LinuxLists.cc - [linus:master] [swap_state] 5649d113ff: vm-scalability.throughput -33.1% regression

2023-03-20 08:30:58

Subject: [linus:master] [swap_state] 5649d113ff: vm-scalability.throughput -33.1% regression

Hello,

We noticed a -33.1%
commit: 5649d113ffce9f532a In addition to that,
+------------------+------ | testcase: change | test machine | test parameters | | nr_pmem=2 | | nr_task=8 | | priority=1 | | test=swap-w-seq | | +------------------+------

If you fix the issue, | Reported-by: kernel | Link: 9ecc5ab96a93e02efbf283 ("swap_state: update shadow_nodes for anonymous page")
el.org/cgit/linux/kernel/git/torvalds/linux.git">https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
88 threads 2 sockets Intel(R) Xeon(R) Gold 6238M CPU @ 2.10GHz (Cascade Lake) with 128G memory
The motivation behind this suite is to exercise functions and regions of the mm/ of the Linux kernel which are of interest to us.
//git.kernel.org/cgit/linux/kernel/git/wfg/vm-scalability.git/">https://git.kernel.org/cgit/linux/kernel/git/wfg/vm-scalability.git/
the commit also has significant impact on the following tests:
------------------------------------------------------------------------------------------+
| vm-scalability: vm-scalability.throughput -11.1% regression |
| 88 threads 2 sockets Intel(R) Xeon(R) Gold 6238M CPU @ 2.10GHz (Cascade Lake) with 128G memory |
| cpufreq_governor=performance |
|
|
|
|
| thp_defrag=never |
| thp_enabled=never |
------------------------------------------------------------------------------------------+
kindly add following tag
test robot <[email protected]>
lore.kernel.org/oe-lkp/202303201529.87356b9e-yujie.liu@intel.com">https://lore.kernel.org/oe-lkp/[email protected]
===============================================================
kconfig/nr_pmem/nr_task/priority/rootfs/tbox_group/test/testcase/thp_defrag/thp_enabled:
4-rhel-8.3/2/128/1/debian-11.1-x86_64-20220510.cgz/lkp-csl-2sp9/swap-w-seq/vm-scalability/never/never
convert get_hwpoison_huge_page() to folios")
update shadow_nodes for anonymous page")
5649d113ffce9f532a9ecc5ab96
---------------------------
%stddev
\
-33.1% 6702748 ? 2% vm-scalability.throughput
50904 vm-scalability.median
1320 ? 14% vm-scalability.stddev%
1.98 ? 6% vm-scalability.free_time
+5.3% 506880 ? 2% vm-scalability.time.maximum_resident_set_size
1522 ? 2% vm-scalability.time.system_time
77.95 vm-scalability.time.user_time
32.00 ? 12% sched_debug.cfs_rq:/.util_est_enqueued.avg
18.28 ? 5% iostat.cpu.idle
+7.1% 77.40 iostat.cpu.system
4.31 ? 2% iostat.cpu.user
11.02 ? 10% mpstat.cpu.all.idle%
0.90 ? 2% mpstat.cpu.all.irq%
4.66 ? 2% mpstat.cpu.all.usr%
10.81 ? 10% turbostat.CPU%c1
-14.4% 19727636 ? 4% turbostat.IRQ
-1.2% 222.24 turbostat.PkgWatt
-5.4% 15.90 turbostat.RAMWatt
1847 ? 21% vmstat.memory.buff
-20.8% 5126808 ? 4% vmstat.memory.free
-21.1% 2497139 ? 2% vmstat.swap.so
7701 ? 3% vmstat.system.cs
-27.4% 432921 ? 5% vmstat.system.in
1873 ? 22% meminfo.Buffers
1740 ? 21% meminfo.Inactive(file)
-37.1% 110933 meminfo.KReclaimable
-21.3% 4371063 ? 6% meminfo.MemAvailable
-20.4% 4474001 ? 6% meminfo.MemFree
-7.8% 96881 meminfo.PageTables
-37.1% 110933 meminfo.SReclaimable
-17.8% 299245 meminfo.Slab
-26.1% 6329722 ? 2% numa-numastat.node0.local_node
+64.1% 5072859 ? 17% numa-numastat.node0.numa_foreign
-25.6% 6382391 ? 2% numa-numastat.node0.numa_hit
-58.3% 1086131 ? 12% numa-numastat.node0.numa_miss
-56.6% 1139437 ? 13% numa-numastat.node0.other_node
-58.3% 1086128 ? 13% numa-numastat.node1.numa_foreign
+64.0% 5073593 ? 17% numa-numastat.node1.numa_miss
+62.2% 5083376 ? 17% numa-numastat.node1.other_node
10921 ? 36% numa-meminfo.node0.Active
10860 ? 36% numa-meminfo.node0.Active(anon)
-13.5% 11786580 numa-meminfo.node0.AnonPages.max
+180.6% 2688329 ? 2% numa-meminfo.node0.FilePages
-34.1% 2083966 ? 8% numa-meminfo.node0.MemFree
31664 ? 4% numa-meminfo.node0.PageTables
+181.4% 2683325 ? 2% numa-meminfo.node0.Unevictable
+20.3% 12605919 numa-meminfo.node1.AnonPages
+13.6% 15039440 numa-meminfo.node1.AnonPages.max
56959 ?118% numa-meminfo.node1.FilePages
+20.4% 12599017 numa-meminfo.node1.Inactive
+20.4% 12597326 numa-meminfo.node1.Inactive(anon)
1690 ? 19% numa-meminfo.node1.Inactive(file)
33731 ? 7% numa-meminfo.node1.KReclaimable
33731 ? 7% numa-meminfo.node1.SReclaimable
-40.4% 114923 ? 3% numa-meminfo.node1.Slab
51307 ?132% numa-meminfo.node1.Unevictable
23321 ? 2% proc-vmstat.allocstall_movable
+5.9% 5600084 proc-vmstat.nr_anon_pages
-22.6% 106496 ? 3% proc-vmstat.nr_dirty_background_threshold
-22.6% 213254 ? 3% proc-vmstat.nr_dirty_threshold
-22.0% 1105772 ? 3% proc-vmstat.nr_free_pages
+6.3% 5614707 proc-vmstat.nr_inactive_anon
434.83 ? 20% proc-vmstat.nr_inactive_file
2069 proc-vmstat.nr_isolated_anon
-7.0% 24371 ? 2% proc-vmstat.nr_page_table_pages
27738 proc-vmstat.nr_slab_reclaimable
-18.7% 7764551 ? 3% proc-vmstat.nr_vmscan_write
-10.2% 14664852 proc-vmstat.nr_written
+6.3% 5612439 proc-vmstat.nr_zone_inactive_anon
433.67 ? 20% proc-vmstat.nr_zone_inactive_file
+36.2% 7282563 ? 3% proc-vmstat.numa_pte_updates
-24.6% 942522 ? 2% proc-vmstat.pgalloc_dma32
+1.2% 23336109 proc-vmstat.pgalloc_normal
5697 ? 17% proc-vmstat.pgmajfault
-45.4% 3808817 ? 7% proc-vmstat.pgscan_kswapd
-10.2% 14665678 proc-vmstat.pgsteal_anon
-9.8% 14224534 proc-vmstat.pgsteal_direct
-22.3% 441711 ? 2% proc-vmstat.pgsteal_kswapd
5744 ? 17% proc-vmstat.pswpin
-10.2% 14664870 proc-vmstat.pswpout
747751 ? 5% proc-vmstat.slabs_scanned
8981 ? 12% proc-vmstat.workingset_nodes
2690 ? 38% numa-vmstat.node0.nr_active_anon
672173 ? 2% numa-vmstat.node0.nr_file_pages
-31.2% 530210 ? 4% numa-vmstat.node0.nr_free_pages
-50.5% 308.83 ? 10% numa-vmstat.node0.nr_isolated_anon
7970 ? 4% numa-vmstat.node0.nr_page_table_pages
418.50 ? 43% numa-vmstat.node0.nr_swapcached
670830 ? 2% numa-vmstat.node0.nr_unevictable
-59.2% 1672046 ? 2% numa-vmstat.node0.nr_vmscan_write
-56.8% 3037254 ? 2% numa-vmstat.node0.nr_written
2706 ? 38% numa-vmstat.node0.nr_zone_active_anon
670829 ? 2% numa-vmstat.node0.nr_zone_unevictable
+64.1% 5072859 ? 17% numa-vmstat.node0.numa_foreign
-25.6% 6381099 ? 2% numa-vmstat.node0.numa_hit
-26.1% 6328430 ? 2% numa-vmstat.node0.numa_local
-58.3% 1086131 ? 12% numa-vmstat.node0.numa_miss
-56.6% 1139437 ? 13% numa-vmstat.node0.numa_other
9305 ? 11% numa-vmstat.node0.workingset_nodes
88.83 ? 27% numa-vmstat.node1.nr_active_file
+19.3% 3144960 numa-vmstat.node1.nr_anon_pages
14145 ?119% numa-vmstat.node1.nr_file_pages
+19.3% 3141695 numa-vmstat.node1.nr_inactive_anon
421.33 ? 18% numa-vmstat.node1.nr_inactive_file
1788 ? 5% numa-vmstat.node1.nr_isolated_anon
8445 ? 7% numa-vmstat.node1.nr_slab_reclaimable
12826 ?132% numa-vmstat.node1.nr_unevictable
+15.5% 6335562 numa-vmstat.node1.nr_vmscan_write
+25.0% 11627597 numa-vmstat.node1.nr_written
88.83 ? 27% numa-vmstat.node1.nr_zone_active_file
+19.3% 3140718 numa-vmstat.node1.nr_zone_inactive_anon
419.83 ? 18% numa-vmstat.node1.nr_zone_inactive_file
12826 ?132% numa-vmstat.node1.nr_zone_unevictable
-58.3% 1086128 ? 13% numa-vmstat.node1.numa_foreign
+64.0% 5073593 ? 17% numa-vmstat.node1.numa_miss
+62.2% 5083376 ? 17% numa-vmstat.node1.numa_other
97.83 ? 19% numa-vmstat.node1.workingset_nodes
0.00 perf-profile.calltrace.cycles-pp.record__mmap_read_evlist.__cmd_record
2.08 ?223% perf-profile.calltrace.cycles-pp.__cmd_record
0.00 perf-profile.calltrace.cycles-pp.perf_mmap__push.record__mmap_read_evlist.__cmd_record
3.47 ?145% perf-profile.calltrace.cycles-pp.do_filp_open.do_sys_openat2.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe
3.47 ?145% perf-profile.calltrace.cycles-pp.path_openat.do_filp_open.do_sys_openat2.__x64_sys_openat.do_syscall_64
1.39 ?223% perf-profile.calltrace.cycles-pp.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe
1.39 ?223% perf-profile.calltrace.cycles-pp.do_sys_openat2.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.record__pushfn.perf_mmap__push.record__mmap_read_evlist.__cmd_record
0.00 perf-profile.calltrace.cycles-pp.vsnprintf.seq_printf.show_interrupts.seq_read_iter.proc_reg_read_iter
0.00 perf-profile.calltrace.cycles-pp.perf_mmap__read_head.perf_mmap__push.record__mmap_read_evlist.__cmd_record
0.00 perf-profile.calltrace.cycles-pp.lookup_fast.walk_component.link_path_walk.path_lookupat.filename_lookup
0.00 perf-profile.calltrace.cycles-pp.walk_component.link_path_walk.path_lookupat.filename_lookup.vfs_statx
0.00 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.open64
0.00 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.open64
0.00 perf-profile.calltrace.cycles-pp.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe.open64
0.00 perf-profile.calltrace.cycles-pp.do_sys_openat2.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe.open64
1.39 ?223% perf-profile.calltrace.cycles-pp.seq_printf.show_interrupts.seq_read_iter.proc_reg_read_iter.vfs_read
1.39 ?223% perf-profile.calltrace.cycles-pp.open64
0.00 perf-profile.calltrace.cycles-pp.lookup_fast.open_last_lookups.path_openat.do_filp_open.do_sys_openat2
0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock.lockref_put_or_lock.dput.step_into.link_path_walk
0.00 perf-profile.calltrace.cycles-pp.up_read.kernfs_dop_revalidate.lookup_fast.walk_component.link_path_walk
2.38 ?223% perf-profile.calltrace.cycles-pp.ring_buffer_read_head.perf_mmap__read_head.perf_mmap__push.record__mmap_read_evlist.__cmd_record
2.38 ?223% perf-profile.calltrace.cycles-pp.asm_exc_page_fault.ring_buffer_read_head.perf_mmap__read_head.perf_mmap__push.record__mmap_read_evlist
2.38 ?223% perf-profile.calltrace.cycles-pp.exc_page_fault.asm_exc_page_fault.ring_buffer_read_head.perf_mmap__read_head.perf_mmap__push
2.38 ?223% perf-profile.calltrace.cycles-pp.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.ring_buffer_read_head.perf_mmap__read_head
2.38 ?223% perf-profile.calltrace.cycles-pp.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.ring_buffer_read_head
0.00 perf-profile.calltrace.cycles-pp.__perf_sw_event.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault
