Hello, kernel test robot noticed a -5.7% regression of fsmark.files_per_sec on: commit: 2edf06a50f5bbe664283f3c55c480fc013221d70 ("xfs: factor xfs_alloc_vextent_this_ag() for _iterate_ags()") https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master testcase: fsmark test machine: 96 threads 2 sockets Intel(R) Xeon(R) Platinum 8260L CPU @ 2.40GHz (Cascade Lake) with 128G memory parameters: iterations: 8 disk: 1SSD nr_threads: 32 fs: xfs filesize: 8K test_size: 50G sync_method: fsyncBeforeClose nr_directories: 16d nr_files_per_directory: 256fpd cpufreq_governor: performance test-description: The fsmark is a file system benchmark to test synchronous write workloads, for example, mail servers workload. test-url: https://sourceforge.net/projects/fsmark/ In addition to that, the commit also has significant impact on the following tests: +------------------+----------------------------------------------------------+ | testcase: change | fsmark: fsmark.files_per_sec -3.7% regression | | test machine | 224 threads 2 sockets (Sapphire Rapids) with 256G memory | | test parameters | cpufreq_governor=performance | | | disk=1SSD | | | filesize=8K | | | fs=xfs | | | iterations=8 | | | nr_directories=16d | | | nr_files_per_directory=256fpd | | | nr_threads=32 | | | sync_method=fsyncBeforeClose | | | test_size=50G | +------------------+----------------------------------------------------------+ If you fix the issue, kindly add following tag | Reported-by: kernel test robot | Link: https://lore.kernel.org/oe-lkp/202305090905.aff4e0e6-oliver.sang@intel.com Details are as below: --------------------------------------------------------------------------------------------------> To reproduce: git clone https://github.com/intel/lkp-tests.git cd lkp-tests sudo bin/lkp install job.yaml # job file is attached in this email bin/lkp split-job --compatible job.yaml # generate the yaml file for lkp run sudo bin/lkp run generated-yaml-file # if come across any failure that blocks the test, # please remove ~/.lkp and /lkp dir to run from a clean state. ========================================================================================= compiler/cpufreq_governor/disk/filesize/fs/iterations/kconfig/nr_directories/nr_files_per_directory/nr_threads/rootfs/sync_method/tbox_group/test_size/testcase: gcc-11/performance/1SSD/8K/xfs/8/x86_64-rhel-8.3/16d/256fpd/32/debian-11.1-x86_64-20220510.cgz/fsyncBeforeClose/lkp-csl-2sp3/50G/fsmark commit: ecd788a924 ("xfs: rework xfs_alloc_vextent()") 2edf06a50f ("xfs: factor xfs_alloc_vextent_this_ag() for _iterate_ags()") ecd788a92460eef4 2edf06a50f5bbe664283f3c55c4 ---------------- --------------------------- %stddev %change %stddev \ | \ 14349 -5.7% 13527 fsmark.files_per_sec 486.29 +5.8% 514.28 fsmark.time.elapsed_time 486.29 +5.8% 514.28 fsmark.time.elapsed_time.max 211.17 ± 2% +7.3% 226.50 fsmark.time.percent_of_cpu_this_job_got 969.01 ± 2% +14.7% 1111 fsmark.time.system_time 32491035 +10.9% 36033196 fsmark.time.voluntary_context_switches 4.63 +6.0% 4.91 iostat.cpu.iowait 427253 ± 6% +10.1% 470526 ± 4% sched_debug.cpu.nr_switches.avg 0.14 ± 2% +20.5% 0.17 ± 2% turbostat.IPC 176400 +3.0% 181615 vmstat.system.cs 4038 ± 2% +5.6% 4262 ± 2% proc-vmstat.nr_active_anon 4038 ± 2% +5.6% 4262 ± 2% proc-vmstat.nr_zone_active_anon 6254 ± 2% +4.3% 6524 proc-vmstat.numa_huge_pte_updates 184421 ± 4% +8.1% 199355 ± 2% proc-vmstat.pgactivate 1474801 +3.6% 1528229 proc-vmstat.pgfault 1.096e+08 +5.1% 1.153e+08 proc-vmstat.pgpgout 3718912 +5.5% 3922432 proc-vmstat.unevictable_pgs_scanned 8.916e+08 +21.2% 1.08e+09 perf-stat.i.branch-instructions 177602 +2.9% 182781 perf-stat.i.context-switches 2.12 ± 3% -17.9% 1.74 ± 3% perf-stat.i.cpi 744.83 ± 2% +4.3% 777.00 perf-stat.i.cycles-between-cache-misses 1.237e+09 +28.8% 1.594e+09 perf-stat.i.dTLB-loads 6.391e+08 +27.6% 8.152e+08 perf-stat.i.dTLB-stores 7441243 +3.6% 7712079 perf-stat.i.iTLB-loads 4.584e+09 +24.7% 5.718e+09 perf-stat.i.instructions 1701 ± 3% +21.6% 2068 perf-stat.i.instructions-per-iTLB-miss 0.48 ± 3% +20.5% 0.58 ± 3% perf-stat.i.ipc 29.02 +25.9% 36.53 perf-stat.i.metric.M/sec 2634 -1.5% 2595 perf-stat.i.minor-faults 2634 -1.5% 2595 perf-stat.i.page-faults 2.04 ± 3% -16.6% 1.70 ± 3% perf-stat.overall.cpi 716.67 +4.3% 747.31 ± 2% perf-stat.overall.cycles-between-cache-misses 1605 ± 3% +23.9% 1989 perf-stat.overall.instructions-per-iTLB-miss 0.49 ± 3% +20.0% 0.59 ± 3% perf-stat.overall.ipc 8.898e+08 +21.2% 1.078e+09 perf-stat.ps.branch-instructions 177212 +2.9% 182411 perf-stat.ps.context-switches 1.234e+09 +28.9% 1.591e+09 perf-stat.ps.dTLB-loads 6.378e+08 +27.6% 8.137e+08 perf-stat.ps.dTLB-stores 7425084 +3.7% 7696575 perf-stat.ps.iTLB-loads 4.574e+09 +24.8% 5.707e+09 perf-stat.ps.instructions 2629 -1.4% 2592 perf-stat.ps.minor-faults 2629 -1.4% 2592 perf-stat.ps.page-faults 2.227e+12 +31.9% 2.938e+12 perf-stat.total.instructions 1.18 ± 5% -0.2 0.97 ± 6% perf-profile.calltrace.cycles-pp.folio_wait_bit_common.folio_wait_writeback.__filemap_fdatawait_range.file_write_and_wait_range.xfs_file_fsync 1.20 ± 5% -0.2 0.98 ± 7% perf-profile.calltrace.cycles-pp.folio_wait_writeback.__filemap_fdatawait_range.file_write_and_wait_range.xfs_file_fsync.__x64_sys_fsync 1.12 ± 5% -0.2 0.91 ± 7% perf-profile.calltrace.cycles-pp.io_schedule.folio_wait_bit_common.folio_wait_writeback.__filemap_fdatawait_range.file_write_and_wait_range 1.26 ± 5% -0.2 1.05 ± 6% perf-profile.calltrace.cycles-pp.__filemap_fdatawait_range.file_write_and_wait_range.xfs_file_fsync.__x64_sys_fsync.do_syscall_64 1.11 ± 5% -0.2 0.91 ± 7% perf-profile.calltrace.cycles-pp.schedule.io_schedule.folio_wait_bit_common.folio_wait_writeback.__filemap_fdatawait_range 1.11 ± 4% -0.2 0.90 ± 7% perf-profile.calltrace.cycles-pp.