Messages in this thread | | | Date | Thu, 7 Aug 2014 18:55:38 +0800 | From | Fengguang Wu <> | Subject | [fs] e691fc4583e: -59.7% dbench.throughput-MB/sec |
| |
Hi Neil,
We noticed the below dbench regression on
git://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master commit e691fc4583e512f936bd04efb870815fe1ed750d ("fs: allow ->d_manage() to declare -EISDIR in rcu_walk mode.")
test case: ivb44/dbench/100%
53bcc0567b8c520 e691fc4583e512f936bd04efb --------------- ------------------------- 25957 ± 0% -59.7% 10460 ± 4% TOTAL dbench.throughput-MB/sec 0.00 +Inf% 24.33 ±31% TOTAL perf-profile.cpu-cycles._raw_spin_lock.lockref_get_not_dead.unlazy_walk.lookup_fast.link_path_walk 15.14 ± 0% -82.1% 2.71 ±25% TOTAL perf-profile.cpu-cycles.copy_user_enhanced_fast_string.copy_page_to_iter.generic_file_read_iter.new_sync_read.vfs_read 0.00 +Inf% 4.33 ±35% TOTAL perf-profile.cpu-cycles.lockref_put_or_lock.dput.link_path_walk.path_openat.do_filp_open 0.00 +Inf% 4.35 ±36% TOTAL perf-profile.cpu-cycles.lockref_get_not_dead.unlazy_walk.lookup_fast.link_path_walk.path_openat 0.00 +Inf% 5.06 ±36% TOTAL perf-profile.cpu-cycles.lockref_put_or_lock.dput.link_path_walk.path_lookupat.filename_lookup 1.11 ± 1% -100.0% 0.00 ± 0% TOTAL perf-profile.cpu-cycles._raw_spin_lock.dcache_readdir.iterate_dir.sys_getdents.system_call_fastpath 0.00 +Inf% 5.10 ±36% TOTAL perf-profile.cpu-cycles.lockref_get_not_dead.unlazy_walk.lookup_fast.link_path_walk.path_lookupat 0.00 +Inf% 11.19 ±30% TOTAL perf-profile.cpu-cycles._raw_spin_lock.lockref_put_or_lock.dput.link_path_walk.path_openat 0.00 +Inf% 12.85 ±31% TOTAL perf-profile.cpu-cycles._raw_spin_lock.lockref_put_or_lock.dput.link_path_walk.path_lookupat 2.00 ± 0% -79.5% 0.41 ±26% TOTAL perf-profile.cpu-cycles.find_get_entry.pagecache_get_page.generic_file_read_iter.new_sync_read.vfs_read 2.01 ± 3% -74.0% 0.52 ±17% TOTAL perf-profile.cpu-cycles.rcu_nocb_kthread.kthread.ret_from_fork 8.70 ± 1% -77.5% 1.96 ±24% TOTAL perf-profile.cpu-cycles.copy_user_enhanced_fast_string.iov_iter_copy_from_user_atomic.generic_perform_write.__generic_file_write_iter.generic_file_write_iter 1.20 ± 0% -75.9% 0.29 ±25% TOTAL perf-profile.cpu-cycles.get_empty_filp.path_openat.do_filp_open.do_sys_open.sys_openat 1.11 ± 3% -72.9% 0.30 ±22% TOTAL perf-profile.cpu-cycles.selinux_inode_permission.security_inode_permission.__inode_permission.inode_permission.link_path_walk 2.03 ± 1% -69.2% 0.63 ±21% TOTAL perf-profile.cpu-cycles.__vdso_gettimeofday 1.52 ± 4% -67.6% 0.49 ±13% TOTAL perf-profile.cpu-cycles.avc_has_perm_noaudit.selinux_inode_permission.security_inode_permission.__inode_permission.inode_permission 47938419 ± 0% -60.6% 18877399 ± 3% TOTAL proc-vmstat.pgalloc_dma32 9.25e+08 ± 0% -59.7% 3.728e+08 ± 3% TOTAL proc-vmstat.pgalloc_normal 22194267 ± 0% -59.3% 9023662 ± 3% TOTAL proc-vmstat.pgactivate 4.53e+08 ± 0% -58.0% 1.904e+08 ± 2% TOTAL numa-numastat.node1.local_node 4.53e+08 ± 0% -58.0% 1.904e+08 ± 2% TOTAL numa-numastat.node1.numa_hit 2987057 ± 0% -56.3% 1306633 ± 2% TOTAL softirqs.RCU 964655 ±14% -47.3% 508785 ±23% TOTAL cpuidle.C3-IVT.time 0.32 ± 5% -37.3% 0.20 ±12% TOTAL turbostat.%c1 69254 ± 6% -32.7% 46602 ± 9% TOTAL cpuidle.C6-IVT.usage 2724 ±15% -32.3% 1843 ± 6% TOTAL cpuidle.C3-IVT.usage 6665511 ±11% -31.9% 4540015 ± 8% TOTAL cpuidle.C1E-IVT.time 26138868 ±11% -27.6% 18932437 ±17% TOTAL cpuidle.C1-IVT.time 6013 ± 9% -28.3% 4311 ±18% TOTAL proc-vmstat.pgmigrate_success 6013 ± 9% -28.3% 4311 ±18% TOTAL proc-vmstat.numa_pages_migrated 48751 ± 4% -18.3% 39837 ± 6% TOTAL proc-vmstat.numa_hint_faults 1849 ± 2% -14.9% 1573 ± 3% TOTAL numa-vmstat.node0.nr_alloc_batch 48570 ± 4% -16.9% 40343 ± 6% TOTAL proc-vmstat.numa_pte_updates 17191 ± 6% -15.3% 14566 ±14% TOTAL cpuidle.C1E-IVT.usage 