0.00 perf-profile.calltrace.cycles-pp.lockref_put_or_lock.dput.step_into.link_path_walk.path_openat
0.00 perf-profile.calltrace.cycles-pp.kernfs_dop_revalidate.lookup_fast.walk_component.link_path_walk.path_lookupat
0.00 perf-profile.calltrace.cycles-pp.folio_add_lru.shmem_get_folio_gfp.shmem_write_begin.generic_perform_write.__generic_file_write_iter
0.00 perf-profile.calltrace.cycles-pp.folio_batch_move_lru.folio_add_lru.shmem_get_folio_gfp.shmem_write_begin.generic_perform_write
0.00 perf-profile.calltrace.cycles-pp.lru_add_fn.folio_batch_move_lru.folio_add_lru.shmem_get_folio_gfp.shmem_write_begin
0.00 perf-profile.calltrace.cycles-pp.mem_cgroup_update_lru_size.lru_add_fn.folio_batch_move_lru.folio_add_lru.shmem_get_folio_gfp
0.00 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.getdents64
0.00 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.getdents64
0.00 perf-profile.calltrace.cycles-pp.asm_exc_page_fault.record__mmap_read_evlist.__cmd_record
0.00 perf-profile.calltrace.cycles-pp.exc_page_fault.asm_exc_page_fault.record__mmap_read_evlist.__cmd_record
0.00 perf-profile.calltrace.cycles-pp.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.record__mmap_read_evlist.__cmd_record
0.00 perf-profile.calltrace.cycles-pp.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.record__mmap_read_evlist
0.00 perf-profile.calltrace.cycles-pp.__do_sys_newstat.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.__check_object_size.strncpy_from_user.getname_flags.vfs_fstatat.__do_sys_newstat
0.00 perf-profile.calltrace.cycles-pp.__check_heap_object.__check_object_size.strncpy_from_user.getname_flags.vfs_fstatat
0.00 perf-profile.calltrace.cycles-pp.lockref_put_return.dput.terminate_walk.path_openat.do_filp_open
0.00 perf-profile.calltrace.cycles-pp.next_tgid.proc_pid_readdir.iterate_dir.__x64_sys_getdents64.do_syscall_64
0.00 perf-profile.calltrace.cycles-pp.__x64_sys_getdents64.do_syscall_64.entry_SYSCALL_64_after_hwframe.getdents64
0.00 perf-profile.calltrace.cycles-pp.__do_fault.do_read_fault.do_fault.__handle_mm_fault.handle_mm_fault
0.00 perf-profile.calltrace.cycles-pp.perf_mmap_fault.__do_fault.do_read_fault.do_fault.__handle_mm_fault
0.00 perf-profile.calltrace.cycles-pp.dput.terminate_walk.path_openat.do_filp_open.do_sys_openat2
0.00 perf-profile.calltrace.cycles-pp.iterate_dir.__x64_sys_getdents64.do_syscall_64.entry_SYSCALL_64_after_hwframe.getdents64
0.00 perf-profile.calltrace.cycles-pp.getname_flags.vfs_fstatat.__do_sys_newstat.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.__rb_insert_augmented.mmap_region.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff
0.00 perf-profile.calltrace.cycles-pp.proc_pid_readdir.iterate_dir.__x64_sys_getdents64.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.terminate_walk.path_openat.do_filp_open.do_sys_openat2.__x64_sys_openat
0.00 perf-profile.calltrace.cycles-pp.vfs_fstatat.__do_sys_newstat.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.strncpy_from_user.getname_flags.vfs_fstatat.__do_sys_newstat.do_syscall_64
0.00 perf-profile.calltrace.cycles-pp.getdents64
0.00 perf-profile.calltrace.cycles-pp.__d_lookup_rcu.lookup_fast.walk_component.link_path_walk.path_lookupat
0.00 perf-profile.calltrace.cycles-pp.alloc_empty_file.path_openat.do_filp_open.do_sys_openat2.__x64_sys_openat
0.00 perf-profile.calltrace.cycles-pp.__alloc_file.alloc_empty_file.path_openat.do_filp_open.do_sys_openat2
0.00 perf-profile.calltrace.cycles-pp.___slab_alloc.kmem_cache_alloc.__alloc_file.alloc_empty_file.path_openat
0.00 perf-profile.calltrace.cycles-pp.kmem_cache_alloc.__alloc_file.alloc_empty_file.path_openat.do_filp_open
0.00 perf-profile.calltrace.cycles-pp.swap_range_free.swapcache_free_entries.free_swap_slot.__swap_entry_free.free_swap_and_cache
1.39 ?223% perf-profile.calltrace.cycles-pp.link_path_walk.path_openat.do_filp_open.do_sys_openat2.__x64_sys_openat
3.33 ?223% perf-profile.calltrace.cycles-pp.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64.entry_SYSCALL_64_after_hwframe
3.33 ?223% perf-profile.calltrace.cycles-pp.mmap_region.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64
4.63 ?145% perf-profile.calltrace.cycles-pp.show_interrupts.seq_read_iter.proc_reg_read_iter.vfs_read.ksys_read
0.00 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.sched_setaffinity.evlist__cpu_begin.__evlist__enable.__cmd_record
0.00 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.sched_setaffinity.evlist__cpu_begin.__evlist__enable
0.00 perf-profile.calltrace.cycles-pp.memcpy_erms.vsnprintf.seq_printf.show_interrupts.seq_read_iter
0.00 perf-profile.calltrace.cycles-pp.vma_interval_tree_insert.mmap_region.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff
0.00 perf-profile.calltrace.cycles-pp.__x64_sys_sched_setaffinity.do_syscall_64.entry_SYSCALL_64_after_hwframe.sched_setaffinity.evlist__cpu_begin
0.00 perf-profile.calltrace.cycles-pp.__sched_setaffinity.sched_setaffinity.__x64_sys_sched_setaffinity.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.__set_cpus_allowed_ptr.__sched_setaffinity.sched_setaffinity.__x64_sys_sched_setaffinity.do_syscall_64
0.00 perf-profile.calltrace.cycles-pp.affine_move_task.__set_cpus_allowed_ptr.__sched_setaffinity.sched_setaffinity.__x64_sys_sched_setaffinity
0.00 perf-profile.calltrace.cycles-pp.__wait_for_common.affine_move_task.__set_cpus_allowed_ptr.__sched_setaffinity.sched_setaffinity
0.00 perf-profile.calltrace.cycles-pp.__cond_resched.__wait_for_common.affine_move_task.__set_cpus_allowed_ptr.__sched_setaffinity
0.00 perf-profile.calltrace.cycles-pp.__schedule.__cond_resched.__wait_for_common.affine_move_task.__set_cpus_allowed_ptr
0.00 perf-profile.calltrace.cycles-pp.__mmdrop.finish_task_switch.__schedule.__cond_resched.__wait_for_common
0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.free_pcppages_bulk.free_unref_page.__mmdrop.finish_task_switch
0.00 perf-profile.calltrace.cycles-pp.evlist__cpu_begin.__evlist__enable.__cmd_record
0.00 perf-profile.calltrace.cycles-pp.sched_setaffinity.evlist__cpu_begin.__evlist__enable.__cmd_record
0.00 perf-profile.calltrace.cycles-pp.sched_setaffinity.__x64_sys_sched_setaffinity.do_syscall_64.entry_SYSCALL_64_after_hwframe.sched_setaffinity
0.00 perf-profile.calltrace.cycles-pp.finish_task_switch.__schedule.__cond_resched.__wait_for_common.affine_move_task
0.00 perf-profile.calltrace.cycles-pp.free_unref_page.__mmdrop.finish_task_switch.__schedule.__cond_resched
0.00 perf-profile.calltrace.cycles-pp.free_pcppages_bulk.free_unref_page.__mmdrop.finish_task_switch.__schedule
3.33 ?223% perf-profile.calltrace.cycles-pp.link_path_walk.path_lookupat.filename_lookup.vfs_statx.vfs_fstatat
0.00 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__xstat64
0.00 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.waitid
0.00 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__xstat64
0.00 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.waitid
0.00 perf-profile.calltrace.cycles-pp.pid_revalidate.lookup_fast.open_last_lookups.path_openat.do_filp_open
0.00 perf-profile.calltrace.cycles-pp.pid_revalidate.lookup_fast.walk_component.path_lookupat.filename_lookup
0.00 perf-profile.calltrace.cycles-pp.do_mas_munmap.mmap_region.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff
0.00 perf-profile.calltrace.cycles-pp.do_mas_munmap.__vm_munmap.elf_map.load_elf_interp.load_elf_binary
0.00 perf-profile.calltrace.cycles-pp.do_mas_align_munmap.do_mas_munmap.mmap_region.do_mmap.vm_mmap_pgoff
0.00 perf-profile.calltrace.cycles-pp.do_mas_align_munmap.do_mas_munmap.__vm_munmap.elf_map.load_elf_interp
0.00 perf-profile.calltrace.cycles-pp.unmap_region.do_mas_align_munmap.do_mas_munmap.mmap_region.do_mmap
0.00 perf-profile.calltrace.cycles-pp.unmap_region.do_mas_align_munmap.do_mas_munmap.__vm_munmap.elf_map
0.00 perf-profile.calltrace.cycles-pp.lookup_fast.walk_component.path_lookupat.filename_lookup.vfs_statx
0.00 perf-profile.calltrace.cycles-pp.__do_sys_newstat.do_syscall_64.entry_SYSCALL_64_after_hwframe.__xstat64
0.00 perf-profile.calltrace.cycles-pp.get_pid_task.proc_pid_permission.inode_permission.link_path_walk.path_openat
0.00 perf-profile.calltrace.cycles-pp.lru_add_fn.folio_batch_move_lru.lru_add_drain_cpu.lru_add_drain.unmap_region
0.00 perf-profile.calltrace.cycles-pp.pid_task.pid_revalidate.lookup_fast.walk_component.path_lookupat
0.00 perf-profile.calltrace.cycles-pp.__do_sys_waitid.do_syscall_64.entry_SYSCALL_64_after_hwframe.waitid
0.00 perf-profile.calltrace.cycles-pp.bprm_execve.do_execveat_common.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.exec_binprm.bprm_execve.do_execveat_common.__x64_sys_execve.do_syscall_64
0.00 perf-profile.calltrace.cycles-pp.elf_map.load_elf_interp.load_elf_binary.search_binary_handler.exec_binprm
0.00 perf-profile.calltrace.cycles-pp.__vm_munmap.elf_map.load_elf_interp.load_elf_binary.search_binary_handler
0.00 perf-profile.calltrace.cycles-pp.filename_lookup.vfs_statx.vfs_fstatat.__do_sys_newstat.do_syscall_64
0.00 perf-profile.calltrace.cycles-pp.folio_batch_move_lru.lru_add_drain_cpu.lru_add_drain.unmap_region.do_mas_align_munmap
0.00 perf-profile.calltrace.cycles-pp.shmem_write_end.generic_perform_write.__generic_file_write_iter.generic_file_write_iter.vfs_write
0.00 perf-profile.calltrace.cycles-pp.inode_permission.link_path_walk.path_openat.do_filp_open.do_sys_openat2
0.00 perf-profile.calltrace.cycles-pp.waitid
0.00 perf-profile.calltrace.cycles-pp.kernel_waitid.__do_sys_waitid.do_syscall_64.entry_SYSCALL_64_after_hwframe.waitid
0.00 perf-profile.calltrace.cycles-pp.find_get_pid.kernel_waitid.__do_sys_waitid.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.ksys_mmap_pgoff.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.search_binary_handler.exec_binprm.bprm_execve.do_execveat_common.__x64_sys_execve
0.00 perf-profile.calltrace.cycles-pp.load_elf_binary.search_binary_handler.exec_binprm.bprm_execve.do_execveat_common
0.00 perf-profile.calltrace.cycles-pp.load_elf_interp.load_elf_binary.search_binary_handler.exec_binprm.bprm_execve
