Greeting, FYI, we noticed a 47.8% improvement of fio.write_iops due to commit: commit: fd25a9e0e23b995fd0ba5e2f00a1099452cbc3cf ("memcg: unify memcg stat flushing") https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master in testcase: fio-basic on test machine: 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 512G memory with following parameters: disk: 2pmem fs: ext4 runtime: 200s nr_task: 50% time_based: tb rw: randrw bs: 4k ioengine: sync test_size: 200G cpufreq_governor: performance ucode: 0x500320a test-description: Fio is a tool that will spawn a number of threads or processes doing a particular type of I/O action as specified by the user. test-url: https://github.com/axboe/fio In addition to that, the commit also has significant impact on the following tests: +------------------+--------------------------------------------------------------------------------+ | testcase: change | fio-basic: fio.write_iops 23.1% improvement | | test machine | 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 512G memory | | test parameters | bs=4k | | | cpufreq_governor=performance | | | disk=2pmem | | | fs=xfs | | | ioengine=mmap | | | nr_task=50% | | | runtime=200s | | | rw=rw | | | test_size=200G | | | time_based=tb | | | ucode=0x500320a | +------------------+--------------------------------------------------------------------------------+ | testcase: change | fio-basic: fio.write_iops 14.0% improvement | | test machine | 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 512G memory | | test parameters | bs=2M | | | cpufreq_governor=performance | | | disk=2pmem | | | fs=ext4 | | | ioengine=libaio | | | nr_task=50% | | | runtime=200s | | | rw=rw | | | test_size=200G | | | time_based=tb | | | ucode=0x500320a | +------------------+--------------------------------------------------------------------------------+ 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. ========================================================================================= bs/compiler/cpufreq_governor/disk/fs/ioengine/kconfig/nr_task/rootfs/runtime/rw/tbox_group/test_size/testcase/time_based/ucode: 4k/gcc-11/performance/2pmem/ext4/sync/x86_64-rhel-8.3/50%/debian-11.1-x86_64-20220510.cgz/200s/randrw/lkp-csl-2sp7/200G/fio-basic/tb/0x500320a commit: 11192d9c12 ("memcg: flush stats only if updated") fd25a9e0e2 ("memcg: unify memcg stat flushing") 11192d9c124d58d6 fd25a9e0e23b995fd0ba5e2f00a ---------------- --------------------------- %stddev %change %stddev \ | \ 0.53 ± 27% -0.5 0.01 fio.latency_1000us% 0.39 ± 87% +2.5 2.85 ± 7% fio.latency_100us% 21.94 ± 4% +3.3 25.24 ± 7% fio.latency_10us% 0.15 ± 66% +0.9 1.04 ± 14% fio.latency_250us% 1.01 ± 19% -1.0 0.01 fio.latency_2ms% 72.48 -6.0 66.44 ± 3% fio.latency_4us% 0.05 ± 87% +0.2 0.24 ± 17% fio.latency_500us% 0.94 ± 16% +1.0 1.94 ± 15% fio.latency_50us% 0.05 ± 32% -0.0 0.01 ± 5% fio.latency_750us% 3556 ± 2% +47.8% 5257 ± 2% fio.read_bw_MBps 5592 +45.9% 8156 ± 8% fio.read_clat_90%_us 8960 +391.4% 44032 ± 9% fio.read_clat_95%_us 32560 ± 49% +280.0% 123712 ± 12% fio.read_clat_99%_us 4827 ± 7% +84.1% 8887 ± 3% fio.read_clat_mean_us 910504 ± 2% +47.8% 1345904 ± 2% fio.read_iops 1.47e+09 ± 2% +48.4% 2.181e+09 ± 2% fio.time.file_system_inputs 1.462e+09 ± 2% +49.0% 2.178e+09 ± 2% fio.time.file_system_outputs 38246 ± 4% +37.3% 52515 fio.time.involuntary_context_switches 4744 -4.0% 4553 fio.time.percent_of_cpu_this_job_got 9279 -5.0% 8818 fio.time.system_time 263.86 ± 3% +57.3% 415.07 ± 14% fio.time.user_time 1465507 ± 23% +231.0% 4850690 ± 19% fio.time.voluntary_context_switches 3.654e+08 ± 2% +49.0% 5.446e+08 ± 2% fio.workload 3556 ± 2% +47.8% 5257 ± 2% fio.write_bw_MBps 6760 +17.9% 7972 ± 6% fio.write_clat_90%_us 17440 ± 23% +116.5% 37760 ± 13% fio.write_clat_95%_us 1068032 ± 2% -87.5% 133952 ± 8% fio.write_clat_99%_us 37376 ± 5% -72.6% 10227 ± 2% fio.write_clat_mean_us 910464 ± 2% +47.8% 1345940 ± 2% fio.write_iops 219862 ± 6% +13.7% 249928 ± 6% numa-meminfo.node1.SUnreclaim 0.02 ± 29% +6915.7% 1.53 ± 28% iostat.cpu.iowait 48.50 -4.7% 46.24 iostat.cpu.system 1.36 ± 3% +55.6% 2.12 ± 14% iostat.cpu.user 7243408 ± 2% +14.3% 8281985 ± 3% meminfo.Active 7196323 ± 2% +14.4% 8235503 ± 3% meminfo.Active(file) 2.035e+08 -14.4% 1.741e+08 ± 3% meminfo.max_used_kB 0.02 ± 30% +1.5 1.55 ± 28% mpstat.cpu.all.iowait% 0.61 +0.1 0.75 ± 3% mpstat.cpu.all.irq% 1.37 ± 3% +0.8 2.14 ± 14% mpstat.cpu.all.usr% 1.175e+08 ± 9% +38.4% 1.626e+08 ± 4% numa-numastat.node0.local_node 1.174e+08 ± 9% +38.0% 1.619e+08 ± 4% numa-numastat.node0.numa_hit 1.858e+08 ± 4% +67.9% 3.121e+08 ± 3% numa-numastat.node1.local_node 1.856e+08 ± 4% +67.5% 3.109e+08 ± 3% numa-numastat.node1.numa_hit 3583655 ± 2% +47.3% 5278691 vmstat.io.bi 3450488 ± 2% +50.6% 5197844 vmstat.io.bo 0.00 +1.4e+102% 1.38 ± 35% vmstat.procs.b 31151 ± 21% +211.7% 97094 ± 21% vmstat.system.cs 424562 ± 13% +20.4% 511014 ± 10% sched_debug.cfs_rq:/.load.avg 362.49 ± 13% +17.1% 424.53 ± 10% sched_debug.cfs_rq:/.util_est_enqueued.avg 21255 ± 25% +266.5% 77891 ± 28% sched_debug.cpu.nr_switches.avg 1307 ± 8% +50.2% 1964 ± 22% sched_debug.cpu.nr_switches.min 39602 ± 26% +156.4% 101541 ± 25% sched_debug.cpu.nr_switches.stddev 0.00 ±173% +0.2 0.18 ±198% turbostat.C1% 0.08 +32.8% 0.11 ± 4% turbostat.IPC 2758150 ± 23% +235.1% 9243143 ± 21% turbostat.POLL 0.04 ± 19% +0.1 0.19 ± 30% turbostat.POLL% 266.30 +3.2% 274.72 turbostat.PkgWatt 47.33 +9.3% 51.73 turbostat.RAMWatt 41422157 ± 10% +50.3% 62254549 ± 7% numa-vmstat.node0.nr_dirtied 39572106 ± 10% +54.6% 61195242 ± 7% numa-vmstat.node0.nr_written 1.174e+08 ± 9% +38.0% 1.619e+08 ± 4% numa-vmstat.node0.numa_hit 1.175e+08 ± 9% +38.4% 1.626e+08 ± 4% numa-vmstat.node0.numa_local 1.413e+08 +48.6% 2.1e+08 numa-vmstat.node1.nr_dirtied 54958 ± 6% +13.7% 62481 ± 6% numa-vmstat.node1.nr_slab_unreclaimable 1.369e+08 +51.3% 2.072e+08 numa-vmstat.node1.nr_written 1.856e+08 ± 4% +67.5% 3.109e+08 ± 3% numa-vmstat.node1.numa_hit 1.858e+08 ± 4% +67.9% 3.121e+08 ± 3% numa-vmstat.node1.numa_local 1798939 ± 2% +14.4% 2058578 ± 3% proc-vmstat.nr_active_file 1.827e+08 ± 2% +49.0% 2.723e+08 ± 2% proc-vmstat.nr_dirtied 26632755 +4.2% 27739758 proc-vmstat.nr_file_pages 49856263 -2.3% 48708764 proc-vmstat.nr_free_pages 24130513 +3.5% 24977194 proc-vmstat.nr_inactive_file 526765 +5.2% 554092 proc-vmstat.nr_slab_reclaimable 106917 ± 3% +5.8% 113170 ± 3% proc-vmstat.nr_slab_unreclaimable 1.764e+08 ± 2% +52.1% 2.683e+08 proc-vmstat.nr_written 1798912 ± 2% +14.4% 2058707 ± 3% proc-vmstat.nr_zone_active_file 24130862 +3.5% 24977575 proc-vmstat.nr_zone_inactive_file 3.031e+08 ± 2% +56.0% 4.729e+08 ± 3% proc-vmstat.numa_hit 3.033e+08 ± 2% +56.5% 4.747e+08 ± 3% proc-vmstat.numa_local 35134068 ± 3% +49.1% 52378797 ± 2% proc-vmstat.pgactivate 3.654e+08 ± 2% +48.7% 5.434e+08 ± 2% proc-vmstat.pgalloc_normal 3.493e+08 ± 2% +54.5% 5.397e+08 ± 2% proc-vmstat.pgfree 7.351e+08 ± 2% +48.4% 1.091e+09 ± 2% proc-vmstat.pgpgin 7.057e+08 ± 2% +52.1% 1.073e+09 proc-vmstat.pgpgout 245935 ± 27% -60.2% 97869 ± 84% proc-vmstat.workingset_refault_file 10.03 +10.4% 11.08 perf-stat.i.MPKI 8.051e+09 +24.5% 1.003e+10 perf-stat.i.branch-instructions 0.51 +0.0 0.54 perf-stat.i.branch-miss-rate% 39726140 ± 2% +32.4% 52583047 ± 2% perf-stat.i.branch-misses 2.78e+08 ± 3% +41.0% 3.918e+08 perf-stat.i.cache-misses 3.942e+08 ± 2% +42.7% 5.626e+08 ± 2% perf-stat.i.cache-references 31675 ± 21% +212.7% 99052 ± 21% perf-stat.i.context-switches 3.60 -24.6% 2.72 perf-stat.i.cpi 1.389e+11 -3.7% 1.338e+11 perf-stat.i.cpu-cycles 121.30 +5.2% 127.62 perf-stat.i.cpu-migrations 545.94 ± 2% -30.4% 379.99 ± 2% perf-stat.i.cycles-between-cache-misses 8673578 ± 7% +40.2% 12162840 ± 9% perf-stat.i.dTLB-load-misses 1.042e+10 +29.4% 1.349e+10 perf-stat.i.dTLB-loads 1200725 ± 8% +53.6% 1844381 ± 5% perf-stat.i.dTLB-store-misses 4.745e+09 ± 2% +46.2% 6.937e+09 perf-stat.i.dTLB-stores 85.85 +3.4 89.29 ± 2% perf-stat.i.iTLB-load-miss-rate% 16340901 ± 3% +35.2% 22091169 ± 5% perf-stat.i.iTLB-load-misses 3.904e+10 +28.5% 5.016e+10 perf-stat.i.instructions 2813 ± 3% -9.2% 2554 ± 5% perf-stat.i.instructions-per-iTLB-miss 0.28 +36.0% 0.38 ± 2% perf-stat.i.ipc 1.45 -3.7% 1.39 perf-stat.i.metric.GHz 1082 ± 2% +34.2% 1452 perf-stat.i.metric.K/sec 245.99 +31.3% 323.05 perf-stat.i.metric.M/sec 52.07 -6.5 45.60 ± 3% perf-stat.i.node-load-miss-rate% 29492421 ± 2% +18.4% 34909755 ± 2% perf-stat.i.node-load-misses 27411662 ± 3% +53.6% 42106608 ± 4% perf-stat.i.node-loads 45.38 ± 2% -11.0 34.37 ± 4% perf-stat.i.node-store-miss-rate% 23596478 ± 3% +63.2% 38509129 ± 3% perf-stat.i.node-stores 10.09 +11.0% 11.20 perf-stat.overall.MPKI 0.49 +0.0 0.52 perf-stat.overall.branch-miss-rate% 3.56 -25.0% 2.67 perf-stat.overall.cpi 500.17 ± 2% -31.6% 342.14 ± 2% perf-stat.overall.cycles-between-cache-misses 87.02 +3.2 90.23 ± 2% perf-stat.overall.iTLB-load-miss-rate% 0.28 +33.3% 0.37 perf-stat.overall.ipc 51.82 -6.5 45.32 ± 3% perf-stat.overall.node-load-miss-rate% 44.95 ± 2% -11.5 33.46 ± 3% perf-stat.overall.node-store-miss-rate% 21580 -13.1% 18743 perf-stat.overall.path-length 8.012e+09 +24.5% 9.974e+09 perf-stat.ps.branch-instructions 39519404 ± 2% +32.2% 52256061 ± 2% perf-stat.ps.branch-misses 2.766e+08 ± 3% +40.8% 3.894e+08 perf-stat.ps.cache-misses 3.922e+08 ± 2% +42.5% 5.591e+08 perf-stat.ps.cache-references 31343 ± 21% +212.5% 97946 ± 21% perf-stat.ps.context-switches 1.382e+11 -3.6% 1.332e+11 perf-stat.ps.cpu-cycles 120.76 +5.1% 126.96 perf-stat.ps.cpu-migrations 8632327 ± 7% +40.1% 12090064 ± 9% perf-stat.ps.dTLB-load-misses 1.037e+10 +29.4% 1.342e+10 perf-stat.ps.dTLB-loads 1195744 ± 8% +53.4% 1834079 ± 5% perf-stat.ps.dTLB-store-misses 4.721e+09 ± 2% +46.0% 6.895e+09 perf-stat.ps.dTLB-stores 16269398 ± 3% +34.9% 21954037 ± 5% perf-stat.ps.iTLB-load-misses 3.885e+10 +28.4% 4.989e+10 perf-stat.ps.instructions 29347557 ± 2% +18.3% 34718132 ± 2% perf-stat.ps.node-load-misses 27282181 ± 2% +53.7% 41931623 ± 4% perf-stat.ps.node-loads 23505686 ± 3% +63.1% 38330434 ± 3% perf-stat.ps.node-stores 7.885e+12 +29.4% 1.021e+13 perf-stat.total.instructions 41.86 ± 30% -41.9 0.00 perf-profile.calltrace.cycles-pp.cgroup_rstat_flush_irqsafe.mem_cgroup_wb_stats.balance_dirty_pages.balance_dirty_pages_ratelimited.generic_perform_write 41.98 ± 30% -41.3 0.69 ± 11% perf-profile.calltrace.cycles-pp.mem_cgroup_wb_stats.balance_dirty_pages.balance_dirty_pages_ratelimited.generic_perform_write.ext4_buffered_write_iter 42.00 ± 30% -41.3 0.70 ± 11% perf-profile.calltrace.cycles-pp.balance_dirty_pages.balance_dirty_pages_ratelimited.generic_perform_write.ext4_buffered_write_iter.new_sync_write 42.11 ± 30% -41.2 0.93 ± 12% perf-profile.calltrace.cycles-pp.balance_dirty_pages_ratelimited.generic_perform_write.ext4_buffered_write_iter.new_sync_write.vfs_write 40.46 ± 31% -40.5 0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.cgroup_rstat_flush_irqsafe.mem_cgroup_wb_stats.balance_dirty_pages.balance_dirty_pages_ratelimited 40.46 ± 31% -40.5 0.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.cgroup_rstat_flush_irqsafe.mem_cgroup_wb_stats.balance_dirty_pages 48.70 ± 25% -34.9 13.76 ± 23% perf-profile.calltrace.cycles-pp.generic_perform_write.ext4_buffered_write_iter.new_sync_write.vfs_write.ksys_write 48.86 ± 25% -34.9 13.93 ± 23% perf-profile.calltrace.cycles-pp.ext4_buffered_write_iter.new_sync_write.vfs_write.ksys_write.do_syscall_64 48.88 ± 25% -34.9 13.96 ± 23% perf-profile.calltrace.cycles-pp.new_sync_write.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe 48.96 ± 25% -34.9 14.06 ± 23% perf-profile.calltrace.cycles-pp.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write 48.98 ± 25% -34.9 14.09 ± 23% perf-profile.calltrace.cycles-pp.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write 49.02 ± 25% -34.9 14.13 ± 23% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__libc_write 49.00 ± 25% -34.9 14.12 ± 23% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write 49.12 ± 25% -34.9 14.27 ± 23% perf-profile.calltrace.cycles-pp.__libc_write 0.00 +0.7 0.68 ± 11% perf-profile.calltrace.cycles-pp.cgroup_rstat_flush_irqsafe.mem_cgroup_flush_stats.mem_cgroup_wb_stats.balance_dirty_pages.balance_dirty_pages_ratelimited 0.00 +0.7 0.68 ± 11% perf-profile.calltrace.cycles-pp.cgroup_rstat_flush_locked.cgroup_rstat_flush_irqsafe.mem_cgroup_flush_stats.mem_cgroup_wb_stats.balance_dirty_pages 0.00 +0.7 0.68 ± 11% perf-profile.calltrace.cycles-pp.mem_cgroup_flush_stats.mem_cgroup_wb_stats.balance_dirty_pages.balance_dirty_pages_ratelimited.generic_perform_write 0.57 ± 42% +0.8 1.35 ± 14% perf-profile.calltrace.cycles-pp.get_page_from_freelist.__alloc_pages.pagecache_get_page.grab_cache_page_write_begin.ext4_da_write_begin 0.60 ± 42% +0.8 1.38 ± 13% perf-profile.calltrace.cycles-pp.__alloc_pages.pagecache_get_page.grab_cache_page_write_begin.ext4_da_write_begin.generic_perform_write 0.39 ± 81% +0.8 1.22 ± 17% perf-profile.calltrace.cycles-pp.rmqueue.get_page_from_freelist.__alloc_pages.pagecache_get_page.grab_cache_page_write_begin 0.18 ±173% +0.9 1.10 ± 25% perf-profile.calltrace.cycles-pp.rmqueue_bulk.rmqueue.get_page_from_freelist.__alloc_pages.page_cache_ra_unbounded 0.39 ±143% +0.9 1.32 ± 26% perf-profile.calltrace.cycles-pp.try_to_free_buffers.invalidate_inode_page.__invalidate_mapping_pages.generic_fadvise.ksys_fadvise64_64 0.18 ±173% +1.0 1.15 ± 20% perf-profile.calltrace.cycles-pp.rmqueue_bulk.rmqueue.get_page_from_freelist.__alloc_pages.pagecache_get_page 0.19 ±173% +1.0 1.16 ± 23% perf-profile.calltrace.cycles-pp.rmqueue.get_page_from_freelist.__alloc_pages.page_cache_ra_unbounded.force_page_cache_ra 0.26 ±133% +1.0 1.26 ± 20% perf-profile.calltrace.cycles-pp.__alloc_pages.page_cache_ra_unbounded.force_page_cache_ra.filemap_get_pages.filemap_read 0.20 ±173% +1.0 1.21 ± 21% perf-profile.calltrace.cycles-pp.get_page_from_freelist.__alloc_pages.page_cache_ra_unbounded.force_page_cache_ra.filemap_get_pages 0.68 ± 84% +1.1 1.74 ± 23% perf-profile.calltrace.cycles-pp.invalidate_inode_page.__invalidate_mapping_pages.generic_fadvise.ksys_fadvise64_64.__x64_sys_fadvise64 0.40 ± 79% +1.3 1.72 ± 23% perf-profile.calltrace.cycles-pp.ext4_end_bio.pmem_submit_bio.__submit_bio.