85893 ± 3% +12.4% 96578 ± 4% TOTAL meminfo.Committed_AS 29718 ± 4% -8.0% 27337 ± 2% TOTAL proc-vmstat.numa_hint_faults_local 13612 ± 3% -10.4% 12195 ± 3% TOTAL slabinfo.kmalloc-192.num_objs 13611 ± 3% -10.4% 12195 ± 3% TOTAL slabinfo.kmalloc-192.active_objs 9469 ± 0% -62.3% 3574 ± 3% TOTAL time.user_time 1365218 ± 1% -44.4% 759678 ± 1% TOTAL time.involuntary_context_switches 4818 ± 1% -38.7% 2954 ± 1% TOTAL vmstat.system.cs ± 0% -37.0% ± 1% TOTAL turbostat.RAM_W 23661 ± 0% +28.8% 30466 ± 0% TOTAL time.system_time 158133 ± 8% -16.6% 131932 ± 9% TOTAL time.voluntary_context_switches 196790 ± 7% -15.8% 165728 ± 8% TOTAL time.minor_page_faults ± 0% -10.5% ± 0% TOTAL turbostat.Cor_W ± 0% -9.8% ± 0% TOTAL turbostat.Pkg_W 4599 ± 0% +2.7% 4724 ± 0% TOTAL time.percent_of_cpu_this_job_got 54951 ± 0% -1.7% 54021 ± 0% TOTAL vmstat.system.in
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.
Thanks, Fengguang echo performance > /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu1/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu10/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu11/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu12/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu13/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu14/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu15/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu16/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu17/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu18/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu19/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu2/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu20/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu21/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu22/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu23/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu24/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu25/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu26/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu27/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu28/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu29/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu3/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu30/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu31/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu32/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu33/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu34/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu35/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu36/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu37/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu38/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu39/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu4/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu40/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu41/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu42/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu43/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu44/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu45/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu46/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu47/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu5/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu6/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu7/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu8/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu9/cpufreq/scaling_governor dbench 48 -c /usr/share/dbench/client.txt _______________________________________________ LKP mailing list LKP@linux.intel.com
| |