0.00 perf-profile.calltrace.cycles-pp.lru_add_drain.unmap_region.do_mas_align_munmap.do_mas_munmap.__vm_munmap
0.00 perf-profile.calltrace.cycles-pp.lru_add_drain_cpu.lru_add_drain.unmap_region.do_mas_align_munmap.do_mas_munmap
0.00 perf-profile.calltrace.cycles-pp.__xstat64
0.00 perf-profile.calltrace.cycles-pp.vfs_fstatat.__do_sys_newstat.do_syscall_64.entry_SYSCALL_64_after_hwframe.__xstat64
0.00 perf-profile.calltrace.cycles-pp.vfs_statx.vfs_fstatat.__do_sys_newstat.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.path_lookupat.filename_lookup.vfs_statx.vfs_fstatat.__do_sys_newstat
0.00 perf-profile.calltrace.cycles-pp.proc_pid_permission.inode_permission.link_path_walk.path_openat.do_filp_open
0.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.lockref_get_not_dead.__legitimize_path.try_to_unlazy
0.00 perf-profile.calltrace.cycles-pp.number.vsnprintf.seq_printf.show_interrupts.seq_read_iter
0.00 perf-profile.calltrace.cycles-pp.page_add_file_rmap.do_set_pte.filemap_map_pages.do_read_fault.do_fault
0.00 perf-profile.calltrace.cycles-pp.fault_in_readable.fault_in_iov_iter_readable.generic_perform_write.__generic_file_write_iter.generic_file_write_iter
0.00 perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt.fault_in_readable.fault_in_iov_iter_readable.generic_perform_write.__generic_file_write_iter
0.00 perf-profile.calltrace.cycles-pp.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.fault_in_readable.fault_in_iov_iter_readable
0.00 perf-profile.calltrace.cycles-pp.__do_softirq.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.fault_in_readable
0.00 perf-profile.calltrace.cycles-pp.asm_exc_page_fault.fopen
0.00 perf-profile.calltrace.cycles-pp.__legitimize_path.try_to_unlazy.lookup_fast.open_last_lookups.path_openat
0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock.lockref_get_not_dead.__legitimize_path.try_to_unlazy.lookup_fast
0.00 perf-profile.calltrace.cycles-pp.fopen
0.00 perf-profile.calltrace.cycles-pp.exc_page_fault.asm_exc_page_fault.fopen
0.00 perf-profile.calltrace.cycles-pp.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.fopen
0.00 perf-profile.calltrace.cycles-pp.filemap_map_pages.do_read_fault.do_fault.__handle_mm_fault.handle_mm_fault
0.00 perf-profile.calltrace.cycles-pp.do_set_pte.filemap_map_pages.do_read_fault.do_fault.__handle_mm_fault
0.00 perf-profile.calltrace.cycles-pp.lockref_get_not_dead.__legitimize_path.try_to_unlazy.lookup_fast.open_last_lookups
0.00 perf-profile.calltrace.cycles-pp.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.fault_in_readable.fault_in_iov_iter_readable.generic_perform_write
0.00 perf-profile.calltrace.cycles-pp.run_rebalance_domains.__do_softirq.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
0.00 perf-profile.calltrace.cycles-pp.try_to_unlazy.lookup_fast.open_last_lookups.path_openat.do_filp_open
0.00 perf-profile.calltrace.cycles-pp.update_blocked_averages.run_rebalance_domains.__do_softirq.__irq_exit_rcu.sysvec_apic_timer_interrupt
0.00 perf-profile.calltrace.cycles-pp.update_rq_clock.update_blocked_averages.run_rebalance_domains.__do_softirq.__irq_exit_rcu
0.00 perf-profile.calltrace.cycles-pp.sched_clock_cpu.update_rq_clock.update_blocked_averages.run_rebalance_domains.__do_softirq
0.00 perf-profile.calltrace.cycles-pp.native_sched_clock.sched_clock_cpu.update_rq_clock.update_blocked_averages.run_rebalance_domains
0.00 perf-profile.calltrace.cycles-pp.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.fopen
2.38 ?223% perf-profile.calltrace.cycles-pp.do_fault.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_page_fault
2.38 ?223% perf-profile.calltrace.cycles-pp.do_read_fault.do_fault.__handle_mm_fault.handle_mm_fault.do_user_addr_fault
1.39 ?223% perf-profile.calltrace.cycles-pp.dput.step_into.link_path_walk.path_openat.do_filp_open
1.39 ?223% perf-profile.calltrace.cycles-pp.step_into.link_path_walk.path_openat.do_filp_open.do_sys_openat2
3.33 ?223% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__mmap
3.33 ?223% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__mmap
3.33 ?223% perf-profile.calltrace.cycles-pp.ksys_mmap_pgoff.do_syscall_64.entry_SYSCALL_64_after_hwframe.__mmap
3.33 ?223% perf-profile.calltrace.cycles-pp.__mmap
3.33 ?223% perf-profile.calltrace.cycles-pp.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64.entry_SYSCALL_64_after_hwframe.__mmap
2.08 ?223% perf-profile.calltrace.cycles-pp.open_last_lookups.path_openat.do_filp_open.do_sys_openat2.__x64_sys_openat
12.66 ? 92% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe
12.66 ? 92% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe
42.54 ? 17% perf-profile.calltrace.cycles-pp.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary
42.54 ? 17% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry
42.54 ? 17% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle
42.54 ? 17% perf-profile.calltrace.cycles-pp.mwait_idle_with_hints.intel_idle.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call
4.76 ?223% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__libc_write.writen.record__pushfn.perf_mmap__push
4.76 ?223% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write.writen.record__pushfn
4.76 ?223% perf-profile.calltrace.cycles-pp.writen.record__pushfn.perf_mmap__push.record__mmap_read_evlist.__cmd_record
4.76 ?223% perf-profile.calltrace.cycles-pp.__libc_write.writen.record__pushfn.perf_mmap__push.record__mmap_read_evlist
4.76 ?223% perf-profile.calltrace.cycles-pp.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write.writen
4.76 ?223% perf-profile.calltrace.cycles-pp.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write
4.76 ?223% perf-profile.calltrace.cycles-pp.generic_file_write_iter.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe
4.76 ?223% perf-profile.calltrace.cycles-pp.__generic_file_write_iter.generic_file_write_iter.vfs_write.ksys_write.do_syscall_64
4.76 ?223% perf-profile.calltrace.cycles-pp.generic_perform_write.__generic_file_write_iter.generic_file_write_iter.vfs_write.ksys_write
2.08 ?223% perf-profile.calltrace.cycles-pp.do_swap_page.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_page_fault
2.08 ?223% perf-profile.calltrace.cycles-pp.__folio_alloc.vma_alloc_folio.do_swap_page.__handle_mm_fault.handle_mm_fault
2.08 ?223% perf-profile.calltrace.cycles-pp.__alloc_pages.__folio_alloc.vma_alloc_folio.do_swap_page.__handle_mm_fault
2.08 ?223% perf-profile.calltrace.cycles-pp.vma_alloc_folio.do_swap_page.__handle_mm_fault.handle_mm_fault.do_user_addr_fault
4.72 ?158% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__lxstat64
4.72 ?158% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__lxstat64
4.72 ?158% perf-profile.calltrace.cycles-pp.__do_sys_newlstat.do_syscall_64.entry_SYSCALL_64_after_hwframe.__lxstat64
4.72 ?158% perf-profile.calltrace.cycles-pp.filename_lookup.vfs_statx.vfs_fstatat.__do_sys_newlstat.do_syscall_64
4.72 ?158% perf-profile.calltrace.cycles-pp.path_lookupat.filename_lookup.vfs_statx.vfs_fstatat.__do_sys_newlstat
4.72 ?158% perf-profile.calltrace.cycles-pp.__lxstat64
4.72 ?158% perf-profile.calltrace.cycles-pp.vfs_fstatat.__do_sys_newlstat.do_syscall_64.entry_SYSCALL_64_after_hwframe.__lxstat64
4.72 ?158% perf-profile.calltrace.cycles-pp.vfs_statx.vfs_fstatat.__do_sys_newlstat.do_syscall_64.entry_SYSCALL_64_after_hwframe
2.38 ?223% perf-profile.calltrace.cycles-pp.shmem_write_begin.generic_perform_write.__generic_file_write_iter.generic_file_write_iter.vfs_write
2.38 ?223% perf-profile.calltrace.cycles-pp.shmem_get_folio_gfp.shmem_write_begin.generic_perform_write.__generic_file_write_iter.generic_file_write_iter
1.39 ?223% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.execve
1.39 ?223% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.execve
1.39 ?223% perf-profile.calltrace.cycles-pp.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe.execve
1.39 ?223% perf-profile.calltrace.cycles-pp.do_execveat_common.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe.execve
1.39 ?223% perf-profile.calltrace.cycles-pp.execve
1.39 ?223% perf-profile.calltrace.cycles-pp.walk_component.path_lookupat.filename_lookup.vfs_statx.vfs_fstatat
4.23 ?143% perf-profile.calltrace.cycles-pp.free_swap_slot.__swap_entry_free.free_swap_and_cache.zap_pte_range.zap_pmd_range
4.23 ?143% perf-profile.calltrace.cycles-pp.swapcache_free_entries.free_swap_slot.__swap_entry_free.free_swap_and_cache.zap_pte_range
2.08 ?223% perf-profile.calltrace.cycles-pp.__evlist__enable.__cmd_record
6.71 ?103% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.read
6.71 ?103% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.read
6.71 ?103% perf-profile.calltrace.cycles-pp.read
6.71 ?103% perf-profile.calltrace.cycles-pp.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.read
6.71 ?103% perf-profile.calltrace.cycles-pp.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.read
6.71 ?103% perf-profile.calltrace.cycles-pp.proc_reg_read_iter.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe
6.71 ?103% perf-profile.calltrace.cycles-pp.seq_read_iter.proc_reg_read_iter.vfs_read.ksys_read.do_syscall_64
2.38 ?223% perf-profile.calltrace.cycles-pp.asm_sysvec_reschedule_ipi
2.38 ?223% perf-profile.calltrace.cycles-pp.irqentry_exit_to_user_mode.asm_sysvec_reschedule_ipi
2.38 ?223% perf-profile.calltrace.cycles-pp.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_reschedule_ipi
2.38 ?223% perf-profile.calltrace.cycles-pp.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_reschedule_ipi
2.38 ?223% perf-profile.calltrace.cycles-pp.arch_do_signal_or_restart.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_reschedule_ipi
2.38 ?223% perf-profile.calltrace.cycles-pp.get_signal.arch_do_signal_or_restart.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode
2.38 ?223% perf-profile.calltrace.cycles-pp.__mem_cgroup_uncharge_swap.swapcache_free_entries.free_swap_slot.__swap_entry_free.free_swap_and_cache
2.38 ?223% perf-profile.calltrace.cycles-pp.swap_cgroup_record.__mem_cgroup_uncharge_swap.swapcache_free_entries.free_swap_slot.__swap_entry_free
