap.__vm_munmap
2.90 -1.8 1.09 ± 2% -1.0 1.87 perf-profile.calltrace.cycles-pp.free_pgd_range.free_pgtables.vms_clear_ptes.vms_complete_munmap_vmas.do_vmi_align_munmap
2.81 -1.8 1.06 ± 2% -1.0 1.81 perf-profile.calltrace.cycles-pp.free_p4d_range.free_pgd_range.free_pgtables.vms_clear_ptes.vms_complete_munmap_vmas
2.78 -1.7 1.05 ± 3% -1.0 1.80 perf-profile.calltrace.cycles-pp.__get_unmapped_area.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64
2.84 -1.7 1.12 ± 3% -0.9 1.91 perf-profile.calltrace.cycles-pp.vms_gather_munmap_vmas.do_vmi_align_munmap.do_vmi_munmap.__vm_munmap.__x64_sys_munmap
2.70 -1.7 1.02 ± 2% -1.0 1.74 perf-profile.calltrace.cycles-pp.shmem_get_unmapped_area.__get_unmapped_area.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff
2.66 -1.7 1.00 ± 3% -1.0 1.71 perf-profile.calltrace.cycles-pp.free_pud_range.free_p4d_range.free_pgd_range.free_pgtables.vms_clear_ptes
2.45 -1.5 0.93 ± 3% -0.9 1.58 perf-profile.calltrace.cycles-pp.arch_get_unmapped_area_topdown.shmem_get_unmapped_area.__get_unmapped_area.do_mmap.vm_mmap_pgoff
1.42 -1.4 0.05 ±300% -0.6 0.80 perf-profile.calltrace.cycles-pp.vm_area_alloc.__mmap_new_vma.__mmap_region.do_mmap.vm_mmap_pgoff
1.97 -1.2 0.76 ± 2% -0.7 1.27 perf-profile.calltrace.cycles-pp.vm_unmapped_area.arch_get_unmapped_area_topdown.shmem_get_unmapped_area.__get_unmapped_area.do_mmap
1.93 -1.2 0.75 ± 3% -0.7 1.25 perf-profile.calltrace.cycles-pp.unmapped_area_topdown.vm_unmapped_area.arch_get_unmapped_area_topdown.shmem_get_unmapped_area.__get_unmapped_area
2.48 -1.1 1.35 ± 2% -0.9 1.60 ± 2% perf-profile.calltrace.cycles-pp.__pi_memcpy.mas_wr_node_store.mas_store_prealloc.__mmap_new_vma.__mmap_region
2.45 ± 2% -1.1 1.34 ± 2% -0.9 1.59 perf-profile.calltrace.cycles-pp.__pi_memcpy.mas_wr_node_store.mas_store_gfp.do_vmi_align_munmap.do_vmi_munmap
0.99 -1.0 0.00 -0.4 0.57 ± 4% perf-profile.calltrace.cycles-pp.d_path.perf_event_mmap_event.perf_event_mmap.__mmap_region.do_mmap
0.87 -0.9 0.00 -0.3 0.57 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64.__mmap
0.85 -0.9 0.00 -0.3 0.56 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64.__munmap
0.85 -0.8 0.00 -0.3 0.58 perf-profile.calltrace.cycles-pp.mas_empty_area_rev.unmapped_area_topdown.vm_unmapped_area.arch_get_unmapped_area_topdown.shmem_get_unmapped_area
0.83 ± 8% -0.8 0.00 -0.2 0.59 ± 11% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.barn_replace_empty_sheaf.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes
0.82 ± 8% -0.8 0.00 -0.2 0.58 ± 12% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.barn_replace_empty_sheaf.__pcs_replace_empty_main.kmem_cache_alloc_noprof
1.02 ± 7% -0.8 0.21 ±122% -0.3 0.70 ± 10% perf-profile.calltrace.cycles-pp.barn_replace_empty_sheaf.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes.mas_preallocate
0.79 -0.8 0.00 -0.2 0.60 perf-profile.calltrace.cycles-pp.zap_pte_range.zap_pmd_range.unmap_page_range.unmap_vmas.vms_clear_ptes
0.74 ± 4% -0.7 0.05 ±300% -0.1 0.63 ± 6% perf-profile.calltrace.cycles-pp.kvfree_call_rcu.mas_wr_node_store.mas_store_prealloc.__mmap_new_vma.__mmap_region
0.00 +0.0 0.00 +0.7 0.69 perf-profile.calltrace.cycles-pp.allocate_slab.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes
0.00 +0.8 0.80 ± 12% +0.0 0.00 perf-profile.calltrace.cycles-pp.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.kmem_cache_free_bulk.kvfree_rcu_bulk
0.00 +0.8 0.80 ± 12% +0.0 0.00 perf-profile.calltrace.cycles-pp.handle_softirqs.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.kmem_cache_free_bulk
0.00 +0.8 0.80 ± 11% +0.0 0.00 perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_monitor.process_one_work
0.00 +0.8 0.80 ± 11% +0.0 0.00 perf-profile.calltrace.cycles-pp.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_monitor
1.02 ± 14% +0.8 1.82 ± 11% -1.0 0.00 perf-profile.calltrace.cycles-pp.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_work.process_one_work.worker_thread
1.02 ± 14% +0.8 1.83 ± 11% -1.0 0.00 perf-profile.calltrace.cycles-pp.kvfree_rcu_bulk.kfree_rcu_work.process_one_work.worker_thread.kthread
1.02 ± 14% +0.8 1.84 ± 11% -1.0 0.00 perf-profile.calltrace.cycles-pp.kfree_rcu_work.process_one_work.worker_thread.kthread.ret_from_fork
0.00 +0.8 0.85 ± 30% +0.0 0.00 perf-profile.calltrace.cycles-pp.rcu_free_sheaf.rcu_do_batch.rcu_core.handle_softirqs.__irq_exit_rcu