__schedule.schedule.io_schedule.folio_wait_bit_common.folio_wait_writeback 1.60 ± 3% -0.2 1.42 ± 8% perf-profile.calltrace.cycles-pp.xfs_end_io.process_one_work.worker_thread.kthread.ret_from_fork 1.58 ± 3% -0.2 1.41 ± 7% perf-profile.calltrace.cycles-pp.xfs_end_ioend.xfs_end_io.process_one_work.worker_thread.kthread 1.77 ± 4% -0.2 1.62 ± 6% perf-profile.calltrace.cycles-pp.__wait_for_common.__flush_workqueue.xlog_cil_push_now.xlog_cil_force_seq.xfs_log_force_seq 0.66 -0.2 0.50 ± 44% perf-profile.calltrace.cycles-pp.folio_end_writeback.iomap_finish_ioend.iomap_finish_ioends.xfs_end_ioend.xfs_end_io 1.93 ± 5% -0.2 1.78 ± 2% perf-profile.calltrace.cycles-pp.xlog_cil_commit.__xfs_trans_commit.xfs_create.xfs_generic_create.lookup_open 1.70 ± 4% -0.2 1.56 ± 6% perf-profile.calltrace.cycles-pp.schedule.schedule_timeout.__wait_for_common.__flush_workqueue.xlog_cil_push_now 1.71 ± 4% -0.2 1.56 ± 6% perf-profile.calltrace.cycles-pp.schedule_timeout.__wait_for_common.__flush_workqueue.xlog_cil_push_now.xlog_cil_force_seq 1.70 ± 3% -0.1 1.55 ± 6% perf-profile.calltrace.cycles-pp.__schedule.schedule.schedule_timeout.__wait_for_common.__flush_workqueue 0.72 ± 2% -0.1 0.63 ± 8% perf-profile.calltrace.cycles-pp.iomap_finish_ioend.iomap_finish_ioends.xfs_end_ioend.xfs_end_io.process_one_work 0.73 ± 2% -0.1 0.64 ± 8% perf-profile.calltrace.cycles-pp.iomap_finish_ioends.xfs_end_ioend.xfs_end_io.process_one_work.worker_thread 0.83 ± 6% -0.1 0.75 ± 10% perf-profile.calltrace.cycles-pp.dequeue_task_fair.__schedule.schedule.schedule_timeout.__wait_for_common 0.98 ± 3% -0.1 0.90 ± 5% perf-profile.calltrace.cycles-pp.xfs_dialloc_ag.xfs_dialloc.xfs_create.xfs_generic_create.lookup_open 0.79 ± 4% +0.3 1.13 ± 6% perf-profile.calltrace.cycles-pp.xlog_cil_commit.__xfs_trans_commit.xfs_bmapi_convert_delalloc.xfs_map_blocks.iomap_writepage_map 0.84 ± 4% +0.4 1.20 ± 6% perf-profile.calltrace.cycles-pp.__xfs_trans_commit.xfs_bmapi_convert_delalloc.xfs_map_blocks.iomap_writepage_map.write_cache_pages 0.00 +0.7 0.66 ± 8% perf-profile.calltrace.cycles-pp.xfs_btree_increment.xfs_alloc_walk_iter.xfs_alloc_ag_vextent_lastblock.xfs_alloc_ag_vextent_near.xfs_alloc_ag_vextent 0.00 +0.7 0.75 ± 6% perf-profile.calltrace.cycles-pp.xfs_btree_get_rec.xfs_alloc_get_rec.xfs_alloc_cur_check.xfs_alloc_walk_iter.xfs_alloc_ag_vextent_lastblock 1.01 ± 6% +0.9 1.88 ± 7% perf-profile.calltrace.cycles-pp.__schedule.schedule.schedule_timeout.___down_common.__down 1.01 ± 7% +0.9 1.89 ± 6% perf-profile.calltrace.cycles-pp.schedule.schedule_timeout.___down_common.__down.down 1.02 ± 7% +0.9 1.91 ± 6% perf-profile.calltrace.cycles-pp.schedule_timeout.___down_common.__down.down.xfs_buf_lock 1.05 ± 7% +0.9 1.95 ± 6% perf-profile.calltrace.cycles-pp.__down.down.xfs_buf_lock.xfs_buf_find_lock.xfs_buf_lookup 1.05 ± 7% +0.9 1.95 ± 6% perf-profile.calltrace.cycles-pp.___down_common.__down.down.xfs_buf_lock.xfs_buf_find_lock 1.08 ± 6% +0.9 1.99 ± 6% perf-profile.calltrace.cycles-pp.down.xfs_buf_lock.xfs_buf_find_lock.xfs_buf_lookup.xfs_buf_get_map 1.08 ± 6% +0.9 2.00 ± 6% perf-profile.calltrace.cycles-pp.xfs_buf_lock.xfs_buf_find_lock.xfs_buf_lookup.xfs_buf_get_map.xfs_buf_read_map 1.10 ± 6% +0.9 2.02 ± 6% perf-profile.calltrace.cycles-pp.xfs_buf_find_lock.xfs_buf_lookup.xfs_buf_get_map.xfs_buf_read_map.xfs_trans_read_buf_map 0.00 +1.0 1.03 ± 4% perf-profile.calltrace.cycles-pp.xfs_alloc_get_rec.xfs_alloc_cur_check.xfs_alloc_walk_iter.xfs_alloc_ag_vextent_lastblock.xfs_alloc_ag_vextent_near 0.00 +1.0 1.04 ± 7% perf-profile.calltrace.cycles-pp.xfs_buf_lookup.xfs_buf_get_map.xfs_buf_read_map.xfs_trans_read_buf_map.xfs_read_agf 0.00 +1.1 1.08 ± 7% perf-profile.calltrace.cycles-pp.xfs_buf_get_map.xfs_buf_read_map.xfs_trans_read_buf_map.xfs_read_agf.xfs_alloc_read_agf 0.00 +1.1 1.09 ± 7% perf-profile.calltrace.cycles-pp.xfs_buf_read_map.xfs_trans_read_buf_map.xfs_read_agf.xfs_alloc_read_agf.xfs_alloc_fix_freelist 0.00 +1.1 1.15 ± 8% perf-profile.calltrace.cycles-pp.xfs_trans_read_buf_map.xfs_read_agf.xfs_alloc_read_agf.xfs_alloc_fix_freelist.__xfs_alloc_vextent_this_ag 0.00 +1.2 1.16 ± 8% perf-profile.calltrace.cycles-pp.xfs_read_agf.xfs_alloc_read_agf.xfs_alloc_fix_freelist.__xfs_alloc_vextent_this_ag.xfs_alloc_vextent_iterate_ags 0.00 +1.2 1.18 ± 7% perf-profile.calltrace.cycles-pp.xfs_alloc_read_agf.xfs_alloc_fix_freelist.__xfs_alloc_vextent_this_ag.xfs_alloc_vextent_iterate_ags.xfs_alloc_vextent 0.00 +1.4 1.36 ± 6% perf-profile.calltrace.cycles-pp.xfs_alloc_fix_freelist.__xfs_alloc_vextent_this_ag.xfs_alloc_vextent_iterate_ags.xfs_alloc_vextent.xfs_bmap_btalloc 0.00 +1.4 1.36 ± 6% perf-profile.calltrace.cycles-pp.__xfs_alloc_vextent_this_ag.xfs_alloc_vextent_iterate_ags.xfs_alloc_vextent.xfs_bmap_btalloc.xfs_bmapi_allocate 0.00 +2.0 1.96 ± 4% perf-profile.calltrace.cycles-pp.xfs_alloc_cur_check.xfs_alloc_walk_iter.xfs_alloc_ag_vextent_lastblock.xfs_alloc_ag_vextent_near.xfs_alloc_ag_vextent 1.18 ± 7% +2.5 3.66 ± 5% perf-profile.calltrace.cycles-pp.xfs_alloc_ag_vextent.xfs_alloc_vextent_iterate_ags.xfs_alloc_vextent.xfs_bmap_btalloc.xfs_bmapi_allocate 1.05 ± 7% +2.6 3.61 ± 5% perf-profile.calltrace.cycles-pp.xfs_alloc_ag_vextent_near.xfs_alloc_ag_vextent.xfs_alloc_vextent_iterate_ags.xfs_alloc_vextent.xfs_bmap_btalloc 0.00 +2.7 2.68 ± 5% perf-profile.calltrace.cycles-pp.xfs_alloc_walk_iter.xfs_alloc_ag_vextent_lastblock.xfs_alloc_ag_vextent_near.xfs_alloc_ag_vextent.xfs_alloc_vextent_iterate_ags 0.00 +2.7 2.71 ± 5% perf-profile.calltrace.cycles-pp.xfs_alloc_ag_vextent_lastblock.xfs_alloc_ag_vextent_near.xfs_alloc_ag_vextent.xfs_alloc_vextent_iterate_ags.xfs_alloc_vextent 1.63 ± 6% +3.4 5.06 ± 5% perf-profile.calltrace.cycles-pp.xfs_alloc_vextent.xfs_bmap_btalloc.xfs_bmapi_allocate.xfs_bmapi_convert_delalloc.xfs_map_blocks 1.61 ± 6% +3.4 5.04 ± 5% perf-profile.calltrace.cycles-pp.xfs_alloc_vextent_iterate_ags.xfs_alloc_vextent.xfs_bmap_btalloc.xfs_bmapi_allocate.xfs_bmapi_convert_delalloc 1.85 ± 6% +3.5 5.30 ± 5% perf-profile.calltrace.cycles-pp.xfs_bmapi_allocate.xfs_bmapi_convert_delalloc.xfs_map_blocks.iomap_writepage_map.write_cache_pages 1.76 ± 6% +3.5 5.22 ± 5% perf-profile.calltrace.cycles-pp.xfs_bmap_btalloc.xfs_bmapi_allocate.xfs_bmapi_convert_delalloc.xfs_map_blocks.iomap_writepage_map 5.14 ± 3% +3.5 8.69 ± 5% perf-profile.calltrace.cycles-pp.file_write_and_wait_range.xfs_file_fsync.