2.38 ?223% perf-profile.calltrace.cycles-pp.fault_in_iov_iter_readable.generic_perform_write.__generic_file_write_iter.generic_file_write_iter.vfs_write
1.39 ?223% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.getxattr
1.39 ?223% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.getxattr
1.39 ?223% perf-profile.calltrace.cycles-pp._raw_spin_trylock.dentry_kill.dput.step_into.link_path_walk
1.39 ?223% perf-profile.calltrace.cycles-pp.strnlen_user.copy_strings.do_execveat_common.__x64_sys_execve.do_syscall_64
1.39 ?223% perf-profile.calltrace.cycles-pp.xattr_resolve_name.__vfs_getxattr.do_getxattr.getxattr.path_getxattr
1.39 ?223% perf-profile.calltrace.cycles-pp.__d_alloc.d_alloc.d_alloc_parallel.__lookup_slow.walk_component
1.39 ?223% perf-profile.calltrace.cycles-pp.__mod_memcg_lruvec_state.mod_objcg_state.memcg_slab_post_alloc_hook.kmem_cache_alloc_lru.__d_alloc
1.39 ?223% perf-profile.calltrace.cycles-pp.do_anonymous_page.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_page_fault
1.39 ?223% perf-profile.calltrace.cycles-pp.__mem_cgroup_charge.do_anonymous_page.__handle_mm_fault.handle_mm_fault.do_user_addr_fault
1.39 ?223% perf-profile.calltrace.cycles-pp.__lookup_slow.walk_component.path_lookupat.filename_lookup.vfs_statx
1.39 ?223% perf-profile.calltrace.cycles-pp.__vfs_getxattr.do_getxattr.getxattr.path_getxattr.do_syscall_64
1.39 ?223% perf-profile.calltrace.cycles-pp.d_alloc_parallel.__lookup_slow.walk_component.path_lookupat.filename_lookup
1.39 ?223% perf-profile.calltrace.cycles-pp.d_alloc.d_alloc_parallel.__lookup_slow.walk_component.path_lookupat
1.39 ?223% perf-profile.calltrace.cycles-pp.copy_strings.do_execveat_common.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
1.39 ?223% perf-profile.calltrace.cycles-pp.do_getxattr.getxattr.path_getxattr.do_syscall_64.entry_SYSCALL_64_after_hwframe
1.39 ?223% perf-profile.calltrace.cycles-pp.dentry_kill.dput.step_into.link_path_walk.path_openat
1.39 ?223% perf-profile.calltrace.cycles-pp.getxattr.path_getxattr.do_syscall_64.entry_SYSCALL_64_after_hwframe.getxattr
1.39 ?223% perf-profile.calltrace.cycles-pp.kmem_cache_alloc_lru.__d_alloc.d_alloc.d_alloc_parallel.__lookup_slow
1.39 ?223% perf-profile.calltrace.cycles-pp.memcg_slab_post_alloc_hook.kmem_cache_alloc_lru.__d_alloc.d_alloc.d_alloc_parallel
1.39 ?223% perf-profile.calltrace.cycles-pp.mod_objcg_state.memcg_slab_post_alloc_hook.kmem_cache_alloc_lru.__d_alloc.d_alloc
1.39 ?223% perf-profile.calltrace.cycles-pp.path_getxattr.do_syscall_64.entry_SYSCALL_64_after_hwframe.getxattr
1.39 ?223% perf-profile.calltrace.cycles-pp.getxattr
7.70 ? 74% perf-profile.calltrace.cycles-pp.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault
44.92 ? 20% perf-profile.calltrace.cycles-pp.secondary_startup_64_no_verify
44.92 ? 20% perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64_no_verify
44.92 ? 20% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
44.92 ? 20% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
44.92 ? 20% perf-profile.calltrace.cycles-pp.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
1.85 ?223% perf-profile.calltrace.cycles-pp._find_next_bit.show_interrupts.seq_read_iter.proc_reg_read_iter.vfs_read
1.85 ?223% perf-profile.calltrace.cycles-pp.copy_mc_fragile.wp_page_copy.__handle_mm_fault.handle_mm_fault.do_user_addr_fault
1.85 ?223% perf-profile.calltrace.cycles-pp.do_swap.sort_r.sort.swapcache_free_entries.free_swap_slot
1.85 ?223% perf-profile.calltrace.cycles-pp.sort.swapcache_free_entries.free_swap_slot.__swap_entry_free.free_swap_and_cache
1.85 ?223% perf-profile.calltrace.cycles-pp.sort_r.sort.swapcache_free_entries.free_swap_slot.__swap_entry_free
1.85 ?223% perf-profile.calltrace.cycles-pp.wp_page_copy.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_page_fault
2.08 ?223% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__open64_nocancel
2.08 ?223% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.write
2.08 ?223% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.sched_setaffinity.evlist_cpu_iterator__next.__evlist__enable.__cmd_record
2.08 ?223% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__open64_nocancel
2.08 ?223% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.write
2.08 ?223% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.sched_setaffinity.evlist_cpu_iterator__next.__evlist__enable
2.08 ?223% perf-profile.calltrace.cycles-pp.__list_del_entry_valid.rmqueue_bulk.rmqueue.get_page_from_freelist.__alloc_pages
2.08 ?223% perf-profile.calltrace.cycles-pp.__x64_sys_sched_setaffinity.do_syscall_64.entry_SYSCALL_64_after_hwframe.sched_setaffinity.evlist_cpu_iterator__next
2.08 ?223% perf-profile.calltrace.cycles-pp.__check_object_size.get_user_cpu_mask.__x64_sys_sched_setaffinity.do_syscall_64.entry_SYSCALL_64_after_hwframe
2.08 ?223% perf-profile.calltrace.cycles-pp.check_heap_object.__check_object_size.get_user_cpu_mask.__x64_sys_sched_setaffinity.do_syscall_64
2.08 ?223% perf-profile.calltrace.cycles-pp.__virt_addr_valid.check_heap_object.__check_object_size.get_user_cpu_mask.__x64_sys_sched_setaffinity
2.08 ?223% perf-profile.calltrace.cycles-pp._raw_spin_lock.d_alloc.d_alloc_parallel.lookup_open.open_last_lookups
2.08 ?223% perf-profile.calltrace.cycles-pp.__wake_up_common_lock.pipe_write.vfs_write.ksys_write.do_syscall_64
2.08 ?223% perf-profile.calltrace.cycles-pp.__wake_up_common.__wake_up_common_lock.pipe_write.vfs_write.ksys_write
2.08 ?223% perf-profile.calltrace.cycles-pp.try_to_wake_up.autoremove_wake_function.__wake_up_common.__wake_up_common_lock.pipe_write
2.08 ?223% perf-profile.calltrace.cycles-pp.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe.__open64_nocancel
2.08 ?223% perf-profile.calltrace.cycles-pp.arch_show_interrupts.seq_read_iter.proc_reg_read_iter.vfs_read.ksys_read
2.08 ?223% perf-profile.calltrace.cycles-pp.vsnprintf.seq_printf.arch_show_interrupts.seq_read_iter.proc_reg_read_iter
2.08 ?223% perf-profile.calltrace.cycles-pp.asm_exc_page_fault.record__pushfn.perf_mmap__push.record__mmap_read_evlist.__cmd_record
2.08 ?223% perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.__wake_up_common_lock.pipe_write.vfs_write
2.08 ?223% perf-profile.calltrace.cycles-pp.do_sys_openat2.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe.__open64_nocancel
2.08 ?223% perf-profile.calltrace.cycles-pp.d_alloc_parallel.lookup_open.open_last_lookups.path_openat.do_filp_open
2.08 ?223% perf-profile.calltrace.cycles-pp.d_alloc.d_alloc_parallel.lookup_open.open_last_lookups.path_openat
2.08 ?223% perf-profile.calltrace.cycles-pp.exc_page_fault.asm_exc_page_fault.record__pushfn.perf_mmap__push.record__mmap_read_evlist
2.08 ?223% perf-profile.calltrace.cycles-pp.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.record__pushfn.perf_mmap__push
2.08 ?223% perf-profile.calltrace.cycles-pp.get_page_from_freelist.__alloc_pages.__folio_alloc.vma_alloc_folio.do_swap_page
2.08 ?223% perf-profile.calltrace.cycles-pp.sched_setaffinity.evlist_cpu_iterator__next.__evlist__enable.__cmd_record
2.08 ?223% perf-profile.calltrace.cycles-pp.get_user_cpu_mask.__x64_sys_sched_setaffinity.do_syscall_64.entry_SYSCALL_64_after_hwframe.sched_setaffinity
2.08 ?223% perf-profile.calltrace.cycles-pp.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.record__pushfn
2.08 ?223% perf-profile.calltrace.cycles-pp.write
2.08 ?223% perf-profile.calltrace.cycles-pp.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.write
2.08 ?223% perf-profile.calltrace.cycles-pp.lookup_open.open_last_lookups.path_openat.do_filp_open.do_sys_openat2
2.08 ?223% perf-profile.calltrace.cycles-pp.pipe_write.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe
2.08 ?223% perf-profile.calltrace.cycles-pp.rmqueue.get_page_from_freelist.__alloc_pages.__folio_alloc.vma_alloc_folio
2.08 ?223% perf-profile.calltrace.cycles-pp.rmqueue_bulk.rmqueue.get_page_from_freelist.__alloc_pages.__folio_alloc
2.08 ?223% perf-profile.calltrace.cycles-pp.seq_printf.arch_show_interrupts.seq_read_iter.proc_reg_read_iter.vfs_read
2.08 ?223% perf-profile.calltrace.cycles-pp.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.write
2.08 ?223% perf-profile.calltrace.cycles-pp.__open64_nocancel
2.08 ?223% perf-profile.calltrace.cycles-pp.evlist_cpu_iterator__next.__evlist__enable.__cmd_record
2.38 ?223% perf-profile.calltrace.cycles-pp._raw_spin_lock.finish_fault.do_read_fault.do_fault.__handle_mm_fault
2.38 ?223% perf-profile.calltrace.cycles-pp.menu_reflect.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary
2.38 ?223% perf-profile.calltrace.cycles-pp.__schedule.schedule.smpboot_thread_fn.kthread.ret_from_fork
2.38 ?223% perf-profile.calltrace.cycles-pp.update_sg_lb_stats.update_sd_lb_stats.find_busiest_group.load_balance.newidle_balance
2.38 ?223% perf-profile.calltrace.cycles-pp.balance_fair.__schedule.schedule.smpboot_thread_fn.kthread
2.38 ?223% perf-profile.calltrace.cycles-pp.free_pcppages_bulk.free_unref_page.exit_mmap.__mmput.exit_mm
2.38 ?223% perf-profile.calltrace.cycles-pp.free_unref_page.exit_mmap.__mmput.exit_mm.do_exit
2.38 ?223% perf-profile.calltrace.cycles-pp.finish_fault.do_read_fault.do_fault.__handle_mm_fault.handle_mm_fault
2.38 ?223% perf-profile.calltrace.cycles-pp.ret_from_fork
2.38 ?223% perf-profile.calltrace.cycles-pp.kthread.ret_from_fork
2.38 ?223% perf-profile.calltrace.cycles-pp.smpboot_thread_fn.kthread.ret_from_fork
2.38 ?223% perf-profile.calltrace.cycles-pp.schedule.smpboot_thread_fn.kthread.ret_from_fork
2.38 ?223% perf-profile.calltrace.cycles-pp.newidle_balance.balance_fair.__schedule.schedule.smpboot_thread_fn
2.38 ?223% perf-profile.calltrace.cycles-pp.load_balance.newidle_balance.balance_fair.__schedule.schedule
2.38 ?223% perf-profile.calltrace.cycles-pp.find_busiest_group.load_balance.newidle_balance.balance_fair.__schedule
2.38 ?223% perf-profile.calltrace.cycles-pp.update_sd_lb_stats.find_busiest_group.load_balance.newidle_balance.balance_fair
2.38 ?223% perf-profile.calltrace.cycles-pp.perf_mmap__read_head.perf_mmap__push.record__mmap_read_evlist.__cmd_record.cmd_record