0.00 +0.9 0.86 ± 30% +0.0 0.00 perf-profile.calltrace.cycles-pp.rcu_core.handle_softirqs.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
0.00 +0.9 0.86 ± 30% +0.0 0.00 perf-profile.calltrace.cycles-pp.rcu_do_batch.rcu_core.handle_softirqs.__irq_exit_rcu.sysvec_apic_timer_interrupt
0.00 +1.5 1.45 ± 14% +0.0 0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__slab_free.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_work
0.00 +1.5 1.48 ± 14% +0.0 0.00 perf-profile.calltrace.cycles-pp.__slab_free.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_work.process_one_work
0.00 +2.7 2.68 ± 15% +0.0 0.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__slab_free.__refill_objects_node.__refill_objects_any
0.00 +2.7 2.69 ± 15% +0.0 0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__slab_free.__refill_objects_node.__refill_objects_any.refill_objects
0.00 +2.7 2.70 ± 15% +0.0 0.00 perf-profile.calltrace.cycles-pp.__slab_free.__refill_objects_node.__refill_objects_any.refill_objects.__pcs_replace_empty_main
0.00 +2.8 2.80 ± 10% +0.0 0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__slab_free.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_monitor
0.00 +2.9 2.87 ± 10% +0.0 0.00 perf-profile.calltrace.cycles-pp.__slab_free.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_monitor.process_one_work
0.56 ± 76% +3.7 4.24 ± 10% -0.6 0.00 perf-profile.calltrace.cycles-pp.kmem_cache_free_bulk.kvfree_rcu_bulk.kfree_rcu_monitor.process_one_work.worker_thread
0.56 ± 76% +3.7 4.24 ± 10% -0.6 0.00 perf-profile.calltrace.cycles-pp.kvfree_rcu_bulk.kfree_rcu_monitor.process_one_work.worker_thread.kthread
0.57 ± 76% +3.7 4.27 ± 10% -0.6 0.00 perf-profile.calltrace.cycles-pp.kfree_rcu_monitor.process_one_work.worker_thread.kthread.ret_from_fork
0.00 +4.2 4.23 ± 9% +0.0 0.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__slab_free.kmem_cache_free_bulk.kvfree_rcu_bulk
1.75 ± 24% +4.4 6.13 ± 7% -1.7 0.00 perf-profile.calltrace.cycles-pp.worker_thread.kthread.ret_from_fork.ret_from_fork_asm
1.74 ± 24% +4.4 6.12 ± 7% -1.7 0.00 perf-profile.calltrace.cycles-pp.process_one_work.worker_thread.kthread.ret_from_fork.ret_from_fork_asm
0.00 +5.7 5.71 ± 7% +2.3 2.32 ± 9% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__slab_free.__refill_objects_node.refill_objects
0.00 +5.7 5.74 ± 7% +2.3 2.34 ± 9% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__slab_free.__refill_objects_node.refill_objects.__pcs_replace_empty_main
0.00 +5.8 5.77 ± 7% +2.4 2.36 ± 9% perf-profile.calltrace.cycles-pp.__slab_free.__refill_objects_node.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof
0.00 +6.0 6.04 ± 7% +7.4 7.38 ± 2% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.alloc_from_new_slab.refill_objects.__pcs_replace_empty_main
0.00 +6.1 6.12 ± 7% +7.5 7.50 ± 2% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.alloc_from_new_slab.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof
0.00 +6.3 6.34 ± 7% +7.9 7.86 ± 2% perf-profile.calltrace.cycles-pp.alloc_from_new_slab.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes
0.00 +13.5 13.47 ± 3% +0.0 0.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__refill_objects_node.__refill_objects_any.refill_objects
0.00 +13.5 13.54 ± 3% +0.0 0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__refill_objects_node.__refill_objects_any.refill_objects.__pcs_replace_empty_main
0.00 +17.2 17.21 ± 5% +0.0 0.00 perf-profile.calltrace.cycles-pp.__refill_objects_node.__refill_objects_any.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof
0.00 +17.4 17.38 ± 5% +0.0 0.00 perf-profile.calltrace.cycles-pp.__refill_objects_any.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes
0.00 +21.2 21.21 +15.2 15.18 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__refill_objects_node.refill_objects.__pcs_replace_empty_main
0.00 +21.3 21.33 +15.3 15.35 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__refill_objects_node.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof
0.00 +29.1 29.07 ± 2% +20.0 19.95 ± 2% perf-profile.calltrace.cycles-pp.__refill_objects_node.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes
33.88 +30.8 64.62 +13.5 47.33 perf-profile.calltrace.cycles-pp.__mmap
31.33 +32.5 63.82 +14.7 45.98 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__mmap
31.21 +32.6 63.78 +14.7 45.90 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__mmap