__submit_bio_noacct.ext4_bio_write_page 0.12 ±264% +1.5 1.57 ± 23% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.lock_page_lruvec_irqsave.pagevec_lru_move_fn.mark_page_accessed 0.12 ±264% +1.5 1.58 ± 23% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.lock_page_lruvec_irqsave.pagevec_lru_move_fn.mark_page_accessed.filemap_read 0.12 ±264% +1.5 1.58 ± 23% perf-profile.calltrace.cycles-pp.lock_page_lruvec_irqsave.pagevec_lru_move_fn.mark_page_accessed.filemap_read.new_sync_read 0.28 ±173% +1.5 1.82 ± 34% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.rmqueue_bulk.rmqueue.get_page_from_freelist 0.28 ±173% +1.5 1.83 ± 34% perf-profile.calltrace.cycles-pp._raw_spin_lock.rmqueue_bulk.rmqueue.get_page_from_freelist.__alloc_pages 0.13 ±264% +1.6 1.74 ± 22% perf-profile.calltrace.cycles-pp.pagevec_lru_move_fn.mark_page_accessed.filemap_read.new_sync_read.vfs_read 0.15 ±264% +1.8 1.94 ± 22% perf-profile.calltrace.cycles-pp.mark_page_accessed.filemap_read.new_sync_read.vfs_read.ksys_read 0.35 ±101% +1.9 2.23 ± 30% perf-profile.calltrace.cycles-pp.__memcpy_flushcache.write_pmem.pmem_do_write.pmem_submit_bio.__submit_bio 0.35 ±101% +1.9 2.25 ± 30% perf-profile.calltrace.cycles-pp.write_pmem.pmem_do_write.pmem_submit_bio.__submit_bio.__submit_bio_noacct 0.36 ±101% +1.9 2.26 ± 30% perf-profile.calltrace.cycles-pp.pmem_do_write.pmem_submit_bio.__submit_bio.__submit_bio_noacct.ext4_bio_write_page 1.24 ± 44% +3.1 4.36 ± 22% perf-profile.calltrace.cycles-pp.pmem_submit_bio.__submit_bio.__submit_bio_noacct.ext4_bio_write_page.mpage_submit_page 1.28 ± 44% +3.2 4.48 ± 22% perf-profile.calltrace.cycles-pp.__submit_bio.__submit_bio_noacct.ext4_bio_write_page.mpage_submit_page.mpage_process_page_bufs 1.29 ± 44% +3.2 4.48 ± 22% perf-profile.calltrace.cycles-pp.__submit_bio_noacct.ext4_bio_write_page.mpage_submit_page.mpage_process_page_bufs.mpage_prepare_extent_to_map 1.70 ± 42% +3.7 5.44 ± 21% perf-profile.calltrace.cycles-pp.ext4_bio_write_page.mpage_submit_page.mpage_process_page_bufs.mpage_prepare_extent_to_map.ext4_writepages 1.96 ± 43% +3.9 5.89 ± 21% perf-profile.calltrace.cycles-pp.mpage_submit_page.mpage_process_page_bufs.mpage_prepare_extent_to_map.ext4_writepages.do_writepages 0.46 ± 85% +4.2 4.63 ± 26% perf-profile.calltrace.cycles-pp.mpage_process_page_bufs.mpage_prepare_extent_to_map.ext4_writepages.do_writepages.filemap_fdatawrite_wbc 1.38 ± 87% +4.3 5.70 ± 37% perf-profile.calltrace.cycles-pp.lru_cache_add.add_to_page_cache_lru.page_cache_ra_unbounded.force_page_cache_ra.filemap_get_pages 1.35 ± 90% +4.3 5.67 ± 37% perf-profile.calltrace.cycles-pp.__pagevec_lru_add.lru_cache_add.add_to_page_cache_lru.page_cache_ra_unbounded.force_page_cache_ra 2.08 ± 55% +4.4 6.44 ± 35% perf-profile.calltrace.cycles-pp.add_to_page_cache_lru.page_cache_ra_unbounded.force_page_cache_ra.filemap_get_pages.filemap_read 0.96 ±134% +4.4 5.33 ± 38% perf-profile.calltrace.cycles-pp.lock_page_lruvec_irqsave.__pagevec_lru_add.lru_cache_add.add_to_page_cache_lru.page_cache_ra_unbounded 1.41 ± 84% +4.4 5.86 ± 39% perf-profile.calltrace.cycles-pp.lru_cache_add.add_to_page_cache_lru.pagecache_get_page.grab_cache_page_write_begin.ext4_da_write_begin 1.38 ± 86% +4.4 5.83 ± 39% perf-profile.calltrace.cycles-pp.__pagevec_lru_add.lru_cache_add.add_to_page_cache_lru.pagecache_get_page.grab_cache_page_write_begin 1.04 ±119% +4.5 5.50 ± 40% perf-profile.calltrace.cycles-pp.lock_page_lruvec_irqsave.__pagevec_lru_add.lru_cache_add.add_to_page_cache_lru.pagecache_get_page 0.52 ± 85% +4.5 4.99 ± 26% perf-profile.calltrace.cycles-pp.mpage_prepare_extent_to_map.ext4_writepages.do_writepages.filemap_fdatawrite_wbc.__filemap_fdatawrite_range 0.52 ± 85% +4.5 5.00 ± 26% perf-profile.calltrace.cycles-pp.do_writepages.filemap_fdatawrite_wbc.__filemap_fdatawrite_range.generic_fadvise.ksys_fadvise64_64 0.52 ± 85% +4.5 5.00 ± 26% perf-profile.calltrace.cycles-pp.ext4_writepages.do_writepages.filemap_fdatawrite_wbc.__filemap_fdatawrite_range.generic_fadvise 0.52 ± 85% +4.5 5.00 ± 26% perf-profile.calltrace.cycles-pp.__filemap_fdatawrite_range.generic_fadvise.ksys_fadvise64_64.__x64_sys_fadvise64.do_syscall_64 0.52 ± 85% +4.5 5.00 ± 26% perf-profile.calltrace.cycles-pp.filemap_fdatawrite_wbc.__filemap_fdatawrite_range.generic_fadvise.ksys_fadvise64_64.__x64_sys_fadvise64 2.12 ± 53% +4.5 6.61 ± 36% perf-profile.calltrace.cycles-pp.add_to_page_cache_lru.pagecache_get_page.grab_cache_page_write_begin.ext4_da_write_begin.generic_perform_write 3.36 ± 34% +5.4 8.80 ± 27% perf-profile.calltrace.cycles-pp.pagecache_get_page.grab_cache_page_write_begin.ext4_da_write_begin.generic_perform_write.ext4_buffered_write_iter 3.36 ± 33% +5.4 8.81 ± 27% perf-profile.calltrace.cycles-pp.grab_cache_page_write_begin.ext4_da_write_begin.generic_perform_write.ext4_buffered_write_iter.new_sync_write 4.55 ± 23% +5.5 10.06 ± 26% perf-profile.calltrace.cycles-pp.ext4_da_write_begin.generic_perform_write.ext4_buffered_write_iter.new_sync_write.vfs_write 4.72 ± 24% +6.2 10.97 ± 24% perf-profile.calltrace.cycles-pp.page_cache_ra_unbounded.force_page_cache_ra.filemap_get_pages.filemap_read.new_sync_read 4.74 ± 24% +6.3 10.99 ± 24% perf-profile.calltrace.cycles-pp.force_page_cache_ra.filemap_get_pages.filemap_read.new_sync_read.vfs_read 5.46 ± 20% +6.6 12.11 ± 23% perf-profile.calltrace.cycles-pp.filemap_get_pages.filemap_read.new_sync_read.vfs_read.ksys_read 6.19 ± 22% +8.3 14.46 ± 22% perf-profile.calltrace.cycles-pp.filemap_read.new_sync_read.vfs_read.ksys_read.do_syscall_64 6.23 ± 22% +8.3 14.50 ± 22% perf-profile.calltrace.cycles-pp.new_sync_read.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe 6.30 ± 21% +8.3 14.59 ± 22% perf-profile.calltrace.cycles-pp.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read 6.32 ± 21% +8.3 14.62 ± 22% perf-profile.calltrace.cycles-pp.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read 6.34 ± 21% +8.3 14.65 ± 22% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read 6.35 ± 21% +8.3 14.66 ± 22% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__libc_read 6.47 ± 21% +8.3 14.82 ± 22% perf-profile.calltrace.cycles-pp.__libc_read 1.98 ±127% +8.8 10.82 ± 39% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.lock_page_lruvec_irqsave.__pagevec_lru_add.lru_cache_add.add_to_page_cache_lru 1.88 ±136% +8.9 10.76 ± 39% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.lock_page_lruvec_irqsave.__pagevec_lru_add.lru_cache_add 41.98 ± 30% -41.3 0.69 ± 11% perf-profile.children.cycles-pp.mem_cgroup_wb_stats 42.00 ± 30% -41.3 0.71 ± 11% perf-profile.children.cycles-pp.balance_dirty_pages 42.11 ± 30% -41.2 0.93 ± 12% perf-profile.children.cycles-pp.balance_dirty_pages_ratelimited 41.87 ± 30% -41.2 0.70 ± 11% perf-profile.children.cycles-pp.cgroup_rstat_flush_irqsafe 48.72 ± 25% -34.9 13.79 ± 23% perf-profile.children.cycles-pp.generic_perform_write 48.86 ± 25% -34.9 13.94 ± 23% perf-profile.children.cycles-pp.ext4_buffered_write_iter 48.90 ± 25% -34.9 13.99 ± 23% perf-profile.children.cycles-pp.new_sync_write 48.97 ± 25% -34.9 14.09 ± 23% perf-profile.children.cycles-pp.vfs_write 49.00 ± 25% -34.9 14.12 ± 23% perf-profile.children.cycles-pp.ksys_write 49.16 ± 25% -34.8 14.32 ± 23% perf-profile.children.cycles-pp.__libc_write 50.76 ± 10% -11.1 39.64 ± 11% perf-profile.children.cycles-pp._raw_spin_lock_irqsave 51.00 ± 10% -9.9 41.11 ± 11% perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath 1.44 ± 13% -0.7 0.72 ± 16% perf-profile.children.cycles-pp.cgroup_rstat_updated 1.40 ± 8% -0.7 0.70 ± 11% perf-profile.children.cycles-pp.cgroup_rstat_flush_locked 0.83 ± 12% -0.3 0.56 ± 11% perf-profile.children.cycles-pp.mem_cgroup_css_rstat_flush 0.42 ± 14% -0.2 0.22 ± 21% perf-profile.children.cycles-pp.mem_cgroup_charge_statistics 0.44 ± 12% -0.2 0.24 ± 19% perf-profile.children.cycles-pp.__count_memcg_events 0.02 ±129% +0.1 0.08 ± 16% perf-profile.children.cycles-pp.__list_add_valid 0.07 ± 17% +0.1 0.13 ± 10% perf-profile.children.cycles-pp.page_mapping 0.01 ±174% +0.1 0.08 ± 17% perf-profile.children.cycles-pp.uncharge_page 0.04 ± 58% +0.1 0.11 ± 16% perf-profile.children.cycles-pp.unlock_page 0.01 ±264% +0.1 0.08 ± 17% perf-profile.children.cycles-pp.drop_buffers 0.00 +0.1 0.08 ± 44% perf-profile.children.cycles-pp.xas_init_marks 0.08 ± 16% +0.1 0.17 ± 27% perf-profile.children.cycles-pp._raw_spin_lock_irq 0.06 ± 19% +0.1 0.15 ± 22% perf-profile.children.cycles-pp.workingset_activation 0.04 ± 79% +0.1 0.13 ± 29% perf-profile.children.cycles-pp.xas_clear_mark 0.01 ±174% +0.1 0.11 ± 20% perf-profile.children.cycles-pp.page_counter_cancel 0.08 ± 44% +0.1 0.18 ± 22% perf-profile.children.cycles-pp.percpu_counter_add_batch 0.06 ± 58% +0.1 0.17 ± 19% perf-profile.children.cycles-pp.uncharge_batch 0.01 ±174% +0.1 0.12 ± 21% perf-profile.children.cycles-pp.page_counter_uncharge 0.03 ±102% +0.1 0.14 ± 23% perf-profile.children.cycles-pp.workingset_age_nonresident 0.00 +0.1 0.13 ± 18% perf-profile.children.cycles-pp.mem_cgroup_wb_domain 0.12 ± 11% +0.1 0.25 ± 12% perf-profile.children.cycles-pp.__mod_node_page_state 0.11 ± 38% +0.1 0.24 ± 18% perf-profile.children.cycles-pp.__mem_cgroup_uncharge_list 0.15 ± 11% +0.2 0.30 ± 13% perf-profile.children.cycles-pp.__mod_lruvec_state 0.15 ± 30% +0.2 0.33 ± 19% perf-profile.children.cycles-pp.pagevec_lookup_range_tag 0.15 ± 30% +0.2 0.33 ± 19% perf-profile.children.cycles-pp.find_get_pages_range_tag 0.06 ±141% +0.2 0.26 ± 54% perf-profile.children.cycles-pp.poll_idle 0.23 ± 36% +0.2 0.44 ± 16% perf-profile.children.cycles-pp.memcg_slab_free_hook 0.22 ± 33% +0.2 0.43 ± 21% perf-profile.children.cycles-pp.clear_page_dirty_for_io 0.16 ± 30% +0.2 0.37 ± 15% perf-profile.children.cycles-pp.jbd2_journal_grab_journal_head 0.16 ± 28% +0.2 0.38 ± 15% perf-profile.children.cycles-pp.jbd2_journal_try_to_free_buffers 0.24 ± 42% +0.2 0.48 ± 27% perf-profile.children.cycles-pp.__free_one_page 0.22 ± 38% +0.2 0.46 ± 20% perf-profile.children.cycles-pp.find_lock_entries 0.22 ± 29% +0.3 0.47 ± 20% perf-profile.children.cycles-pp.__test_set_page_writeback 0.24 ± 10% +0.3 0.55 ± 25% perf-profile.children.cycles-pp.xas_store 0.24 ± 35% +0.4 0.66 ± 23% perf-profile.children.cycles-pp.ext4_put_io_end_defer 0.46 ± 37% +0.4 0.88 ± 23% perf-profile.children.cycles-pp.__delete_from_page_cache 0.37 ± 47% +0.4 0.80 ± 29% perf-profile.children.cycles-pp.free_pcppages_bulk 0.64 ± 24% +0.5 1.14 ± 14% perf-profile.children.cycles-pp.__list_del_entry_valid 0.42 ± 45% +0.5 0.93 ± 28% perf-profile.children.cycles-pp.free_unref_page_list 0.37 ± 31% +0.6 0.92 ± 20% perf-profile.children.cycles-pp.test_clear_page_writeback 0.52 ± 37% +0.6 1.09 ± 23% perf-profile.children.cycles-pp.__remove_mapping 0.38 ± 31% +0.6 1.01 ± 22% perf-profile.children.cycles-pp.end_page_writeback 0.57 ± 69% +0.7 1.27 ± 26% perf-profile.children.cycles-pp.free_buffer_head 0.00 +0.7 0.70 ± 11% perf-profile.children.cycles-pp.mem_cgroup_flush_stats 0.63 ± 62% +0.7 1.38 ± 24% perf-profile.children.cycles-pp.kmem_cache_free 0.45 ± 31% +0.8 1.20 ± 21% perf-profile.children.cycles-pp.ext4_finish_bio 0.61 ± 66% +0.8 1.38 ± 25% perf-profile.children.cycles-pp.try_to_free_buffers 0.80 ± 57% +0.9 1.75 ± 23% perf-profile.children.cycles-pp.invalidate_inode_page 1.16 ± 23% +1.2 2.31 ± 24% perf-profile.children.cycles-pp._raw_spin_lock 0.70 ± 32% +1.2 1.87 ± 21% perf-profile.children.cycles-pp.ext4_end_bio 0.91 ± 35% +1.4 2.26 ± 22% perf-profile.children.cycles-pp.rmqueue_bulk 1.02 ± 31% +1.4 2.40 ± 20% perf-profile.children.cycles-pp.rmqueue 1.12 ± 27% +1.4 2.57 ± 17% perf-profile.children.cycles-pp.get_page_from_freelist 1.18 ± 25% +1.5 2.66 ± 16% perf-profile.children.cycles-pp.__alloc_pages 0.30 ± 93% +1.5 1.78 ± 22% perf-profile.children.cycles-pp.pagevec_lru_move_fn 0.40 ± 72% +1.5 1.94 ± 22% perf-profile.children.cycles-pp.mark_page_accessed 0.70 ± 33% +1.7 2.35 ± 25% perf-profile.children.cycles-pp.__memcpy_flushcache 0.70 ± 33% +1.7 2.37 ± 24% perf-profile.children.cycles-pp.write_pmem 0.70 ± 33% +1.7 2.38 ± 25% perf-profile.children.cycles-pp.pmem_do_write 1.92 ± 31% +3.5 5.45 ± 21% perf-profile.children.cycles-pp.ext4_bio_write_page 2.14 ± 32% +3.8 5.88 ± 21% perf-profile.children.cycles-pp.mpage_submit_page 2.98 ± 19% +3.9 6.88 ± 20% perf-profile.children.cycles-pp.pmem_submit_bio 2.25 ± 32% +4.0 6.25 ± 21% perf-profile.children.cycles-pp.mpage_process_page_bufs 3.15 ± 19% +4.0 7.15 ± 20% perf-profile.children.cycles-pp.__submit_bio 3.16 ± 19% +4.0 7.16 ± 20% perf-profile.children.cycles-pp.__submit_bio_noacct 2.48 ± 32% +4.3 6.80 ± 20% perf-profile.children.cycles-pp.do_writepages 2.48 ± 32% +4.3 6.80 ± 20% perf-profile.children.cycles-pp.ext4_writepages 2.48 ± 32% +4.3 6.80 ± 20% perf-profile.children.cycles-pp.mpage_prepare_extent_to_map 0.65 ± 50% +4.3 5.00 ± 26% perf-profile.children.cycles-pp.__filemap_fdatawrite_range 0.65 ± 50% +4.3 5.00 ± 26% perf-profile.children.cycles-pp.filemap_fdatawrite_wbc 3.36 ± 34% +5.4 8.80 ± 27% perf-profile.children.cycles-pp.pagecache_get_page 3.37 ± 33% +5.4 8.81 ± 27% perf-profile.children.cycles-pp.grab_cache_page_write_begin 4.55 ± 23% +5.5 10.06 ± 26% perf-profile.children.cycles-pp.ext4_da_write_begin 4.74 ± 24% +6.3 10.99 ± 24% perf-profile.children.cycles-pp.force_page_cache_ra 4.73 ± 24% +6.3 11.07 ± 24% perf-profile.children.cycles-pp.page_cache_ra_unbounded 5.46 ± 20% +6.7 12.12 ± 23% perf-profile.children.cycles-pp.filemap_get_pages 6.20 ± 22% +8.3 14.46 ± 22% perf-profile.children.cycles-pp.filemap_read 6.24 ± 21% +8.3 14.51 ± 22% perf-profile.children.cycles-pp.new_sync_read 6.34 ± 21% +8.3 14.63 ± 22% perf-profile.children.cycles-pp.vfs_read 6.36 ± 21% +8.3 14.66 ± 22% perf-profile.children.cycles-pp.ksys_read 6.49 ± 21% +8.4 14.84 ± 22% perf-profile.children.cycles-pp.__libc_read 2.80 ± 86% +8.8 11.62 ± 38% perf-profile.children.cycles-pp.lru_cache_add 2.74 ± 88% +8.8 11.57 ± 38% perf-profile.children.cycles-pp.__pagevec_lru_add 4.20 ± 54% +8.9 13.13 ± 36% perf-profile.children.cycles-pp.add_to_page_cache_lru 51.00 ± 10% -9.9 41.11 ± 11% perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath 1.12 ± 15% -0.6 0.55 ± 17% perf-profile.self.cycles-pp.cgroup_rstat_updated 0.82 ± 12% -0.3 0.55 ± 12% perf-profile.self.cycles-pp.mem_cgroup_css_rstat_flush 0.51 ± 17% -0.2 0.30 ± 15% perf-profile.self.cycles-pp._raw_spin_lock 0.19 ± 14% -0.1 0.08 ± 10% perf-profile.self.cycles-pp.cgroup_rstat_flush_locked 0.06 ± 16% +0.1 0.11 ± 14% perf-profile.self.cycles-pp.xas_store 0.00 +0.1 0.05 ± 8% perf-profile.self.cycles-pp.submit_bio_checks 0.01 ±174% +0.1 0.07 ± 20% perf-profile.self.cycles-pp.