2.38 ?223% perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.shmem_add_to_page_cache.shmem_get_folio_gfp.shmem_write_begin.generic_perform_write
2.38 ?223% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.swap_cgroup_record.__mem_cgroup_uncharge_swap.swapcache_free_entries.free_swap_slot
2.38 ?223% perf-profile.calltrace.cycles-pp.shmem_add_to_page_cache.shmem_get_folio_gfp.shmem_write_begin.generic_perform_write.__generic_file_write_iter
2.78 ?223% perf-profile.calltrace.cycles-pp.__entry_text_start.open64
5.18 ?149% perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe
5.18 ?149% perf-profile.calltrace.cycles-pp.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe
5.18 ?149% perf-profile.calltrace.cycles-pp.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe
5.18 ?149% perf-profile.calltrace.cycles-pp.arch_do_signal_or_restart.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64
5.18 ?149% perf-profile.calltrace.cycles-pp.get_signal.arch_do_signal_or_restart.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode
3.24 ?143% perf-profile.calltrace.cycles-pp.asm_exc_page_fault
3.24 ?143% perf-profile.calltrace.cycles-pp.exc_page_fault.asm_exc_page_fault
3.24 ?143% perf-profile.calltrace.cycles-pp.do_user_addr_fault.exc_page_fault.asm_exc_page_fault
3.24 ?143% perf-profile.calltrace.cycles-pp.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault
7.57 ?105% perf-profile.calltrace.cycles-pp.do_group_exit.get_signal.arch_do_signal_or_restart.exit_to_user_mode_loop.exit_to_user_mode_prepare
7.57 ?105% perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.get_signal.arch_do_signal_or_restart.exit_to_user_mode_loop
7.57 ?105% perf-profile.calltrace.cycles-pp.exit_mm.do_exit.do_group_exit.get_signal.arch_do_signal_or_restart
7.57 ?105% perf-profile.calltrace.cycles-pp.__mmput.exit_mm.do_exit.do_group_exit.get_signal
7.57 ?105% perf-profile.calltrace.cycles-pp.free_swap_and_cache.zap_pte_range.zap_pmd_range.unmap_page_range.unmap_vmas
7.57 ?105% perf-profile.calltrace.cycles-pp.__swap_entry_free.free_swap_and_cache.zap_pte_range.zap_pmd_range.unmap_page_range
3.33 ?223% perf-profile.calltrace.cycles-pp.mas_replace.mas_wmb_replace.mas_split.mas_wr_bnode.mas_store_prealloc
3.33 ?223% perf-profile.calltrace.cycles-pp.lockref_put_return.dput.step_into.link_path_walk.path_lookupat
3.33 ?223% perf-profile.calltrace.cycles-pp.dput.step_into.link_path_walk.path_lookupat.filename_lookup
3.33 ?223% perf-profile.calltrace.cycles-pp.mas_store_prealloc.mmap_region.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff
3.33 ?223% perf-profile.calltrace.cycles-pp.mas_wr_bnode.mas_store_prealloc.mmap_region.do_mmap.vm_mmap_pgoff
3.33 ?223% perf-profile.calltrace.cycles-pp.mas_split.mas_wr_bnode.mas_store_prealloc.mmap_region.do_mmap
3.33 ?223% perf-profile.calltrace.cycles-pp.mas_wmb_replace.mas_split.mas_wr_bnode.mas_store_prealloc.mmap_region
3.33 ?223% perf-profile.calltrace.cycles-pp.step_into.link_path_walk.path_lookupat.filename_lookup.vfs_statx
4.76 ?223% perf-profile.calltrace.cycles-pp.asm_exc_page_fault.fault_in_readable.fault_in_iov_iter_readable.generic_perform_write.__generic_file_write_iter
6.08 ?146% perf-profile.calltrace.cycles-pp.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
6.08 ?146% perf-profile.calltrace.cycles-pp.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
6.08 ?146% perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
6.08 ?146% perf-profile.calltrace.cycles-pp.__mmput.exit_mm.do_exit.do_group_exit.__x64_sys_exit_group
6.08 ?146% perf-profile.calltrace.cycles-pp.exit_mm.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64
6.84 ?156% perf-profile.calltrace.cycles-pp.record__pushfn.perf_mmap__push.record__mmap_read_evlist.__cmd_record.cmd_record
11.27 ?111% perf-profile.calltrace.cycles-pp.unmap_vmas.exit_mmap.__mmput.exit_mm.do_exit
11.27 ?111% perf-profile.calltrace.cycles-pp.unmap_page_range.unmap_vmas.exit_mmap.__mmput.exit_mm
11.27 ?111% perf-profile.calltrace.cycles-pp.zap_pmd_range.unmap_page_range.unmap_vmas.exit_mmap.__mmput
11.27 ?111% perf-profile.calltrace.cycles-pp.zap_pte_range.zap_pmd_range.unmap_page_range.unmap_vmas.exit_mmap
9.23 ?114% perf-profile.calltrace.cycles-pp.__libc_start_main
9.23 ?114% perf-profile.calltrace.cycles-pp.main.__libc_start_main
9.23 ?114% perf-profile.calltrace.cycles-pp.run_builtin.main.__libc_start_main
9.23 ?114% perf-profile.calltrace.cycles-pp.cmd_record.run_builtin.main.__libc_start_main
9.23 ?114% perf-profile.calltrace.cycles-pp.__cmd_record.cmd_record.run_builtin.main.__libc_start_main
9.23 ?114% perf-profile.calltrace.cycles-pp.record__mmap_read_evlist.__cmd_record.cmd_record.run_builtin.main
9.23 ?114% perf-profile.calltrace.cycles-pp.perf_mmap__push.record__mmap_read_evlist.__cmd_record.cmd_record.run_builtin
13.65 ? 84% perf-profile.calltrace.cycles-pp.exit_mmap.__mmput.exit_mm.do_exit.do_group_exit
5.11 perf-stat.i.MPKI
-10.1% 1.254e+10 perf-stat.i.branch-instructions
-15.0% 31322172 ? 3% perf-stat.i.branch-misses
47.68 perf-stat.i.cache-miss-rate%
-35.0% 1.199e+08 perf-stat.i.cache-misses
-38.6% 2.5e+08 perf-stat.i.cache-references
6921 ? 3% perf-stat.i.context-switches
5.03 perf-stat.i.cpi
+2.3% 2.327e+11 perf-stat.i.cpu-cycles
-27.9% 3277727 ? 7% perf-stat.i.dTLB-load-misses
-10.6% 1.243e+10 perf-stat.i.dTLB-loads
0.15 perf-stat.i.dTLB-store-miss-rate%
-17.0% 5961634 perf-stat.i.dTLB-store-misses
-19.1% 3.739e+09 perf-stat.i.dTLB-stores
-23.5% 2359116 ? 4% perf-stat.i.iTLB-load-misses
-22.0% 268269 ? 8% perf-stat.i.iTLB-loads
-10.3% 4.908e+10 perf-stat.i.instructions
0.24 perf-stat.i.ipc
2.63 perf-stat.i.metric.GHz
-25.8% 560.80 ? 3% perf-stat.i.metric.K/sec
-12.0% 325.72 perf-stat.i.metric.M/sec
-16.7% 1231768 perf-stat.i.minor-faults
78.11 perf-stat.i.node-load-miss-rate%
-21.3% 19080033 perf-stat.i.node-load-misses
-25.8% 4687351 perf-stat.i.node-loads
81.97 perf-stat.i.node-store-miss-rate%
-19.8% 10971410 ? 2% perf-stat.i.node-store-misses
-17.7% 2580680 perf-stat.i.node-stores
-16.7% 1232012 perf-stat.i.page-faults
5.09 perf-stat.overall.MPKI
48.04 perf-stat.overall.cache-miss-rate%
4.77 perf-stat.overall.cpi
1950 perf-stat.overall.cycles-between-cache-misses
0.16 perf-stat.overall.dTLB-store-miss-rate%
20810 ? 3% perf-stat.overall.instructions-per-iTLB-miss
0.21 perf-stat.overall.ipc
80.26 perf-stat.overall.node-load-miss-rate%
8948 perf-stat.overall.path-length
-8.7% 1.179e+10 perf-stat.ps.branch-instructions
-12.9% 29096714 ? 3% perf-stat.ps.branch-misses
-34.2% 1.129e+08 perf-stat.ps.cache-misses
-37.6% 2.351e+08 perf-stat.ps.cache-references
6516 ? 3% perf-stat.ps.context-switches
+2.7% 2.203e+11 perf-stat.ps.cpu-cycles
-27.0% 3088833 ? 7% perf-stat.ps.dTLB-load-misses
-9.2% 1.169e+10 perf-stat.ps.dTLB-loads
-14.9% 5567126 perf-stat.ps.dTLB-store-misses
-17.7% 3.51e+09 perf-stat.ps.dTLB-stores
-22.5% 2221945 ? 4% perf-stat.ps.iTLB-load-misses
-18.7% 247129 ? 8% perf-stat.ps.iTLB-loads
-8.9% 4.617e+10 perf-stat.ps.instructions
-14.6% 1150302 perf-stat.ps.minor-faults
-20.5% 17991646 perf-stat.ps.node-load-misses
-25.1% 4424982 perf-stat.ps.node-loads
-18.9% 10364793 ? 2% perf-stat.ps.node-store-misses
-14.6% 2395268 ? 2% perf-stat.ps.node-stores
-14.6% 1150544 perf-stat.ps.page-faults
+6.6% 9.599e+11 perf-stat.total.instructions
*************************************************************************
threads 2 sockets Intel(R) Xeon(R) Gold 6238M CPU @ 2.10GHz (Cascade Lake) with 128G memory
===============================================================
kconfig/nr_pmem/nr_task/priority/rootfs/tbox_group/test/testcase/thp_defrag/thp_enabled:
4-rhel-8.3/2/8/1/debian-11.1-x86_64-20220510.cgz/lkp-csl-2sp9/swap-w-seq/vm-scalability/never/never
convert get_hwpoison_huge_page() to folios")
update shadow_nodes for anonymous page")
5649d113ffce9f532a9ecc5ab96
---------------------------
%stddev
\
-11.1% 492012 ? 2% vm-scalability.median
-11.1% 3909359 ? 2% vm-scalability.throughput
+9.6% 211.73 ? 3% vm-scalability.time.system_time
9.23 iostat.cpu.system
1.78 ? 2% iostat.cpu.user
13992 ? 70% numa-vmstat.node0.workingset_nodes
8805 ?147% numa-vmstat.node1.workingset_nodes
-8.8% 24.17 perf-node.node-local-load-ratio
+10.8% 2.239e+08 ? 2% perf-node.node-store-misses
-6.5% 1831262 ? 2% vmstat.swap.so
-4.7% 573111 vmstat.system.in
-31.3% 115939 ? 2% meminfo.KReclaimable
-31.3% 115939 ? 2% meminfo.SReclaimable
-14.8% 300764 meminfo.Slab
5027 ? 8% sched_debug.cfs_rq:/.min_vruntime.stddev
-4580 sched_debug.cfs_rq:/.spread0.avg
5032 ? 8% sched_debug.cfs_rq:/.spread0.stddev
28973 ? 2% proc-vmstat.nr_slab_reclaimable
-3.5% 6413330 proc-vmstat.pgsteal_kswapd
324728 ? 20% proc-vmstat.slabs_scanned
+6.6% 390656 proc-vmstat.unevictable_pgs_scanned
22725 ? 22% proc-vmstat.workingset_nodes
8.31 ? 2% perf-stat.i.MPKI
-5.9% 5.471e+09 ? 2% perf-stat.i.branch-instructions
0.34 ? 2% perf-stat.i.branch-miss-rate%
-4.3% 85261416 ? 2% perf-stat.i.cache-misses
-5.5% 5.502e+09 ? 2% perf-stat.i.dTLB-loads
-4.7% 3768612 ? 2% perf-stat.i.dTLB-store-misses
-5.2% 2.365e+09 ? 2% perf-stat.i.dTLB-stores
-5.5% 2.16e+10 ? 2% perf-stat.i.instructions
9519 ? 2% perf-stat.i.instructions-per-iTLB-miss
0.73 ? 2% perf-stat.i.ipc
-5.6% 153.17 ? 2% perf-stat.i.metric.M/sec
-6.7% 698091 ? 2% perf-stat.i.minor-faults
-9.5% 3804975 ? 3% perf-stat.i.node-loads
-6.7% 698208 ? 2% perf-stat.i.page-faults
7.54 ? 2% perf-stat.overall.MPKI
0.33 ? 2% perf-stat.overall.branch-miss-rate%
52.57 ? 2% perf-stat.overall.cache-miss-rate%
1.37 ? 2% perf-stat.overall.cpi
7180 ? 2% perf-stat.overall.instructions-per-iTLB-miss
0.73 ? 2% perf-stat.overall.ipc
75.64 perf-stat.overall.node-load-miss-rate%
82.91 perf-stat.overall.node-store-miss-rate%
6921 perf-stat.overall.path-length
-6.2% 5.292e+09 ? 2% perf-stat.ps.branch-instructions
-4.5% 82733559 ? 2% perf-stat.ps.cache-misses
-5.7% 5.325e+09 ? 2% perf-stat.ps.dTLB-loads
-5.0% 3640859 ? 2% perf-stat.ps.dTLB-store-misses
-5.5% 2.29e+09 ? 2% perf-stat.ps.dTLB-stores
-5.7% 2.09e+10 ? 2% perf-stat.ps.instructions
-7.0% 674261 ? 2% perf-stat.ps.minor-faults