30.70 +32.9 63.58 +14.9 45.55 perf-profile.calltrace.cycles-pp.ksys_mmap_pgoff.do_syscall_64.entry_SYSCALL_64_after_hwframe.__mmap
30.10 +33.2 63.34 +15.0 45.14 perf-profile.calltrace.cycles-pp.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64.entry_SYSCALL_64_after_hwframe.__mmap
29.42 +33.6 63.06 +15.3 44.68 perf-profile.calltrace.cycles-pp.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64.entry_SYSCALL_64_after_hwframe
26.13 +35.7 61.80 +16.3 42.47 perf-profile.calltrace.cycles-pp.__mmap_region.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64
15.61 ± 3% +42.5 58.07 +20.5 36.07 perf-profile.calltrace.cycles-pp.__mmap_new_vma.__mmap_region.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff
7.84 ± 6% +46.5 54.35 ± 2% +23.1 30.93 perf-profile.calltrace.cycles-pp.mas_preallocate.__mmap_new_vma.__mmap_region.do_mmap.vm_mmap_pgoff
6.97 ± 7% +47.0 54.02 ± 2% +23.4 30.36 perf-profile.calltrace.cycles-pp.mas_alloc_nodes.mas_preallocate.__mmap_new_vma.__mmap_region.do_mmap
6.93 ± 7% +47.1 54.00 ± 2% +23.4 30.34 perf-profile.calltrace.cycles-pp.kmem_cache_alloc_noprof.mas_alloc_nodes.mas_preallocate.__mmap_new_vma.__mmap_region
6.66 ± 7% +47.1 53.78 ± 2% +23.3 29.93 perf-profile.calltrace.cycles-pp.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes.mas_preallocate.__mmap_new_vma
0.00 +53.1 53.08 ± 2% +28.7 28.73 perf-profile.calltrace.cycles-pp.refill_objects.__pcs_replace_empty_main.kmem_cache_alloc_noprof.mas_alloc_nodes.mas_preallocate
36.02 ± 2% -18.9 17.16 ± 4% -2.5 33.54 ± 2% perf-profile.children.cycles-pp.smpboot_thread_fn
36.01 ± 2% -18.9 17.16 ± 4% -2.5 33.53 ± 2% perf-profile.children.cycles-pp.run_ksoftirqd
38.02 -18.0 20.05 ± 4% -3.5 34.57 ± 2% perf-profile.children.cycles-pp.handle_softirqs
38.00 -18.0 20.04 ± 4% -3.4 34.56 ± 2% perf-profile.children.cycles-pp.rcu_core
38.00 -18.0 20.04 ± 4% -3.4 34.56 ± 2% perf-profile.children.cycles-pp.rcu_do_batch
37.17 -17.6 19.57 ± 4% -3.3 33.87 ± 2% perf-profile.children.cycles-pp.rcu_free_sheaf
36.67 -17.3 19.34 ± 4% -3.2 33.47 ± 2% perf-profile.children.cycles-pp.__kmem_cache_free_bulk
28.93 -17.0 11.92 ± 2% -10.0 18.97 perf-profile.children.cycles-pp.__munmap
26.06 -15.3 10.81 ± 2% -9.0 17.08 perf-profile.children.cycles-pp.__x64_sys_munmap
26.06 -15.2 10.80 ± 2% -9.0 17.07 perf-profile.children.cycles-pp.__vm_munmap
25.26 -14.8 10.49 ± 2% -8.7 16.54 perf-profile.children.cycles-pp.do_vmi_munmap
37.77 -14.5 23.29 ± 3% -4.1 33.62 ± 2% perf-profile.children.cycles-pp.kthread
37.77 -14.5 23.29 ± 3% -4.1 33.62 ± 2% perf-profile.children.cycles-pp.ret_from_fork
37.77 -14.5 23.29 ± 3% -4.1 33.62 ± 2% perf-profile.children.cycles-pp.ret_from_fork_asm
24.76 -14.5 10.30 ± 2% -8.5 16.23 perf-profile.children.cycles-pp.do_vmi_align_munmap
13.10 -13.1 0.00 -13.1 0.00 perf-profile.children.cycles-pp.__put_partials
15.38 -9.6 5.81 ± 2% -5.7 9.71 perf-profile.children.cycles-pp.vms_complete_munmap_vmas
13.20 -8.2 5.00 ± 2% -4.8 8.42 perf-profile.children.cycles-pp.vms_clear_ptes
8.37 -5.2 3.19 ± 3% -3.0 5.38 perf-profile.children.cycles-pp.unmap_vmas
8.11 -5.0 3.09 ± 3% -2.9 5.23 perf-profile.children.cycles-pp.unmap_page_range
7.74 -4.7 3.01 ± 3% -2.7 5.08 perf-profile.children.cycles-pp.zap_pmd_range
9.51 -4.3 5.24 ± 2% -2.9 6.65 ± 2% perf-profile.children.cycles-pp.mas_wr_node_store
6.64 -3.2 3.43 ± 2% -2.0 4.68 perf-profile.children.cycles-pp.mas_store_gfp
4.41 -2.9 1.55 ± 2% -1.7 2.70 perf-profile.children.cycles-pp.perf_event_mmap
4.22 -2.6 1.57 ± 2% -1.6 2.64 perf-profile.children.cycles-pp.free_pgtables
5.44 -2.6 2.89 ± 2% -1.7 3.76 ± 2% perf-profile.children.cycles-pp.mas_store_prealloc
3.90 -2.5 1.35 ± 2% -1.5 2.36 perf-profile.children.cycles-pp.perf_event_mmap_event
3.80 -2.4 1.44 ± 3% -1.4 2.44 perf-profile.children.cycles-pp.__cond_resched
4.96 -2.3 2.71 ± 2% -1.8 3.21 ± 2% perf-profile.children.cycles-pp.__pi_memcpy
2.91 -1.8 1.10 ± 3% -1.0 1.88 perf-profile.children.cycles-pp.free_pgd_range
2.81 -1.8 1.06 ± 2% -1.0 1.81 perf-profile.children.cycles-pp.free_p4d_range
2.78 -1.7 1.05 ± 3% -1.0 1.80 perf-profile.children.cycles-pp.__get_unmapped_area
2.86 -1.7 1.13 ± 2% -0.9 1.94 perf-profile.children.cycles-pp.vms_gather_munmap_vmas
2.70 -1.7 1.02 ± 3% -1.0 1.74 perf-profile.children.cycles-pp.shmem_get_unmapped_area
2.68 -1.7 1.01 ± 2% -1.0 1.72 perf-profile.children.cycles-pp.free_pud_range