__x64_sys_fsync.do_syscall_64.entry_SYSCALL_64_after_hwframe 3.86 ± 3% +3.8 7.62 ± 5% perf-profile.calltrace.cycles-pp.__filemap_fdatawrite_range.file_write_and_wait_range.xfs_file_fsync.__x64_sys_fsync.do_syscall_64 3.85 ± 3% +3.8 7.62 ± 5% perf-profile.calltrace.cycles-pp.filemap_fdatawrite_wbc.__filemap_fdatawrite_range.file_write_and_wait_range.xfs_file_fsync.__x64_sys_fsync 3.81 ± 3% +3.8 7.58 ± 5% perf-profile.calltrace.cycles-pp.do_writepages.filemap_fdatawrite_wbc.__filemap_fdatawrite_range.file_write_and_wait_range.xfs_file_fsync 3.80 ± 3% +3.8 7.57 ± 5% perf-profile.calltrace.cycles-pp.xfs_vm_writepages.do_writepages.filemap_fdatawrite_wbc.__filemap_fdatawrite_range.file_write_and_wait_range 3.47 ± 3% +3.8 7.25 ± 5% perf-profile.calltrace.cycles-pp.write_cache_pages.iomap_writepages.xfs_vm_writepages.do_writepages.filemap_fdatawrite_wbc 3.47 ± 3% +3.8 7.26 ± 5% perf-profile.calltrace.cycles-pp.iomap_writepages.xfs_vm_writepages.do_writepages.filemap_fdatawrite_wbc.__filemap_fdatawrite_range 2.92 ± 3% +3.8 6.71 ± 5% perf-profile.calltrace.cycles-pp.xfs_map_blocks.iomap_writepage_map.write_cache_pages.iomap_writepages.xfs_vm_writepages 2.88 ± 4% +3.8 6.67 ± 5% perf-profile.calltrace.cycles-pp.xfs_bmapi_convert_delalloc.xfs_map_blocks.iomap_writepage_map.write_cache_pages.iomap_writepages 3.21 ± 3% +3.8 7.01 ± 5% perf-profile.calltrace.cycles-pp.iomap_writepage_map.write_cache_pages.iomap_writepages.xfs_vm_writepages.do_writepages 1.20 ± 5% -0.2 0.98 ± 7% perf-profile.children.cycles-pp.folio_wait_writeback 1.18 ± 5% -0.2 0.97 ± 6% perf-profile.children.cycles-pp.folio_wait_bit_common 1.12 ± 5% -0.2 0.91 ± 7% perf-profile.children.cycles-pp.io_schedule 1.26 ± 5% -0.2 1.05 ± 6% perf-profile.children.cycles-pp.__filemap_fdatawait_range 1.60 ± 3% -0.2 1.42 ± 8% perf-profile.children.cycles-pp.xfs_end_io 1.58 ± 3% -0.2 1.41 ± 7% perf-profile.children.cycles-pp.xfs_end_ioend 0.50 ± 28% -0.1 0.35 ± 14% perf-profile.children.cycles-pp.syscall_exit_to_user_mode 0.72 ± 2% -0.1 0.63 ± 8% perf-profile.children.cycles-pp.iomap_finish_ioend 0.73 ± 2% -0.1 0.64 ± 8% perf-profile.children.cycles-pp.iomap_finish_ioends 0.66 -0.1 0.58 ± 9% perf-profile.children.cycles-pp.folio_end_writeback 0.98 ± 4% -0.1 0.90 ± 5% perf-profile.children.cycles-pp.xfs_dialloc_ag 0.24 ± 10% -0.1 0.18 ± 11% perf-profile.children.cycles-pp.xfs_perag_get 0.40 ± 5% -0.1 0.35 ± 10% perf-profile.children.cycles-pp.folio_wake_bit 0.37 ± 4% -0.1 0.32 ± 9% perf-profile.children.cycles-pp.wake_page_function 0.32 ± 5% -0.0 0.28 ± 10% perf-profile.children.cycles-pp.lapic_next_deadline 0.18 ± 10% -0.0 0.14 ± 12% perf-profile.children.cycles-pp.xfs_inode_item_push 0.17 ± 10% -0.0 0.13 ± 11% perf-profile.children.cycles-pp.xfs_iflush_cluster 0.12 ± 9% -0.0 0.08 ± 7% perf-profile.children.cycles-pp.xfs_iflush 0.11 ± 9% -0.0 0.08 ± 10% perf-profile.children.cycles-pp.memcg_slab_post_alloc_hook 0.07 ± 18% -0.0 0.04 ± 45% perf-profile.children.cycles-pp.try_to_unlazy 0.15 ± 10% -0.0 0.12 ± 10% perf-profile.children.cycles-pp.d_alloc 0.11 ± 6% -0.0 0.09 ± 10% perf-profile.children.cycles-pp.__intel_pmu_enable_all 0.11 ± 4% -0.0 0.09 ± 7% perf-profile.children.cycles-pp.set_next_entity 0.17 ± 12% +0.0 0.21 ± 9% perf-profile.children.cycles-pp.xfs_alloc_lookup_eq 0.12 ± 17% +0.1 0.17 ± 6% perf-profile.children.cycles-pp.xfs_ialloc_ag_alloc 0.03 ±100% +0.1 0.09 ± 18% perf-profile.children.cycles-pp.xfs_dir2_node_add_datablk 0.00 +0.1 0.07 ± 16% perf-profile.children.cycles-pp.xfs_allocbt_get_maxrecs 0.02 ±141% +0.1 0.08 ± 14% perf-profile.children.cycles-pp.xfs_dir2_grow_inode 0.02 ±141% +0.1 0.08 ± 14% perf-profile.children.cycles-pp.xfs_da_grow_inode_int 0.02 ±141% +0.1 0.09 ± 13% perf-profile.children.cycles-pp.xfs_allocbt_init_key_from_rec 0.09 ± 10% +0.1 0.17 ± 13% perf-profile.children.cycles-pp.xfs_lookup_get_search_key 0.01 ±223% +0.1 0.09 ± 8% perf-profile.children.cycles-pp.xfs_btree_rec_offset 0.35 ± 10% +0.1 0.45 ± 10% perf-profile.children.cycles-pp.xfs_alloc_cur_finish 0.00 +0.1 0.10 ± 15% perf-profile.children.cycles-pp.xfs_btree_check_block 0.07 ± 24% +0.1 0.20 ± 12% perf-profile.children.cycles-pp.xfs_errortag_test 3.16 ± 3% +0.2 3.31 ± 2% perf-profile.children.cycles-pp.xlog_cil_commit 3.35 ± 2% +0.2 3.54 ± 2% perf-profile.children.cycles-pp.__xfs_trans_commit 0.00 +0.2 0.22 ± 10% perf-profile.children.cycles-pp.xfs_alloc_compute_diff 1.58 ± 4% +0.3 1.83 ± 6% perf-profile.children.cycles-pp.__orc_find 0.57 ± 4% +0.3 0.85 ± 7% perf-profile.children.cycles-pp.up 1.24 ± 4% +0.3 1.52 ± 6% perf-profile.children.cycles-pp.orc_find 0.59 ± 5% +0.3 0.88 ± 6% perf-profile.children.cycles-pp.xfs_buf_unlock 0.62 ± 5% +0.3 0.92 ± 7% perf-profile.children.cycles-pp.xfs_buf_item_release 0.02 ±141% +0.3 0.36 ± 7% perf-profile.children.cycles-pp.xfs_extent_busy_trim 0.06 ± 6% +0.4 0.46 ± 6% perf-profile.children.cycles-pp.xfs_alloc_compute_aligned 3.91 ± 3% +0.4 4.33 ± 4% perf-profile.children.cycles-pp.unwind_next_frame 6.16 ± 2% +0.6 6.73 ± 5% perf-profile.children.cycles-pp.perf_callchain 0.10 ± 31% +0.6 0.72 ± 7% perf-profile.children.cycles-pp.xfs_btree_increment 0.14 ± 9% +0.7 0.84 ± 5% perf-profile.children.cycles-pp.xfs_btree_get_rec 0.16 ± 11% +0.7 0.88 ± 6% perf-profile.children.cycles-pp.__xfs_btree_check_sblock 2.80 ± 3% +0.7 3.53 ± 5% perf-profile.children.cycles-pp.schedule_timeout 2.18 ± 5% +0.8 2.98 ± 6% perf-profile.children.cycles-pp.xfs_buf_get_map 2.20 ± 5% +0.8 3.00 ± 6% perf-profile.children.cycles-pp.xfs_buf_read_map 2.84 ± 5% +0.8 3.66 ± 6% perf-profile.children.cycles-pp.xfs_trans_read_buf_map 1.72 ± 4% +0.9 2.58 ± 7% perf-profile.children.cycles-pp.xfs_buf_lookup 1.32 ± 5% +0.9 2.21 ± 6% perf-profile.children.cycles-pp.xfs_buf_find_lock 1.10 ± 6% +0.9 1.99 ± 6% perf-profile.children.cycles-pp.