__slab_free 0.06 ± 15% +0.1 0.12 ± 14% perf-profile.self.cycles-pp.page_mapping 0.04 ± 58% +0.1 0.10 ± 16% perf-profile.self.cycles-pp.unlock_page 0.02 ±129% +0.1 0.08 ± 21% perf-profile.self.cycles-pp.clear_page_dirty_for_io 0.01 ±173% +0.1 0.08 ± 17% perf-profile.self.cycles-pp.uncharge_page 0.01 ±174% +0.1 0.08 ± 21% perf-profile.self.cycles-pp.__remove_mapping 0.01 ±264% +0.1 0.08 ± 19% perf-profile.self.cycles-pp.drop_buffers 0.00 +0.1 0.08 ± 17% perf-profile.self.cycles-pp.__list_add_valid 0.01 ±173% +0.1 0.09 ± 20% perf-profile.self.cycles-pp.__test_set_page_writeback 0.01 ±264% +0.1 0.09 ± 23% perf-profile.self.cycles-pp.mpage_prepare_extent_to_map 0.04 ± 79% +0.1 0.12 ± 30% perf-profile.self.cycles-pp.xas_clear_mark 0.04 ± 79% +0.1 0.13 ± 22% perf-profile.self.cycles-pp.percpu_counter_add_batch 0.04 ± 79% +0.1 0.13 ± 23% perf-profile.self.cycles-pp.test_clear_page_writeback 0.10 ± 24% +0.1 0.20 ± 18% perf-profile.self.cycles-pp.kmem_cache_free 0.01 ±174% +0.1 0.11 ± 21% perf-profile.self.cycles-pp.page_counter_cancel 0.06 ± 62% +0.1 0.17 ± 17% perf-profile.self.cycles-pp.ext4_bio_write_page 0.03 ±102% +0.1 0.14 ± 23% perf-profile.self.cycles-pp.workingset_age_nonresident 0.03 ±104% +0.1 0.15 ± 40% perf-profile.self.cycles-pp.___slab_alloc 0.12 ± 12% +0.1 0.24 ± 13% perf-profile.self.cycles-pp.__mod_node_page_state 0.00 +0.1 0.13 ± 18% perf-profile.self.cycles-pp.mem_cgroup_wb_domain 0.12 ± 29% +0.2 0.28 ± 19% perf-profile.self.cycles-pp.find_get_pages_range_tag 0.06 ±141% +0.2 0.25 ± 54% perf-profile.self.cycles-pp.poll_idle 0.16 ± 25% +0.2 0.36 ± 14% perf-profile.self.cycles-pp.memcg_slab_free_hook 0.16 ± 30% +0.2 0.37 ± 15% perf-profile.self.cycles-pp.jbd2_journal_grab_journal_head 0.18 ± 35% +0.2 0.40 ± 22% perf-profile.self.cycles-pp.__free_one_page 0.20 ± 37% +0.2 0.42 ± 20% perf-profile.self.cycles-pp.find_lock_entries 0.12 ± 35% +0.2 0.36 ± 23% perf-profile.self.cycles-pp.mpage_process_page_bufs 0.24 ± 35% +0.4 0.65 ± 23% perf-profile.self.cycles-pp.ext4_put_io_end_defer 0.40 ± 14% +0.4 0.83 ± 18% perf-profile.self.cycles-pp.__mod_memcg_lruvec_state 0.64 ± 24% +0.5 1.14 ± 14% perf-profile.self.cycles-pp.__list_del_entry_valid 0.69 ± 33% +1.6 2.33 ± 25% perf-profile.self.cycles-pp.__memcpy_flushcache *************************************************************************************************** lkp-csl-2sp7: 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 512G memory ========================================================================================= bs/compiler/cpufreq_governor/disk/fs/ioengine/kconfig/nr_task/rootfs/runtime/rw/tbox_group/test_size/testcase/time_based/ucode: 4k/gcc-11/performance/2pmem/xfs/mmap/x86_64-rhel-8.3/50%/debian-11.1-x86_64-20220510.cgz/200s/rw/lkp-csl-2sp7/200G/fio-basic/tb/0x500320a commit: 11192d9c12 ("memcg: flush stats only if updated") fd25a9e0e2 ("memcg: unify memcg stat flushing") 11192d9c124d58d6 fd25a9e0e23b995fd0ba5e2f00a ---------------- --------------------------- %stddev %change %stddev \ | \ 0.09 ± 23% -0.1 0.01 fio.latency_1000us% 0.14 ± 12% -0.1 0.03 ± 5% fio.latency_100us% 2.90 ± 5% +1.2 4.14 ± 9% fio.latency_10us% 0.39 ± 8% +0.2 0.63 ± 10% fio.latency_20us% 0.29 ± 6% -0.2 0.09 ± 4% fio.latency_250us% 0.02 ± 30% -0.0 0.01 fio.latency_2ms% 24.32 ± 2% +9.1 33.40 ± 2% fio.latency_2us% 25.64 ± 2% -6.1 19.51 ± 4% fio.latency_4us% 0.33 ± 30% -0.3 0.03 ± 9% fio.latency_500us% 0.33 ± 9% -0.2 0.14 ± 9% fio.latency_50us% 0.30 ± 18% -0.3 0.01 ± 6% fio.latency_750us% 4684 +23.1% 5766 ± 3% fio.read_bw_MBps 982.67 ± 2% +43.8% 1413 ± 6% fio.read_clat_90%_us 1533 ± 3% +64.5% 2522 fio.read_clat_95%_us 3637 +46.0% 5312 ± 4% fio.read_clat_99%_us 883.59 +29.1% 1140 fio.read_clat_mean_us 6383 +103.3% 12981 ± 4% fio.read_clat_stddev 1199278 +23.1% 1476197 ± 3% fio.read_iops 1.927e+09 +23.1% 2.371e+09 ± 3% fio.time.file_system_outputs 10231 ± 4% -51.0% 5013 ± 10% fio.time.involuntary_context_switches 2.413e+08 +23.0% 2.968e+08 ± 3% fio.time.major_page_faults 22094002 ± 5% +27.1% 28086792 fio.time.maximum_resident_set_size 7057976 ± 3% +23.6% 8725928 ± 3% fio.time.minor_page_faults 1747 ± 4% -51.8% 842.83 ± 5% fio.time.percent_of_cpu_this_job_got 3064 ± 4% -66.2% 1036 ± 6% fio.time.system_time 467.32 ± 2% +43.4% 669.93 ± 5% fio.time.user_time 4.818e+08 +23.1% 5.928e+08 ± 3% fio.workload 4685 +23.1% 5767 ± 3% fio.write_bw_MBps 6112 ± 3% -9.2% 5546 ± 4% fio.write_clat_95%_us 444757 ± 21% -97.1% 12800 ± 5% fio.write_clat_99%_us 37030 -19.7% 29718 ± 4% fio.write_clat_mean_us 1199468 +23.1% 1476390 ± 3% fio.write_iops 1.549e+10 +11.8% 1.732e+10 cpuidle..time 32060523 +11.9% 35866344 cpuidle..usage 49.50 +2.3% 50.64 iostat.cpu.idle 30.40 ± 2% +26.3% 38.40 iostat.cpu.iowait 17.77 ± 4% -56.1% 7.80 ± 4% iostat.cpu.system 2.33 ± 2% +35.9% 3.17 ± 5% iostat.cpu.user 29.83 ± 2% +26.8% 37.83 vmstat.cpu.wa 4519957 +23.7% 5592929 ± 3% vmstat.io.bo 29.00 ± 4% +30.5% 37.83 vmstat.procs.b 19.00 ± 7% -47.4% 10.00 ± 5% vmstat.procs.r 47466 ± 49% -46.7% 25314 ± 12% meminfo.Active 45586 ± 51% -48.6% 23440 ± 13% meminfo.Active(anon) 929526 +26.9% 1179873 ± 3% meminfo.PageTables 66821 ± 40% -38.3% 41242 ± 8% meminfo.Shmem 1420637 ± 2% +19.0% 1691263 ± 5% meminfo.Writeback 30.69 ± 2% +8.1 38.77 mpstat.cpu.all.iowait% 0.75 ± 3% +0.1 0.87 ± 4% mpstat.cpu.all.irq% 0.05 ± 3% +0.0 0.05 mpstat.cpu.all.soft% 17.13 ± 4% -10.2 6.93 ± 5% mpstat.cpu.all.sys% 2.35 ± 2% +0.8 3.20 ± 5% mpstat.cpu.all.usr% 595.00 ± 3% -42.6% 341.33 ± 4% turbostat.Avg_MHz 21.33 ± 3% -9.1 12.27 ± 4% turbostat.Busy% 78.60 +11.5% 87.64 turbostat.CPU%c1 0.13 ± 4% +87.2% 0.24 turbostat.IPC 209.63 -5.7% 197.72 turbostat.PkgWatt 42.66 +8.1% 46.14 turbostat.RAMWatt 11378737 ± 2% +18.3% 13462900 ± 3% numa-meminfo.node0.Dirty 308669 ± 4% +32.9% 410318 ± 6% numa-meminfo.node0.Writeback 44240 ± 53% -48.0% 22983 ± 16% numa-meminfo.node1.Active 44240 ± 53% -49.4% 22365 ± 14% numa-meminfo.node1.Active(anon) 374833 ± 5% +7.8% 403918 ± 4% numa-meminfo.node1.KReclaimable 757587 ± 2% +32.1% 1000984 ± 3% numa-meminfo.node1.PageTables 374833 ± 5% +7.8% 403918 ± 4% numa-meminfo.node1.SReclaimable 1088573 ± 5% +16.3% 1265938 ± 5% numa-meminfo.node1.Writeback 52202136 ± 4% +22.7% 64049942 ± 3% numa-vmstat.node0.nr_dirtied 2844482 ± 2% +18.4% 3367535 ± 3% numa-vmstat.node0.nr_dirty 165.83 ± 56% -79.6% 33.83 ±199% numa-vmstat.node0.nr_mlock 78282 ± 3% +30.9% 102434 ± 4% numa-vmstat.node0.nr_writeback 49430955 ± 6% +27.2% 62887052 ± 3% numa-vmstat.node0.nr_written 2922973 ± 2% +18.7% 3470209 ± 3% numa-vmstat.node0.nr_zone_write_pending 11061 ± 53% -49.6% 5575 ± 15% numa-vmstat.node1.nr_active_anon 1.887e+08 +23.1% 2.324e+08 ± 4% numa-vmstat.node1.nr_dirtied 66.17 ±141% +156.2% 169.50 ± 43% numa-vmstat.node1.nr_mlock 188820 ± 2% +32.7% 250515 ± 3% numa-vmstat.node1.nr_page_table_pages 93715 ± 5% +7.7% 100970 ± 4% numa-vmstat.node1.nr_slab_reclaimable 271839 ± 5% +16.6% 316969 ± 6% numa-vmstat.node1.nr_writeback 1.809e+08 ± 2% +23.9% 2.241e+08 ± 4% numa-vmstat.node1.nr_written 11061 ± 53% -49.6% 5575 ± 15% numa-vmstat.node1.nr_zone_active_anon 0.27 ± 27% -49.2% 0.14 ± 10% sched_debug.cfs_rq:/.h_nr_running.avg 0.40 ± 5% -14.2% 0.34 ± 5% sched_debug.cfs_rq:/.h_nr_running.stddev 239463 ± 32% -57.7% 101381 ± 10% sched_debug.cfs_rq:/.load.avg 335444 ± 6% -20.0% 268223 ± 5% sched_debug.cfs_rq:/.load.stddev 0.27 ± 27% -49.3% 0.14 ± 10% sched_debug.cfs_rq:/.nr_running.avg 0.40 ± 5% -14.3% 0.34 ± 5% sched_debug.cfs_rq:/.nr_running.stddev 317.95 ± 24% -48.2% 164.84 ± 5% sched_debug.cfs_rq:/.runnable_avg.avg 315.63 ± 7% -29.5% 222.58 ± 3% sched_debug.cfs_rq:/.runnable_avg.stddev 317.60 ± 24% -48.3% 164.20 ± 6% sched_debug.cfs_rq:/.util_avg.avg 315.18 ± 7% -29.5% 222.09 ± 3% sched_debug.cfs_rq:/.util_avg.stddev 136.90 ± 40% -76.8% 31.76 ± 6% sched_debug.cfs_rq:/.util_est_enqueued.avg 221.23 ± 12% -41.1% 130.35 ± 4% sched_debug.cfs_rq:/.util_est_enqueued.stddev 1060 ± 27% -51.7% 512.30 ± 8% sched_debug.cpu.curr->pid.avg 1940 ± 9% -18.8% 1575 ± 4% sched_debug.cpu.curr->pid.stddev 0.21 ± 24% -46.7% 0.11 ± 8% sched_debug.cpu.nr_running.avg 0.38 ± 8% -17.0% 0.31 ± 4% sched_debug.cpu.nr_running.stddev 11396 ± 51% -48.6% 5859 ± 13% proc-vmstat.nr_active_anon 2.409e+08 +23.1% 2.964e+08 ± 3% proc-vmstat.nr_dirtied 12452137 +2.0% 12702886 proc-vmstat.nr_dirty 47083061 +2.5% 48263837 proc-vmstat.nr_file_pages 29628750 -4.2% 28385431 proc-vmstat.nr_free_pages 46383516 +2.6% 47570490 proc-vmstat.nr_inactive_file 45891563 +2.4% 47011143 proc-vmstat.nr_mapped 232389 +26.9% 294974 ± 3% proc-vmstat.nr_page_table_pages 16705 ± 40% -38.3% 10310 ± 8% proc-vmstat.nr_shmem 132773 +2.0% 135451 proc-vmstat.nr_slab_reclaimable 355004 ± 2% +19.0% 422599 ± 5% proc-vmstat.nr_writeback 2.305e+08 +24.5% 2.87e+08 ± 4% proc-vmstat.nr_written 11396 ± 51% -48.6% 5859 ± 13% proc-vmstat.nr_zone_active_anon 46383514 +2.6% 47570490 proc-vmstat.nr_zone_inactive_file 12808434 +2.5% 13126679 proc-vmstat.nr_zone_write_pending 1.265e+08 ± 9% -28.2% 90860448 ± 4% proc-vmstat.numa_pte_updates 4.903e+08 +23.0% 6.03e+08 ± 3% proc-vmstat.pgfault 1127515 +6.8% 1204554 proc-vmstat.pgfree 9.227e+08 +24.5% 1.149e+09 ± 4% proc-vmstat.pgpgout 1.112e+08 +23.7% 1.375e+08 ± 3% proc-vmstat.pgreuse 13.87 -1.6% 13.66 perf-stat.i.MPKI 0.35 +0.0 0.37 ± 2% perf-stat.i.branch-miss-rate% 17443526 +9.6% 19118078 ± 5% perf-stat.i.branch-misses 76.66 +4.3 80.98 perf-stat.i.cache-miss-rate% 2.696e+08 +12.4% 3.029e+08 ± 3% perf-stat.i.cache-misses 3.554e+08 ± 2% +6.8% 3.797e+08 ± 3% perf-stat.i.cache-references 1.94 ± 2% -43.1% 1.10 perf-stat.i.cpi 5.592e+10 ± 3% -43.3% 3.171e+10 ± 3% perf-stat.i.cpu-cycles 212.99 ± 3% -44.8% 117.47 perf-stat.i.cycles-between-cache-misses 0.01 ± 14% +0.0 0.02 ± 14% perf-stat.i.dTLB-load-miss-rate% 861200 ± 16% +58.0% 1360994 ± 14% perf-stat.i.dTLB-load-misses 7.16e+09 +7.8% 7.719e+09 ± 3% perf-stat.i.dTLB-loads 4970815 +21.8% 6053776 ± 3% perf-stat.i.dTLB-store-misses 3.702e+09 +20.6% 4.466e+09 ± 3% perf-stat.i.dTLB-stores 0.63 ± 2% +53.0% 0.97 perf-stat.i.ipc 1192159 +22.0% 1454052 ± 3% perf-stat.i.major-faults 0.58 ± 3% -43.4% 0.33 ± 3% perf-stat.i.metric.GHz 1271 +8.5% 1380 ± 3% perf-stat.i.metric.K/sec 170.68 +9.0% 186.00 ± 3% perf-stat.i.metric.M/sec 37395 ± 3% +20.8% 45168 ± 3% perf-stat.i.minor-faults 34087239 ± 3% +8.9% 37130893 perf-stat.i.node-load-misses 25769456 ± 4% +20.3% 30992977 ± 10% perf-stat.i.node-loads 1229555 +21.9% 1499220 ± 3% perf-stat.i.page-faults 0.33 +0.0 0.36 ± 2% perf-stat.overall.branch-miss-rate% 75.82 +3.9 79.69 perf-stat.overall.cache-miss-rate% 2.17 ± 2% -46.6% 1.16 perf-stat.overall.cpi 209.90 ± 3% -49.8% 105.47 perf-stat.overall.cycles-between-cache-misses 0.01 ± 15% +0.0 0.02 ± 13% perf-stat.overall.dTLB-load-miss-rate% 0.46 ± 2% +87.1% 0.86 perf-stat.overall.ipc 11007 -13.6% 9506 perf-stat.overall.path-length 17381310 +9.7% 19064131 ± 5% perf-stat.ps.branch-misses 2.689e+08 +12.5% 3.025e+08 ± 3% perf-stat.ps.cache-misses 3.547e+08 ± 2% +7.1% 3.797e+08 ± 3% perf-stat.ps.cache-references 5.643e+10 ± 3% -43.4% 3.192e+10 ± 4% perf-stat.ps.cpu-cycles 856351 ± 16% +58.2% 1354654 ± 14% perf-stat.ps.dTLB-load-misses 7.162e+09 +7.8% 7.719e+09 ± 4% perf-stat.ps.dTLB-loads 4936410 +22.4% 6041559 ± 3% perf-stat.ps.dTLB-store-misses 3.696e+09 +20.7% 4.461e+09 ± 3% perf-stat.ps.dTLB-stores 1183837 +22.6% 1451485 ± 3% perf-stat.ps.major-faults 37115 ± 3% +21.5% 45089 ± 3% perf-stat.ps.minor-faults 33988476 ± 3% +9.0% 37042889 perf-stat.ps.node-load-misses 25570492 ± 4% +21.0% 30930343 ± 10% perf-stat.ps.node-loads 1220952 +22.6% 1496575 ± 3% perf-stat.ps.page-faults 30.04 ± 29% -30.0 0.00 perf-profile.calltrace.cycles-pp.cgroup_rstat_flush_irqsafe.mem_cgroup_wb_stats.balance_dirty_pages.balance_dirty_pages_ratelimited.fault_dirty_shared_page 26.77 ± 37% -26.8 0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.cgroup_rstat_flush_irqsafe.mem_cgroup_wb_stats.balance_dirty_pages.balance_dirty_pages_ratelimited 26.75 ± 37% -26.8 0.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.cgroup_rstat_flush_irqsafe.mem_cgroup_wb_stats.balance_dirty_pages 37.94 ± 16% -24.0 13.94 ± 8% perf-profile.calltrace.cycles-pp.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault 36.20 ± 18% -23.9 12.33 ± 8% perf-profile.calltrace.cycles-pp.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault 38.68 ± 15% -23.1 15.58 ± 9% perf-profile.calltrace.cycles-pp.do_user_addr_fault.exc_page_fault.asm_exc_page_fault 38.76 ± 15% -23.0 15.74 ± 9% perf-profile.calltrace.cycles-pp.exc_page_fault.asm_exc_page_fault 38.95 ± 15% -22.9 16.05 ± 9% perf-profile.calltrace.cycles-pp.asm_exc_page_fault 15.58 ± 29% -15.4 0.17 ±141% perf-profile.calltrace.cycles-pp.mem_cgroup_wb_stats.balance_dirty_pages.balance_dirty_pages_ratelimited.fault_dirty_shared_page.do_fault 15.64 ± 29% -15.3 0.39 ± 71% perf-profile.calltrace.cycles-pp.balance_dirty_pages.balance_dirty_pages_ratelimited.fault_dirty_shared_page.do_fault.__handle_mm_fault 15.78 ± 28% -14.9 0.86 ± 12% perf-profile.calltrace.cycles-pp.balance_dirty_pages_ratelimited.fault_dirty_shared_page.do_fault.__handle_mm_fault.handle_mm_fault 16.00 ± 27% -14.7 1.33 ± 12% perf-profile.calltrace.cycles-pp.fault_dirty_shared_page.do_fault.__handle_mm_fault.handle_mm_fault.do_user_addr_fault 14.63 ± 29% -14.6 0.00 perf-profile.calltrace.cycles-pp.mem_cgroup_wb_stats.balance_dirty_pages.balance_dirty_pages_ratelimited.fault_dirty_shared_page.do_wp_page 14.69 ± 29% -14.4 0.27 ±100% perf-profile.calltrace.cycles-pp.balance_dirty_pages.balance_dirty_pages_ratelimited.fault_dirty_shared_page.do_wp_page.__handle_mm_fault 14.81 ± 28% -14.1 0.75 ± 12% perf-profile.calltrace.cycles-pp.balance_dirty_pages_ratelimited.fault_dirty_shared_page.do_wp_page.__handle_mm_fault.handle_mm_fault 14.99 ± 28% -13.9 1.13 ± 11% perf-profile.calltrace.cycles-pp.fault_dirty_shared_page.do_wp_page.__handle_mm_fault.handle_mm_fault.do_user_addr_fault 16.86 ± 20% -12.4 4.41 ± 11% perf-profile.calltrace.cycles-pp.do_wp_page.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_page_fault 18.72 ± 18% -12.1 6.61 ± 7% perf-profile.calltrace.cycles-pp.do_fault.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_page_fault 1.30 ± 28% -0.6 0.68 ± 14% perf-profile.calltrace.cycles-pp.__count_memcg_events.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault 0.18 ±141% +0.6 0.74 ± 12% perf-profile.calltrace.cycles-pp.filemap_fault.__do_fault.do_fault.__handle_mm_fault.handle_mm_fault 0.18 ±142% +0.6 0.80 ± 12% perf-profile.calltrace.cycles-pp.iomap_iter.iomap_page_mkwrite.__xfs_filemap_fault.do_page_mkwrite.do_wp_page 0.22 ±141% +0.7 0.89 ± 11% perf-profile.calltrace.cycles-pp.