2.46 -1.5 0.93 ± 2% -0.9 1.59 perf-profile.children.cycles-pp.arch_get_unmapped_area_topdown
2.23 -1.4 0.82 ± 3% -0.8 1.41 perf-profile.children.cycles-pp.mas_find
1.97 -1.2 0.76 ± 2% -0.7 1.27 perf-profile.children.cycles-pp.vm_unmapped_area
1.94 -1.2 0.75 ± 2% -0.7 1.25 perf-profile.children.cycles-pp.unmapped_area_topdown
1.78 -1.1 0.69 ± 3% -0.6 1.17 perf-profile.children.cycles-pp.entry_SYSCALL_64
1.24 ± 2% -1.0 0.28 ± 7% -0.5 0.70 perf-profile.children.cycles-pp.allocate_slab
1.43 -0.9 0.48 ± 3% -0.6 0.80 perf-profile.children.cycles-pp.vm_area_alloc
1.42 -0.9 0.56 ± 2% -0.5 0.95 perf-profile.children.cycles-pp.entry_SYSRETQ_unsafe_stack
1.34 ± 7% -0.8 0.54 ± 12% -0.5 0.88 ± 10% perf-profile.children.cycles-pp.barn_get_empty_sheaf
0.96 ± 2% -0.8 0.19 ± 6% -0.4 0.52 perf-profile.children.cycles-pp.shuffle_freelist
1.16 -0.7 0.46 ± 3% -0.4 0.77 perf-profile.children.cycles-pp.rcu_all_qs
1.06 -0.7 0.39 ± 2% -0.4 0.64 perf-profile.children.cycles-pp.mas_walk
1.00 -0.7 0.33 ± 3% -0.4 0.58 ± 4% perf-profile.children.cycles-pp.d_path
0.78 ± 2% -0.6 0.15 ± 7% -0.4 0.41 ± 2% perf-profile.children.cycles-pp.setup_object
0.88 -0.6 0.27 ± 3% -0.4 0.48 perf-profile.children.cycles-pp.__build_id_parse
0.87 ± 4% -0.6 0.32 ± 3% -0.4 0.49 ± 2% perf-profile.children.cycles-pp.kmem_cache_free
1.15 ± 5% -0.5 0.60 ± 8% -0.2 0.90 ± 7% perf-profile.children.cycles-pp.__kfree_rcu_sheaf
1.02 ± 7% -0.5 0.48 ± 11% -0.3 0.70 ± 11% perf-profile.children.cycles-pp.barn_replace_empty_sheaf
0.77 -0.5 0.24 ± 4% -0.4 0.41 ± 5% perf-profile.children.cycles-pp.prepend_path
0.84 -0.5 0.32 ± 2% -0.3 0.54 perf-profile.children.cycles-pp.mas_prev_slot
1.06 ± 4% -0.5 0.54 ± 3% -0.2 0.81 ± 3% perf-profile.children.cycles-pp.mas_update_gap
0.86 -0.5 0.35 ± 2% -0.3 0.59 perf-profile.children.cycles-pp.mas_empty_area_rev
0.67 -0.5 0.19 ± 3% -0.4 0.31 ± 2% perf-profile.children.cycles-pp.__memcg_slab_post_alloc_hook
0.76 -0.5 0.29 ± 3% -0.3 0.50 perf-profile.children.cycles-pp.mas_next_slot
0.68 -0.5 0.21 ± 3% -0.3 0.38 perf-profile.children.cycles-pp.shmem_mmap_prepare
0.96 ± 4% -0.5 0.50 ± 3% -0.2 0.74 ± 3% perf-profile.children.cycles-pp.mas_leaf_max_gap
0.80 -0.4 0.36 ± 3% -0.2 0.61 perf-profile.children.cycles-pp.zap_pte_range
0.72 ± 2% -0.4 0.28 ± 3% -0.2 0.48 perf-profile.children.cycles-pp.mas_wr_store_type
0.60 -0.4 0.18 ± 3% -0.3 0.34 perf-profile.children.cycles-pp.touch_atime
0.68 -0.4 0.27 ± 3% -0.2 0.45 perf-profile.children.cycles-pp.__vma_start_write
0.61 -0.4 0.23 ± 2% -0.2 0.39 perf-profile.children.cycles-pp.up_write
1.58 ± 5% -0.4 1.20 ± 3% -0.2 1.33 ± 5% perf-profile.children.cycles-pp.kvfree_call_rcu
0.51 -0.4 0.15 ± 4% -0.2 0.28 perf-profile.children.cycles-pp.atime_needs_update
0.62 -0.4 0.26 ± 4% -0.2 0.44 perf-profile.children.cycles-pp.perf_iterate_sb
0.58 -0.4 0.23 ± 3% -0.2 0.40 perf-profile.children.cycles-pp.unlink_file_vma_batch_process
0.59 -0.3 0.25 ± 3% -0.2 0.41 perf-profile.children.cycles-pp.mas_rev_awalk
0.54 -0.3 0.21 ± 2% -0.2 0.36 perf-profile.children.cycles-pp.arch_exit_to_user_mode_prepare
0.43 -0.3 0.12 ± 4% -0.2 0.23 ± 2% perf-profile.children.cycles-pp.security_vm_enough_memory_mm
0.49 -0.3 0.18 ± 3% -0.2 0.31 perf-profile.children.cycles-pp.down_write_killable
0.51 ± 3% -0.3 0.20 ± 3% -0.2 0.31 ± 2% perf-profile.children.cycles-pp.kfree
0.41 -0.3 0.12 ± 3% -0.2 0.19 ± 2% perf-profile.children.cycles-pp.__kmalloc_cache_noprof
0.41 -0.3 0.12 ± 4% -0.2 0.20 ± 2% perf-profile.children.cycles-pp.freader_fetch
0.46 -0.3 0.18 ± 4% -0.1 0.31 perf-profile.children.cycles-pp.__vma_enter_locked
0.45 ± 9% -0.3 0.19 ± 5% -0.2 0.28 ± 2% perf-profile.children.cycles-pp.__memcg_slab_free_hook
0.44 -0.3 0.19 ± 3% -0.1 0.31 perf-profile.children.cycles-pp.down_write
0.65 ± 9% -0.3 0.40 ± 11% -0.1 0.55 ± 12% perf-profile.children.cycles-pp.barn_put_full_sheaf
0.54 ± 2% -0.2 0.29 ± 5% -0.2 0.31 ± 5% perf-profile.children.cycles-pp.build_detached_freelist
0.24 ± 2% -0.2 0.00 -0.2 0.08 ± 6% perf-profile.children.cycles-pp.mas_next_range
0.45 ± 2% -0.2 0.21 ± 3% -0.1 0.38 ± 2% perf-profile.children.cycles-pp.__rcu_free_sheaf_prepare
0.36 -0.2 0.14 ± 3% -0.1 0.23 perf-profile.children.cycles-pp.up_read
0.36 -0.2 0.15 ± 3% -0.1 0.25 perf-profile.children.cycles-pp.fput
0.31 -0.2 0.10 ± 4% -0.1 0.17 perf-profile.children.cycles-pp.remove_vma