-9.6% 3693484 ? 3% perf-stat.ps.node-loads
-7.0% 674377 ? 2% perf-stat.ps.page-faults
+1.4% 7.424e+11 perf-stat.total.instructions
18.08 ? 17% perf-profile.calltrace.cycles-pp.__munmap
18.08 ? 17% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__munmap
18.08 ? 17% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__munmap
18.08 ? 17% perf-profile.calltrace.cycles-pp.do_mas_munmap.__vm_munmap.__x64_sys_munmap.do_syscall_64.entry_SYSCALL_64_after_hwframe
18.08 ? 17% perf-profile.calltrace.cycles-pp.do_mas_align_munmap.do_mas_munmap.__vm_munmap.__x64_sys_munmap.do_syscall_64
18.08 ? 17% perf-profile.calltrace.cycles-pp.unmap_region.do_mas_align_munmap.do_mas_munmap.__vm_munmap.__x64_sys_munmap
18.08 ? 17% perf-profile.calltrace.cycles-pp.__x64_sys_munmap.do_syscall_64.entry_SYSCALL_64_after_hwframe.__munmap
18.08 ? 17% perf-profile.calltrace.cycles-pp.__vm_munmap.__x64_sys_munmap.do_syscall_64.entry_SYSCALL_64_after_hwframe.__munmap
18.08 ? 17% perf-profile.calltrace.cycles-pp.unmap_vmas.unmap_region.do_mas_align_munmap.do_mas_munmap.__vm_munmap
18.08 ? 17% perf-profile.calltrace.cycles-pp.unmap_page_range.unmap_vmas.unmap_region.do_mas_align_munmap.do_mas_munmap
18.08 ? 17% perf-profile.calltrace.cycles-pp.zap_pmd_range.unmap_page_range.unmap_vmas.unmap_region.do_mas_align_munmap
18.08 ? 17% perf-profile.calltrace.cycles-pp.zap_pte_range.zap_pmd_range.unmap_page_range.unmap_vmas.unmap_region
18.03 ? 17% perf-profile.calltrace.cycles-pp.free_swap_and_cache.zap_pte_range.zap_pmd_range.unmap_page_range.unmap_vmas
17.85 ? 17% perf-profile.calltrace.cycles-pp.__swap_entry_free.free_swap_and_cache.zap_pte_range.zap_pmd_range.unmap_page_range
17.73 ? 17% perf-profile.calltrace.cycles-pp.free_swap_slot.__swap_entry_free.free_swap_and_cache.zap_pte_range.zap_pmd_range
17.45 ? 17% perf-profile.calltrace.cycles-pp.swapcache_free_entries.free_swap_slot.__swap_entry_free.free_swap_and_cache.zap_pte_range
12.34 ? 18% perf-profile.calltrace.cycles-pp._raw_spin_lock.swapcache_free_entries.free_swap_slot.__swap_entry_free.free_swap_and_cache
12.24 ? 18% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.swapcache_free_entries.free_swap_slot.__swap_entry_free
3.86 ? 14% perf-profile.calltrace.cycles-pp.__mem_cgroup_uncharge_swap.swapcache_free_entries.free_swap_slot.__swap_entry_free.free_swap_and_cache
1.51 ? 14% perf-profile.calltrace.cycles-pp.swap_cgroup_record.__mem_cgroup_uncharge_swap.swapcache_free_entries.free_swap_slot.__swap_entry_free
1.36 ? 14% perf-profile.calltrace.cycles-pp.page_counter_uncharge.__mem_cgroup_uncharge_swap.swapcache_free_entries.free_swap_slot.__swap_entry_free
0.98 ? 13% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.swap_cgroup_record.__mem_cgroup_uncharge_swap.swapcache_free_entries.free_swap_slot
0.75 ? 16% perf-profile.calltrace.cycles-pp.mem_cgroup_id_put_many.__mem_cgroup_uncharge_swap.swapcache_free_entries.free_swap_slot.__swap_entry_free
1.80 ? 13% perf-profile.calltrace.cycles-pp.folio_end_writeback.pmem_rw_page.bdev_write_page.__swap_writepage.pageout
3.16 ? 17% perf-profile.calltrace.cycles-pp.do_rw_once
4.36 ? 14% perf-profile.calltrace.cycles-pp.pmem_rw_page.bdev_write_page.__swap_writepage.pageout.shrink_folio_list
1.62 ? 33% perf-profile.calltrace.cycles-pp.rmap_walk_anon.folio_referenced.shrink_folio_list.shrink_inactive_list.shrink_lruvec
1.84 ? 24% perf-profile.calltrace.cycles-pp.folio_referenced.shrink_folio_list.shrink_inactive_list.shrink_lruvec.shrink_node_memcgs
1.75 ? 29% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.list_lru_add.workingset_update_node.xas_store
1.82 ? 31% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.list_lru_del.workingset_update_node.xas_store
5.96 ? 19% perf-profile.calltrace.cycles-pp.smp_call_function_many_cond.on_each_cpu_cond_mask.arch_tlbbatch_flush.try_to_unmap_flush_dirty.shrink_folio_list
6.03 ? 19% perf-profile.calltrace.cycles-pp.on_each_cpu_cond_mask.arch_tlbbatch_flush.try_to_unmap_flush_dirty.shrink_folio_list.shrink_inactive_list
6.10 ? 19% perf-profile.calltrace.cycles-pp.try_to_unmap_flush_dirty.shrink_folio_list.shrink_inactive_list.shrink_lruvec.shrink_node_memcgs
6.09 ? 19% perf-profile.calltrace.cycles-pp.arch_tlbbatch_flush.try_to_unmap_flush_dirty.shrink_folio_list.shrink_inactive_list.shrink_lruvec
2.32 ? 31% perf-profile.calltrace.cycles-pp._raw_spin_lock.list_lru_del.workingset_update_node.xas_store.add_to_swap_cache
2.38 ? 31% perf-profile.calltrace.cycles-pp._raw_spin_lock.list_lru_add.workingset_update_node.xas_store.__delete_from_swap_cache
2.91 ? 25% perf-profile.calltrace.cycles-pp.list_lru_del.workingset_update_node.xas_store.add_to_swap_cache.add_to_swap
3.00 ? 21% perf-profile.calltrace.cycles-pp.list_lru_add.workingset_update_node.xas_store.__delete_from_swap_cache.__remove_mapping
3.01 ? 21% perf-profile.calltrace.cycles-pp.workingset_update_node.xas_store.__delete_from_swap_cache.__remove_mapping.shrink_folio_list
3.04 ? 21% perf-profile.calltrace.cycles-pp.workingset_update_node.xas_store.add_to_swap_cache.add_to_swap.shrink_folio_list
3.12 ? 21% perf-profile.calltrace.cycles-pp.xas_store.add_to_swap_cache.add_to_swap.shrink_folio_list.shrink_inactive_list
3.12 ? 21% perf-profile.calltrace.cycles-pp.xas_store.__delete_from_swap_cache.__remove_mapping.shrink_folio_list.shrink_inactive_list
3.72 ? 20% perf-profile.calltrace.cycles-pp.__delete_from_swap_cache.__remove_mapping.shrink_folio_list.shrink_inactive_list.shrink_lruvec
8.13 ? 17% perf-profile.calltrace.cycles-pp.add_to_swap.shrink_folio_list.shrink_inactive_list.shrink_lruvec.shrink_node_memcgs
3.97 ? 20% perf-profile.calltrace.cycles-pp.add_to_swap_cache.add_to_swap.shrink_folio_list.shrink_inactive_list.shrink_lruvec
4.66 ? 19% perf-profile.calltrace.cycles-pp.__remove_mapping.shrink_folio_list.shrink_inactive_list.shrink_lruvec.shrink_node_memcgs
11.24 ? 28% perf-profile.calltrace.cycles-pp.ret_from_fork
11.24 ? 28% perf-profile.calltrace.cycles-pp.kthread.ret_from_fork
10.90 ? 29% perf-profile.calltrace.cycles-pp.shrink_inactive_list.shrink_lruvec.shrink_node_memcgs.shrink_node.balance_pgdat
10.90 ? 29% perf-profile.calltrace.cycles-pp.shrink_lruvec.shrink_node_memcgs.shrink_node.balance_pgdat.kswapd
10.92 ? 30% perf-profile.calltrace.cycles-pp.shrink_node_memcgs.shrink_node.balance_pgdat.kswapd.kthread
10.92 ? 30% perf-profile.calltrace.cycles-pp.shrink_node.balance_pgdat.kswapd.kthread.ret_from_fork
10.92 ? 30% perf-profile.calltrace.cycles-pp.kswapd.kthread.ret_from_fork
10.92 ? 30% perf-profile.calltrace.cycles-pp.balance_pgdat.kswapd.kthread.ret_from_fork
20.95 ? 20% perf-profile.calltrace.cycles-pp.shrink_inactive_list.shrink_lruvec.shrink_node_memcgs.shrink_node.shrink_zones
21.04 ? 20% perf-profile.calltrace.cycles-pp.shrink_lruvec.shrink_node_memcgs.shrink_node.shrink_zones.do_try_to_free_pages
21.68 ? 20% perf-profile.calltrace.cycles-pp.shrink_node_memcgs.shrink_node.shrink_zones.do_try_to_free_pages.try_to_free_pages
21.79 ? 19% perf-profile.calltrace.cycles-pp.do_try_to_free_pages.try_to_free_pages.__alloc_pages_slowpath.__alloc_pages.__folio_alloc
21.78 ? 19% perf-profile.calltrace.cycles-pp.shrink_zones.do_try_to_free_pages.try_to_free_pages.__alloc_pages_slowpath.__alloc_pages
21.78 ? 19% perf-profile.calltrace.cycles-pp.shrink_node.shrink_zones.do_try_to_free_pages.try_to_free_pages.__alloc_pages_slowpath
21.83 ? 19% perf-profile.calltrace.cycles-pp.try_to_free_pages.__alloc_pages_slowpath.__alloc_pages.__folio_alloc.vma_alloc_folio
22.79 ? 19% perf-profile.calltrace.cycles-pp.__alloc_pages_slowpath.__alloc_pages.__folio_alloc.vma_alloc_folio.do_anonymous_page
23.20 ? 19% perf-profile.calltrace.cycles-pp.__alloc_pages.__folio_alloc.vma_alloc_folio.do_anonymous_page.__handle_mm_fault
23.22 ? 19% perf-profile.calltrace.cycles-pp.__folio_alloc.vma_alloc_folio.do_anonymous_page.__handle_mm_fault.handle_mm_fault
23.28 ? 19% perf-profile.calltrace.cycles-pp.vma_alloc_folio.do_anonymous_page.__handle_mm_fault.handle_mm_fault.do_user_addr_fault
25.84 ? 19% perf-profile.calltrace.cycles-pp.do_anonymous_page.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_page_fault
26.21 ? 19% perf-profile.calltrace.cycles-pp.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.do_access
26.14 ? 19% perf-profile.calltrace.cycles-pp.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault
26.63 ? 19% perf-profile.calltrace.cycles-pp.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.do_access
26.68 ? 19% perf-profile.calltrace.cycles-pp.exc_page_fault.asm_exc_page_fault.do_access
28.66 ? 19% perf-profile.calltrace.cycles-pp.asm_exc_page_fault.do_access
31.02 ? 19% perf-profile.calltrace.cycles-pp.do_access
29.52 ? 18% perf-profile.calltrace.cycles-pp.shrink_folio_list.shrink_inactive_list.shrink_lruvec.shrink_node_memcgs.shrink_node
18.08 ? 17% perf-profile.children.cycles-pp.__munmap
18.08 ? 17% perf-profile.children.cycles-pp.unmap_region
18.08 ? 17% perf-profile.children.cycles-pp.unmap_page_range
18.08 ? 17% perf-profile.children.cycles-pp.zap_pmd_range
18.08 ? 17% perf-profile.children.cycles-pp.do_mas_munmap
18.08 ? 17% perf-profile.children.cycles-pp.do_mas_align_munmap
18.08 ? 17% perf-profile.children.cycles-pp.__x64_sys_munmap
18.08 ? 17% perf-profile.children.cycles-pp.__vm_munmap
18.08 ? 17% perf-profile.children.cycles-pp.zap_pte_range
18.09 ? 17% perf-profile.children.cycles-pp.unmap_vmas
18.56 ? 15% perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
18.56 ? 15% perf-profile.children.cycles-pp.do_syscall_64
18.03 ? 17% perf-profile.children.cycles-pp.free_swap_and_cache
17.86 ? 17% perf-profile.children.cycles-pp.__swap_entry_free
17.73 ? 17% perf-profile.children.cycles-pp.free_swap_slot
17.45 ? 17% perf-profile.children.cycles-pp.swapcache_free_entries
18.53 ? 9% perf-profile.children.cycles-pp._raw_spin_lock
18.52 ? 9% perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath
3.86 ? 14% perf-profile.children.cycles-pp.__mem_cgroup_uncharge_swap
2.02 ? 9% perf-profile.children.cycles-pp.swap_cgroup_record