__do_fault.do_fault.__handle_mm_fault.handle_mm_fault.do_user_addr_fault 0.23 ±141% +0.7 0.90 ± 25% perf-profile.calltrace.cycles-pp.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt 0.70 ± 72% +0.7 1.39 ± 6% perf-profile.calltrace.cycles-pp.__set_page_dirty_nobuffers.iomap_page_mkwrite.__xfs_filemap_fault.do_page_mkwrite.do_fault 0.20 ±142% +0.7 0.95 ± 8% perf-profile.calltrace.cycles-pp.iomap_iter.iomap_page_mkwrite.__xfs_filemap_fault.do_page_mkwrite.do_fault 0.10 ±223% +0.8 0.87 ± 15% perf-profile.calltrace.cycles-pp.td_io_queue 0.51 ± 75% +0.8 1.28 ± 7% perf-profile.calltrace.cycles-pp.sync_regs.error_entry 0.82 ± 74% +0.8 1.63 ± 9% perf-profile.calltrace.cycles-pp.page_vma_mapped_walk.page_mkclean_one.rmap_walk_file.page_mkclean.clear_page_dirty_for_io 0.55 ± 74% +0.9 1.43 ± 10% perf-profile.calltrace.cycles-pp.get_io_u 0.60 ± 75% +0.9 1.50 ± 8% perf-profile.calltrace.cycles-pp.error_entry 0.73 ± 76% +0.9 1.64 ± 24% perf-profile.calltrace.cycles-pp.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state 0.74 ± 76% +0.9 1.66 ± 24% perf-profile.calltrace.cycles-pp.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter 1.24 ± 76% +1.0 2.22 ± 16% perf-profile.calltrace.cycles-pp.end_page_writeback.iomap_finish_ioend.pmem_submit_bio.__submit_bio.__submit_bio_noacct 1.37 ± 41% +1.0 2.34 ± 13% perf-profile.calltrace.cycles-pp.iomap_page_mkwrite.__xfs_filemap_fault.do_page_mkwrite.do_wp_page.__handle_mm_fault 0.55 ± 73% +1.0 1.55 ± 8% perf-profile.calltrace.cycles-pp.io_completed 1.29 ± 76% +1.0 2.31 ± 16% perf-profile.calltrace.cycles-pp.iomap_finish_ioend.pmem_submit_bio.__submit_bio.__submit_bio_noacct.iomap_submit_ioend 1.13 ± 41% +1.0 2.17 ± 6% perf-profile.calltrace.cycles-pp.fio_gettime 0.08 ±223% +1.2 1.29 ± 9% perf-profile.calltrace.cycles-pp.xfs_buffered_write_iomap_begin.iomap_iter.iomap_page_mkwrite.__xfs_filemap_fault.do_page_mkwrite 1.65 ± 40% +1.2 2.89 ± 11% perf-profile.calltrace.cycles-pp.__xfs_filemap_fault.do_page_mkwrite.do_wp_page.__handle_mm_fault.handle_mm_fault 1.67 ± 40% +1.3 2.93 ± 11% perf-profile.calltrace.cycles-pp.do_page_mkwrite.do_wp_page.__handle_mm_fault.handle_mm_fault.do_user_addr_fault 1.16 ± 78% +1.3 2.44 ± 24% perf-profile.calltrace.cycles-pp.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call 1.47 ± 37% +1.3 2.77 ± 5% perf-profile.calltrace.cycles-pp.iomap_page_mkwrite.__xfs_filemap_fault.do_page_mkwrite.do_fault.__handle_mm_fault 1.30 ± 78% +1.4 2.73 ± 25% perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle 1.78 ± 37% +1.6 3.38 ± 5% perf-profile.calltrace.cycles-pp.__xfs_filemap_fault.do_page_mkwrite.do_fault.__handle_mm_fault.handle_mm_fault 1.79 ± 37% +1.6 3.41 ± 5% perf-profile.calltrace.cycles-pp.do_page_mkwrite.do_fault.__handle_mm_fault.handle_mm_fault.do_user_addr_fault 3.66 ± 34% +2.2 5.84 ± 11% perf-profile.calltrace.cycles-pp.rmap_walk_file.page_mkclean.clear_page_dirty_for_io.write_cache_pages.iomap_writepages 3.82 ± 35% +2.4 6.18 ± 12% perf-profile.calltrace.cycles-pp.page_mkclean.clear_page_dirty_for_io.write_cache_pages.iomap_writepages.xfs_vm_writepages 4.97 ± 31% +2.5 7.51 ± 12% perf-profile.calltrace.cycles-pp.clear_page_dirty_for_io.write_cache_pages.iomap_writepages.xfs_vm_writepages.do_writepages 0.51 ±146% +3.2 3.69 ± 47% perf-profile.calltrace.cycles-pp.__invalidate_mapping_pages.generic_fadvise.ksys_fadvise64_64.__x64_sys_fadvise64.do_syscall_64 7.34 ± 43% +4.9 12.24 ± 13% perf-profile.calltrace.cycles-pp.iomap_writepage_map.write_cache_pages.iomap_writepages.xfs_vm_writepages.do_writepages 1.47 ±143% +7.9 9.40 ± 49% perf-profile.calltrace.cycles-pp.__x64_sys_fadvise64.do_syscall_64.entry_SYSCALL_64_after_hwframe.posix_fadvise 1.47 ±143% +7.9 9.40 ± 49% perf-profile.calltrace.cycles-pp.ksys_fadvise64_64.__x64_sys_fadvise64.do_syscall_64.entry_SYSCALL_64_after_hwframe.posix_fadvise 1.47 ±143% +7.9 9.40 ± 49% perf-profile.calltrace.cycles-pp.generic_fadvise.ksys_fadvise64_64.__x64_sys_fadvise64.do_syscall_64.entry_SYSCALL_64_after_hwframe 1.47 ±143% +8.0 9.42 ± 49% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.posix_fadvise 1.47 ±143% +8.0 9.42 ± 49% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.posix_fadvise 1.47 ±143% +8.0 9.42 ± 49% perf-profile.calltrace.cycles-pp.posix_fadvise 30.21 ± 29% -29.4 0.80 ± 14% perf-profile.children.cycles-pp.mem_cgroup_wb_stats 30.33 ± 29% -29.3 1.04 ± 13% perf-profile.children.cycles-pp.balance_dirty_pages 30.04 ± 29% -29.3 0.77 ± 14% perf-profile.children.cycles-pp.cgroup_rstat_flush_irqsafe 30.59 ± 28% -29.0 1.62 ± 11% perf-profile.children.cycles-pp.balance_dirty_pages_ratelimited 31.00 ± 27% -28.5 2.48 ± 10% perf-profile.children.cycles-pp.fault_dirty_shared_page 27.21 ± 36% -26.7 0.52 ± 68% perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath 27.89 ± 34% -26.4 1.53 ± 31% perf-profile.children.cycles-pp._raw_spin_lock_irqsave 37.97 ± 16% -24.0 14.00 ± 8% perf-profile.children.cycles-pp.handle_mm_fault 36.21 ± 18% -23.8 12.36 ± 8% perf-profile.children.cycles-pp.__handle_mm_fault 38.71 ± 15% -23.1 15.65 ± 9% perf-profile.children.cycles-pp.do_user_addr_fault 38.77 ± 15% -23.0 15.76 ± 9% perf-profile.children.cycles-pp.exc_page_fault 38.98 ± 15% -22.9 16.12 ± 9% perf-profile.children.cycles-pp.asm_exc_page_fault 16.87 ± 20% -12.4 4.42 ± 11% perf-profile.children.cycles-pp.do_wp_page 18.73 ± 18% -12.1 6.63 ± 7% perf-profile.children.cycles-pp.do_fault 3.27 ± 36% -2.5 0.76 ± 14% perf-profile.children.cycles-pp.cgroup_rstat_flush_locked 2.14 ± 28% -1.5 0.68 ± 16% perf-profile.children.cycles-pp.cgroup_rstat_updated 1.51 ± 30% -0.7 0.78 ± 14% perf-profile.children.cycles-pp.__mod_memcg_lruvec_state 1.20 ± 34% -0.7 0.49 ± 18% perf-profile.children.cycles-pp.mem_cgroup_css_rstat_flush 1.33 ± 28% -0.6 0.71 ± 15% perf-profile.children.cycles-pp.__count_memcg_events 0.03 ±100% +0.0 0.07 ± 11% perf-profile.children.cycles-pp.inc_node_page_state 0.04 ± 71% +0.0 0.09 ± 18% perf-profile.children.cycles-pp.task_tick_fair 0.01 ±223% +0.1 0.06 ± 13% perf-profile.children.cycles-pp.perf_exclude_event 0.00 +0.1 0.05 ± 13% perf-profile.children.cycles-pp.iomap_do_writepage 0.02 ±143% +0.1 0.08 ± 16% perf-profile.children.cycles-pp.vma_interval_tree_iter_next 0.02 ±141% +0.1 0.08 ± 12% perf-profile.children.cycles-pp.__irqentry_text_end 0.05 ± 76% +0.1 0.11 ± 15% perf-profile.children.cycles-pp.fput_many 0.01 ±223% +0.1 0.08 ± 12% perf-profile.children.cycles-pp.__radix_tree_lookup 0.02 ±144% +0.1 0.08 ± 32% perf-profile.children.cycles-pp.get_start_offset 0.02 ±141% +0.1 0.08 ± 12% perf-profile.children.cycles-pp.io_bytes_exceeded 0.00 +0.1 0.06 ± 11% perf-profile.children.cycles-pp.xfs_filemap_fault 0.05 ± 74% +0.1 0.12 ± 6% perf-profile.children.cycles-pp.in_ramp_time 0.01 ±223% +0.1 0.08 ± 31% perf-profile.children.cycles-pp.rcu_pending 0.04 ± 72% +0.1 0.11 ± 9% perf-profile.children.cycles-pp.ntime_since 0.07 ± 77% +0.1 0.14 ± 11% perf-profile.children.cycles-pp.xas_find_marked 0.06 ± 75% +0.1 0.13 ± 21% perf-profile.children.cycles-pp.irqtime_account_irq 0.07 ± 73% +0.1 0.14 ± 10% perf-profile.children.cycles-pp.rcu_read_unlock_strict 0.00 +0.1 0.07 ± 12% perf-profile.children.cycles-pp.__inc_zone_page_state 0.02 ±142% +0.1 0.10 ± 14% perf-profile.children.cycles-pp.utime_since 0.11 ± 17% +0.1 0.19 ± 25% perf-profile.children.cycles-pp.exit_to_user_mode_prepare 0.05 ± 73% +0.1 0.13 ± 15% perf-profile.children.cycles-pp.up_write 0.06 ± 78% +0.1 0.14 ± 14% perf-profile.children.cycles-pp.xas_set_mark 0.02 ±142% +0.1 0.10 ± 37% perf-profile.children.cycles-pp.__schedule 0.01 ±223% +0.1 0.09 ± 12% perf-profile.children.cycles-pp.zbd_unaligned_write 0.10 ± 40% +0.1 0.18 ± 16% perf-profile.children.cycles-pp.__mark_inode_dirty 0.03 ±100% +0.1 0.11 ± 20% perf-profile.children.cycles-pp.log_io_u 0.04 ± 73% +0.1 0.13 ± 19% perf-profile.children.cycles-pp.xfs_iext_lookup_extent 0.06 ± 74% +0.1 0.15 ± 9% perf-profile.children.cycles-pp.finish_mkwrite_fault 0.06 ± 78% +0.1 0.16 ± 31% perf-profile.children.cycles-pp.perf_mux_hrtimer_handler 0.08 ± 56% +0.1 0.18 ± 9% perf-profile.children.cycles-pp.unlock_page_memcg 0.06 ± 74% +0.1 0.16 ± 13% perf-profile.children.cycles-pp.down_read_trylock 0.06 ± 74% +0.1 0.17 ± 12% perf-profile.children.cycles-pp.page_add_file_rmap 0.07 ± 78% +0.1 0.17 ± 27% perf-profile.children.cycles-pp.calc_global_load_tick 0.06 ± 75% +0.1 0.16 ± 11% perf-profile.children.cycles-pp.utime_since_now 0.02 ±145% +0.1 0.13 ± 24% perf-profile.children.cycles-pp.llist_reverse_order 0.09 ± 44% +0.1 0.20 ± 19% perf-profile.children.cycles-pp.init_icd 0.14 ± 43% +0.1 0.25 ± 10% perf-profile.children.cycles-pp.next_uptodate_page 0.10 ± 57% +0.1 0.21 ± 11% perf-profile.children.cycles-pp.rcu_all_qs 0.07 ± 77% +0.1 0.18 ± 10% perf-profile.children.cycles-pp.xas_start 0.06 ± 76% +0.1 0.18 ± 9% perf-profile.children.cycles-pp.vmacache_find 0.10 ± 77% +0.1 0.22 ± 16% perf-profile.children.cycles-pp.check_pte 0.09 ± 64% +0.1 0.21 ± 11% perf-profile.children.cycles-pp.put_io_u 0.09 ± 58% +0.1 0.21 ± 11% perf-profile.children.cycles-pp.td_io_prep 0.06 ± 75% +0.1 0.18 ± 15% perf-profile.children.cycles-pp.xfs_bmbt_to_iomap 0.12 ± 42% +0.1 0.24 ± 5% perf-profile.children.cycles-pp.__get_io_u 0.08 ± 59% +0.1 0.21 ± 8% perf-profile.children.cycles-pp.find_vma 0.12 ± 43% +0.1 0.25 ± 13% perf-profile.children.cycles-pp.__fprop_inc_percpu 0.09 ± 75% +0.1 0.22 ± 22% perf-profile.children.cycles-pp.tick_nohz_irq_exit 0.14 ± 49% +0.1 0.27 ± 11% perf-profile.children.cycles-pp.down_write 0.14 ± 62% +0.1 0.27 ± 6% perf-profile.children.cycles-pp.PageHuge 0.04 ±105% +0.1 0.17 ± 11% perf-profile.children.cycles-pp.io_u_mark_depth 0.23 ± 27% +0.1 0.37 ± 10% perf-profile.children.cycles-pp.file_update_time 0.14 ± 43% +0.2 0.30 ± 10% perf-profile.children.cycles-pp.rand_between 0.18 ± 48% +0.2 0.34 ± 11% perf-profile.children.cycles-pp.__might_sleep 0.18 ± 48% +0.2 0.34 ± 10% perf-profile.children.cycles-pp.__xa_clear_mark 0.07 ± 75% +0.2 0.23 ± 19% perf-profile.children.cycles-pp.io_u_mark_submit 0.14 ± 39% +0.2 0.30 ± 9% perf-profile.children.cycles-pp.do_set_pte 0.08 ± 83% +0.2 0.23 ± 31% perf-profile.children.cycles-pp.get_next_seq_offset 0.13 ± 47% +0.2 0.30 ± 10% perf-profile.children.cycles-pp.finish_fault 0.21 ± 46% +0.2 0.38 ± 9% perf-profile.children.cycles-pp.find_get_pages_range_tag 0.22 ± 47% +0.2 0.39 ± 8% perf-profile.children.cycles-pp.pagevec_lookup_range_tag 0.16 ± 43% +0.2 0.34 ± 20% perf-profile.children.cycles-pp.lock_page_memcg 0.11 ± 40% +0.2 0.30 ± 13% perf-profile.children.cycles-pp.io_u_sync_complete 0.16 ± 43% +0.2 0.35 ± 13% perf-profile.children.cycles-pp.set_page_dirty 0.16 ± 49% +0.2 0.36 ± 9% perf-profile.children.cycles-pp.___perf_sw_event 0.02 ±141% +0.2 0.23 ± 42% perf-profile.children.cycles-pp.xas_find 0.13 ± 48% +0.2 0.34 ± 15% perf-profile.children.cycles-pp.io_queue_event 0.15 ± 45% +0.2 0.36 ± 22% perf-profile.children.cycles-pp.fio_mmapio_queue 0.16 ± 37% +0.2 0.38 ± 10% perf-profile.children.cycles-pp.xfs_iunlock 0.23 ± 44% +0.2 0.44 ± 9% perf-profile.children.cycles-pp.__xa_set_mark 0.22 ± 49% +0.2 0.43 ± 18% perf-profile.children.cycles-pp.scheduler_tick 0.10 ± 59% +0.2 0.32 ± 4% perf-profile.children.cycles-pp.io_u_mark_complete 0.26 ± 47% +0.2 0.48 ± 7% perf-profile.children.cycles-pp.__mod_node_page_state 0.23 ± 42% +0.2 0.46 ± 12% perf-profile.children.cycles-pp.handle_pte_fault 0.22 ± 44% +0.2 0.46 ± 7% perf-profile.children.cycles-pp.account_io_completion 0.22 ± 44% +0.2 0.46 ± 7% perf-profile.children.cycles-pp.__cond_resched 0.31 ± 47% +0.2 0.56 ± 6% perf-profile.children.cycles-pp.__mod_lruvec_state 0.28 ± 43% +0.2 0.53 ± 10% perf-profile.children.cycles-pp.filemap_map_pages 0.29 ± 43% +0.2 0.54 ± 10% perf-profile.children.cycles-pp.xfs_filemap_map_pages 0.27 ± 41% +0.3 0.53 ± 12% perf-profile.children.cycles-pp.pagecache_get_page 0.32 ± 42% +0.3 0.59 ± 10% perf-profile.children.cycles-pp.do_read_fault 0.26 ± 46% +0.3 0.53 ± 5% perf-profile.children.cycles-pp.add_lat_sample 0.24 ± 36% +0.3 0.54 ± 14% perf-profile.children.cycles-pp.up_read 0.10 ± 27% +0.3 0.40 ± 33% perf-profile.children.cycles-pp.__pagevec_release 0.28 ± 45% +0.3 0.58 ± 12% perf-profile.children.cycles-pp.add_clat_sample 0.32 ± 45% +0.3 0.63 ± 11% perf-profile.children.cycles-pp.xfs_ilock 0.32 ± 59% +0.3 0.65 ± 16% perf-profile.children.cycles-pp.clockevents_program_event 0.36 ± 43% +0.3 0.70 ± 10% perf-profile.children.cycles-pp.down_read 0.26 ± 38% +0.4 0.62 ± 16% perf-profile.children.cycles-pp.page_mapping 0.27 ± 48% +0.4 0.62 ± 8% perf-profile.children.cycles-pp.__perf_sw_event 0.22 ± 41% +0.4 0.58 ± 29% perf-profile.children.cycles-pp.flush_smp_call_function_queue 0.38 ± 41% +0.4 0.74 ± 12% perf-profile.children.cycles-pp.filemap_fault 0.22 ± 42% +0.4 0.58 ± 29% perf-profile.children.cycles-pp.__sysvec_call_function_single 0.19 ± 21% +0.4 0.56 ± 20% perf-profile.children.cycles-pp.unlock_page 0.29 ± 45% +0.4 0.65 ± 8% perf-profile.children.cycles-pp.___might_sleep 0.24 ± 40% +0.4 0.64 ± 28% perf-profile.children.cycles-pp.sysvec_call_function_single 0.38 ± 48% +0.4 0.78 ± 18% perf-profile.children.cycles-pp.update_process_times 0.32 ± 46% +0.4 0.73 ± 8% perf-profile.children.cycles-pp.thread_main 0.38 ± 47% +0.4 0.80 ± 18% perf-profile.children.cycles-pp.tick_sched_handle 0.45 ± 41% +0.4 0.89 ± 11% perf-profile.children.cycles-pp.__do_fault 0.42 ± 48% +0.4 0.86 ± 19% perf-profile.children.cycles-pp.tick_sched_timer 0.33 ± 33% +0.5 0.84 ± 31% perf-profile.children.cycles-pp.asm_sysvec_call_function_single 0.13 ± 52% +0.6 0.68 ± 39% perf-profile.children.cycles-pp.release_pages 1.09 ± 41% +0.6 1.65 ± 9% perf-profile.children.cycles-pp.__test_set_page_writeback 0.45 ± 43% +0.6 1.02 ± 13% perf-profile.children.cycles-pp.td_io_queue 0.56 ± 50% +0.6 1.17 ± 20% perf-profile.children.cycles-pp.__hrtimer_run_queues 0.59 ± 42% +0.6 1.20 ± 9% perf-profile.children.cycles-pp.xas_load 0.45 ± 46% +0.6 1.10 ± 14% perf-profile.children.cycles-pp.percpu_counter_add_batch 0.63 ± 43% +0.7 1.34 ± 7% perf-profile.children.cycles-pp.sync_regs 0.59 ± 41% +0.7 1.31 ± 9% perf-profile.children.cycles-pp.xfs_buffered_write_iomap_begin 0.00 +0.8 0.78 ± 13% perf-profile.children.cycles-pp.mem_cgroup_flush_stats 0.68 ± 42% +0.8 1.48 ± 9% perf-profile.children.cycles-pp.get_io_u 0.78 ± 42% +0.9 1.66 ± 8% perf-profile.children.cycles-pp.error_entry 0.66 ± 41% +0.9 1.56 ± 8% perf-profile.children.cycles-pp.io_completed 0.13 ±141% +0.9 1.07 ± 44% perf-profile.children.cycles-pp.find_lock_entries 0.80 ± 40% +1.0 1.76 ± 8% perf-profile.children.cycles-pp.iomap_iter 0.96 ± 51% +1.0 1.96 ± 6% perf-profile.children.cycles-pp.page_vma_mapped_walk 1.01 ± 54% +1.0 2.03 ± 20% perf-profile.children.cycles-pp.hrtimer_interrupt 1.02 ± 55% +1.0 2.05 ± 20% perf-profile.children.cycles-pp.