0.30 ± 3% -0.2 0.09 ± 3% -0.1 0.17 ± 2% perf-profile.children.cycles-pp.current_time
0.36 ± 2% -0.2 0.15 ± 2% -0.1 0.25 ± 2% perf-profile.children.cycles-pp.fget
0.33 -0.2 0.14 ± 3% -0.1 0.25 perf-profile.children.cycles-pp.__pte_offset_map_lock
0.29 -0.2 0.10 ± 2% -0.1 0.16 ± 2% perf-profile.children.cycles-pp.freader_get_folio
0.29 -0.2 0.12 ± 4% -0.1 0.22 perf-profile.children.cycles-pp.prepend_copy
0.28 -0.2 0.11 ± 3% -0.1 0.18 perf-profile.children.cycles-pp.freader_init_from_file
0.27 ± 2% -0.2 0.10 ± 4% -0.1 0.17 perf-profile.children.cycles-pp.tlb_gather_mmu
0.26 ± 2% -0.2 0.10 ± 4% -0.1 0.18 ± 2% perf-profile.children.cycles-pp.tlb_finish_mmu
0.25 -0.2 0.09 ± 10% -0.1 0.16 ± 3% perf-profile.children.cycles-pp.__alloc_frozen_pages_noprof
0.25 -0.2 0.09 ± 5% -0.2 0.10 ± 3% perf-profile.children.cycles-pp.may_expand_vm
0.24 -0.2 0.08 ± 9% -0.1 0.16 perf-profile.children.cycles-pp.get_page_from_freelist
0.23 -0.2 0.08 ± 3% -0.1 0.13 perf-profile.children.cycles-pp.__filemap_get_folio_mpol
0.25 -0.1 0.10 ± 4% -0.1 0.19 perf-profile.children.cycles-pp.copy_from_kernel_nofault
0.24 -0.1 0.10 ± 5% -0.1 0.16 perf-profile.children.cycles-pp.entry_SYSCALL_64_safe_stack
0.20 ± 3% -0.1 0.06 ± 5% -0.1 0.11 ± 4% perf-profile.children.cycles-pp.ktime_get_coarse_real_ts64_mg
0.23 -0.1 0.09 ± 5% -0.1 0.14 ± 2% perf-profile.children.cycles-pp.mas_prev
0.20 ± 3% -0.1 0.06 ± 6% -0.1 0.10 ± 4% perf-profile.children.cycles-pp.percpu_counter_add_batch
0.26 ± 3% -0.1 0.12 -0.1 0.16 ± 3% perf-profile.children.cycles-pp.hrtimer_interrupt
0.26 ± 3% -0.1 0.13 ± 3% -0.1 0.17 ± 3% perf-profile.children.cycles-pp.__sysvec_apic_timer_interrupt
0.21 ± 2% -0.1 0.07 ± 6% -0.1 0.13 ± 3% perf-profile.children.cycles-pp.vma_mark_detached
0.13 ± 2% -0.1 0.00 -0.0 0.08 ± 4% perf-profile.children.cycles-pp.rmqueue_bulk
0.23 ± 2% -0.1 0.11 ± 4% -0.1 0.17 perf-profile.children.cycles-pp.downgrade_write
0.20 ± 2% -0.1 0.08 ± 5% -0.1 0.14 ± 3% perf-profile.children.cycles-pp.__vm_enough_memory
0.13 ± 5% -0.1 0.00 -0.0 0.09 perf-profile.children.cycles-pp.vma_set_page_prot
0.17 ± 4% -0.1 0.04 ± 51% -0.1 0.07 ± 4% perf-profile.children.cycles-pp.free_frozen_page_commit
0.14 ± 5% -0.1 0.02 ±152% -0.1 0.06 ± 7% perf-profile.children.cycles-pp.free_pcppages_bulk
0.12 ± 7% -0.1 0.01 ±300% -0.1 0.07 ± 7% perf-profile.children.cycles-pp.obj_cgroup_charge_account
0.19 ± 2% -0.1 0.07 ± 4% -0.1 0.12 perf-profile.children.cycles-pp.mas_wr_store_entry
0.19 -0.1 0.08 ± 3% -0.1 0.13 ± 2% perf-profile.children.cycles-pp.vma_merge_new_range
0.11 ± 4% -0.1 0.00 -0.0 0.08 ± 5% perf-profile.children.cycles-pp.cap_capable
0.18 ± 2% -0.1 0.07 ± 4% -0.1 0.12 ± 2% perf-profile.children.cycles-pp.testcase
0.14 ± 3% -0.1 0.04 ± 50% -0.0 0.10 ± 2% perf-profile.children.cycles-pp.rmqueue
0.13 ± 3% -0.1 0.03 ± 81% -0.0 0.09 ± 5% perf-profile.children.cycles-pp.__rmqueue_pcplist
0.10 -0.1 0.00 -0.0 0.05 ± 9% perf-profile.children.cycles-pp.unlink_anon_vmas
0.19 ± 17% -0.1 0.09 ± 15% -0.0 0.15 ± 13% perf-profile.children.cycles-pp.strlen
0.10 ± 4% -0.1 0.00 -0.0 0.06 ± 6% perf-profile.children.cycles-pp.refill_obj_stock
0.09 -0.1 0.00 -0.0 0.06 perf-profile.children.cycles-pp.security_mmap_file
0.22 ± 3% -0.1 0.13 ± 10% -0.1 0.16 ± 4% perf-profile.children.cycles-pp.__free_frozen_pages
0.08 ± 5% -0.1 0.00 -0.0 0.05 ± 7% perf-profile.children.cycles-pp.static_key_count
0.14 ± 3% -0.1 0.05 ± 7% -0.0 0.09 ± 3% perf-profile.children.cycles-pp.perf_event_mmap_output
0.08 ± 3% -0.1 0.00 -0.0 0.06 ± 5% perf-profile.children.cycles-pp.copy_from_kernel_nofault_allowed
0.08 -0.1 0.00 -0.0 0.03 ± 81% perf-profile.children.cycles-pp.vm_get_page_prot
0.14 ± 4% -0.1 0.06 ± 8% -0.0 0.09 ± 9% perf-profile.children.cycles-pp.syscall_return_via_sysret
0.08 ± 5% -0.1 0.00 -0.0 0.05 perf-profile.children.cycles-pp.__account_obj_stock
0.15 ± 4% -0.1 0.08 -0.1 0.10 perf-profile.children.cycles-pp.tick_nohz_handler
0.12 -0.1 0.05 -0.0 0.08 perf-profile.children.cycles-pp.mas_prev_setup
0.12 -0.1 0.05 -0.0 0.08 ± 3% perf-profile.children.cycles-pp.filemap_get_entry
0.14 ± 3% -0.1 0.07 -0.1 0.09 ± 4% perf-profile.children.cycles-pp.update_process_times
0.16 ± 3% -0.1 0.09 -0.1 0.11 ± 3% perf-profile.children.cycles-pp.__hrtimer_run_queues
0.06 ± 4% -0.1 0.00 -0.0 0.04 ± 33% perf-profile.children.cycles-pp.__free_one_page
0.06 ± 8% -0.1 0.00 +0.0 0.08 ± 5% perf-profile.children.cycles-pp.memfd_check_seals_mmap