1.42 ? 13% perf-profile.children.cycles-pp.page_counter_uncharge
1.41 ? 9% perf-profile.children.cycles-pp._raw_spin_lock_irqsave
0.43 ? 17% perf-profile.children.cycles-pp.clear_shadow_from_swap_cache
0.76 ? 16% perf-profile.children.cycles-pp.mem_cgroup_id_put_many
0.12 ?114% perf-profile.children.cycles-pp.kcompactd
0.12 ?114% perf-profile.children.cycles-pp.compact_zone
0.12 ?116% perf-profile.children.cycles-pp.proactive_compact_node
0.11 ?118% perf-profile.children.cycles-pp.migrate_pages
0.11 ?118% perf-profile.children.cycles-pp.unmap_and_move
0.10 ?116% perf-profile.children.cycles-pp.__unmap_and_move
0.09 ?119% perf-profile.children.cycles-pp.try_to_migrate
0.09 ?119% perf-profile.children.cycles-pp.try_to_migrate_one
0.11 ? 98% perf-profile.children.cycles-pp.ptep_clear_flush
0.11 ? 98% perf-profile.children.cycles-pp.flush_tlb_mm_range
0.12 ? 18% perf-profile.children.cycles-pp.xas_find
0.11 ? 20% perf-profile.children.cycles-pp.xas_load
0.32 ? 12% perf-profile.children.cycles-pp.sort
0.32 ? 12% perf-profile.children.cycles-pp.sort_r
0.36 ? 15% perf-profile.children.cycles-pp.swap_range_free
0.37 ? 17% perf-profile.children.cycles-pp._swap_info_get
0.24 ? 4% perf-profile.children.cycles-pp.propagate_protected_usage
0.24 ? 11% perf-profile.children.cycles-pp.__mod_memcg_state
0.10 ? 8% perf-profile.children.cycles-pp.swp_entry_cmp
0.11 ? 17% perf-profile.children.cycles-pp.__perf_sw_event
0.06 perf-profile.children.cycles-pp.memset_erms
0.08 ? 14% perf-profile.children.cycles-pp.folio_mapping
0.09 ? 12% perf-profile.children.cycles-pp.__cond_resched
0.16 ? 14% perf-profile.children.cycles-pp.workingset_age_nonresident
0.19 ? 19% perf-profile.children.cycles-pp._find_next_bit
0.10 ? 21% perf-profile.children.cycles-pp.folio_unlock
0.33 ? 16% perf-profile.children.cycles-pp.llist_reverse_order
0.22 ? 21% perf-profile.children.cycles-pp.__mod_node_page_state
0.31 ? 13% perf-profile.children.cycles-pp.sync_regs
0.25 ? 19% perf-profile.children.cycles-pp.up_read
0.15 ? 26% perf-profile.children.cycles-pp.count_shadow_nodes
0.12 ? 26% perf-profile.children.cycles-pp.__list_lru_walk_one
0.12 ? 26% perf-profile.children.cycles-pp.shadow_lru_isolate
0.12 ? 26% perf-profile.children.cycles-pp.list_lru_walk_one_irq
0.34 ? 20% perf-profile.children.cycles-pp.task_numa_work
0.34 ? 20% perf-profile.children.cycles-pp.change_prot_numa
0.34 ? 20% perf-profile.children.cycles-pp.change_protection_range
0.34 ? 20% perf-profile.children.cycles-pp.change_pmd_range
0.34 ? 21% perf-profile.children.cycles-pp.task_work_run
0.34 ? 20% perf-profile.children.cycles-pp.change_pte_range
0.34 ? 20% perf-profile.children.cycles-pp.exit_to_user_mode_loop
0.35 ? 20% perf-profile.children.cycles-pp.exit_to_user_mode_prepare
0.58 ? 15% perf-profile.children.cycles-pp.flush_tlb_func
0.35 ? 21% perf-profile.children.cycles-pp.irqentry_exit_to_user_mode
0.26 ? 19% perf-profile.children.cycles-pp.xas_create_range
0.32 ? 18% perf-profile.children.cycles-pp.xas_create
0.24 ? 21% perf-profile.children.cycles-pp.noop_dirty_folio
0.24 ? 39% perf-profile.children.cycles-pp.move_folios_to_lru
0.42 ? 20% perf-profile.children.cycles-pp.shrink_slab
0.51 ? 20% perf-profile.children.cycles-pp.do_shrink_slab
0.32 ? 18% perf-profile.children.cycles-pp.__list_add_valid
1.47 ? 14% perf-profile.children.cycles-pp.__flush_smp_call_function_queue
1.49 ? 14% perf-profile.children.cycles-pp.__sysvec_call_function_single
0.55 ? 24% perf-profile.children.cycles-pp.isolate_lru_folios
0.77 ? 22% perf-profile.children.cycles-pp.folio_lock_anon_vma_read
1.62 ? 14% perf-profile.children.cycles-pp.sysvec_call_function_single
0.69 ? 18% perf-profile.children.cycles-pp.__list_del_entry_valid
1.82 ? 13% perf-profile.children.cycles-pp.folio_end_writeback
1.08 ? 22% perf-profile.children.cycles-pp.page_vma_mapped_walk
2.37 ? 14% perf-profile.children.cycles-pp.asm_sysvec_call_function_single
1.03 ? 25% perf-profile.children.cycles-pp.folio_referenced_one
0.63 ? 83% perf-profile.children.cycles-pp.lru_note_cost
3.65 ? 17% perf-profile.children.cycles-pp.do_rw_once
4.38 ? 14% perf-profile.children.cycles-pp.pmem_rw_page
1.90 ? 25% perf-profile.children.cycles-pp.folio_referenced
1.78 ? 61% perf-profile.children.cycles-pp.start_kernel
1.78 ? 61% perf-profile.children.cycles-pp.arch_call_rest_init
1.78 ? 61% perf-profile.children.cycles-pp.rest_init
6.15 ? 17% perf-profile.children.cycles-pp.smp_call_function_many_cond
6.16 ? 17% perf-profile.children.cycles-pp.on_each_cpu_cond_mask
2.15 ? 77% perf-profile.children.cycles-pp._raw_spin_lock_irq
6.14 ? 19% perf-profile.children.cycles-pp.try_to_unmap_flush_dirty
6.14 ? 19% perf-profile.children.cycles-pp.arch_tlbbatch_flush
3.01 ? 21% perf-profile.children.cycles-pp.list_lru_del
3.02 ? 21% perf-profile.children.cycles-pp.list_lru_add
3.74 ? 20% perf-profile.children.cycles-pp.__delete_from_swap_cache
4.09 ? 20% perf-profile.children.cycles-pp.add_to_swap_cache
8.19 ? 17% perf-profile.children.cycles-pp.add_to_swap
4.76 ? 19% perf-profile.children.cycles-pp.__remove_mapping
11.24 ? 28% perf-profile.children.cycles-pp.kthread
11.24 ? 28% perf-profile.children.cycles-pp.ret_from_fork
10.92 ? 30% perf-profile.children.cycles-pp.kswapd
10.92 ? 30% perf-profile.children.cycles-pp.balance_pgdat
6.34 ? 21% perf-profile.children.cycles-pp.xas_store
6.10 ? 21% perf-profile.children.cycles-pp.workingset_update_node
22.44 ? 21% perf-profile.children.cycles-pp.shrink_zones
22.44 ? 21% perf-profile.children.cycles-pp.do_try_to_free_pages
22.48 ? 21% perf-profile.children.cycles-pp.try_to_free_pages
23.57 ? 20% perf-profile.children.cycles-pp.__folio_alloc
23.64 ? 20% perf-profile.children.cycles-pp.vma_alloc_folio
23.46 ? 20% perf-profile.children.cycles-pp.__alloc_pages_slowpath
23.88 ? 20% perf-profile.children.cycles-pp.__alloc_pages
26.00 ? 19% perf-profile.children.cycles-pp.do_anonymous_page
26.34 ? 19% perf-profile.children.cycles-pp.__handle_mm_fault
26.54 ? 19% perf-profile.children.cycles-pp.handle_mm_fault
26.92 ? 19% perf-profile.children.cycles-pp.do_user_addr_fault
26.96 ? 19% perf-profile.children.cycles-pp.exc_page_fault
28.29 ? 19% perf-profile.children.cycles-pp.asm_exc_page_fault
31.56 ? 18% perf-profile.children.cycles-pp.do_access
29.78 ? 18% perf-profile.children.cycles-pp.shrink_folio_list
32.50 ? 21% perf-profile.children.cycles-pp.shrink_inactive_list
32.59 ? 21% perf-profile.children.cycles-pp.shrink_lruvec
33.25 ? 20% perf-profile.children.cycles-pp.shrink_node_memcgs
33.35 ? 20% perf-profile.children.cycles-pp.shrink_node
18.46 ? 9% perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath
1.30 ? 13% perf-profile.self.cycles-pp.page_counter_uncharge
1.32 ? 10% perf-profile.self.cycles-pp._raw_spin_lock_irqsave
0.74 ? 16% perf-profile.self.cycles-pp.mem_cgroup_id_put_many
0.71 ? 8% perf-profile.self.cycles-pp.swap_cgroup_record
0.08 ? 24% perf-profile.self.cycles-pp.xas_load
0.21 ? 14% perf-profile.self.cycles-pp.free_swap_slot
0.36 ? 16% perf-profile.self.cycles-pp._swap_info_get
0.30 ? 15% perf-profile.self.cycles-pp.swap_range_free
0.20 ? 18% perf-profile.self.cycles-pp.sort_r
0.08 ? 8% perf-profile.self.cycles-pp.xas_store
0.04 ? 71% perf-profile.self.cycles-pp.zap_pte_range
0.22 ? 5% perf-profile.self.cycles-pp.propagate_protected_usage
0.04 ? 75% perf-profile.self.cycles-pp.clear_shadow_from_swap_cache
0.38 ? 10% perf-profile.self.cycles-pp._raw_spin_lock_irq
0.21 ? 12% perf-profile.self.cycles-pp.__mod_memcg_state
0.08 ? 19% perf-profile.self.cycles-pp.__mem_cgroup_uncharge_swap
0.09 ? 10% perf-profile.self.cycles-pp.swp_entry_cmp
0.05 perf-profile.self.cycles-pp.swapcache_free_entries
0.08 ? 19% perf-profile.self.cycles-pp.rmqueue_bulk
0.15 ? 18% perf-profile.self.cycles-pp._find_next_bit
0.16 ? 14% perf-profile.self.cycles-pp.workingset_age_nonresident
0.09 ? 24% perf-profile.self.cycles-pp.folio_unlock
0.13 ? 23% perf-profile.self.cycles-pp.__might_resched
0.12 ? 17% perf-profile.self.cycles-pp.__remove_mapping
0.33 ? 16% perf-profile.self.cycles-pp.llist_reverse_order
0.17 ? 30% perf-profile.self.cycles-pp.rmap_walk_anon
0.21 ? 19% perf-profile.self.cycles-pp.__mod_node_page_state
0.29 ? 14% perf-profile.self.cycles-pp.flush_tlb_func
0.31 ? 13% perf-profile.self.cycles-pp.sync_regs
0.21 ? 19% perf-profile.self.cycles-pp.add_to_swap_cache
0.23 ? 18% perf-profile.self.cycles-pp.up_read
0.10 ? 28% perf-profile.self.cycles-pp.count_shadow_nodes
0.30 ? 19% perf-profile.self.cycles-pp.change_pte_range
0.14 ? 45% perf-profile.self.cycles-pp.move_folios_to_lru
0.58 ? 13% perf-profile.self.cycles-pp.__flush_smp_call_function_queue
0.34 ? 18% perf-profile.self.cycles-pp.folio_lock_anon_vma_read
0.25 ? 24% perf-profile.self.cycles-pp.folio_referenced_one
0.28 ? 21% perf-profile.self.cycles-pp.xas_create
0.21 ? 21% perf-profile.self.cycles-pp.list_lru_del
0.32 ? 26% perf-profile.self.cycles-pp.isolate_lru_folios
0.22 ? 20% perf-profile.self.cycles-pp.list_lru_add
0.41 ? 20% perf-profile.self.cycles-pp.shrink_folio_list
0.22 ? 23% perf-profile.self.cycles-pp.noop_dirty_folio
0.53 ? 21% perf-profile.self.cycles-pp.page_vma_mapped_walk
0.32 ? 19% perf-profile.self.cycles-pp.__list_add_valid
0.40 ? 19% perf-profile.self.cycles-pp.__delete_from_swap_cache
1.36 ? 14% perf-profile.self.cycles-pp.folio_end_writeback
0.68 ? 19% perf-profile.self.cycles-pp.__list_del_entry_valid
3.18 ? 16% perf-profile.self.cycles-pp.do_access
3.41 ? 17% perf-profile.self.cycles-pp.do_rw_once
1.83 ? 18% perf-profile.self.cycles-pp._raw_spin_lock
5.28 ? 18% perf-profile.self.cycles-pp.smp_call_function_many_cond
href="https://github.com/intel/lkp-tests.git">https://github.com/intel/lkp-tests.git
# job file is attached in this email
--compatible job.yaml # generate the yaml file for lkp run
run generated-yaml-file
across any failure that blocks the test,
remove ~/.lkp and /lkp dir to run from a clean state.
estimated based on internal Intel analysis and are provided
purposes only. Any difference in system hardware or software
may affect actual performance.
om/intel/lkp-tests">https://github.com/intel/lkp-tests