__sysvec_apic_timer_interrupt 1.36 ± 41% +1.0 2.40 ± 14% perf-profile.children.cycles-pp.test_clear_page_writeback 1.14 ± 41% +1.0 2.18 ± 6% perf-profile.children.cycles-pp.fio_gettime 1.92 ± 38% +1.1 3.04 ± 8% perf-profile.children.cycles-pp.__set_page_dirty_nobuffers 1.51 ± 41% +1.1 2.65 ± 12% perf-profile.children.cycles-pp.end_page_writeback 0.99 ± 41% +1.2 2.16 ± 8% perf-profile.children.cycles-pp.native_irq_return_iret 0.25 ±157% +1.2 1.44 ± 58% perf-profile.children.cycles-pp.pagevec_lru_move_fn 1.56 ± 41% +1.2 2.76 ± 12% perf-profile.children.cycles-pp.iomap_finish_ioend 1.51 ± 58% +1.4 2.88 ± 20% perf-profile.children.cycles-pp.sysvec_apic_timer_interrupt 0.30 ±154% +1.6 1.86 ± 54% perf-profile.children.cycles-pp.deactivate_file_page 1.72 ± 58% +1.6 3.29 ± 20% perf-profile.children.cycles-pp.asm_sysvec_apic_timer_interrupt 1.84 ± 36% +1.7 3.54 ± 21% perf-profile.children.cycles-pp.iomap_submit_ioend 3.82 ± 28% +2.1 5.92 ± 11% perf-profile.children.cycles-pp.rmap_walk_file 3.97 ± 29% +2.2 6.20 ± 11% perf-profile.children.cycles-pp.page_mkclean 2.85 ± 39% +2.3 5.14 ± 8% perf-profile.children.cycles-pp.iomap_page_mkwrite 4.98 ± 31% +2.6 7.55 ± 11% perf-profile.children.cycles-pp.clear_page_dirty_for_io 3.45 ± 38% +2.9 6.31 ± 7% perf-profile.children.cycles-pp.__xfs_filemap_fault 3.47 ± 38% +2.9 6.34 ± 7% perf-profile.children.cycles-pp.do_page_mkwrite 0.52 ±146% +3.2 3.72 ± 47% perf-profile.children.cycles-pp.__invalidate_mapping_pages 0.95 ±142% +4.7 5.69 ± 52% perf-profile.children.cycles-pp.__filemap_fdatawrite_range 0.95 ±142% +4.7 5.69 ± 52% perf-profile.children.cycles-pp.filemap_fdatawrite_wbc 7.34 ± 43% +4.9 12.29 ± 13% perf-profile.children.cycles-pp.iomap_writepage_map 1.47 ±143% +7.9 9.40 ± 49% perf-profile.children.cycles-pp.__x64_sys_fadvise64 1.47 ±143% +7.9 9.40 ± 49% perf-profile.children.cycles-pp.ksys_fadvise64_64 1.47 ±143% +7.9 9.40 ± 49% perf-profile.children.cycles-pp.generic_fadvise 1.47 ±143% +8.0 9.42 ± 49% perf-profile.children.cycles-pp.posix_fadvise 12.85 ± 38% +8.0 20.81 ± 11% perf-profile.children.cycles-pp.iomap_writepages 12.85 ± 38% +8.0 20.81 ± 11% perf-profile.children.cycles-pp.write_cache_pages 1.62 ±129% +8.0 9.63 ± 48% perf-profile.children.cycles-pp.do_syscall_64 1.62 ±129% +8.0 9.63 ± 48% perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe 14.58 ± 39% +8.9 23.44 ± 10% perf-profile.children.cycles-pp.do_writepages 14.58 ± 39% +8.9 23.44 ± 10% perf-profile.children.cycles-pp.xfs_vm_writepages 27.20 ± 36% -26.7 0.51 ± 68% perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath 1.58 ± 27% -1.2 0.42 ± 17% perf-profile.self.cycles-pp.cgroup_rstat_updated 1.17 ± 34% -0.7 0.47 ± 19% perf-profile.self.cycles-pp.mem_cgroup_css_rstat_flush 0.42 ± 31% -0.3 0.09 ± 12% perf-profile.self.cycles-pp.cgroup_rstat_flush_locked 0.02 ±141% +0.0 0.06 ± 13% perf-profile.self.cycles-pp.finish_fault 0.04 ± 76% +0.0 0.09 ± 12% perf-profile.self.cycles-pp.write_pmem 0.04 ± 76% +0.0 0.09 ± 9% perf-profile.self.cycles-pp.in_ramp_time 0.02 ±142% +0.0 0.07 ± 11% perf-profile.self.cycles-pp.pmem_do_write 0.00 +0.1 0.05 perf-profile.self.cycles-pp.do_fault 0.04 ± 73% +0.1 0.09 ± 9% perf-profile.self.cycles-pp.xfs_ilock 0.05 ± 73% +0.1 0.11 ± 15% perf-profile.self.cycles-pp.iomap_finish_ioend 0.05 ± 76% +0.1 0.10 ± 15% perf-profile.self.cycles-pp.fput_many 0.02 ±141% +0.1 0.07 ± 14% perf-profile.self.cycles-pp.__irqentry_text_end 0.04 ± 73% +0.1 0.10 ± 10% perf-profile.self.cycles-pp.rcu_read_unlock_strict 0.03 ±102% +0.1 0.08 ± 21% perf-profile.self.cycles-pp.do_set_pte 0.04 ± 72% +0.1 0.09 ± 14% perf-profile.self.cycles-pp.file_update_time 0.01 ±223% +0.1 0.06 ± 7% perf-profile.self.cycles-pp.memset_erms 0.02 ±142% +0.1 0.08 ± 20% perf-profile.self.cycles-pp.vma_interval_tree_iter_next 0.05 ± 76% +0.1 0.11 ± 16% perf-profile.self.cycles-pp.balance_dirty_pages 0.04 ± 73% +0.1 0.10 ± 14% perf-profile.self.cycles-pp.exc_page_fault 0.02 ±141% +0.1 0.08 ± 10% perf-profile.self.cycles-pp.io_bytes_exceeded 0.03 ±100% +0.1 0.09 ± 7% perf-profile.self.cycles-pp.ntime_since 0.00 +0.1 0.06 ± 13% perf-profile.self.cycles-pp.xfs_filemap_fault 0.02 ±144% +0.1 0.08 ± 17% perf-profile.self.cycles-pp.utime_since 0.06 ± 74% +0.1 0.12 ± 12% perf-profile.self.cycles-pp.__bio_try_merge_page 0.01 ±223% +0.1 0.07 ± 15% perf-profile.self.cycles-pp.__radix_tree_lookup 0.05 ± 72% +0.1 0.11 ± 9% perf-profile.self.cycles-pp.up_write 0.03 ±100% +0.1 0.10 ± 18% perf-profile.self.cycles-pp.pmem_submit_bio 0.07 ± 77% +0.1 0.14 ± 11% perf-profile.self.cycles-pp.xas_find_marked 0.00 +0.1 0.07 ± 10% perf-profile.self.cycles-pp.__inc_zone_page_state 0.06 ± 76% +0.1 0.13 ± 14% perf-profile.self.cycles-pp.iomap_add_to_ioend 0.04 ± 75% +0.1 0.12 ± 10% perf-profile.self.cycles-pp.page_add_file_rmap 0.04 ± 73% +0.1 0.11 ± 12% perf-profile.self.cycles-pp.xfs_iext_lookup_extent 0.00 +0.1 0.07 ± 9% perf-profile.self.cycles-pp.__set_page_dirty 0.06 ± 74% +0.1 0.13 ± 11% perf-profile.self.cycles-pp.set_page_dirty 0.06 ± 76% +0.1 0.13 ± 11% perf-profile.self.cycles-pp.xas_set_mark 0.06 ± 81% +0.1 0.14 ± 11% perf-profile.self.cycles-pp.fio_mmapio_prep 0.13 ± 39% +0.1 0.22 ± 10% perf-profile.self.cycles-pp.pagecache_get_page 0.02 ±142% +0.1 0.10 ± 21% perf-profile.self.cycles-pp.log_io_u 0.07 ± 74% +0.1 0.15 ± 13% perf-profile.self.cycles-pp.rcu_all_qs 0.06 ± 74% +0.1 0.15 ± 11% perf-profile.self.cycles-pp.unlock_page_memcg 0.03 ±105% +0.1 0.12 ± 18% perf-profile.self.cycles-pp.xfs_iunlock 0.09 ± 74% +0.1 0.17 ± 13% perf-profile.self.cycles-pp.flush_tlb_mm_range 0.09 ± 56% +0.1 0.18 ± 15% perf-profile.self.cycles-pp.__mark_inode_dirty 0.05 ± 74% +0.1 0.14 ± 8% perf-profile.self.cycles-pp.down_read_trylock 0.07 ± 81% +0.1 0.16 ± 11% perf-profile.self.cycles-pp.down_write 0.03 ±101% +0.1 0.12 ± 11% perf-profile.self.cycles-pp.xfs_bmbt_to_iomap 0.09 ± 42% +0.1 0.18 ± 15% perf-profile.self.cycles-pp.init_icd 0.09 ± 76% +0.1 0.19 ± 18% perf-profile.self.cycles-pp.check_pte 0.09 ± 55% +0.1 0.19 ± 13% perf-profile.self.cycles-pp.asm_exc_page_fault 0.06 ± 75% +0.1 0.16 ± 13% perf-profile.self.cycles-pp.xas_start 0.06 ± 73% +0.1 0.16 ± 9% perf-profile.self.cycles-pp.utime_since_now 0.07 ± 78% +0.1 0.17 ± 27% perf-profile.self.cycles-pp.calc_global_load_tick 0.10 ± 40% +0.1 0.21 ± 11% perf-profile.self.cycles-pp.filemap_fault 0.06 ± 78% +0.1 0.16 ± 10% perf-profile.self.cycles-pp.vmacache_find 0.05 ± 77% +0.1 0.16 ± 18% perf-profile.self.cycles-pp.account_page_dirtied 0.02 ±145% +0.1 0.13 ± 24% perf-profile.self.cycles-pp.llist_reverse_order 0.14 ± 43% +0.1 0.25 ± 11% perf-profile.self.cycles-pp.next_uptodate_page 0.07 ± 78% +0.1 0.18 ± 13% perf-profile.self.cycles-pp.fault_dirty_shared_page 0.14 ± 44% +0.1 0.25 ± 12% perf-profile.self.cycles-pp.find_get_pages_range_tag 0.14 ± 47% +0.1 0.26 ± 5% perf-profile.self.cycles-pp.end_page_writeback 0.08 ± 75% +0.1 0.20 ± 13% perf-profile.self.cycles-pp.td_io_prep 0.08 ± 80% +0.1 0.19 ± 11% perf-profile.self.cycles-pp.put_io_u 0.07 ± 82% +0.1 0.18 ± 19% perf-profile.self.cycles-pp.update_process_times 0.11 ± 40% +0.1 0.24 ± 9% perf-profile.self.cycles-pp.__cond_resched 0.10 ± 36% +0.1 0.23 ± 9% perf-profile.self.cycles-pp.__xfs_filemap_fault 0.10 ± 76% +0.1 0.22 ± 6% perf-profile.self.cycles-pp.PageHuge 0.04 ±105% +0.1 0.16 ± 12% perf-profile.self.cycles-pp.io_u_mark_depth 0.12 ± 60% +0.1 0.25 ± 5% perf-profile.self.cycles-pp.rmap_walk_file 0.10 ± 59% +0.1 0.24 ± 7% perf-profile.self.cycles-pp.__get_io_u 0.11 ± 44% +0.1 0.24 ± 10% perf-profile.self.cycles-pp.handle_pte_fault 0.16 ± 48% +0.1 0.29 ± 13% perf-profile.self.cycles-pp.write_cache_pages 0.17 ± 48% +0.1 0.32 ± 10% perf-profile.self.cycles-pp.__might_sleep 0.07 ± 74% +0.1 0.21 ± 21% perf-profile.self.cycles-pp.io_u_mark_submit 0.15 ± 43% +0.2 0.30 ± 20% perf-profile.self.cycles-pp.lock_page_memcg 0.14 ± 45% +0.2 0.29 ± 8% perf-profile.self.cycles-pp.rand_between 0.07 ± 85% +0.2 0.22 ± 30% perf-profile.self.cycles-pp.get_next_seq_offset 0.10 ± 59% +0.2 0.26 ± 12% perf-profile.self.cycles-pp.__perf_sw_event 0.10 ± 60% +0.2 0.26 ± 11% perf-profile.self.cycles-pp.___perf_sw_event 0.19 ± 48% +0.2 0.35 ± 6% perf-profile.self.cycles-pp.clear_page_dirty_for_io 0.08 ± 58% +0.2 0.24 ± 32% perf-profile.self.cycles-pp.flush_smp_call_function_queue 0.14 ± 37% +0.2 0.30 ± 11% perf-profile.self.cycles-pp.error_entry 0.24 ± 44% +0.2 0.41 ± 10% perf-profile.self.cycles-pp.__test_set_page_writeback 0.10 ± 61% +0.2 0.27 ± 16% perf-profile.self.cycles-pp.io_queue_event 0.10 ± 59% +0.2 0.28 ± 14% perf-profile.self.cycles-pp.io_u_sync_complete 0.17 ± 42% +0.2 0.36 ± 10% perf-profile.self.cycles-pp.do_user_addr_fault 0.08 ± 74% +0.2 0.28 ± 5% perf-profile.self.cycles-pp.io_u_mark_complete 0.17 ± 37% +0.2 0.37 ± 8% perf-profile.self.cycles-pp.iomap_iter 0.15 ± 46% +0.2 0.35 ± 22% perf-profile.self.cycles-pp.fio_mmapio_queue 0.25 ± 46% +0.2 0.46 ± 8% perf-profile.self.cycles-pp.__mod_node_page_state 0.22 ± 42% +0.2 0.44 ± 9% perf-profile.self.cycles-pp.down_read 0.20 ± 47% +0.2 0.42 ± 11% perf-profile.self.cycles-pp.iomap_page_mkwrite 0.15 ± 37% +0.2 0.38 ± 7% perf-profile.self.cycles-pp.xfs_buffered_write_iomap_begin 0.25 ± 44% +0.2 0.48 ± 12% perf-profile.self.cycles-pp.__set_page_dirty_nobuffers 0.22 ± 45% +0.2 0.45 ± 7% perf-profile.self.cycles-pp.account_io_completion 0.26 ± 55% +0.3 0.51 ± 6% perf-profile.self.cycles-pp.page_mkclean_one 0.25 ± 46% +0.3 0.51 ± 5% perf-profile.self.cycles-pp.add_lat_sample 0.26 ± 46% +0.3 0.52 ± 14% perf-profile.self.cycles-pp.test_clear_page_writeback 0.21 ± 51% +0.3 0.48 ± 10% perf-profile.self.cycles-pp.balance_dirty_pages_ratelimited 0.24 ± 35% +0.3 0.52 ± 14% perf-profile.self.cycles-pp.up_read 0.24 ± 42% +0.3 0.54 ± 13% perf-profile.self.cycles-pp.handle_mm_fault 0.26 ± 44% +0.3 0.56 ± 13% perf-profile.self.cycles-pp.add_clat_sample 0.34 ± 67% +0.3 0.67 ± 14% perf-profile.self.cycles-pp.ktime_get 0.24 ± 37% +0.3 0.57 ± 16% perf-profile.self.cycles-pp.page_mapping 0.18 ± 20% +0.3 0.52 ± 20% perf-profile.self.cycles-pp.unlock_page 0.27 ± 44% +0.3 0.62 ± 7% perf-profile.self.cycles-pp.___might_sleep 0.26 ± 46% +0.4 0.63 ± 9% perf-profile.self.cycles-pp.thread_main 0.04 ±142% +0.4 0.41 ± 40% perf-profile.self.cycles-pp.deactivate_file_page 0.36 ± 44% +0.4 0.81 ± 11% perf-profile.self.cycles-pp.__handle_mm_fault 0.31 ± 43% +0.5 0.78 ± 14% perf-profile.self.cycles-pp.percpu_counter_add_batch 0.51 ± 42% +0.5 1.02 ± 8% perf-profile.self.cycles-pp.xas_load 0.12 ± 53% +0.5 0.64 ± 39% perf-profile.self.cycles-pp.release_pages 0.43 ± 43% +0.5 0.98 ± 13% perf-profile.self.cycles-pp.td_io_queue 0.08 ±142% +0.6 0.66 ± 43% perf-profile.self.cycles-pp.pagevec_lru_move_fn 0.62 ± 51% +0.6 1.22 ± 5% perf-profile.self.cycles-pp.page_vma_mapped_walk 0.62 ± 44% +0.7 1.31 ± 7% perf-profile.self.cycles-pp.sync_regs 0.66 ± 42% +0.8 1.43 ± 9% perf-profile.self.cycles-pp.get_io_u 0.10 ±141% +0.8 0.88 ± 43% perf-profile.self.cycles-pp.find_lock_entries 0.64 ± 41% +0.9 1.52 ± 8% perf-profile.self.cycles-pp.io_completed 1.09 ± 42% +1.0 2.07 ± 6% perf-profile.self.cycles-pp.fio_gettime 0.99 ± 41% +1.2 2.16 ± 8% perf-profile.self.cycles-pp.native_irq_return_iret *************************************************************************************************** lkp-csl-2sp7: 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 512G memory ========================================================================================= bs/compiler/cpufreq_governor/disk/fs/ioengine/kconfig/nr_task/rootfs/runtime/rw/tbox_group/test_size/testcase/time_based/ucode: 2M/gcc-11/performance/2pmem/ext4/libaio/x86_64-rhel-8.3/50%/debian-11.1-x86_64-20220510.cgz/200s/rw/lkp-csl-2sp7/200G/fio-basic/tb/0x500320a commit: 11192d9c12 ("memcg: flush stats only if updated") fd25a9e0e2 ("memcg: unify memcg stat flushing") 11192d9c124d58d6 fd25a9e0e23b995fd0ba5e2f00a ---------------- --------------------------- %stddev %change %stddev \ | \ 0.39 ± 88% +7.0 7.36 ± 29% fio.latency_100ms% 0.01 +0.0 0.02 ± 19% fio.latency_10ms% 0.02 ± 22% +0.0 0.02 ± 28% fio.latency_20ms% 35.65 ± 6% +7.2 42.90 ± 11% fio.latency_250ms% 0.01 +0.0 0.01 ± 9% fio.latency_4ms% 59.44 ± 4% -15.7 43.78 ± 18% fio.latency_500ms% 0.13 ± 46% +0.7 0.84 ± 30% fio.latency_50ms% 4734 ± 2% +14.0% 5398 ± 2% fio.read_bw_MBps 2.916e+08 ± 2% -13.8% 2.512e+08 ± 3% fio.read_clat_mean_us 98714844 ± 7% +32.8% 1.311e+08 ± 13% fio.read_clat_stddev 2367 ± 2% +14.0% 2699 ± 2% fio.read_iops 1290688 ± 2% +40.6% 1814909 ± 9% fio.read_slat_mean_us 755337 ± 5% +128.4% 1724856 ± 7% fio.read_slat_stddev 9.728e+08 ± 2% +15.0% 1.119e+09 ± 2% fio.time.file_system_inputs 1.942e+09 ± 2% +14.1% 2.217e+09 ± 2% fio.time.file_system_outputs 22091 ± 6% -47.7% 11558 ± 12% fio.time.involuntary_context_switches 3556 ± 7% -56.4% 1549 ± 11% fio.time.percent_of_cpu_this_job_got 7057 ± 7% -58.3% 2943 ± 11% fio.time.system_time 121.36 ± 4% +56.0% 189.33 ± 13% fio.time.user_time 948915 ± 2% +14.2% 1083277 ± 2% fio.workload 4729 ± 2% +14.0% 5391 ± 2% fio.write_bw_MBps 2.917e+08 ± 2% -13.2% 2.531e+08 ± 3% fio.write_clat_mean_us 99135535 ± 7% +32.5% 1.313e+08 ± 14% fio.write_clat_stddev 2364 ± 2% +14.0% 2695 ± 2% fio.write_iops 17604710 ± 2% -17.6% 14505732 ± 4% fio.write_slat_mean_us 9434086 ± 9% +61.3% 15215638 fio.write_slat_stddev 1.186e+10 ± 4% +33.7% 1.585e+10 ± 2% cpuidle..time 25720587 ± 3% +33.6% 34354696 ± 3% cpuidle..usage 12.02 ± 23% +165.9% 31.95 ± 6% iostat.cpu.iowait 37.83 ± 7% -54.1% 17.37 ± 10% iostat.cpu.system 0.66 ± 4% +48.9% 0.98 ± 13% iostat.cpu.user 67142307 ± 4% +16.5% 78254209 ± 6% numa-numastat.node0.local_node 67142230 ± 4% +16.6% 78263094 ± 6% numa-numastat.node0.numa_hit 65625 ±101% -95.1% 3227 ±221% numa-numastat.node0.numa_miss 65750 ±101% -95.1% 3227 ±221% numa-numastat.node1.numa_foreign 65497 ± 6% -35.7% 42093 ± 17% meminfo.Active 41592 ± 9% -55.0% 18731 ± 39% meminfo.Active(anon) 42484071 +17.8% 50038961 meminfo.Dirty 55262 ± 11% -16.9% 45947 meminfo.Mapped 65949 ± 15% -49.1% 33545 ± 22% meminfo.Shmem 12.13 ± 23% +20.1 32.26 ± 6% mpstat.cpu.all.iowait% 0.65 ± 2% +0.2 0.83 ± 