___down_common 1.10 ± 6% +0.9 1.99 ± 6% perf-profile.children.cycles-pp.__down 1.24 ± 5% +0.9 2.14 ± 6% perf-profile.children.cycles-pp.xfs_buf_lock 1.19 ± 5% +0.9 2.09 ± 6% perf-profile.children.cycles-pp.down 0.21 ± 13% +0.9 1.14 ± 7% perf-profile.children.cycles-pp.xfs_btree_check_sblock 0.11 ± 10% +1.0 1.08 ± 3% perf-profile.children.cycles-pp.xfs_alloc_get_rec 0.41 ± 6% +1.0 1.41 ± 6% perf-profile.children.cycles-pp.xfs_alloc_fix_freelist 0.22 ± 8% +1.0 1.22 ± 7% perf-profile.children.cycles-pp.xfs_alloc_read_agf 0.19 ± 9% +1.0 1.20 ± 7% perf-profile.children.cycles-pp.xfs_read_agf 0.00 +1.4 1.41 ± 6% perf-profile.children.cycles-pp.__xfs_alloc_vextent_this_ag 0.22 ± 12% +1.9 2.07 ± 4% perf-profile.children.cycles-pp.xfs_alloc_cur_check 0.29 ± 12% +2.5 2.80 ± 5% perf-profile.children.cycles-pp.xfs_alloc_ag_vextent_lastblock 0.31 ± 10% +2.5 2.82 ± 5% perf-profile.children.cycles-pp.xfs_alloc_walk_iter 1.24 ± 7% +2.6 3.79 ± 5% perf-profile.children.cycles-pp.xfs_alloc_ag_vextent 1.09 ± 7% +2.6 3.74 ± 5% perf-profile.children.cycles-pp.xfs_alloc_ag_vextent_near 1.63 ± 6% +3.5 5.11 ± 5% perf-profile.children.cycles-pp.xfs_alloc_vextent_iterate_ags 1.89 ± 6% +3.5 5.38 ± 5% perf-profile.children.cycles-pp.xfs_bmapi_allocate 1.79 ± 6% +3.5 5.29 ± 5% perf-profile.children.cycles-pp.xfs_bmap_btalloc 5.14 ± 3% +3.6 8.70 ± 5% perf-profile.children.cycles-pp.file_write_and_wait_range 1.65 ± 6% +3.6 5.24 ± 5% perf-profile.children.cycles-pp.xfs_alloc_vextent 3.86 ± 3% +3.8 7.62 ± 5% perf-profile.children.cycles-pp.__filemap_fdatawrite_range 3.86 ± 3% +3.8 7.62 ± 5% perf-profile.children.cycles-pp.filemap_fdatawrite_wbc 3.90 ± 3% +3.8 7.66 ± 5% perf-profile.children.cycles-pp.do_writepages 3.86 ± 3% +3.8 7.63 ± 5% perf-profile.children.cycles-pp.xfs_vm_writepages 3.50 ± 3% +3.8 7.28 ± 5% perf-profile.children.cycles-pp.write_cache_pages 2.92 ± 3% +3.8 6.71 ± 5% perf-profile.children.cycles-pp.xfs_map_blocks 3.50 ± 3% +3.8 7.29 ± 5% perf-profile.children.cycles-pp.iomap_writepages 2.88 ± 4% +3.8 6.67 ± 5% perf-profile.children.cycles-pp.xfs_bmapi_convert_delalloc 3.21 ± 3% +3.8 7.02 ± 5% perf-profile.children.cycles-pp.iomap_writepage_map 0.21 ± 9% -0.1 0.15 ± 10% perf-profile.self.cycles-pp.xfs_perag_get 0.32 ± 5% -0.0 0.28 ± 10% perf-profile.self.cycles-pp.lapic_next_deadline 0.06 ± 7% -0.0 0.04 ± 71% perf-profile.self.cycles-pp.memcg_slab_post_alloc_hook 0.11 ± 6% -0.0 0.09 ± 10% perf-profile.self.cycles-pp.__intel_pmu_enable_all 0.12 ± 6% -0.0 0.10 ± 8% perf-profile.self.cycles-pp.down_read 0.01 ±223% +0.1 0.08 ± 10% perf-profile.self.cycles-pp.xfs_allocbt_init_key_from_rec 0.00 +0.1 0.07 ± 10% perf-profile.self.cycles-pp.xfs_btree_rec_offset 0.00 +0.1 0.09 ± 15% perf-profile.self.cycles-pp.xfs_extent_busy_trim 0.00 +0.1 0.09 ± 18% perf-profile.self.cycles-pp.xfs_btree_check_block 0.00 +0.1 0.10 ± 16% perf-profile.self.cycles-pp.xfs_alloc_compute_aligned 0.06 ± 51% +0.1 0.17 ± 12% perf-profile.self.cycles-pp.xfs_errortag_test 0.00 +0.1 0.12 ± 11% perf-profile.self.cycles-pp.xfs_btree_get_rec 0.00 +0.1 0.14 ± 15% perf-profile.self.cycles-pp.xfs_btree_check_sblock 0.00 +0.2 0.16 ± 6% perf-profile.self.cycles-pp.xfs_btree_increment 0.00 +0.2 0.20 ± 3% perf-profile.self.cycles-pp.xfs_alloc_compute_diff 1.10 ± 6% +0.2 1.33 ± 3% perf-profile.self.cycles-pp._raw_spin_lock 1.57 ± 4% +0.3 1.82 ± 6% perf-profile.self.cycles-pp.__orc_find 0.00 +0.3 0.27 ± 6% perf-profile.self.cycles-pp.xfs_alloc_get_rec 0.00 +0.3 0.28 ± 9% perf-profile.self.cycles-pp.xfs_alloc_cur_check 0.13 ± 12% +0.7 0.79 ± 7% perf-profile.self.cycles-pp.__xfs_btree_check_sblock *************************************************************************************************** lkp-spr-r02: 224 threads 2 sockets (Sapphire Rapids) with 256G memory ========================================================================================= compiler/cpufreq_governor/disk/filesize/fs/iterations/kconfig/nr_directories/nr_files_per_directory/nr_threads/rootfs/sync_method/tbox_group/test_size/testcase: gcc-11/performance/1SSD/8K/xfs/8/x86_64-rhel-8.3/16d/256fpd/32/debian-11.1-x86_64-20220510.cgz/fsyncBeforeClose/lkp-spr-r02/50G/fsmark commit: ecd788a924 ("xfs: rework xfs_alloc_vextent()") 2edf06a50f ("xfs: factor xfs_alloc_vextent_this_ag() for _iterate_ags()") ecd788a92460eef4 2edf06a50f5bbe664283f3c55c4 ---------------- --------------------------- %stddev %change %stddev \ | \ 71096 -3.7% 68492 fsmark.files_per_sec 537.67 +5.8% 569.00 fsmark.time.percent_of_cpu_this_job_got 606.41 +9.1% 661.29 fsmark.time.system_time 33180376 +4.8% 34779903 fsmark.time.voluntary_context_switches 3.76 +3.2% 3.88 iostat.cpu.system 0.34 +21.6% 0.41 turbostat.IPC 809115 ± 17% +19.6% 968066 ± 15% turbostat.POLL 921206 +3.5% 953292 vmstat.io.bo 728873 +1.5% 739648 vmstat.system.cs 1178 -3.3% 1140 ± 2% proc-vmstat.direct_map_level2_splits 1118 +5.9% 1184 proc-vmstat.nr_active_anon 1118 +5.9% 1184 proc-vmstat.nr_zone_active_anon 5814 -6.9% 5411 proc-vmstat.numa_huge_pte_updates 3078897 -6.8% 2870651 proc-vmstat.numa_pte_updates 7925 -4.3% 7584 proc-vmstat.pgactivate 1.099e+08 +6.4% 1.169e+08 proc-vmstat.pgpgout 976896 +2.5% 1000960 proc-vmstat.unevictable_pgs_scanned 13064 ± 11% +32.7% 17334 ± 12% sched_debug.cfs_rq:/.min_vruntime.stddev 122.15 ± 15% -32.4% 82.52 ± 16% sched_debug.cfs_rq:/.runnable_avg.avg 1169 ± 10% -24.5% 883.06 ± 2% sched_debug.cfs_rq:/.runnable_avg.max 211.40 ± 12% -27.4% 153.52 ± 8% sched_debug.cfs_rq:/.runnable_avg.stddev 13115 ± 11% +32.2% 17337 ± 12% sched_debug.cfs_rq:/.spread0.stddev 121.88 ± 15% -32.5% 82.25 ± 16% sched_debug.cfs_rq:/.util_avg.avg 1168 ± 10% -25.8% 867.28 sched_debug.cfs_rq:/.util_avg.max 210.78 ± 13% -27.6% 152.65 ± 8% sched_debug.cfs_rq:/.util_avg.stddev 10.49 ± 19% -39.8% 6.31 ± 28% sched_debug.cfs_rq:/.util_est_enqueued.avg 69.83 ± 19% -35.0% 45.42 ± 22% sched_debug.cfs_rq:/.util_est_enqueued.stddev 72309 ± 19% +41.5% 102337 ± 13% sched_debug.cpu.clock.avg 72324 ± 19% +41.5% 102351 ± 13% sched_debug.cpu.clock.max 72292 ± 19% +41.5% 102321 ± 13% sched_debug.cpu.clock.min 71938 ± 19% +41.3% 101681 ± 13% sched_debug.cpu.clock_task.avg 72187 ± 19% +41.2% 101961 ± 13% sched_debug.cpu.clock_task.max 5843 ± 18% +33.2% 7781 ± 7% sched_debug.cpu.curr->pid.max 41950 ±137% +275.6% 157567 ± 27% sched_debug.cpu.nr_switches.avg 177180 ± 91% +190.6% 514800 ± 11% sched_debug.cpu.nr_switches.max 31279 ±121% +264.0% 113856 ± 16% sched_debug.cpu.nr_switches.stddev 72293 ± 19% +41.5% 102321 ± 13% sched_debug.cpu_clk 71040 ± 19% +42.3% 101061 ± 14% sched_debug.ktime 73365 ± 19% +40.9% 103391 ± 13% sched_debug.sched_clk 13.52 -16.6% 11.27 ± 3% perf-stat.i.MPKI 3.816e+09 +22.2% 4.663e+09 perf-stat.i.branch-instructions 0.75 -0.1 0.63 perf-stat.i.branch-miss-rate% 73259837 +6.2% 77781445 perf-stat.i.cache-misses 750409 +1.7% 763405 perf-stat.i.context-switches 1.76 -15.9% 1.48 perf-stat.i.cpi 3.367e+10 +2.5% 3.451e+10 perf-stat.i.cpu-cycles 5.662e+09 +27.5% 7.218e+09 perf-stat.i.dTLB-loads 220394 ± 4% +18.5% 261185 ± 9% perf-stat.i.dTLB-store-misses 2.753e+09 +28.0% 3.523e+09 perf-stat.i.dTLB-stores 2.019e+10 +24.8% 2.52e+10 perf-stat.i.instructions 0.59 +19.5% 0.70 perf-stat.i.ipc 0.15 +2.5% 0.15 perf-stat.i.metric.GHz 55.82 +25.5% 70.04 perf-stat.i.metric.M/sec 18252836 ± 2% +4.4% 19056718 perf-stat.i.node-load-misses 14.22 -16.9% 11.82 ± 4% perf-stat.overall.MPKI 0.81 -0.1 0.67 perf-stat.overall.branch-miss-rate% 1.67 -17.9% 1.37 perf-stat.overall.cpi 459.70 -3.5% 443.70 ± 2% perf-stat.overall.cycles-between-cache-misses 0.01 ± 4% -0.0 0.01 ± 7% perf-stat.overall.dTLB-store-miss-rate% 0.60 +21.8% 0.73 perf-stat.overall.ipc 3.783e+09 +22.1% 4.62e+09 perf-stat.ps.branch-instructions 72614040 +6.2% 77102710 perf-stat.ps.cache-misses 743776 +1.7% 756437 perf-stat.ps.context-switches 3.337e+10 +2.5% 3.42e+10 perf-stat.ps.cpu-cycles 5.612e+09 +27.5% 7.153e+09 ± 2% perf-stat.ps.dTLB-loads 218292 ± 4% +18.5% 258697 ± 9% perf-stat.ps.dTLB-store-misses 2.729e+09 +27.9% 3.491e+09 perf-stat.ps.dTLB-stores 2.002e+10 +24.8% 2.498e+10 perf-stat.ps.instructions 18090782 ± 2% +4.4% 18888511 perf-stat.ps.node-load-misses 2.339e+12 +28.0% 2.994e+12 perf-stat.total.instructions 22.79 -3.6 19.20 ± 7% perf-profile.calltrace.cycles-pp.evict_inodes.generic_shutdown_super.kill_block_super.deactivate_locked_super.cleanup_mnt 20.93 -3.3 17.65 ± 7% perf-profile.calltrace.cycles-pp.dispose_list.evict_inodes.generic_shutdown_super.kill_block_super.deactivate_locked_super 9.63 -2.5 7.10 ± 24% perf-profile.calltrace.cycles-pp.__do_softirq.run_ksoftirqd.smpboot_thread_fn.kthread.ret_from_fork 9.63 -2.5 7.10 ± 24% perf-profile.calltrace.cycles-pp.run_ksoftirqd.smpboot_thread_fn.kthread.ret_from_fork 9.79 -2.5 7.27 ± 23% perf-profile.calltrace.cycles-pp.smpboot_thread_fn.kthread.ret_from_fork 9.62 -2.5 7.10 ± 24% perf-profile.calltrace.cycles-pp.rcu_core.__do_softirq.run_ksoftirqd.smpboot_thread_fn.kthread 9.60 -2.5 7.08 ± 24% perf-profile.calltrace.cycles-pp.rcu_do_batch.rcu_core.__do_softirq.run_ksoftirqd.smpboot_thread_fn 13.98 -2.1 11.91 ± 6% perf-profile.calltrace.cycles-pp.evict.dispose_list.evict_inodes.generic_shutdown_super.kill_block_super 22.80 -2.1 20.73 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.umount2 22.80 -2.1 20.73 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.umount2 22.80 -2.1 20.73 perf-profile.calltrace.cycles-pp.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.umount2 22.80 -2.1 20.73 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 22.80 -2.1 20.73 perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.umount2 22.80 -2.1 20.73 perf-profile.calltrace.cycles-pp.task_work_run.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64 22.80 -2.1 20.73 perf-profile.calltrace.cycles-pp.umount2 22.80 -2.1 20.73 perf-profile.calltrace.cycles-pp.cleanup_mnt.task_work_run.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode 22.80 -2.1 20.73 perf-profile.calltrace.cycles-pp.deactivate_locked_super.cleanup_mnt.task_work_run.exit_to_user_mode_loop.exit_to_user_mode_prepare 22.80 -2.1 20.73 perf-profile.calltrace.cycles-pp.kill_block_super.deactivate_locked_super.cleanup_mnt.task_work_run.exit_to_user_mode_loop 22.80 -2.1 20.73 perf-profile.calltrace.cycles-pp.generic_shutdown_super.kill_block_super.deactivate_locked_super.cleanup_mnt.task_work_run 11.34 -1.6 9.74 ± 7% perf-profile.calltrace.cycles-pp.truncate_inode_pages_range.evict.dispose_list.evict_inodes.generic_shutdown_super 5.64 ± 3% -1.0 4.59 ± 8% perf-profile.calltrace.cycles-pp.destroy_inode.dispose_list.evict_inodes.generic_shutdown_super.kill_block_super 4.09 ± 4% -0.7 3.35 ± 19% perf-profile.calltrace.cycles-pp.xfs_icwalk_ag.xfs_icwalk.xfs_reclaim_worker.process_one_work.worker_thread 4.10 ± 4% -0.7 3.36 ± 19% perf-profile.calltrace.cycles-pp.xfs_reclaim_worker.process_one_work.worker_thread.kthread.ret_from_fork 4.10 ± 4% -0.7 3.36 ± 19% perf-profile.calltrace.cycles-pp.xfs_icwalk.xfs_reclaim_worker.process_one_work.worker_thread.kthread 1.81 -0.7 1.10 ± 43% perf-profile.calltrace.cycles-pp.xfs_inode_item_destroy.xfs_inode_free_callback.rcu_do_batch.rcu_core.__do_softirq 4.05 -0.6 3.42 ± 8% perf-profile.calltrace.cycles-pp.find_lock_entries.truncate_inode_pages_range.evict.dispose_list.evict_inodes 3.72 -0.5 3.27 ± 6% perf-profile.calltrace.cycles-pp.__pagevec_release.truncate_inode_pages_range.evict.dispose_list.evict_inodes 3.64 -0.4 3.19 ± 7% perf-profile.calltrace.cycles-pp.release_pages.