Attachments:

(No filename) (100.62 kB)
config-6.2.0-rc4-00343-g5649d113ffce (163.59 kB)
job-script (8.35 kB)
job.yaml (6.09 kB)
reproduce (998.00 B)
Download all attachments

2023-03-20 12:47:28

by Yang Yang

[permalink] [raw]

Subject: [linus:master] [swap_state] 5649d113ff: vm-scalability.throughput -33.1% regression

> 04bac040bc71b4b3 5649d113ffce9f532a9ecc5ab96
> ---------------- ---------------------------
> %stddev %change %stddev
> \ | \
> 10026093 ± 3% -33.1% 6702748 ± 2% vm-scalability.throughput

I try to reproduce this and see vm-scalability.throughput really decrease.
And I use ftrace found that functions related to this patch
add_to_swap_cache()/__delete_from_swap_cache()/clear_shadow_from_swap_cache()
consume more time while workingset_update_node() be called much more times.

Since the patch result in consuming much more resource, and the problem this
patch try to solve is not apparent to user, we may abandon this patch.

By the way, as what this test result shows, mapping_set_update() should also
consume much time. Should we care about this?

Thanks.

Reproduce before this patch:
/vm-scalability-master # cat /sys/kernel/debug/tracing/trace_stat/function0
Function Hit Time Avg s^2
-------- --- ---- --- ---
add_to_swap_cache 26108 476290.6 us 18.243 us 487.762 us
__delete_from_swap_cache 26117 462492.6 us 17.708 us 77.801 us
clear_shadow_from_swap_cache 27840 199925.1 us 7.181 us 313.126 us

Reproduce after this patch:
/vm-scalability-master # cat /sys/kernel/debug/tracing/trace_stat/function*
Function Hit Time Avg s^2
-------- --- ---- --- ---
add_to_swap_cache 51268 1371819 us 26.757 us 676.311 us
__delete_from_swap_cache 51260 1322712 us 25.803 us 123.010 us
workingset_update_node 157455 770064.9 us 4.890 us 15.108 us
clear_shadow_from_swap_cache 52928 563597.4 us 10.648 us 199.766 us

2023-03-21 07:56:39

by Yang Yang

[permalink] [raw]

Subject: [linus:master] [swap_state] 5649d113ff: vm-scalability.throughput -33.1% regression

> commit:
> 04bac040bc ("mm/hugetlb: convert get_hwpoison_huge_page() to folios")
> 5649d113ff ("swap_state: update shadow_nodes for anonymous page")
> 04bac040bc71b4b3 5649d113ffce9f532a9ecc5ab96
> ---------------- ---------------------------
> %stddev %change %stddev
> \ | \
> 10026093 ± 3% -33.1% 6702748 ± 2% vm-scalability.throughput

> 04bac040bc71b4b3 5649d113ffce9f532a9ecc5ab96
> ---------------- ---------------------------
> %stddev %change %stddev
> \ | \
> 553378 -11.1% 492012 ± 2% vm-scalability.median

I see the two results are much different, one is -33.1%, another is -11.1%.
So I tried more times to reproduce on my machine, and see a 8% of regression
of vm-scalability.throughput.

As this test add/delete/clear swap cache frequently, the impact of commit
5649d113ff might be magnified ?

Commit 5649d113ff tried to fix the problem that if swap space is huge and
apps are using many shadow entries, shadow nodes may waste much space in
memory. So the shadow nodes should be reclaimed when it's number is huge while
memory is in tense.

I reviewed commit 5649d113ff carefully, and didn't found any obviously
problem. If we want to correctly update shadow_nodes for anonymous page,
we have to update them when add/delete/clear swap cache.

Thanks.

2023-04-04 02:46:17

by Yin, Fengwei

[permalink] [raw]

Subject: Re: [linus:master] [swap_state] 5649d113ff: vm-scalability.throughput -33.1% regression

Hi Yang,

On Tue, 2023-03-21 at 07:56 +0000, Yang Yang wrote:
> > commit:
> > 04bac040bc ("mm/hugetlb: convert get_hwpoison_huge_page() to
> > folios")
> > 5649d113ff ("swap_state: update shadow_nodes for anonymous page")
> > 04bac040bc71b4b3 5649d113ffce9f532a9ecc5ab96
> > ---------------- ---------------------------
> >          %stddev     %change         %stddev
> >              \          |                \
> > 10026093 ± 3%     -33.1%    6702748 ± 2% vm-
> > scalability.throughput
>
> > 04bac040bc71b4b3 5649d113ffce9f532a9ecc5ab96
> > ---------------- ---------------------------
> >          %stddev     %change         %stddev
> >              \          |                \
> >    553378           -11.1%     492012 ± 2% vm-scalability.median
>
> I see the two results are much different, one is -33.1%, another is -
> 11.1%.
> So I tried more times to reproduce on my machine, and see a 8% of
> regression
> of vm-scalability.throughput.
>
> As this test add/delete/clear swap cache frequently, the impact of
> commit
> 5649d113ff might be magnified ?
>
> Commit 5649d113ff tried to fix the problem that if swap space is huge
> and
> apps are using many shadow entries, shadow nodes may waste much space
> in
> memory. So the shadow nodes should be reclaimed when it's number is
> huge while
> memory is in tense.
>
> I reviewed commit 5649d113ff carefully, and didn't found any
> obviously
> problem. If we want to correctly update shadow_nodes for anonymous
> page,
> we have to update them when add/delete/clear swap cache.
Thanks for the info and sorry for delayed response. We didn't get your
replies in our compony Inbox (don't know why). Just noticed you replies
on lore.kernel.org when revise the ticket. We will mark this regression
won't fix as it's for functional fixing.

Regards
Yin, Fengwei

>
> Thanks.