5% mpstat.cpu.all.irq% 0.04 ± 4% +0.0 0.06 ± 4% mpstat.cpu.all.soft% 37.49 ± 7% -20.9 16.63 ± 10% mpstat.cpu.all.sys% 0.66 ± 4% +0.3 0.99 ± 13% mpstat.cpu.all.usr% 10357776 ± 5% +21.8% 12610998 ± 7% numa-meminfo.node0.Dirty 40985 ± 7% -56.9% 17672 ± 42% numa-meminfo.node1.Active 40225 ± 8% -57.6% 17072 ± 44% numa-meminfo.node1.Active(anon) 32140823 ± 3% +16.5% 37431157 numa-meminfo.node1.Dirty 55579 ± 16% -52.5% 26413 ± 27% numa-meminfo.node1.Shmem 11.50 ± 24% +172.5% 31.33 ± 6% vmstat.cpu.wa 2365172 ± 2% +15.0% 2720925 ± 2% vmstat.io.bi 4547031 ± 2% +14.7% 5217062 ± 2% vmstat.io.bo 11.17 ± 20% +177.6% 31.00 ± 6% vmstat.procs.b 36.67 ± 6% -53.6% 17.00 ± 11% vmstat.procs.r 177032 +10.0% 194738 vmstat.system.in 1110 ± 7% -50.8% 546.33 ± 9% turbostat.Avg_MHz 39.76 ± 7% -20.2 19.59 ± 9% turbostat.Busy% 15959953 ± 17% +43.8% 22954567 ± 11% turbostat.C1E 60.04 ± 4% +33.7% 80.26 ± 2% turbostat.CPU%c1 0.06 ± 7% +74.4% 0.11 ± 6% turbostat.IPC 66.17 ± 2% -7.8% 61.00 turbostat.PkgTmp 246.81 ± 2% -13.0% 214.63 ± 2% turbostat.PkgWatt 50.88 +2.7% 52.24 turbostat.RAMWatt 63017097 ± 4% +16.6% 73450071 ± 6% numa-vmstat.node0.nr_dirtied 2589553 ± 5% +21.8% 3153493 ± 7% numa-vmstat.node0.nr_dirty 2594037 ± 5% +21.7% 3157643 ± 7% numa-vmstat.node0.nr_zone_write_pending 67142279 ± 4% +16.6% 78263002 ± 6% numa-vmstat.node0.numa_hit 67142356 ± 4% +16.5% 78254117 ± 6% numa-vmstat.node0.numa_local 65625 ±101% -95.1% 3227 ±221% numa-vmstat.node0.numa_miss 10020 ± 8% -57.4% 4266 ± 44% numa-vmstat.node1.nr_active_anon 1.798e+08 ± 2% +13.3% 2.037e+08 ± 2% numa-vmstat.node1.nr_dirtied 8033999 ± 3% +16.5% 9358569 numa-vmstat.node1.nr_dirty 13850 ± 16% -52.2% 6617 ± 27% numa-vmstat.node1.nr_shmem 1.718e+08 ± 2% +14.0% 1.958e+08 ± 2% numa-vmstat.node1.nr_written 10020 ± 8% -57.4% 4265 ± 44% numa-vmstat.node1.nr_zone_active_anon 8043027 ± 3% +16.5% 9368655 numa-vmstat.node1.nr_zone_write_pending 65750 ±101% -95.1% 3227 ±221% numa-vmstat.node1.numa_foreign 10398 ± 9% -55.0% 4682 ± 39% proc-vmstat.nr_active_anon 2.428e+08 ± 2% +14.1% 2.771e+08 ± 2% proc-vmstat.nr_dirtied 10620936 +17.8% 12509591 proc-vmstat.nr_dirty 13815 ± 11% -16.9% 11486 proc-vmstat.nr_mapped 16487 ± 15% -49.1% 8386 ± 22% proc-vmstat.nr_shmem 153539 +3.2% 158400 proc-vmstat.nr_slab_unreclaimable 2.326e+08 ± 2% +15.0% 2.675e+08 ± 2% proc-vmstat.nr_written 10398 ± 9% -55.0% 4682 ± 39% proc-vmstat.nr_zone_active_anon 10634810 +17.8% 12523439 proc-vmstat.nr_zone_write_pending 34272 ± 9% +14.0% 39081 ± 9% proc-vmstat.numa_hint_faults_local 1.853e+08 ± 3% +10.8% 2.054e+08 ± 6% proc-vmstat.numa_hit 22095 ± 2% -21.4% 17362 ± 7% proc-vmstat.numa_huge_pte_updates 1.854e+08 ± 3% +10.9% 2.056e+08 ± 6% proc-vmstat.numa_local 11378151 ± 2% -21.5% 8937282 ± 7% proc-vmstat.numa_pte_updates 72660 ± 42% -62.3% 27417 ± 14% proc-vmstat.pgactivate 2.505e+08 ± 2% +14.0% 2.857e+08 ± 2% proc-vmstat.pgalloc_normal 714616 -3.8% 687146 proc-vmstat.pgfault 2.265e+08 ± 3% +15.6% 2.62e+08 ± 3% proc-vmstat.pgfree 4.864e+08 ± 2% +15.0% 5.596e+08 ± 2% proc-vmstat.pgpgin 9.305e+08 ± 2% +15.0% 1.07e+09 ± 2% proc-vmstat.pgpgout 0.41 ± 17% -58.0% 0.17 ± 16% sched_debug.cfs_rq:/.h_nr_running.avg 0.42 ± 5% -11.8% 0.37 ± 5% sched_debug.cfs_rq:/.h_nr_running.stddev 387722 ± 19% -61.8% 148125 ± 22% sched_debug.cfs_rq:/.load.avg 365.68 ± 12% -60.0% 146.17 ± 24% sched_debug.cfs_rq:/.load_avg.avg 20212 ± 10% -26.3% 14891 ± 5% sched_debug.cfs_rq:/.min_vruntime.stddev 0.41 ± 17% -58.0% 0.17 ± 16% sched_debug.cfs_rq:/.nr_running.avg 0.42 ± 5% -11.8% 0.37 ± 5% sched_debug.cfs_rq:/.nr_running.stddev 432.73 ± 10% -52.9% 203.67 ± 6% sched_debug.cfs_rq:/.runnable_avg.avg 352.26 ± 6% -29.4% 248.64 ± 7% sched_debug.cfs_rq:/.runnable_avg.stddev -115117 -22.1% -89637 sched_debug.cfs_rq:/.spread0.min 20214 ± 10% -26.3% 14892 ± 5% sched_debug.cfs_rq:/.spread0.stddev 432.38 ± 11% -53.0% 203.20 ± 6% sched_debug.cfs_rq:/.util_avg.avg 352.08 ± 6% -29.6% 248.00 ± 7% sched_debug.cfs_rq:/.util_avg.stddev 253.21 ± 15% -81.2% 47.55 ± 35% sched_debug.cfs_rq:/.util_est_enqueued.avg 256.67 ± 10% -51.8% 123.79 ± 19% sched_debug.cfs_rq:/.util_est_enqueued.stddev 2.73 ± 4% -12.3% 2.40 ± 3% sched_debug.cpu.clock.stddev 1785 ± 12% -63.7% 647.66 ± 19% sched_debug.cpu.curr->pid.avg 2199 ± 7% -22.7% 1700 ± 8% sched_debug.cpu.curr->pid.stddev 0.34 ± 11% -59.9% 0.14 ± 15% sched_debug.cpu.nr_running.avg 0.42 ± 6% -20.1% 0.34 ± 6% sched_debug.cpu.nr_running.stddev 1585 ± 15% +36.7% 2166 ± 12% sched_debug.cpu.nr_switches.min 23.07 ± 3% +25.6% 28.98 perf-stat.i.MPKI 4.901e+09 ± 4% -24.1% 3.718e+09 ± 2% perf-stat.i.branch-instructions 0.28 ± 3% +0.1 0.35 perf-stat.i.branch-miss-rate% 82.75 +1.8 84.52 perf-stat.i.cache-miss-rate% 4.468e+08 ± 2% +8.9% 4.865e+08 ± 2% perf-stat.i.cache-misses 5.38e+08 ± 2% +7.3% 5.774e+08 ± 3% perf-stat.i.cache-references 4.19 ± 4% -41.7% 2.44 ± 8% perf-stat.i.cpi 1.063e+11 ± 7% -51.6% 5.147e+10 ± 10% perf-stat.i.cpu-cycles 252.03 ± 6% -52.7% 119.14 ± 5% perf-stat.i.cycles-between-cache-misses 0.02 ± 9% +0.0 0.03 ± 17% perf-stat.i.dTLB-load-miss-rate% 6.43e+09 ± 4% -15.1% 5.46e+09 ± 2% perf-stat.i.dTLB-loads 0.01 ± 6% +0.0 0.01 ± 12% perf-stat.i.dTLB-store-miss-rate% 197331 ± 7% +38.3% 272925 ± 14% perf-stat.i.dTLB-store-misses 2.947e+09 ± 2% +12.6% 3.319e+09 ± 2% perf-stat.i.dTLB-stores 57.78 -1.8 56.01 perf-stat.i.iTLB-load-miss-rate% 1859795 ± 4% +9.6% 2038933 ± 3% perf-stat.i.iTLB-load-misses 1306943 +16.1% 1517987 perf-stat.i.iTLB-loads 2.433e+10 ± 4% -16.7% 2.027e+10 ± 2% perf-stat.i.instructions 13333 ± 2% -23.5% 10195 perf-stat.i.instructions-per-iTLB-miss 0.27 ± 6% +75.7% 0.47 ± 7% perf-stat.i.ipc 1.11 ± 7% -51.6% 0.54 ± 10% perf-stat.i.metric.GHz 1654 ± 2% +6.7% 1765 ± 2% perf-stat.i.metric.K/sec 154.33 ± 3% -11.7% 136.24 ± 2% perf-stat.i.metric.M/sec 2892 ± 2% -4.9% 2749 perf-stat.i.minor-faults 51092281 ± 3% +10.9% 56656127 ± 6% perf-stat.i.node-loads 56739346 ± 2% +15.3% 65443819 ± 3% perf-stat.i.node-stores 2906 ± 2% -4.9% 2764 perf-stat.i.page-faults 22.11 ± 2% +28.8% 28.48 perf-stat.overall.MPKI 0.26 ± 2% +0.1 0.34 perf-stat.overall.branch-miss-rate% 83.02 +1.2 84.20 perf-stat.overall.cache-miss-rate% 4.37 ± 3% -41.6% 2.55 ± 8% perf-stat.overall.cpi 238.26 ± 5% -55.3% 106.40 ± 7% perf-stat.overall.cycles-between-cache-misses 0.02 ± 10% +0.0 0.03 ± 18% perf-stat.overall.dTLB-load-miss-rate% 0.01 ± 6% +0.0 0.01 ± 12% perf-stat.overall.dTLB-store-miss-rate% 13066 -24.2% 9902 perf-stat.overall.instructions-per-iTLB-miss 0.23 ± 3% +71.9% 0.39 ± 7% perf-stat.overall.ipc 5198153 ± 2% -26.9% 3798948 perf-stat.overall.path-length 4.884e+09 ± 4% -24.1% 3.704e+09 ± 2% perf-stat.ps.branch-instructions 4.447e+08 ± 2% +8.9% 4.844e+08 ± 2% perf-stat.ps.cache-misses 5.356e+08 ± 2% +7.4% 5.754e+08 ± 3% perf-stat.ps.cache-references 1.06e+11 ± 7% -51.3% 5.162e+10 ± 10% perf-stat.ps.cpu-cycles 1154804 ± 11% +33.7% 1543751 ± 20% perf-stat.ps.dTLB-load-misses 6.409e+09 ± 4% -15.1% 5.441e+09 ± 2% perf-stat.ps.dTLB-loads 196129 ± 7% +38.3% 271181 ± 14% perf-stat.ps.dTLB-store-misses 2.937e+09 ± 2% +12.6% 3.306e+09 ± 2% perf-stat.ps.dTLB-stores 1856169 ± 4% +9.9% 2040184 ± 2% perf-stat.ps.iTLB-load-misses 1299290 +16.1% 1509095 perf-stat.ps.iTLB-loads 2.425e+10 ± 4% -16.7% 2.02e+10 ± 2% perf-stat.ps.instructions 2869 ± 2% -4.8% 2731 perf-stat.ps.minor-faults 50805618 ± 3% +11.0% 56380675 ± 6% perf-stat.ps.node-loads 56364613 ± 2% +15.5% 65091234 ± 3% perf-stat.ps.node-stores 2883 ± 2% -4.8% 2745 perf-stat.ps.page-faults 4.935e+12 ± 4% -16.6% 4.115e+12 ± 2% perf-stat.total.instructions 55.92 ± 8% -55.9 0.00 perf-profile.calltrace.cycles-pp.cgroup_rstat_flush_irqsafe.mem_cgroup_wb_stats.balance_dirty_pages.balance_dirty_pages_ratelimited.generic_perform_write 56.10 ± 8% -55.2 0.88 ± 11% perf-profile.calltrace.cycles-pp.mem_cgroup_wb_stats.balance_dirty_pages.balance_dirty_pages_ratelimited.generic_perform_write.ext4_buffered_write_iter 56.13 ± 8% -55.1 1.03 ± 9% perf-profile.calltrace.cycles-pp.balance_dirty_pages.balance_dirty_pages_ratelimited.generic_perform_write.ext4_buffered_write_iter.aio_write 54.48 ± 8% -54.5 0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.cgroup_rstat_flush_irqsafe.mem_cgroup_wb_stats.balance_dirty_pages.balance_dirty_pages_ratelimited 54.47 ± 8% -54.5 0.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.cgroup_rstat_flush_irqsafe.mem_cgroup_wb_stats.balance_dirty_pages 56.29 ± 8% -54.3 2.02 ± 39% perf-profile.calltrace.cycles-pp.balance_dirty_pages_ratelimited.generic_perform_write.ext4_buffered_write_iter.aio_write.io_submit_one 62.95 ± 8% -39.1 23.82 ± 3% perf-profile.calltrace.cycles-pp.generic_perform_write.ext4_buffered_write_iter.aio_write.io_submit_one.__x64_sys_io_submit 62.97 ± 8% -39.1 23.88 ± 3% perf-profile.calltrace.cycles-pp.ext4_buffered_write_iter.aio_write.io_submit_one.__x64_sys_io_submit.do_syscall_64 62.97 ± 8% -39.1 23.89 ± 3% perf-profile.calltrace.cycles-pp.aio_write.io_submit_one.__x64_sys_io_submit.do_syscall_64.entry_SYSCALL_64_after_hwframe 68.26 ± 8% -24.6 43.68 ± 5% perf-profile.calltrace.cycles-pp.io_submit_one.__x64_sys_io_submit.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall 68.26 ± 8% -24.6 43.69 ± 5% perf-profile.calltrace.cycles-pp.__x64_sys_io_submit.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall 68.26 ± 8% -24.5 43.72 ± 5% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.syscall 68.26 ± 8% -24.5 43.72 ± 5% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall 68.27 ± 8% -24.5 43.74 ± 5% perf-profile.calltrace.cycles-pp.syscall 0.00 +0.8 0.83 ± 10% perf-profile.calltrace.cycles-pp.cgroup_rstat_flush_locked.cgroup_rstat_flush_irqsafe.mem_cgroup_flush_stats.mem_cgroup_wb_stats.balance_dirty_pages 0.53 ± 46% +0.8 1.36 ± 57% perf-profile.calltrace.cycles-pp.account_page_dirtied.__set_page_dirty.mark_buffer_dirty.__block_commit_write.generic_write_end 0.00 +0.8 0.83 ± 10% perf-profile.calltrace.cycles-pp.cgroup_rstat_flush_irqsafe.mem_cgroup_flush_stats.mem_cgroup_wb_stats.balance_dirty_pages.balance_dirty_pages_ratelimited 0.00 +0.8 0.85 ± 9% perf-profile.calltrace.cycles-pp.mem_cgroup_flush_stats.mem_cgroup_wb_stats.balance_dirty_pages.balance_dirty_pages_ratelimited.generic_perform_write 0.00 +0.9 0.86 ± 32% perf-profile.calltrace.cycles-pp.__test_set_page_writeback.ext4_bio_write_page.mpage_submit_page.mpage_process_page_bufs.mpage_prepare_extent_to_map 0.00 +1.0 0.96 ± 22% perf-profile.calltrace.cycles-pp.get_page_from_freelist.__alloc_pages.pagecache_get_page.grab_cache_page_write_begin.ext4_da_write_begin 0.71 ± 13% +1.0 1.69 ± 42% perf-profile.calltrace.cycles-pp.__set_page_dirty.mark_buffer_dirty.__block_commit_write.generic_write_end.generic_perform_write 0.51 ± 45% +1.0 1.52 ± 54% perf-profile.calltrace.cycles-pp.__add_to_page_cache_locked.add_to_page_cache_lru.pagecache_get_page.grab_cache_page_write_begin.ext4_da_write_begin 0.00 +1.0 1.00 ± 14% perf-profile.calltrace.cycles-pp.kmem_cache_alloc.alloc_buffer_head.alloc_page_buffers.create_empty_buffers.ext4_block_write_begin 0.00 +1.0 1.03 ± 22% perf-profile.calltrace.cycles-pp.__alloc_pages.pagecache_get_page.grab_cache_page_write_begin.ext4_da_write_begin.generic_perform_write 0.00 +1.1 1.07 ± 15% perf-profile.calltrace.cycles-pp.alloc_buffer_head.alloc_page_buffers.create_empty_buffers.ext4_block_write_begin.ext4_da_write_begin 0.00 +1.1 1.11 ± 15% perf-profile.calltrace.cycles-pp.alloc_page_buffers.create_empty_buffers.ext4_block_write_begin.ext4_da_write_begin.generic_perform_write 0.00 +1.2 1.20 ± 72% perf-profile.calltrace.cycles-pp.lru_cache_add.add_to_page_cache_lru.pagecache_get_page.grab_cache_page_write_begin.ext4_da_write_begin 0.00 +1.2 1.24 ± 22% perf-profile.calltrace.cycles-pp.invalidate_inode_page.__invalidate_mapping_pages.generic_fadvise.ksys_fadvise64_64.__x64_sys_fadvise64 0.00 +1.4 1.36 ± 65% perf-profile.calltrace.cycles-pp.__add_to_page_cache_locked.add_to_page_cache_lru.page_cache_ra_unbounded.filemap_get_pages.filemap_read 0.65 ± 10% +1.4 2.02 ± 19% perf-profile.calltrace.cycles-pp.ext4_block_write_begin.ext4_da_write_begin.generic_perform_write.ext4_buffered_write_iter.aio_write 0.09 ±223% +1.4 1.47 ± 16% perf-profile.calltrace.cycles-pp.create_empty_buffers.ext4_block_write_begin.ext4_da_write_begin.generic_perform_write.ext4_buffered_write_iter 0.00 +1.4 1.39 ± 75% perf-profile.calltrace.cycles-pp.__remove_mapping.remove_mapping.__invalidate_mapping_pages.generic_fadvise.ksys_fadvise64_64 0.00 +1.4 1.40 ± 75% perf-profile.calltrace.cycles-pp.remove_mapping.__invalidate_mapping_pages.generic_fadvise.ksys_fadvise64_64.__x64_sys_fadvise64 0.08 ±223% +1.4 1.52 ± 17% perf-profile.calltrace.cycles-pp.iov_iter_fault_in_readable.generic_perform_write.ext4_buffered_write_iter.aio_write.io_submit_one 0.00 +1.5 1.46 ± 17% perf-profile.calltrace.cycles-pp.__get_user_nocheck_1.iov_iter_fault_in_readable.generic_perform_write.ext4_buffered_write_iter.aio_write 0.09 ±223% +1.5 1.63 ± 16% perf-profile.calltrace.cycles-pp.mark_page_accessed.filemap_read.aio_read.io_submit_one.__x64_sys_io_submit 0.97 ± 11% +1.7 2.68 ± 24% perf-profile.calltrace.cycles-pp.mark_buffer_dirty.__block_commit_write.generic_write_end.generic_perform_write.ext4_buffered_write_iter 0.47 ± 45% +1.7 2.22 ± 67% perf-profile.calltrace.cycles-pp.add_to_page_cache_lru.page_cache_ra_unbounded.filemap_get_pages.filemap_read.aio_read 0.87 ± 13% +1.9 2.73 ± 60% perf-profile.calltrace.cycles-pp.add_to_page_cache_lru.pagecache_get_page.grab_cache_page_write_begin.ext4_da_write_begin.generic_perform_write 0.10 ±223% +2.0 2.12 ± 51% perf-profile.calltrace.cycles-pp.test_clear_page_writeback.end_page_writeback.ext4_finish_bio.ext4_end_bio.pmem_submit_bio 0.10 ±223% +2.2 2.34 ± 45% perf-profile.calltrace.cycles-pp.end_page_writeback.ext4_finish_bio.ext4_end_bio.pmem_submit_bio.__submit_bio 0.98 ± 10% +2.5 3.51 ± 17% perf-profile.calltrace.cycles-pp.get_io_u 1.40 ± 11% +2.7 4.07 ± 7% perf-profile.calltrace.cycles-pp.__block_commit_write.generic_write_end.generic_perform_write.ext4_buffered_write_iter.aio_write 0.64 ± 19% +2.8 3.41 ± 24% perf-profile.calltrace.cycles-pp.ext4_end_bio.pmem_submit_bio.__submit_bio.__submit_bio_noacct.ext4_bio_write_page 1.46 ± 10% +2.8 4.24 ± 6% perf-profile.calltrace.cycles-pp.generic_write_end.generic_perform_write.ext4_buffered_write_iter.aio_write.io_submit_one 0.56 ± 48% +2.8 3.39 ± 25% perf-profile.calltrace.cycles-pp.ext4_finish_bio.ext4_end_bio.pmem_submit_bio.__submit_bio.