__pagevec_release.truncate_inode_pages_range.evict.dispose_list 2.32 ± 4% -0.4 1.89 ± 16% perf-profile.calltrace.cycles-pp.xfs_reclaim_inode.xfs_icwalk_ag.xfs_icwalk.xfs_reclaim_worker.process_one_work 2.30 -0.4 1.88 ± 8% perf-profile.calltrace.cycles-pp.xfs_inode_mark_reclaimable.destroy_inode.dispose_list.evict_inodes.generic_shutdown_super 2.63 ± 3% -0.4 2.25 ± 6% perf-profile.calltrace.cycles-pp.delete_from_page_cache_batch.truncate_inode_pages_range.evict.dispose_list.evict_inodes 2.59 -0.4 2.24 ± 9% perf-profile.calltrace.cycles-pp.xas_find.find_lock_entries.truncate_inode_pages_range.evict.dispose_list 1.93 ± 2% -0.4 1.58 ± 8% perf-profile.calltrace.cycles-pp.xfs_can_free_eofblocks.xfs_inode_mark_reclaimable.destroy_inode.dispose_list.evict_inodes 1.73 ± 2% -0.3 1.39 ± 10% perf-profile.calltrace.cycles-pp.__destroy_inode.destroy_inode.dispose_list.evict_inodes.generic_shutdown_super 1.62 ± 2% -0.3 1.28 ± 9% perf-profile.calltrace.cycles-pp.fsnotify_destroy_marks.__destroy_inode.destroy_inode.dispose_list.evict_inodes 1.58 ± 3% -0.3 1.25 ± 9% perf-profile.calltrace.cycles-pp.fsnotify_grab_connector.fsnotify_destroy_marks.__destroy_inode.destroy_inode.dispose_list 3.79 ± 7% -0.3 3.46 ± 5% perf-profile.calltrace.cycles-pp.timekeeping_max_deferment.tick_nohz_next_event.tick_nohz_get_sleep_length.menu_select.cpuidle_idle_call 1.61 ± 2% -0.3 1.34 ± 5% perf-profile.calltrace.cycles-pp.free_unref_page_list.release_pages.__pagevec_release.truncate_inode_pages_range.evict 1.31 -0.3 1.04 ± 9% perf-profile.calltrace.cycles-pp.kmem_cache_free.rcu_do_batch.rcu_core.__do_softirq.run_ksoftirqd 0.58 -0.2 0.36 ± 70% perf-profile.calltrace.cycles-pp.xas_store.delete_from_page_cache_batch.truncate_inode_pages_range.evict.dispose_list 0.91 ± 3% -0.2 0.73 ± 10% perf-profile.calltrace.cycles-pp.down_read.xfs_can_free_eofblocks.xfs_inode_mark_reclaimable.destroy_inode.dispose_list 0.95 ± 9% -0.2 0.77 ± 9% perf-profile.calltrace.cycles-pp.xfs_inodegc_set_reclaimable.destroy_inode.dispose_list.evict_inodes.generic_shutdown_super 0.89 -0.2 0.74 ± 3% perf-profile.calltrace.cycles-pp.free_pcppages_bulk.free_unref_page_list.release_pages.__pagevec_release.truncate_inode_pages_range 0.81 ± 5% -0.2 0.66 perf-profile.calltrace.cycles-pp.inode_io_list_del.evict.dispose_list.evict_inodes.generic_shutdown_super 0.73 ± 2% -0.1 0.60 ± 4% perf-profile.calltrace.cycles-pp.__free_one_page.free_pcppages_bulk.free_unref_page_list.release_pages.__pagevec_release 0.70 ± 4% -0.1 0.61 ± 6% perf-profile.calltrace.cycles-pp.xfs_bmapi_read.xfs_can_free_eofblocks.xfs_inode_mark_reclaimable.destroy_inode.dispose_list 0.63 ± 5% +0.1 0.68 ± 4% perf-profile.calltrace.cycles-pp.native_apic_msr_eoi_write.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state 0.60 ± 4% +0.1 0.66 perf-profile.calltrace.cycles-pp.perf_mux_hrtimer_handler.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt 0.86 ± 4% +0.1 0.93 perf-profile.calltrace.cycles-pp.lapic_next_deadline.clockevents_program_event.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt 0.74 ± 2% +0.1 0.88 ± 4% perf-profile.calltrace.cycles-pp.ktime_get_update_offsets_now.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt 0.17 ±141% +0.4 0.53 ± 2% perf-profile.calltrace.cycles-pp.check_cpu_stall.rcu_pending.rcu_sched_clock_irq.update_process_times.tick_sched_handle 11.33 +0.8 12.17 ± 7% perf-profile.calltrace.cycles-pp.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter 22.97 +1.2 24.14 ± 4% perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle 26.56 +1.2 27.80 ± 2% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle 26.55 +1.2 27.79 ± 2% perf-profile.calltrace.cycles-pp.mwait_idle_with_hints.intel_idle.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call 0.00 +1.6 1.58 ± 46% perf-profile.calltrace.cycles-pp.do_writepages.__writeback_single_inode.writeback_sb_inodes.wb_writeback.wb_do_writeback 0.00 +1.6 1.64 ± 47% perf-profile.calltrace.cycles-pp.__writeback_single_inode.writeback_sb_inodes.wb_writeback.wb_do_writeback.wb_workfn 60.97 +2.1 63.10 perf-profile.calltrace.cycles-pp.secondary_startup_64_no_verify 60.68 +2.1 62.83 perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify 60.68 +2.1 62.83 perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64_no_verify 60.66 +2.1 62.81 perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify 59.73 +2.2 61.88 perf-profile.calltrace.cycles-pp.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify 50.48 +2.3 52.79 ± 3% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry 51.52 +2.4 53.92 ± 3% perf-profile.calltrace.cycles-pp.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary 4.18 ± 4% +2.6 6.73 ± 17% perf-profile.calltrace.cycles-pp.worker_thread.kthread.ret_from_fork 4.15 ± 4% +2.6 6.70 ± 17% perf-profile.calltrace.cycles-pp.process_one_work.worker_thread.kthread.ret_from_fork 0.00 +2.6 2.61 ± 47% perf-profile.calltrace.cycles-pp.writeback_sb_inodes.wb_writeback.wb_do_writeback.wb_workfn.process_one_work 0.00 +3.3 3.28 ± 51% perf-profile.calltrace.cycles-pp.wb_workfn.process_one_work.worker_thread.kthread.ret_from_fork 0.00 +3.3 3.28 ± 51% perf-profile.calltrace.cycles-pp.wb_do_writeback.wb_workfn.process_one_work.worker_thread.kthread 0.00 +3.3 3.28 ± 51% perf-profile.calltrace.cycles-pp.wb_writeback.wb_do_writeback.wb_workfn.process_one_work.worker_thread 22.80 -3.6 19.21 ± 7% perf-profile.children.cycles-pp.evict_inodes 20.95 -3.3 17.67 ± 7% perf-profile.children.cycles-pp.dispose_list 9.63 -2.5 7.10 ± 24% perf-profile.children.cycles-pp.run_ksoftirqd 9.79 -2.5 7.27 ± 23% perf-profile.children.cycles-pp.smpboot_thread_fn 11.02 -2.4 8.61 ± 19% perf-profile.children.cycles-pp.rcu_core 23.93 -2.1 21.80 perf-profile.children.cycles-pp.do_syscall_64 23.93 -2.1 21.81 perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe 14.02 -2.1 11.94 ± 6% perf-profile.children.cycles-pp.evict 22.87 -2.1 20.80 perf-profile.children.cycles-pp.exit_to_user_mode_prepare 22.86 -2.1 20.79 perf-profile.children.cycles-pp.exit_to_user_mode_loop 22.86 -2.1 20.79 perf-profile.children.cycles-pp.task_work_run 22.80 -2.1 20.73 