0.10 -0.0 0.06 ± 9% -0.0 0.08 ± 3% perf-profile.children.cycles-pp.mas_data_end
0.10 ± 4% -0.0 0.06 ± 7% -0.0 0.10 perf-profile.children.cycles-pp.mas_prev_range
0.06 ± 4% -0.0 0.04 ± 50% -0.0 0.05 perf-profile.children.cycles-pp.sched_tick
0.26 ± 5% -0.0 0.24 ± 3% -0.1 0.14 ± 2% perf-profile.children.cycles-pp._raw_spin_lock
0.12 -0.0 0.10 ± 5% +0.1 0.25 perf-profile.children.cycles-pp.__kmalloc_noprof
0.18 ± 4% +0.0 0.20 ± 5% +0.0 0.20 ± 2% perf-profile.children.cycles-pp.perf_session__process_events
0.18 ± 4% +0.0 0.20 ± 5% +0.0 0.20 ± 2% perf-profile.children.cycles-pp.reader__read_event
0.18 ± 4% +0.0 0.20 ± 5% +0.0 0.20 ± 2% perf-profile.children.cycles-pp.record__finish_output
0.05 ± 33% +0.0 0.08 ± 10% +0.0 0.09 ± 5% perf-profile.children.cycles-pp._raw_spin_trylock
0.00 +0.1 0.06 ± 5% +0.0 0.00 perf-profile.children.cycles-pp.get_state_synchronize_rcu_full
0.00 +0.1 0.10 ± 6% +0.3 0.26 perf-profile.children.cycles-pp.__alloc_empty_sheaf
2.26 ± 9% +0.8 3.04 ± 5% -1.1 1.18 ± 2% perf-profile.children.cycles-pp.asm_sysvec_apic_timer_interrupt
2.24 ± 9% +0.8 3.02 ± 5% -1.1 1.16 ± 2% perf-profile.children.cycles-pp.sysvec_apic_timer_interrupt
1.02 ± 14% +0.8 1.84 ± 11% -1.0 0.07 ± 10% perf-profile.children.cycles-pp.kfree_rcu_work
1.97 ± 11% +0.9 2.90 ± 6% -1.0 0.99 ± 3% perf-profile.children.cycles-pp.__irq_exit_rcu
0.69 ± 43% +3.6 4.27 ± 10% -0.7 0.00 perf-profile.children.cycles-pp.kfree_rcu_monitor
1.70 ± 24% +4.4 6.07 ± 7% -1.6 0.07 ± 10% perf-profile.children.cycles-pp.kmem_cache_free_bulk
1.70 ± 24% +4.4 6.07 ± 7% -1.6 0.07 ± 10% perf-profile.children.cycles-pp.kvfree_rcu_bulk
1.75 ± 24% +4.4 6.13 ± 7% -1.7 0.08 ± 6% perf-profile.children.cycles-pp.worker_thread
1.74 ± 24% +4.4 6.12 ± 7% -1.7 0.08 ± 6% perf-profile.children.cycles-pp.process_one_work
0.00 +6.3 6.34 ± 7% +7.9 7.87 ± 2% perf-profile.children.cycles-pp.alloc_from_new_slab
24.08 ± 2% +7.8 31.92 +11.2 35.32 perf-profile.children.cycles-pp.__slab_free
57.97 +16.9 74.89 +5.5 63.47 perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
57.76 +17.1 74.81 +5.6 63.33 perf-profile.children.cycles-pp.do_syscall_64
0.00 +17.4 17.38 ± 5% +0.2 0.23 ± 3% perf-profile.children.cycles-pp.__refill_objects_any
33.78 +31.0 64.80 +13.8 47.62 perf-profile.children.cycles-pp.__mmap
30.70 +32.9 63.58 +14.9 45.55 perf-profile.children.cycles-pp.ksys_mmap_pgoff
30.11 +33.2 63.34 +15.0 45.15 perf-profile.children.cycles-pp.vm_mmap_pgoff
29.42 +33.6 63.07 +15.3 44.68 perf-profile.children.cycles-pp.do_mmap
39.14 +33.8 72.98 +19.3 58.42 perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath
39.50 +34.0 73.54 +19.7 59.20 perf-profile.children.cycles-pp._raw_spin_lock_irqsave
26.20 +35.6 61.82 +16.3 42.53 perf-profile.children.cycles-pp.__mmap_region
15.64 ± 3% +42.4 58.08 +20.5 36.09 perf-profile.children.cycles-pp.__mmap_new_vma
8.11 ± 6% +46.3 54.41 ± 2% +22.9 31.01 perf-profile.children.cycles-pp.kmem_cache_alloc_noprof
0.00 +46.4 46.38 ± 3% +20.2 20.17 ± 2% perf-profile.children.cycles-pp.__refill_objects_node
7.85 ± 6% +46.5 54.35 ± 2% +23.1 30.93 perf-profile.children.cycles-pp.mas_preallocate
6.97 ± 7% +47.0 54.02 ± 2% +23.4 30.36 perf-profile.children.cycles-pp.mas_alloc_nodes
6.66 ± 7% +47.2 53.87 ± 2% +23.5 30.14 perf-profile.children.cycles-pp.__pcs_replace_empty_main
0.00 +53.2 53.16 ± 2% +28.9 28.94 perf-profile.children.cycles-pp.refill_objects
4.11 -2.6 1.53 ± 3% -1.5 2.59 perf-profile.self.cycles-pp.zap_pmd_range
4.84 -2.2 2.64 ± 2% -1.7 3.13 ± 2% perf-profile.self.cycles-pp.__pi_memcpy
2.71 -1.7 1.02 ± 3% -1.0 1.74 perf-profile.self.cycles-pp.__mmap_region
2.63 -1.6 0.98 ± 2% -0.9 1.69 perf-profile.self.cycles-pp.free_pud_range
2.09 -1.3 0.79 ± 3% -0.7 1.35 perf-profile.self.cycles-pp.__cond_resched
1.92 -1.1 0.78 ± 3% -0.6 1.29 perf-profile.self.cycles-pp.mas_wr_node_store
1.43 ± 2% -0.9 0.56 ± 3% -0.5 0.91 perf-profile.self.cycles-pp.__slab_free
1.42 -0.9 0.55 ± 3% -0.5 0.94 perf-profile.self.cycles-pp.entry_SYSRETQ_unsafe_stack
0.88 ± 2% -0.7 0.18 ± 7% -0.4 0.47 perf-profile.self.cycles-pp.shuffle_freelist
1.20 -0.7 0.50 ± 3% -0.4 0.82 perf-profile.self.cycles-pp.mas_store_gfp
1.03 -0.7 0.37 ± 3% -0.4 0.62 perf-profile.self.cycles-pp.mas_walk
0.98 -0.6 0.39 ± 3% -0.3 0.65 perf-profile.self.cycles-pp.rcu_all_qs
0.90 -0.5 0.35 ± 3% -0.3 0.60 perf-profile.self.cycles-pp.entry_SYSCALL_64