__submit_bio_noacct 1.33 ± 12% +3.1 4.38 ± 30% perf-profile.calltrace.cycles-pp.pagecache_get_page.grab_cache_page_write_begin.ext4_da_write_begin.generic_perform_write.ext4_buffered_write_iter 1.34 ± 12% +3.1 4.42 ± 30% perf-profile.calltrace.cycles-pp.grab_cache_page_write_begin.ext4_da_write_begin.generic_perform_write.ext4_buffered_write_iter.aio_write 0.00 +3.1 3.12 ±111% perf-profile.calltrace.cycles-pp.release_pages.__pagevec_release.__invalidate_mapping_pages.generic_fadvise.ksys_fadvise64_64 0.00 +3.1 3.13 ±111% perf-profile.calltrace.cycles-pp.__pagevec_release.__invalidate_mapping_pages.generic_fadvise.ksys_fadvise64_64.__x64_sys_fadvise64 1.26 ± 9% +3.9 5.14 ± 12% perf-profile.calltrace.cycles-pp.copy_mc_fragile.pmem_do_read.pmem_submit_bio.__submit_bio.__submit_bio_noacct 1.27 ± 8% +3.9 5.18 ± 12% perf-profile.calltrace.cycles-pp.pmem_do_read.pmem_submit_bio.__submit_bio.__submit_bio_noacct.ext4_mpage_readpages 1.30 ± 9% +4.0 5.34 ± 12% perf-profile.calltrace.cycles-pp.pmem_submit_bio.__submit_bio.__submit_bio_noacct.ext4_mpage_readpages.read_pages 1.31 ± 8% +4.0 5.35 ± 12% perf-profile.calltrace.cycles-pp.__submit_bio_noacct.ext4_mpage_readpages.read_pages.page_cache_ra_unbounded.filemap_get_pages 1.31 ± 8% +4.0 5.35 ± 12% perf-profile.calltrace.cycles-pp.__submit_bio.__submit_bio_noacct.ext4_mpage_readpages.read_pages.page_cache_ra_unbounded 1.37 ± 9% +4.2 5.59 ± 13% perf-profile.calltrace.cycles-pp.ext4_mpage_readpages.read_pages.page_cache_ra_unbounded.filemap_get_pages.filemap_read 1.37 ± 9% +4.2 5.60 ± 13% perf-profile.calltrace.cycles-pp.read_pages.page_cache_ra_unbounded.filemap_get_pages.filemap_read.aio_read 2.04 ± 11% +4.5 6.57 ± 14% perf-profile.calltrace.cycles-pp.ext4_da_write_begin.generic_perform_write.ext4_buffered_write_iter.aio_write.io_submit_one 0.00 +5.7 5.73 ± 48% perf-profile.calltrace.cycles-pp.mpage_process_page_bufs.mpage_prepare_extent_to_map.ext4_writepages.do_writepages.filemap_fdatawrite_wbc 0.00 +6.1 6.06 ± 45% perf-profile.calltrace.cycles-pp.mpage_prepare_extent_to_map.ext4_writepages.do_writepages.filemap_fdatawrite_wbc.__filemap_fdatawrite_range 0.00 +6.1 6.07 ± 45% perf-profile.calltrace.cycles-pp.do_writepages.filemap_fdatawrite_wbc.__filemap_fdatawrite_range.generic_fadvise.ksys_fadvise64_64 0.00 +6.1 6.07 ± 45% perf-profile.calltrace.cycles-pp.ext4_writepages.do_writepages.filemap_fdatawrite_wbc.__filemap_fdatawrite_range.generic_fadvise 0.00 +6.1 6.07 ± 45% perf-profile.calltrace.cycles-pp.__filemap_fdatawrite_range.generic_fadvise.ksys_fadvise64_64.__x64_sys_fadvise64.do_syscall_64 0.00 +6.1 6.07 ± 45% perf-profile.calltrace.cycles-pp.filemap_fdatawrite_wbc.__filemap_fdatawrite_range.generic_fadvise.ksys_fadvise64_64.__x64_sys_fadvise64 2.48 ± 11% +6.2 8.65 ± 22% perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string.copyout.copy_page_to_iter.filemap_read.aio_read 2.49 ± 11% +6.2 8.70 ± 22% perf-profile.calltrace.cycles-pp.copyout.copy_page_to_iter.filemap_read.aio_read.io_submit_one 0.00 +6.3 6.25 ± 75% perf-profile.calltrace.cycles-pp.__invalidate_mapping_pages.generic_fadvise.ksys_fadvise64_64.__x64_sys_fadvise64.do_syscall_64 2.54 ± 10% +6.3 8.86 ± 22% perf-profile.calltrace.cycles-pp.copy_page_to_iter.filemap_read.aio_read.io_submit_one.__x64_sys_io_submit 2.09 ± 9% +6.4 8.54 ± 11% perf-profile.calltrace.cycles-pp.page_cache_ra_unbounded.filemap_get_pages.filemap_read.aio_read.io_submit_one 2.60 ± 10% +6.5 9.08 ± 21% perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string.copyin.copy_page_from_iter_atomic.generic_perform_write.ext4_buffered_write_iter 2.62 ± 10% +6.5 9.14 ± 21% perf-profile.calltrace.cycles-pp.copyin.copy_page_from_iter_atomic.generic_perform_write.ext4_buffered_write_iter.aio_write 2.66 ± 10% +6.6 9.25 ± 21% perf-profile.calltrace.cycles-pp.copy_page_from_iter_atomic.generic_perform_write.ext4_buffered_write_iter.aio_write.io_submit_one 2.76 ± 26% +6.7 9.43 ± 32% perf-profile.calltrace.cycles-pp.mpage_process_page_bufs.mpage_prepare_extent_to_map.ext4_writepages.do_writepages.__writeback_single_inode 2.20 ± 9% +6.8 8.95 ± 10% perf-profile.calltrace.cycles-pp.filemap_get_pages.filemap_read.aio_read.io_submit_one.__x64_sys_io_submit 1.39 ± 33% +6.9 8.27 ± 22% perf-profile.calltrace.cycles-pp.__memcpy_flushcache.write_pmem.pmem_do_write.pmem_submit_bio.__submit_bio 1.40 ± 33% +6.9 8.32 ± 22% perf-profile.calltrace.cycles-pp.write_pmem.pmem_do_write.pmem_submit_bio.__submit_bio.__submit_bio_noacct 1.41 ± 33% +7.0 8.36 ± 22% perf-profile.calltrace.cycles-pp.pmem_do_write.pmem_submit_bio.__submit_bio.__submit_bio_noacct.ext4_bio_write_page 2.93 ± 27% +7.1 10.02 ± 32% perf-profile.calltrace.cycles-pp.mpage_prepare_extent_to_map.ext4_writepages.do_writepages.__writeback_single_inode.writeback_sb_inodes 2.95 ± 27% +7.1 10.06 ± 32% perf-profile.calltrace.cycles-pp.do_writepages.__writeback_single_inode.writeback_sb_inodes.__writeback_inodes_wb.wb_writeback 2.95 ± 27% +7.1 10.06 ± 32% perf-profile.calltrace.cycles-pp.ext4_writepages.do_writepages.__writeback_single_inode.writeback_sb_inodes.__writeback_inodes_wb 2.95 ± 27% +7.1 10.06 ± 32% perf-profile.calltrace.cycles-pp.wb_workfn.process_one_work.worker_thread.kthread.ret_from_fork 2.95 ± 27% +7.1 10.06 ± 32% perf-profile.calltrace.cycles-pp.wb_do_writeback.wb_workfn.process_one_work.worker_thread.kthread 2.95 ± 27% +7.1 10.06 ± 32% perf-profile.calltrace.cycles-pp.wb_writeback.wb_do_writeback.wb_workfn.process_one_work.worker_thread 2.95 ± 27% +7.1 10.06 ± 32% perf-profile.calltrace.cycles-pp.__writeback_inodes_wb.wb_writeback.wb_do_writeback.wb_workfn.process_one_work 2.95 ± 27% +7.1 10.06 ± 32% perf-profile.calltrace.cycles-pp.writeback_sb_inodes.__writeback_inodes_wb.wb_writeback.wb_do_writeback.wb_workfn 2.95 ± 27% +7.1 10.06 ± 32% perf-profile.calltrace.cycles-pp.__writeback_single_inode.writeback_sb_inodes.__writeback_inodes_wb.wb_writeback.wb_do_writeback 2.96 ± 27% +7.1 10.08 ± 32% perf-profile.calltrace.cycles-pp.process_one_work.worker_thread.kthread.ret_from_fork 2.96 ± 27% +7.1 10.09 ± 32% perf-profile.calltrace.cycles-pp.worker_thread.kthread.ret_from_fork 3.00 ± 25% +7.1 10.13 ± 32% perf-profile.calltrace.cycles-pp.ret_from_fork 3.00 ± 25% +7.1 10.13 ± 32% perf-profile.calltrace.cycles-pp.kthread.ret_from_fork 2.07 ± 29% +9.8 11.83 ± 13% perf-profile.calltrace.cycles-pp.pmem_submit_bio.__submit_bio.__submit_bio_noacct.ext4_bio_write_page.mpage_submit_page 2.07 ± 29% +9.8 11.84 ± 13% perf-profile.calltrace.cycles-pp.__submit_bio.__submit_bio_noacct.ext4_bio_write_page.mpage_submit_page.mpage_process_page_bufs 2.07 ± 29% +9.8 11.84 ± 13% perf-profile.calltrace.cycles-pp.__submit_bio_noacct.ext4_bio_write_page.mpage_submit_page.mpage_process_page_bufs.mpage_prepare_extent_to_map 2.45 ± 27% +11.3 13.71 ± 11% perf-profile.calltrace.cycles-pp.ext4_bio_write_page.mpage_submit_page.mpage_process_page_bufs.mpage_prepare_extent_to_map.ext4_writepages 2.66 ± 27% +12.0 14.66 ± 10% perf-profile.calltrace.cycles-pp.mpage_submit_page.mpage_process_page_bufs.mpage_prepare_extent_to_map.ext4_writepages.do_writepages 0.00 +12.3 12.33 ± 58% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.posix_fadvise 0.00 +12.3 12.33 ± 58% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.posix_fadvise 0.00 +12.3 12.33 ± 58% perf-profile.calltrace.cycles-pp.__x64_sys_fadvise64.do_syscall_64.entry_SYSCALL_64_after_hwframe.posix_fadvise 0.00 +12.3 12.33 ± 58% perf-profile.calltrace.cycles-pp.ksys_fadvise64_64.__x64_sys_fadvise64.do_syscall_64.entry_SYSCALL_64_after_hwframe.posix_fadvise 0.00 +12.3 12.33 ± 58% perf-profile.calltrace.cycles-pp.generic_fadvise.ksys_fadvise64_64.__x64_sys_fadvise64.do_syscall_64.entry_SYSCALL_64_after_hwframe 0.00 +12.3 12.33 ± 58% perf-profile.calltrace.cycles-pp.posix_fadvise 5.26 ± 10% +14.5 19.74 ± 9% perf-profile.calltrace.cycles-pp.filemap_read.aio_read.io_submit_one.__x64_sys_io_submit.do_syscall_64 5.27 ± 9% +14.5 19.76 ± 9% perf-profile.calltrace.cycles-pp.aio_read.io_submit_one.__x64_sys_io_submit.do_syscall_64.entry_SYSCALL_64_after_hwframe 56.10 ± 8% -55.2 0.88 ± 11% perf-profile.children.cycles-pp.mem_cgroup_wb_stats 56.13 ± 8% -55.1 1.03 ± 9% perf-profile.children.cycles-pp.balance_dirty_pages 55.92 ± 8% -55.1 0.83 ± 10% perf-profile.children.cycles-pp.cgroup_rstat_flush_irqsafe 56.29 ± 8% -54.3 2.02 ± 39% perf-profile.children.cycles-pp.balance_dirty_pages_ratelimited 54.78 ± 8% -51.7 3.06 ±134% perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath 55.13 ± 8% -51.2 3.93 ±100% perf-profile.children.cycles-pp._raw_spin_lock_irqsave 62.97 ± 8% -39.1 23.86 ± 3% perf-profile.children.cycles-pp.generic_perform_write 62.97 ± 8% -39.1 23.88 ± 3% perf-profile.children.cycles-pp.ext4_buffered_write_iter 62.97 ± 8% -39.1 23.89 ± 3% perf-profile.children.cycles-pp.aio_write 68.26 ± 8% -24.6 43.68 ± 5% perf-profile.children.cycles-pp.io_submit_one 68.26 ± 8% -24.6 43.69 ± 5% perf-profile.children.cycles-pp.__x64_sys_io_submit 68.27 ± 8% -24.5 43.75 ± 5% perf-profile.children.cycles-pp.syscall 68.43 ± 8% -12.2 56.19 ± 10% perf-profile.children.cycles-pp.do_syscall_64 68.43 ± 8% -12.2 56.19 ± 10% perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe 1.44 ± 9% -0.6 0.83 ± 10% perf-profile.children.cycles-pp.cgroup_rstat_flush_locked 0.00 +0.1 0.06 ± 9% perf-profile.children.cycles-pp.__wake_up_common 0.00 +0.1 0.06 ± 13% perf-profile.children.cycles-pp.wait_on_page_bit_common 0.00 +0.1 0.06 ± 19% perf-profile.children.cycles-pp.wake_up_page_bit 0.04 ± 71% +0.1 0.10 ± 10% perf-profile.children.cycles-pp.task_tick_fair 0.00 +0.1 0.06 ± 11% perf-profile.children.cycles-pp.schedule 0.00 +0.1 0.06 ± 19% perf-profile.children.cycles-pp.try_to_wake_up 0.00 +0.1 0.07 ± 14% perf-profile.children.cycles-pp.xas_init_marks 0.00 +0.1 0.07 ± 33% perf-profile.children.cycles-pp.obj_cgroup_charge 0.00 +0.1 0.08 ± 24% perf-profile.children.cycles-pp.ext4_da_write_end 0.00 +0.1 0.08 ± 19% perf-profile.children.cycles-pp.mem_cgroup_update_lru_size 0.00 +0.1 0.08 ± 16% perf-profile.children.cycles-pp.__list_add_valid 0.00 +0.1 0.08 ± 27% perf-profile.children.cycles-pp.xas_find_marked 0.00 +0.1 0.09 ± 23% perf-profile.children.cycles-pp.xas_create 0.00 +0.1 0.09 ± 23% perf-profile.children.cycles-pp.__cond_resched 0.04 ± 45% +0.1 0.14 ± 21% perf-profile.children.cycles-pp.__mark_inode_dirty 0.01 ±223% +0.1 0.10 ± 23% perf-profile.children.cycles-pp.node_page_state 0.00 +0.1 0.10 ± 9% perf-profile.children.cycles-pp.__schedule 0.00 +0.1 0.11 ± 25% perf-profile.children.cycles-pp.rcu_read_unlock_strict 0.00 +0.1 0.11 ± 24% perf-profile.children.cycles-pp.page_counter_try_charge 0.00 +0.1 0.11 ± 87% perf-profile.children.cycles-pp.unlock_page_memcg 0.00 +0.1 0.12 ± 18% perf-profile.children.cycles-pp.__read_end_io 0.00 +0.1 0.12 ± 25% perf-profile.children.cycles-pp.serial8250_console_putchar 0.00 +0.1 0.12 ± 25% perf-profile.children.cycles-pp.wait_for_xmitr 0.06 ± 13% +0.1 0.18 ± 21% perf-profile.children.cycles-pp.get_obj_cgroup_from_current 0.00 +0.1 0.12 ± 26% perf-profile.children.cycles-pp.serial8250_console_write 0.00 +0.1 0.12 ± 26% perf-profile.children.cycles-pp.uart_console_write 0.00 +0.1 0.12 ± 28% perf-profile.children.cycles-pp.irq_work_run 0.00 +0.1 0.12 ± 28% perf-profile.children.cycles-pp._printk 0.00 +0.1 0.12 ± 28% perf-profile.children.cycles-pp.vprintk_emit 0.00 +0.1 0.12 ± 28% perf-profile.children.cycles-pp.console_unlock 0.00 +0.1 0.12 ± 28% perf-profile.children.cycles-pp.call_console_drivers 0.00 +0.1 0.12 ± 29% perf-profile.children.cycles-pp.irq_work_run_list 0.00 +0.1 0.12 ± 29% perf-profile.children.cycles-pp.asm_sysvec_irq_work 0.00 +0.1 0.12 ± 29% perf-profile.children.cycles-pp.sysvec_irq_work 0.00 +0.1 0.12 ± 29% perf-profile.children.cycles-pp.__sysvec_irq_work 0.00 +0.1 0.12 ± 29% perf-profile.children.cycles-pp.irq_work_single 0.04 ± 71% +0.1 0.16 ± 27% perf-profile.children.cycles-pp.___might_sleep 0.04 ± 71% +0.1 0.16 ± 25% perf-profile.children.cycles-pp.xa_load 0.04 ± 45% +0.1 0.17 ± 28% perf-profile.children.cycles-pp.xa_get_order 0.04 ± 79% +0.1 0.17 ± 16% perf-profile.children.cycles-pp._raw_spin_lock_irq 0.06 ± 13% +0.1 0.20 ± 26% perf-profile.children.cycles-pp.ext4_es_lookup_extent 0.00 +0.1 0.14 ± 19% perf-profile.children.cycles-pp.__slab_free 0.06 ± 14% +0.1 0.21 ± 25% perf-profile.children.cycles-pp.__xa_set_mark 0.06 ± 8% +0.1 0.20 ± 16% perf-profile.children.cycles-pp.try_charge_memcg 0.06 ± 11% +0.2 0.21 ± 25% perf-profile.children.cycles-pp.xas_start 0.00 +0.2 0.16 ± 31% perf-profile.children.cycles-pp.mod_objcg_state 0.00 +0.2 0.16 ± 56% perf-profile.children.cycles-pp.lock_page_memcg 0.07 ± 9% +0.2 0.24 ± 26% perf-profile.children.cycles-pp.ext4_da_map_blocks 0.00 +0.2 0.17 ± 19% perf-profile.children.cycles-pp.xas_clear_mark 0.06 ± 11% +0.2 0.24 ± 26% perf-profile.children.cycles-pp.page_mapping 0.00 +0.2 0.19 ± 60% perf-profile.children.cycles-pp.page_counter_cancel 0.08 ± 14% +0.2 0.28 ± 25% perf-profile.children.cycles-pp.node_dirty_ok 0.01 ±223% +0.2 0.20 ± 22% perf-profile.children.cycles-pp.unlock_page 0.09 ± 10% +0.2 0.28 ± 30% perf-profile.children.cycles-pp.___slab_alloc 0.01 ±223% +0.2 0.21 ± 49% perf-profile.children.cycles-pp.__fprop_inc_percpu 0.03 ±223% +0.2 0.24 ± 36% perf-profile.children.cycles-pp.__irq_exit_rcu 0.11 ± 14% +0.2 0.32 ± 23% perf-profile.children.cycles-pp.memcg_slab_post_alloc_hook 0.00 +0.2 0.21 ± 20% perf-profile.children.cycles-pp.drop_buffers 0.04 ± 75% +0.2 0.26 ± 27% perf-profile.children.cycles-pp.__xa_clear_mark 0.00 +0.2 0.22 ± 29% perf-profile.children.cycles-pp.__softirqentry_text_start 0.07 ± 14% +0.2 0.29 ± 25% perf-profile.children.cycles-pp.scheduler_tick 0.00 +0.2 0.23 ±106% perf-profile.children.cycles-pp.mem_cgroup_wb_domain 0.00 +0.2 0.23 ± 80% perf-profile.children.cycles-pp.page_counter_uncharge 0.00 +0.3 0.26 ± 42% perf-profile.children.cycles-pp.memcg_slab_free_hook 0.10 ± 13% +0.3 0.40 ± 21% perf-profile.children.cycles-pp.filemap_get_read_batch 0.14 ± 12% +0.3 0.45 ± 27% perf-profile.children.cycles-pp.ext4_da_get_block_prep 0.02 ±223% +0.3 0.34 ± 40% perf-profile.children.cycles-pp.tick_nohz_get_sleep_length 0.00 +0.3 0.32 ± 16% perf-profile.children.cycles-pp.__free_one_page 0.00 +0.3 0.33 ± 97% perf-profile.children.cycles-pp.uncharge_batch 0.07 ± 17% +0.4 0.42 ± 17% perf-profile.children.cycles-pp.__list_del_entry_valid 0.06 ± 13% +0.4 0.42 ± 13% perf-profile.children.cycles-pp.xas_store 0.09 ± 12% +0.4 0.46 ± 13% perf-profile.children.cycles-pp.