perf-profile.children.cycles-pp.umount2 22.80 -2.1 20.73 perf-profile.children.cycles-pp.cleanup_mnt 22.80 -2.1 20.73 perf-profile.children.cycles-pp.deactivate_locked_super 22.80 -2.1 20.73 perf-profile.children.cycles-pp.kill_block_super 22.80 -2.1 20.73 perf-profile.children.cycles-pp.generic_shutdown_super 22.86 -2.1 20.80 perf-profile.children.cycles-pp.syscall_exit_to_user_mode 11.40 -1.6 9.77 ± 7% perf-profile.children.cycles-pp.truncate_inode_pages_range 5.66 ± 3% -1.0 4.61 ± 8% perf-profile.children.cycles-pp.destroy_inode 4.10 ± 4% -0.7 3.36 ± 19% perf-profile.children.cycles-pp.xfs_reclaim_worker 4.10 ± 4% -0.7 3.36 ± 19% perf-profile.children.cycles-pp.xfs_icwalk 4.10 ± 4% -0.7 3.36 ± 19% perf-profile.children.cycles-pp.xfs_icwalk_ag 1.82 -0.7 1.11 ± 43% perf-profile.children.cycles-pp.xfs_inode_item_destroy 4.08 -0.6 3.45 ± 8% perf-profile.children.cycles-pp.find_lock_entries 3.74 -0.5 3.27 ± 6% perf-profile.children.cycles-pp.__pagevec_release 3.70 -0.5 3.25 ± 6% perf-profile.children.cycles-pp.release_pages 2.35 ± 4% -0.4 1.92 ± 16% perf-profile.children.cycles-pp.xfs_reclaim_inode 2.31 -0.4 1.89 ± 8% perf-profile.children.cycles-pp.xfs_inode_mark_reclaimable 2.82 -0.4 2.42 ± 9% perf-profile.children.cycles-pp.xas_find 2.69 ± 3% -0.4 2.29 ± 6% perf-profile.children.cycles-pp.delete_from_page_cache_batch 1.96 ± 2% -0.4 1.61 ± 8% perf-profile.children.cycles-pp.xfs_can_free_eofblocks 1.76 ± 3% -0.4 1.41 ± 10% perf-profile.children.cycles-pp.__destroy_inode 1.63 ± 3% -0.3 1.29 ± 9% perf-profile.children.cycles-pp.fsnotify_destroy_marks 1.60 ± 2% -0.3 1.26 ± 8% perf-profile.children.cycles-pp.fsnotify_grab_connector 3.81 ± 7% -0.3 3.48 ± 5% perf-profile.children.cycles-pp.timekeeping_max_deferment 1.72 -0.3 1.43 ± 10% perf-profile.children.cycles-pp.kmem_cache_free 1.27 -0.3 0.99 ± 9% perf-profile.children.cycles-pp.free_pcppages_bulk 1.65 ± 2% -0.3 1.38 ± 5% perf-profile.children.cycles-pp.free_unref_page_list 1.20 ± 4% -0.2 1.01 ± 7% perf-profile.children.cycles-pp._raw_spin_lock_irqsave 0.92 ± 3% -0.2 0.75 ± 9% perf-profile.children.cycles-pp.down_read 0.97 ± 9% -0.2 0.80 ± 8% perf-profile.children.cycles-pp.xfs_inodegc_set_reclaimable 0.43 ± 2% -0.2 0.27 ± 23% perf-profile.children.cycles-pp.free_unref_page 0.90 -0.2 0.74 ± 9% perf-profile.children.cycles-pp.__free_one_page 0.84 ± 5% -0.2 0.69 perf-profile.children.cycles-pp.inode_io_list_del 0.33 ± 10% -0.1 0.18 ± 18% perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath 0.72 ± 4% -0.1 0.63 ± 6% perf-profile.children.cycles-pp.xfs_bmapi_read 0.46 ± 6% -0.1 0.36 ± 10% perf-profile.children.cycles-pp.down_write_trylock 0.48 ± 5% -0.1 0.39 ± 7% perf-profile.children.cycles-pp.xfs_ilock_nowait 0.62 -0.1 0.53 ± 10% perf-profile.children.cycles-pp.xas_store 0.24 ± 9% -0.1 0.17 ± 10% perf-profile.children.cycles-pp.free_unref_page_commit 0.41 ± 10% -0.1 0.34 ± 4% perf-profile.children.cycles-pp.xfs_perag_set_inode_tag 0.31 ± 6% -0.1 0.24 ± 14% perf-profile.children.cycles-pp.read 0.18 ± 2% -0.1 0.12 ± 15% perf-profile.children.cycles-pp.clear_inode 0.17 ± 4% -0.1 0.11 ± 19% perf-profile.children.cycles-pp.get_slabinfo 0.17 ± 4% -0.1 0.11 ± 23% perf-profile.children.cycles-pp.slab_show 0.30 ± 2% -0.1 0.24 ± 13% perf-profile.children.cycles-pp.xfs_iunlock 0.41 ± 6% -0.1 0.36 ± 6% perf-profile.children.cycles-pp.filemap_unaccount_folio 0.14 ± 5% -0.1 0.09 ± 9% perf-profile.children.cycles-pp.find_get_entries 0.14 ± 8% -0.1 0.09 ± 5% perf-profile.children.cycles-pp.truncate_cleanup_folio 0.18 ± 7% -0.0 0.13 ± 15% perf-profile.children.cycles-pp.xfs_ilock 0.48 ± 5% -0.0 0.43 ± 5% perf-profile.children.cycles-pp.ksys_read 0.47 ± 5% -0.0 0.43 ± 3% perf-profile.children.cycles-pp.vfs_read 0.19 ± 15% -0.0 0.14 ± 20% perf-profile.children.cycles-pp.xfs_perag_put 0.16 ± 3% -0.0 0.11 ± 8% perf-profile.children.cycles-pp.obj_cgroup_uncharge_pages 0.38 ± 4% -0.0 0.34 ± 4% perf-profile.children.cycles-pp.seq_read_iter 0.33 ± 5% -0.0 0.29 ± 5% perf-profile.children.cycles-pp.seq_read 0.17 -0.0 0.14 ± 6% perf-profile.children.cycles-pp.filemap_free_folio 0.12 ± 8% -0.0 0.08 ± 20% perf-profile.children.cycles-pp.down_write 0.10 ± 9% -0.0 0.07 ± 11% perf-profile.children.cycles-pp.__srcu_read_lock 0.12 ± 6% -0.0 0.09 ± 10% perf-profile.children.cycles-pp.__mod_zone_page_state 0.18 ± 5% -0.0 0.15 ± 11% perf-profile.children.cycles-pp.radix_tree_node_rcu_free 0.07 ± 7% -0.0 0.04 ± 71% perf-profile.children.cycles-pp.__inode_wait_for_writeback 0.50 -0.0 0.48 perf-profile.children.cycles-pp.irq_enter_rcu 0.46 -0.0 0.43 perf-profile.children.cycles-pp.tick_irq_enter 0.12 ± 3% -0.0 0.10 ± 14% perf-profile.children.cycles-pp.cgroup_rstat_updated 0.09 ± 9% -0.0 0.07 ± 7% perf-profile.children.cycles-pp.up_read 0.08 ± 10% -0.0 0.06 ± 16% perf-profile.children.cycles-pp.node_tag_clear 0.30 ± 3% -0.0 0.28 ± 5% perf-profile.children.cycles-pp.radix_tree_tag_set 0.10 ± 12% -0.0 0.08 ± 12% perf-profile.children.cycles-pp.force_qs_rnp 0.07 ± 11% -0.0 0.05 perf-profile.children.cycles-pp.refill_stock 0.07 ± 17% -0.0 0.05 ± 8% perf-profile.children.cycles-pp.dyntick_save_progress_counter 0.14 ± 3% -0.0 0.12 ± 6% perf-profile.children.cycles-pp.timerqueue_del 0.10 ± 8% -0.0 0.08 ± 5% perf-profile.children.cycles-pp.rcu_all_qs 0.08 ± 10% -0.0 0.06 ± 7% perf-profile.children.cycles-pp.__count_memcg_events 0.08 ± 5% -0.0 0.07 perf-profile.children.cycles-pp.filename_lookup 0.08 ± 5% -0.0 0.07 perf-profile.children.cycles-pp.path_lookupat 0.28 ± 2% +0.0 0.31 ± 5% perf-profile.children.cycles-pp.__intel_pmu_enable_all 0.43 ± 2% +0.0 0.46 ± 3% perf-profile.children.cycles-pp.perf_rotate_context 0.49 ± 3% +0.0 0.53 ± 3% perf-profile.children.cycles-pp.perf_session__deliver_event 0.02 ±141% +0.0 0.06 ± 8% perf-profile.children.cycles-pp.get_cpu_device 0.31 ± 9% +0.0 0.35 ± 2% perf-profile.children.cycles-pp.hrtimer_next_event_without 0.23 ± 6% +0.0 0.28 ± 4% perf-profile.children.cycles-pp.