0.78 -0.5 0.31 ± 2% -0.3 0.52 perf-profile.self.cycles-pp.mas_prev_slot
0.74 -0.5 0.28 ± 2% -0.3 0.48 perf-profile.self.cycles-pp.mas_next_slot
0.94 ± 5% -0.4 0.49 ± 3% -0.2 0.73 ± 3% perf-profile.self.cycles-pp.mas_leaf_max_gap
0.69 ± 2% -0.4 0.26 ± 2% -0.2 0.46 perf-profile.self.cycles-pp.mas_wr_store_type
0.64 -0.4 0.25 ± 3% -0.2 0.42 perf-profile.self.cycles-pp.do_vmi_align_munmap
0.54 -0.4 0.14 ± 3% -0.3 0.24 perf-profile.self.cycles-pp.__memcg_slab_post_alloc_hook
0.66 -0.4 0.27 ± 3% -0.2 0.45 perf-profile.self.cycles-pp.__mmap
0.70 -0.4 0.32 ± 3% -0.2 0.51 perf-profile.self.cycles-pp.kmem_cache_alloc_noprof
0.53 -0.4 0.15 ± 8% -0.3 0.25 ± 10% perf-profile.self.cycles-pp.prepend_path
0.61 -0.4 0.23 ± 2% -0.2 0.40 perf-profile.self.cycles-pp.__munmap
0.59 -0.4 0.22 ± 2% -0.2 0.36 perf-profile.self.cycles-pp.unmapped_area_topdown
0.59 -0.4 0.22 ± 3% -0.2 0.38 perf-profile.self.cycles-pp.up_write
0.54 -0.3 0.20 ± 2% -0.2 0.34 perf-profile.self.cycles-pp.mas_find
0.54 -0.3 0.20 ± 3% -0.2 0.35 perf-profile.self.cycles-pp.mas_preallocate
0.55 -0.3 0.22 ± 3% -0.2 0.38 perf-profile.self.cycles-pp.mas_store_prealloc
0.53 -0.3 0.21 ± 3% -0.2 0.35 perf-profile.self.cycles-pp.arch_exit_to_user_mode_prepare
0.52 -0.3 0.21 ± 3% -0.2 0.35 perf-profile.self.cycles-pp.__vm_munmap
0.52 -0.3 0.20 ± 3% -0.2 0.34 perf-profile.self.cycles-pp.perf_event_mmap
0.49 -0.3 0.19 ± 4% -0.2 0.32 perf-profile.self.cycles-pp.vm_area_alloc
0.49 -0.3 0.19 ± 2% -0.2 0.33 perf-profile.self.cycles-pp.mas_rev_awalk
0.44 ± 2% -0.3 0.15 ± 3% -0.2 0.26 perf-profile.self.cycles-pp.__mmap_new_vma
0.45 ± 4% -0.3 0.16 ± 3% -0.2 0.28 ± 2% perf-profile.self.cycles-pp.kfree
0.42 ± 3% -0.3 0.13 ± 4% -0.2 0.20 perf-profile.self.cycles-pp.kmem_cache_free
0.43 -0.3 0.15 ± 2% -0.2 0.28 perf-profile.self.cycles-pp.arch_get_unmapped_area_topdown
0.37 -0.3 0.10 ± 4% -0.2 0.17 perf-profile.self.cycles-pp.__kmalloc_cache_noprof
0.46 -0.3 0.20 ± 5% -0.1 0.34 perf-profile.self.cycles-pp.perf_iterate_sb
0.41 -0.3 0.15 ± 3% -0.1 0.26 perf-profile.self.cycles-pp.down_write_killable
0.43 -0.3 0.17 ± 2% -0.1 0.29 perf-profile.self.cycles-pp.__vma_enter_locked
0.40 ± 8% -0.3 0.14 ± 12% -0.1 0.29 ± 8% perf-profile.self.cycles-pp.perf_event_mmap_event
0.50 -0.2 0.25 ± 3% -0.2 0.31 ± 4% perf-profile.self.cycles-pp.build_detached_freelist
0.45 -0.2 0.21 ± 3% -0.1 0.38 ± 3% perf-profile.self.cycles-pp.__rcu_free_sheaf_prepare
0.37 ± 2% -0.2 0.14 ± 3% -0.1 0.27 perf-profile.self.cycles-pp.__kfree_rcu_sheaf
0.30 -0.2 0.07 ± 4% -0.2 0.12 perf-profile.self.cycles-pp.unmap_page_range
0.38 -0.2 0.16 ± 3% -0.1 0.27 perf-profile.self.cycles-pp.down_write
0.35 ± 12% -0.2 0.13 ± 6% -0.1 0.21 ± 3% perf-profile.self.cycles-pp.__memcg_slab_free_hook
0.35 -0.2 0.13 ± 4% -0.1 0.23 perf-profile.self.cycles-pp.up_read
0.35 -0.2 0.14 ± 3% -0.1 0.24 perf-profile.self.cycles-pp.fput
0.35 ± 2% -0.2 0.15 ± 2% -0.1 0.24 ± 2% perf-profile.self.cycles-pp.fget
0.19 ± 2% -0.2 0.00 -0.1 0.09 perf-profile.self.cycles-pp.__build_id_parse
0.38 ± 2% -0.2 0.19 ± 3% -0.1 0.32 perf-profile.self.cycles-pp.zap_pte_range
0.28 ± 2% -0.2 0.11 ± 6% -0.1 0.17 ± 3% perf-profile.self.cycles-pp.do_mmap
0.27 -0.2 0.10 ± 4% -0.1 0.18 perf-profile.self.cycles-pp.freader_init_from_file
0.26 -0.2 0.10 ± 3% -0.1 0.17 ± 2% perf-profile.self.cycles-pp.tlb_gather_mmu
0.26 -0.2 0.10 ± 4% -0.1 0.17 perf-profile.self.cycles-pp.mas_empty_area_rev
0.27 -0.2 0.11 ± 4% -0.1 0.20 perf-profile.self.cycles-pp.vms_gather_munmap_vmas
0.26 -0.2 0.10 ± 4% -0.1 0.17 perf-profile.self.cycles-pp.tlb_finish_mmu
0.26 -0.2 0.10 ± 4% -0.1 0.17 ± 2% perf-profile.self.cycles-pp.do_syscall_64
0.24 -0.2 0.09 ± 6% -0.1 0.09 ± 5% perf-profile.self.cycles-pp.may_expand_vm
0.22 ± 2% -0.1 0.08 ± 7% -0.1 0.14 ± 7% perf-profile.self.cycles-pp.shmem_get_unmapped_area
0.20 ± 4% -0.1 0.06 ± 5% -0.1 0.10 ± 4% perf-profile.self.cycles-pp.ktime_get_coarse_real_ts64_mg
0.19 ± 3% -0.1 0.05 ± 8% -0.1 0.09 perf-profile.self.cycles-pp.percpu_counter_add_batch
0.20 -0.1 0.07 ± 5% -0.1 0.12 perf-profile.self.cycles-pp.vma_mark_detached
0.23 -0.1 0.10 ± 4% -0.1 0.16 ± 2% perf-profile.self.cycles-pp.downgrade_write
0.21 ± 2% -0.1 0.08 ± 3% -0.1 0.14 ± 3% perf-profile.self.cycles-pp.__vma_start_write
0.18 ± 2% -0.1 0.06 ± 7% -0.1 0.11 ± 3% perf-profile.self.cycles-pp.free_pgtables