__mod_node_page_state 0.00 +0.4 0.38 ± 14% perf-profile.children.cycles-pp.poll_idle 0.00 +0.4 0.40 ± 22% perf-profile.children.cycles-pp.jbd2_journal_grab_journal_head 0.36 ± 14% +0.4 0.76 ± 68% perf-profile.children.cycles-pp.charge_memcg 0.11 ± 13% +0.4 0.51 ± 28% perf-profile.children.cycles-pp.update_process_times 0.00 +0.4 0.40 ± 17% perf-profile.children.cycles-pp.find_lock_entries 0.11 ± 14% +0.4 0.52 ± 28% perf-profile.children.cycles-pp.tick_sched_handle 0.08 ±223% +0.4 0.50 ± 39% perf-profile.children.cycles-pp.menu_select 0.13 ± 29% +0.4 0.55 ± 27% perf-profile.children.cycles-pp.find_get_pages_range_tag 0.00 +0.4 0.42 ± 22% perf-profile.children.cycles-pp.jbd2_journal_try_to_free_buffers 0.13 ± 29% +0.4 0.55 ± 27% perf-profile.children.cycles-pp.pagevec_lookup_range_tag 0.00 +0.4 0.43 ± 17% perf-profile.children.cycles-pp.free_pcppages_bulk 0.12 ± 13% +0.4 0.55 ± 28% perf-profile.children.cycles-pp.tick_sched_timer 0.11 ± 14% +0.4 0.54 ± 14% perf-profile.children.cycles-pp.__mod_lruvec_state 0.06 ± 21% +0.4 0.51 ±113% perf-profile.children.cycles-pp.get_mem_cgroup_from_mm 0.31 ± 11% +0.4 0.76 ± 15% perf-profile.children.cycles-pp.__pagevec_lru_add_fn 0.00 +0.5 0.47 ± 28% perf-profile.children.cycles-pp.free_buffer_head 0.00 +0.5 0.49 ± 28% perf-profile.children.cycles-pp.kmem_cache_free 0.00 +0.5 0.49 ±107% perf-profile.children.cycles-pp.__mem_cgroup_uncharge_list 0.13 ± 18% +0.5 0.67 ± 24% perf-profile.children.cycles-pp.percpu_counter_add_batch 0.01 ±223% +0.5 0.55 ± 16% perf-profile.children.cycles-pp.free_unref_page_list 0.16 ± 17% +0.7 0.83 ± 17% perf-profile.children.cycles-pp.rmqueue_bulk 0.22 ± 17% +0.7 0.92 ± 43% perf-profile.children.cycles-pp.clear_page_dirty_for_io 0.34 ± 10% +0.7 1.07 ± 15% perf-profile.children.cycles-pp.alloc_buffer_head 0.35 ± 10% +0.7 1.08 ± 15% perf-profile.children.cycles-pp.kmem_cache_alloc 0.01 ±223% +0.7 0.75 ± 24% perf-profile.children.cycles-pp.try_to_free_buffers 0.36 ± 9% +0.8 1.11 ± 15% perf-profile.children.cycles-pp.alloc_page_buffers 0.61 ± 14% +0.8 1.37 ± 57% perf-profile.children.cycles-pp.account_page_dirtied 0.25 ± 15% +0.8 1.09 ± 18% perf-profile.children.cycles-pp.rmqueue 0.00 +0.8 0.85 ± 9% perf-profile.children.cycles-pp.mem_cgroup_flush_stats 0.01 ±223% +0.9 0.87 ±121% perf-profile.children.cycles-pp.unaccount_page_cache_page 0.31 ± 15% +0.9 1.20 ± 24% perf-profile.children.cycles-pp.xas_load 0.28 ± 19% +0.9 1.20 ± 26% perf-profile.children.cycles-pp.__test_set_page_writeback 0.71 ± 13% +1.0 1.70 ± 41% perf-profile.children.cycles-pp.__set_page_dirty 0.48 ± 9% +1.0 1.48 ± 16% perf-profile.children.cycles-pp.create_empty_buffers 0.44 ± 12% +1.0 1.48 ± 96% perf-profile.children.cycles-pp.__mem_cgroup_charge 0.42 ± 9% +1.1 1.49 ± 17% perf-profile.children.cycles-pp.__get_user_nocheck_1 0.44 ± 9% +1.1 1.53 ± 17% perf-profile.children.cycles-pp.iov_iter_fault_in_readable 0.38 ± 13% +1.1 1.49 ± 19% perf-profile.children.cycles-pp.get_page_from_freelist 0.44 ± 11% +1.2 1.63 ± 16% perf-profile.children.cycles-pp.mark_page_accessed 0.02 ±223% +1.2 1.22 ± 85% perf-profile.children.cycles-pp.__delete_from_page_cache 0.42 ± 13% +1.2 1.63 ± 19% perf-profile.children.cycles-pp.__alloc_pages 0.02 ±223% +1.2 1.25 ± 23% perf-profile.children.cycles-pp.invalidate_inode_page 0.66 ± 9% +1.4 2.03 ± 19% perf-profile.children.cycles-pp.ext4_block_write_begin 0.02 ±223% +1.4 1.39 ± 75% perf-profile.children.cycles-pp.__remove_mapping 0.02 ±223% +1.4 1.40 ± 74% perf-profile.children.cycles-pp.remove_mapping 0.42 ± 11% +1.5 1.94 ± 77% perf-profile.children.cycles-pp.__pagevec_lru_add 0.46 ± 11% +1.6 2.04 ± 72% perf-profile.children.cycles-pp.lru_cache_add 0.97 ± 11% +1.7 2.68 ± 24% perf-profile.children.cycles-pp.mark_buffer_dirty 0.44 ± 17% +1.9 2.29 ± 43% perf-profile.children.cycles-pp.test_clear_page_writeback 0.95 ± 11% +1.9 2.88 ± 58% perf-profile.children.cycles-pp.__add_to_page_cache_locked 0.47 ± 17% +2.0 2.43 ± 40% perf-profile.children.cycles-pp.end_page_writeback 1.09 ± 14% +2.0 3.13 ±102% perf-profile.children.cycles-pp.__mod_memcg_lruvec_state 0.08 ± 16% +2.5 2.55 ±158% perf-profile.children.cycles-pp.lock_page_lruvec_irqsave 0.98 ± 10% +2.5 3.52 ± 17% perf-profile.children.cycles-pp.get_io_u 1.40 ± 11% +2.7 4.07 ± 7% perf-profile.children.cycles-pp.__block_commit_write 0.66 ± 19% +2.8 3.41 ± 24% perf-profile.children.cycles-pp.ext4_finish_bio 0.66 ± 19% +2.8 3.42 ± 24% perf-profile.children.cycles-pp.ext4_end_bio 1.46 ± 10% +2.8 4.24 ± 6% perf-profile.children.cycles-pp.generic_write_end 1.23 ± 12% +3.0 4.18 ± 86% perf-profile.children.cycles-pp.__mod_lruvec_page_state 1.33 ± 12% +3.1 4.39 ± 30% perf-profile.children.cycles-pp.pagecache_get_page 1.34 ± 12% +3.1 4.42 ± 30% perf-profile.children.cycles-pp.grab_cache_page_write_begin 0.03 ±161% +3.2 3.21 ±107% perf-profile.children.cycles-pp.__pagevec_release 0.04 ±120% +3.2 3.30 ±104% perf-profile.children.cycles-pp.release_pages 1.42 ± 11% +3.5 4.95 ± 63% perf-profile.children.cycles-pp.add_to_page_cache_lru 1.26 ± 8% +3.9 5.14 ± 12% perf-profile.children.cycles-pp.copy_mc_fragile 1.27 ± 8% +3.9 5.18 ± 12% perf-profile.children.cycles-pp.pmem_do_read 1.37 ± 9% +4.2 5.60 ± 13% perf-profile.children.cycles-pp.ext4_mpage_readpages 1.37 ± 9% +4.2 5.60 ± 13% perf-profile.children.cycles-pp.read_pages 2.04 ± 11% +4.5 6.58 ± 14% perf-profile.children.cycles-pp.ext4_da_write_begin 0.03 ±157% +6.0 6.07 ± 45% perf-profile.children.cycles-pp.__filemap_fdatawrite_range 0.03 ±157% +6.0 6.07 ± 45% perf-profile.children.cycles-pp.filemap_fdatawrite_wbc 0.08 ±162% +6.2 6.25 ± 75% perf-profile.children.cycles-pp.__invalidate_mapping_pages 2.49 ± 11% +6.2 8.70 ± 22% perf-profile.children.cycles-pp.copyout 2.54 ± 11% +6.3 8.86 ± 22% perf-profile.children.cycles-pp.copy_page_to_iter 2.10 ± 9% +6.4 8.54 ± 11% perf-profile.children.cycles-pp.page_cache_ra_unbounded 2.62 ± 10% +6.5 9.14 ± 21% perf-profile.children.cycles-pp.copyin 2.66 ± 10% +6.6 9.26 ± 21% perf-profile.children.cycles-pp.copy_page_from_iter_atomic 2.20 ± 9% +6.8 8.95 ± 10% perf-profile.children.cycles-pp.filemap_get_pages 1.42 ± 32% +6.9 8.30 ± 22% perf-profile.children.cycles-pp.__memcpy_flushcache 1.43 ± 32% +6.9 8.35 ± 22% perf-profile.children.cycles-pp.write_pmem 1.44 ± 32% +6.9 8.38 ± 22% perf-profile.children.cycles-pp.pmem_do_write 2.95 ± 27% +7.1 10.06 ± 32% perf-profile.children.cycles-pp.wb_workfn 2.95 ± 27% +7.1 10.06 ± 32% perf-profile.children.cycles-pp.wb_do_writeback 2.95 ± 27% +7.1 10.06 ± 32% perf-profile.children.cycles-pp.wb_writeback 2.95 ± 27% +7.1 10.06 ± 32% perf-profile.children.cycles-pp.__writeback_inodes_wb 2.95 ± 27% +7.1 10.06 ± 32% perf-profile.children.cycles-pp.writeback_sb_inodes 2.95 ± 27% +7.1 10.06 ± 32% perf-profile.children.cycles-pp.__writeback_single_inode 2.96 ± 27% +7.1 10.08 ± 32% perf-profile.children.cycles-pp.process_one_work 2.96 ± 27% +7.1 10.09 ± 32% perf-profile.children.cycles-pp.worker_thread 3.00 ± 25% +7.1 10.13 ± 32% perf-profile.children.cycles-pp.ret_from_fork 3.00 ± 25% +7.1 10.13 ± 32% perf-profile.children.cycles-pp.kthread 2.47 ± 26% +11.2 13.72 ± 11% perf-profile.children.cycles-pp.ext4_bio_write_page 2.69 ± 26% +12.0 14.66 ± 10% perf-profile.children.cycles-pp.mpage_submit_page 0.11 ±160% +12.2 12.33 ± 58% perf-profile.children.cycles-pp.__x64_sys_fadvise64 0.11 ±160% +12.2 12.33 ± 58% perf-profile.children.cycles-pp.ksys_fadvise64_64 0.11 ±160% +12.2 12.33 ± 58% perf-profile.children.cycles-pp.generic_fadvise 0.11 ±160% +12.2 12.33 ± 58% perf-profile.children.cycles-pp.posix_fadvise 2.79 ± 25% +12.4 15.16 ± 10% perf-profile.children.cycles-pp.mpage_process_page_bufs 5.11 ± 10% +12.7 17.83 ± 21% perf-profile.children.cycles-pp.copy_user_enhanced_fast_string 2.97 ± 25% +13.1 16.10 ± 10% perf-profile.children.cycles-pp.mpage_prepare_extent_to_map 2.98 ± 25% +13.1 16.13 ± 10% perf-profile.children.cycles-pp.do_writepages 2.98 ± 25% +13.1 16.13 ± 10% perf-profile.children.cycles-pp.ext4_writepages 3.42 ± 20% +13.8 17.21 ± 11% perf-profile.children.cycles-pp.pmem_submit_bio 3.42 ± 20% +13.8 17.22 ± 11% perf-profile.children.cycles-pp.__submit_bio_noacct 3.42 ± 20% +13.8 17.22 ± 11% perf-profile.children.cycles-pp.__submit_bio 5.26 ± 9% +14.5 19.75 ± 9% perf-profile.children.cycles-pp.filemap_read 5.27 ± 9% +14.5 19.76 ± 9% perf-profile.children.cycles-pp.aio_read 54.78 ± 8% -51.7 3.06 ±134% perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath 0.73 ± 4% -0.3 0.44 ± 19% perf-profile.self.cycles-pp._raw_spin_lock 0.01 ±223% +0.1 0.07 ± 15% perf-profile.self.cycles-pp._raw_spin_unlock_irqrestore 0.00 +0.1 0.08 ± 24% perf-profile.self.cycles-pp.ext4_da_write_end 0.00 +0.1 0.08 ± 14% perf-profile.self.cycles-pp.__list_add_valid 0.00 +0.1 0.08 ± 19% perf-profile.self.cycles-pp.mem_cgroup_update_lru_size 0.00 +0.1 0.08 ± 14% perf-profile.self.cycles-pp.account_page_dirtied 0.00 +0.1 0.08 ± 19% perf-profile.self.cycles-pp.__read_end_io 0.00 +0.1 0.08 ± 27% perf-profile.self.cycles-pp.xas_find_marked 0.00 +0.1 0.09 ± 18% perf-profile.self.cycles-pp.__mark_inode_dirty 0.00 +0.1 0.09 ± 21% perf-profile.self.cycles-pp.ext4_da_write_begin 0.00 +0.1 0.09 ± 18% perf-profile.self.cycles-pp.try_charge_memcg 0.00 +0.1 0.09 ± 17% perf-profile.self.cycles-pp.node_page_state 0.00 +0.1 0.10 ± 16% perf-profile.self.cycles-pp.page_counter_try_charge 0.00 +0.1 0.10 ± 15% perf-profile.self.cycles-pp.__remove_mapping 0.00 +0.1 0.10 ± 22% perf-profile.self.cycles-pp.lru_cache_add 0.00 +0.1 0.10 ± 21% perf-profile.self.cycles-pp.__mod_lruvec_state 0.00 +0.1 0.10 ± 17% perf-profile.self.cycles-pp.mod_objcg_state 0.00 +0.1 0.10 ± 93% perf-profile.self.cycles-pp.unlock_page_memcg 0.01 ±223% +0.1 0.11 ± 30% perf-profile.self.cycles-pp.generic_write_end 0.00 +0.1 0.11 ± 27% perf-profile.self.cycles-pp.ext4_block_write_begin 0.02 ±141% +0.1 0.12 ± 30% perf-profile.self.cycles-pp.ext4_da_get_block_prep 0.06 ± 13% +0.1 0.17 ± 24% perf-profile.self.cycles-pp.get_obj_cgroup_from_current 0.04 ± 71% +0.1 0.15 ± 24% perf-profile.self.cycles-pp.kmem_cache_alloc 0.01 ±223% +0.1 0.12 ± 28% perf-profile.self.cycles-pp.copy_page_from_iter_atomic 0.06 ± 13% +0.1 0.17 ± 28% perf-profile.self.cycles-pp.rmqueue 0.04 ± 71% +0.1 0.15 ± 85% perf-profile.self.cycles-pp.__count_memcg_events 0.04 ± 71% +0.1 0.16 ± 26% perf-profile.self.cycles-pp.___might_sleep 0.00 +0.1 0.12 ± 31% perf-profile.self.cycles-pp.get_page_from_freelist 0.06 ± 11% +0.1 0.19 ± 26% perf-profile.self.cycles-pp.__add_to_page_cache_locked 0.02 ±141% +0.1 0.14 ± 24% perf-profile.self.cycles-pp.ext4_mpage_readpages 0.02 ±146% +0.1 0.15 ± 18% perf-profile.self.cycles-pp._raw_spin_lock_irq 0.00 +0.1 0.13 ± 31% perf-profile.self.cycles-pp.end_page_writeback 0.05 ± 45% +0.1 0.18 ± 27% perf-profile.self.cycles-pp.create_empty_buffers 0.05 ± 45% +0.1 0.18 ± 28% perf-profile.self.cycles-pp.node_dirty_ok 0.00 +0.1 0.14 ± 19% perf-profile.self.cycles-pp.__slab_free 0.00 +0.1 0.14 ± 25% perf-profile.self.cycles-pp.memcg_slab_free_hook 0.05 ± 45% +0.1 0.19 ± 25% perf-profile.self.cycles-pp.xas_start 0.00 +0.2 0.16 ± 58% perf-profile.self.cycles-pp.lock_page_memcg 0.00 +0.2 0.16 ± 19% perf-profile.self.cycles-pp.xas_clear_mark 0.00 +0.2 0.16 ± 14% perf-profile.self.cycles-pp.xas_store 0.03 ±100% +0.2 0.20 ± 26% perf-profile.self.cycles-pp.clear_page_dirty_for_io 0.06 ± 13% +0.2 0.23 ± 26% perf-profile.self.cycles-pp.page_mapping 0.00 +0.2 0.18 ± 30% perf-profile.self.cycles-pp.mpage_prepare_extent_to_map 0.01 ±223% +0.2 0.19 ± 21% perf-profile.self.cycles-pp.unlock_page 0.00 +0.2 0.19 ± 60% perf-profile.self.cycles-pp.page_counter_cancel 0.10 ± 14% +0.2 0.29 ± 24% perf-profile.self.cycles-pp.rmqueue_bulk 0.08 ± 12% +0.2 0.28 ± 21% perf-profile.self.cycles-pp.filemap_read 0.04 ± 75% +0.2 0.24 ± 28% perf-profile.self.cycles-pp.__test_set_page_writeback 0.00 +0.2 0.21 ± 21% perf-profile.self.cycles-pp.drop_buffers 0.11 ± 12% +0.2 0.32 ± 16% perf-profile.self.cycles-pp.__pagevec_lru_add_fn 0.07 ± 18% +0.2 0.29 ± 22% perf-profile.self.cycles-pp.pagecache_get_page 0.00 +0.2 0.23 ±106% perf-profile.self.cycles-pp.mem_cgroup_wb_domain 0.08 ± 16% +0.3 0.34 ± 22% perf-profile.self.cycles-pp.filemap_get_read_batch 0.07 ± 29% +0.3 0.33 ± 29% perf-profile.self.cycles-pp.ext4_bio_write_page 0.00 +0.3 0.27 ±135% perf-profile.self.cycles-pp.charge_memcg 0.06 ± 15% +0.3 0.33 ± 20% perf-profile.self.cycles-pp.test_clear_page_writeback 0.02 ±223% +0.3 0.29 ± 36% perf-profile.self.cycles-pp.ktime_get 0.00 +0.3 0.28 ± 14% perf-profile.self.cycles-pp.__free_one_page 0.08 ± 34% +0.3 0.40 ± 28% perf-profile.self.cycles-pp.ext4_finish_bio 0.02 ±141% +0.3 0.34 ± 15% perf-profile.self.cycles-pp.release_pages 0.07 ± 17% +0.3 0.42 ± 17% perf-profile.self.cycles-pp.__list_del_entry_valid 0.09 ± 12% +0.3 0.44 ± 13% perf-profile.self.cycles-pp.__mod_node_page_state 0.11 ± 33% +0.4 0.46 ± 27% perf-profile.self.cycles-pp.find_get_pages_range_tag 0.00 +0.4 0.36 ± 17% perf-profile.self.cycles-pp.find_lock_entries 0.00 +0.4 0.37 ± 14% perf-profile.self.cycles-pp.poll_idle 0.00 +0.4 0.40 ± 22% perf-profile.self.cycles-pp.jbd2_journal_grab_journal_head 0.10 ± 18% +0.4 0.50 ± 15% perf-profile.self.cycles-pp.mpage_process_page_bufs 0.09 ± 23% +0.4 0.49 ± 27% perf-profile.self.cycles-pp.percpu_counter_add_batch 0.06 ± 19% +0.4 0.49 ±118% perf-profile.self.cycles-pp.get_mem_cgroup_from_mm 0.20 ± 8% +0.6 0.77 ± 21% perf-profile.self.cycles-pp.mark_buffer_dirty 0.26 ± 15% +0.8 1.01 ± 24% perf-profile.self.cycles-pp.xas_load 0.14 ± 6% +0.8 0.94 ± 76% perf-profile.self.cycles-pp.balance_dirty_pages_ratelimited 0.36 ± 10% +0.8 1.18 ± 14% perf-profile.self.cycles-pp._raw_spin_lock_irqsave 0.40 ± 14% +0.9 1.29 ± 25% perf-profile.self.cycles-pp.__block_commit_write 0.19 ± 8% +0.9 1.12 ± 84% perf-profile.self.cycles-pp.__mod_lruvec_page_state 0.42 ± 10% +1.1 1.47 ± 17% perf-profile.self.cycles-pp.__get_user_nocheck_1 0.44 ± 10% +1.2 1.62 ± 15% perf-profile.self.cycles-pp.mark_page_accessed 0.29 ± 7% +2.0 2.32 ±120% perf-profile.self.cycles-pp.__mod_memcg_lruvec_state 0.97 ± 9% +2.5 3.47 ± 17% perf-profile.self.cycles-pp.get_io_u 1.25 ± 8% +3.8 5.09 ± 12% perf-profile.self.cycles-pp.copy_mc_fragile 1.41 ± 32% +6.8 8.24 ± 22% perf-profile.self.cycles-pp.__memcpy_flushcache 5.08 ± 10% +12.6 17.68 ± 21% perf-profile.self.cycles-pp.copy_user_enhanced_fast_string 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://01.org/lkp