__hrtimer_next_event_base 0.63 ± 3% +0.0 0.68 perf-profile.children.cycles-pp.perf_mux_hrtimer_handler 0.78 ± 5% +0.1 0.84 perf-profile.children.cycles-pp.native_apic_msr_eoi_write 0.91 ± 4% +0.1 0.98 perf-profile.children.cycles-pp.lapic_next_deadline 0.43 ± 14% +0.1 0.54 ± 2% perf-profile.children.cycles-pp.check_cpu_stall 0.77 +0.1 0.91 ± 3% perf-profile.children.cycles-pp.ktime_get_update_offsets_now 0.00 +0.2 0.21 ± 50% perf-profile.children.cycles-pp.xas_find_marked 0.00 +0.2 0.22 ± 47% perf-profile.children.cycles-pp.tag_pages_for_writeback 0.00 +0.3 0.34 ± 49% perf-profile.children.cycles-pp.write_cache_pages 0.00 +0.3 0.34 ± 49% perf-profile.children.cycles-pp.iomap_writepages 0.00 +0.5 0.48 ± 47% perf-profile.children.cycles-pp.inode_cgwb_move_to_attached 1.90 ± 4% +0.8 2.73 ± 16% perf-profile.children.cycles-pp.__list_del_entry_valid 19.09 +1.0 20.14 ± 5% perf-profile.children.cycles-pp.sysvec_apic_timer_interrupt 21.60 +1.2 22.81 ± 5% perf-profile.children.cycles-pp.asm_sysvec_apic_timer_interrupt 0.00 +1.2 1.21 ± 48% perf-profile.children.cycles-pp.xfs_vm_writepages 26.70 +1.2 27.92 ± 2% perf-profile.children.cycles-pp.intel_idle 26.68 +1.3 27.97 ± 2% perf-profile.children.cycles-pp.mwait_idle_with_hints 0.00 +1.6 1.58 ± 46% perf-profile.children.cycles-pp.do_writepages 0.00 +1.6 1.64 ± 47% perf-profile.children.cycles-pp.__writeback_single_inode 60.97 +2.1 63.10 perf-profile.children.cycles-pp.secondary_startup_64_no_verify 60.97 +2.1 63.10 perf-profile.children.cycles-pp.cpu_startup_entry 60.97 +2.1 63.10 perf-profile.children.cycles-pp.do_idle 60.05 +2.1 62.19 perf-profile.children.cycles-pp.cpuidle_idle_call 60.68 +2.1 62.83 perf-profile.children.cycles-pp.start_secondary 51.71 +2.4 54.09 ± 3% perf-profile.children.cycles-pp.cpuidle_enter_state 51.77 +2.4 54.16 ± 3% perf-profile.children.cycles-pp.cpuidle_enter 4.18 ± 4% +2.6 6.73 ± 17% perf-profile.children.cycles-pp.worker_thread 4.15 ± 4% +2.6 6.70 ± 17% perf-profile.children.cycles-pp.process_one_work 0.00 +2.6 2.64 ± 47% perf-profile.children.cycles-pp.writeback_sb_inodes 0.00 +3.3 3.28 ± 51% perf-profile.children.cycles-pp.wb_workfn 0.00 +3.3 3.28 ± 51% perf-profile.children.cycles-pp.wb_do_writeback 0.00 +3.3 3.28 ± 51% perf-profile.children.cycles-pp.wb_writeback 3.81 ± 7% -0.3 3.48 ± 5% perf-profile.self.cycles-pp.timekeeping_max_deferment 2.06 ± 2% -0.3 1.75 ± 8% perf-profile.self.cycles-pp.xas_find 1.36 ± 2% -0.3 1.09 ± 10% perf-profile.self.cycles-pp.fsnotify_grab_connector 1.41 -0.2 1.17 ± 5% perf-profile.self.cycles-pp.find_lock_entries 1.19 ± 5% -0.2 0.97 ± 6% perf-profile.self.cycles-pp.delete_from_page_cache_batch 0.82 ± 3% -0.2 0.63 ± 7% perf-profile.self.cycles-pp.down_read 0.88 ± 2% -0.2 0.72 ± 12% perf-profile.self.cycles-pp.kmem_cache_free 1.17 -0.2 1.01 ± 9% perf-profile.self.cycles-pp.evict_inodes 0.33 ± 10% -0.1 0.18 ± 18% perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath 0.65 ± 6% -0.1 0.53 ± 9% perf-profile.self.cycles-pp.evict 0.68 ± 2% -0.1 0.57 ± 8% perf-profile.self.cycles-pp.__free_one_page 0.97 ± 5% -0.1 0.88 ± 5% perf-profile.self.cycles-pp._raw_spin_lock_irqsave 0.45 ± 6% -0.1 0.36 ± 11% perf-profile.self.cycles-pp.down_write_trylock 0.17 ± 4% -0.1 0.11 ± 19% perf-profile.self.cycles-pp.get_slabinfo 0.49 ± 3% -0.1 0.43 ± 6% perf-profile.self.cycles-pp.release_pages 0.20 ± 6% -0.1 0.14 ± 11% perf-profile.self.cycles-pp.free_unref_page_commit 0.14 ± 10% -0.1 0.09 ± 28% perf-profile.self.cycles-pp.xfs_inode_item_destroy 0.14 ± 3% -0.0 0.10 ± 24% perf-profile.self.cycles-pp.xfs_icwalk_ag 0.15 ± 8% -0.0 0.10 ± 12% perf-profile.self.cycles-pp.xfs_can_free_eofblocks 0.14 ± 6% -0.0 0.10 ± 25% perf-profile.self.cycles-pp.rcu_do_batch 0.18 ± 16% -0.0 0.14 ± 18% perf-profile.self.cycles-pp.xfs_perag_put 0.12 ± 10% -0.0 0.09 ± 5% perf-profile.self.cycles-pp.truncate_cleanup_folio 0.10 ± 14% -0.0 0.06 ± 19% perf-profile.self.cycles-pp.__srcu_read_lock 0.18 ± 4% -0.0 0.15 ± 8% perf-profile.self.cycles-pp.radix_tree_node_rcu_free 0.14 ± 5% -0.0 0.11 ± 4% perf-profile.self.cycles-pp.xfs_bmapi_read 0.16 ± 3% -0.0 0.13 ± 7% perf-profile.self.cycles-pp.filemap_free_folio 0.15 ± 3% -0.0 0.12 ± 3% perf-profile.self.cycles-pp.sched_clock_cpu 0.10 ± 12% -0.0 0.07 ± 6% perf-profile.self.cycles-pp.filemap_unaccount_folio 0.08 ± 5% -0.0 0.06 ± 13% perf-profile.self.cycles-pp.up_read 0.17 ± 7% -0.0 0.14 ± 3% perf-profile.self.cycles-pp.truncate_inode_pages_range 0.29 ± 4% -0.0 0.27 ± 3% perf-profile.self.cycles-pp.radix_tree_tag_set 0.07 ± 17% -0.0 0.05 ± 8% perf-profile.self.cycles-pp.dyntick_save_progress_counter 0.17 ± 5% -0.0 0.15 ± 8% perf-profile.self.cycles-pp.__mod_memcg_lruvec_state 0.10 ± 9% -0.0 0.08 ± 6% perf-profile.self.cycles-pp.__mod_zone_page_state 0.12 ± 8% -0.0 0.10 ± 4% perf-profile.self.cycles-pp.free_pcp_prepare 0.12 ± 4% -0.0 0.10 perf-profile.self.cycles-pp.__list_add_valid 0.06 ± 7% -0.0 0.05 perf-profile.self.cycles-pp.rcu_all_qs 0.07 ± 7% +0.0 0.08 perf-profile.self.cycles-pp.tick_sched_timer 0.28 ± 2% +0.0 0.31 ± 5% perf-profile.self.cycles-pp.__intel_pmu_enable_all 0.81 +0.0 0.85 ± 3% perf-profile.self.cycles-pp.read_tsc 0.77 ± 5% +0.1 0.83 perf-profile.self.cycles-pp.native_apic_msr_eoi_write 0.90 ± 4% +0.1 0.98 perf-profile.self.cycles-pp.lapic_next_deadline 0.43 ± 14% +0.1 0.54 ± 2% perf-profile.self.cycles-pp.check_cpu_stall 0.61 ± 3% +0.1 0.73 ± 3% perf-profile.self.cycles-pp.ktime_get_update_offsets_now 3.20 ± 2% +0.2 3.38 ± 2% perf-profile.self.cycles-pp.cpuidle_enter_state 0.00 +0.2 0.20 ± 49% perf-profile.self.cycles-pp.xas_find_marked 0.00 +0.2 0.23 ± 50% perf-profile.self.cycles-pp.writeback_sb_inodes 0.00 +0.4 0.37 ± 42% perf-profile.self.cycles-pp.do_writepages 1.79 ± 3% +0.8 2.60 ± 16% perf-profile.self.cycles-pp.__list_del_entry_valid 26.67 +1.3 27.96 ± 2% perf-profile.self.cycles-pp.mwait_idle_with_hints 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