0.19 -0.1 0.07 -0.1 0.13 ± 3% perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe
0.17 -0.1 0.05 -0.1 0.09 ± 4% perf-profile.self.cycles-pp.atime_needs_update
0.19 -0.1 0.07 -0.1 0.12 ± 3% perf-profile.self.cycles-pp.entry_SYSCALL_64_safe_stack
0.11 ± 3% -0.1 0.00 -0.0 0.07 ± 7% perf-profile.self.cycles-pp.d_path
0.11 -0.1 0.00 -0.1 0.05 perf-profile.self.cycles-pp.__filemap_get_folio_mpol
0.11 -0.1 0.00 -0.0 0.07 ± 4% perf-profile.self.cycles-pp.__vm_enough_memory
0.10 ± 4% -0.1 0.00 -0.0 0.08 ± 6% perf-profile.self.cycles-pp.cap_capable
0.11 ± 2% -0.1 0.01 ±200% -0.0 0.07 perf-profile.self.cycles-pp.mas_prev_setup
0.15 ± 2% -0.1 0.05 -0.1 0.09 ± 4% perf-profile.self.cycles-pp.security_vm_enough_memory_mm
0.10 ± 4% -0.1 0.00 -0.0 0.06 perf-profile.self.cycles-pp.current_time
0.16 -0.1 0.06 ± 7% -0.0 0.13 ± 2% perf-profile.self.cycles-pp.copy_from_kernel_nofault
0.16 ± 2% -0.1 0.06 ± 6% -0.0 0.11 perf-profile.self.cycles-pp.mas_wr_store_entry
0.18 ± 15% -0.1 0.09 ± 15% -0.0 0.15 ± 15% perf-profile.self.cycles-pp.strlen
0.14 ± 2% -0.1 0.04 ± 33% -0.1 0.08 perf-profile.self.cycles-pp.vms_complete_munmap_vmas
0.09 ± 5% -0.1 0.00 -0.0 0.06 ± 4% perf-profile.self.cycles-pp.free_pgd_range
0.13 ± 2% -0.1 0.04 ± 50% -0.1 0.08 perf-profile.self.cycles-pp.mas_prev
0.09 ± 3% -0.1 0.00 -0.0 0.06 ± 7% perf-profile.self.cycles-pp.mas_update_gap
0.09 ± 3% -0.1 0.00 -0.0 0.07 ± 6% perf-profile.self.cycles-pp.unlink_file_vma_batch_process
0.14 -0.1 0.05 -0.0 0.10 ± 4% perf-profile.self.cycles-pp.free_p4d_range
0.09 -0.1 0.00 -0.0 0.06 ± 6% perf-profile.self.cycles-pp.touch_atime
0.09 ± 3% -0.1 0.00 -0.0 0.06 perf-profile.self.cycles-pp.vms_clear_ptes
0.13 -0.1 0.04 ± 33% -0.1 0.07 perf-profile.self.cycles-pp.unmap_vmas
0.16 -0.1 0.08 ± 6% -0.0 0.12 perf-profile.self.cycles-pp.__pcs_replace_empty_main
0.13 ± 3% -0.1 0.05 ± 8% -0.0 0.09 perf-profile.self.cycles-pp.vma_merge_new_range
0.08 -0.1 0.00 -0.0 0.05 perf-profile.self.cycles-pp.static_key_count
0.08 -0.1 0.00 -0.0 0.05 ± 7% perf-profile.self.cycles-pp.copy_from_kernel_nofault_allowed
0.13 ± 4% -0.1 0.05 -0.0 0.09 ± 4% perf-profile.self.cycles-pp.perf_event_mmap_output
0.14 ± 4% -0.1 0.06 ± 8% -0.0 0.09 ± 9% perf-profile.self.cycles-pp.syscall_return_via_sysret
0.14 ± 2% -0.1 0.06 ± 7% -0.0 0.11 perf-profile.self.cycles-pp.vm_mmap_pgoff
0.12 ± 3% -0.1 0.05 -0.0 0.09 ± 5% perf-profile.self.cycles-pp.testcase
0.07 ± 5% -0.1 0.00 -0.0 0.04 ± 50% perf-profile.self.cycles-pp.__account_obj_stock
0.07 ± 4% -0.1 0.00 -0.0 0.05 perf-profile.self.cycles-pp.unlink_anon_vmas
0.07 -0.1 0.00 -0.0 0.04 ± 65% perf-profile.self.cycles-pp.filemap_get_entry
0.10 -0.1 0.04 ± 65% -0.0 0.09 perf-profile.self.cycles-pp.__pte_offset_map_lock
0.06 ± 4% -0.1 0.00 -0.0 0.05 perf-profile.self.cycles-pp.vma_set_page_prot
0.06 -0.1 0.00 -0.0 0.03 ± 81% perf-profile.self.cycles-pp.__free_one_page
0.10 ± 4% -0.0 0.05 -0.0 0.08 perf-profile.self.cycles-pp.mas_data_end
0.04 ± 50% -0.0 0.00 +0.0 0.08 ± 6% perf-profile.self.cycles-pp.memfd_check_seals_mmap
0.08 ± 5% -0.0 0.05 ± 9% +0.0 0.09 perf-profile.self.cycles-pp.mas_prev_range
0.24 ± 4% -0.0 0.23 ± 3% -0.1 0.13 ± 3% perf-profile.self.cycles-pp._raw_spin_lock
0.00 +0.0 0.00 +0.1 0.06 ± 8% perf-profile.self.cycles-pp.__call_rcu_common
0.33 ± 2% +0.0 0.38 ± 2% +0.1 0.41 perf-profile.self.cycles-pp.kvfree_call_rcu
0.00 +0.1 0.05 +0.0 0.00 perf-profile.self.cycles-pp.kmem_cache_free_bulk
0.00 +0.1 0.06 ± 5% +0.0 0.00 perf-profile.self.cycles-pp.get_state_synchronize_rcu_full
0.00 +0.1 0.07 ± 8% +0.1 0.09 ± 5% perf-profile.self.cycles-pp._raw_spin_trylock
0.00 +0.1 0.10 ± 8% +0.3 0.28 perf-profile.self.cycles-pp.alloc_from_new_slab
0.00 +0.2 0.16 ± 7% +0.2 0.20 ± 2% perf-profile.self.cycles-pp.__refill_objects_any
0.39 +0.2 0.58 ± 3% +0.4 0.78 perf-profile.self.cycles-pp._raw_spin_lock_irqsave
0.00 +2.3 2.33 +2.3 2.26 ± 2% perf-profile.self.cycles-pp.__refill_objects_node
39.14 +33.8 72.98 +19.3 58.42 perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2026-01-30 1:24 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-01-13 13:57 [vbabka:b4/sheaves-for-all-rebased] [slab] aa8fdb9e25: will-it-scale.per_process_ops 46.5% regression kernel test robot
2026-01-28 10:31 ` Vlastimil Babka
2026-01-29 7:05 ` Hao Li
2026-01-29 8:47 ` Vlastimil Babka
2026-01-29 14:49 ` Hao Li
2026-01-30 1:24 ` Oliver Sang
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox