Hello,
kernel test robot noticed a 50.0% regression of filebench.sum_operations/s on:
commit: e2d46f2ec332533816417b60933954173f602121 ("netfs: Change the read result collector to only use one work item")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
[test failed on linus/master 69b54314c975f4dfd3a29d6b9211ab68fff46682]
[test failed on linux-next/master ed58d103e6da15a442ff87567898768dc3a66987]
testcase: filebench
config: x86_64-rhel-9.4
compiler: gcc-12
test machine: 128 threads 2 sockets Intel(R) Xeon(R) Platinum 8358 CPU @ 2.60GHz (Ice Lake) with 128G memory
parameters:
disk: 1HDD
fs: ext4
fs2: cifs
test: copyfiles.f
cpufreq_governor: performance
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <oliver.sang@xxxxxxxxx>
| Closes: https://lore.kernel.org/oe-lkp/202502121520.d583b3c3-lkp@xxxxxxxxx
Details are as below:
-------------------------------------------------------------------------------------------------->
The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20250212/202502121520.d583b3c3-lkp@xxxxxxxxx
=========================================================================================
compiler/cpufreq_governor/disk/fs2/fs/kconfig/rootfs/tbox_group/test/testcase:
gcc-12/performance/1HDD/cifs/ext4/x86_64-rhel-9.4/debian-12-x86_64-20240206.cgz/lkp-icl-2sp6/copyfiles.f/filebench
commit:
eddf51f2bb ("afs: Make {Y,}FS.FetchData an asynchronous operation")
e2d46f2ec3 ("netfs: Change the read result collector to only use one work item")
eddf51f2bb2c28b0 e2d46f2ec332533816417b60933
---------------- ---------------------------
%stddev %change %stddev
\ | \
111550 ± 17% -17.2% 92328 sched_debug.cfs_rq:/.load.stddev
4.52 ±218% -98.3% 0.08 ± 12% perf-sched.sch_delay.avg.ms.schedule_timeout.kcompactd.kthread.ret_from_fork
0.15 ± 2% +22.0% 0.19 ± 12% perf-sched.sch_delay.max.ms.schedule_timeout.rcu_gp_fqs_loop.rcu_gp_kthread.kthread
9404404 +5.0% 9871609 perf-stat.i.cache-references
4.82 +0.1 4.87 perf-stat.overall.branch-miss-rate%
4.88 -0.3 4.63 ± 2% perf-stat.overall.cache-miss-rate%
9290667 +5.0% 9751847 perf-stat.ps.cache-references
31.30 -50.2% 15.60 filebench.sum_bytes_mb/s
6001 -50.0% 3000 filebench.sum_operations/s
1001 -50.0% 500.00 filebench.sum_reads/s
0.11 ± 14% +104.3% 0.23 ± 8% filebench.sum_time_ms/op
1000 -50.0% 500.00 filebench.sum_writes/s
5.89 ± 8% -1.4 4.52 ± 45% perf-profile.calltrace.cycles-pp.ast_primary_plane_helper_atomic_update.drm_atomic_helper_commit_planes.drm_atomic_helper_commit_tail.ast_mode_config_helper_atomic_commit_tail.commit_tail
5.88 ± 8% -1.4 4.52 ± 45% perf-profile.calltrace.cycles-pp.drm_fb_memcpy.ast_primary_plane_helper_atomic_update.drm_atomic_helper_commit_planes.drm_atomic_helper_commit_tail.ast_mode_config_helper_atomic_commit_tail
5.81 ± 8% -1.3 4.47 ± 45% perf-profile.calltrace.cycles-pp.memcpy_toio.drm_fb_memcpy.ast_primary_plane_helper_atomic_update.drm_atomic_helper_commit_planes.drm_atomic_helper_commit_tail
6.58 ± 7% -0.5 6.10 ± 4% perf-profile.calltrace.cycles-pp.process_one_work.worker_thread.kthread.ret_from_fork.ret_from_fork_asm
6.74 ± 8% -0.5 6.26 ± 4% perf-profile.calltrace.cycles-pp.worker_thread.kthread.ret_from_fork.ret_from_fork_asm
2.35 ± 9% -0.5 1.88 ± 10% perf-profile.calltrace.cycles-pp.getxattr
5.35 ± 3% -0.3 5.08 ± 2% perf-profile.calltrace.cycles-pp.console_flush_all.console_unlock.vprintk_emit.devkmsg_emit.devkmsg_write
5.35 ± 3% -0.3 5.08 ± 2% perf-profile.calltrace.cycles-pp.console_unlock.vprintk_emit.devkmsg_emit.devkmsg_write.vfs_write
5.35 ± 3% -0.3 5.09 ± 2% perf-profile.calltrace.cycles-pp.devkmsg_emit.devkmsg_write.vfs_write.ksys_write.do_syscall_64
5.35 ± 3% -0.3 5.09 ± 2% perf-profile.calltrace.cycles-pp.vprintk_emit.devkmsg_emit.devkmsg_write.vfs_write.ksys_write
5.35 ± 3% -0.3 5.10 ± 2% perf-profile.calltrace.cycles-pp.devkmsg_write.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.59 ± 10% +0.1 0.66 ± 6% perf-profile.calltrace.cycles-pp.do_sys_openat2.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe.openat64
2.16 ± 2% +0.2 2.32 ± 2% perf-profile.calltrace.cycles-pp.do_execveat_common.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe.execve
2.17 ± 3% +0.2 2.33 ± 2% perf-profile.calltrace.cycles-pp.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe.execve
2.17 ± 3% +0.2 2.33 ± 2% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.execve
2.17 ± 3% +0.2 2.33 ± 2% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.execve
2.17 ± 3% +0.2 2.33 ± 2% perf-profile.calltrace.cycles-pp.execve
1.39 ± 10% +0.2 1.62 ± 11% perf-profile.calltrace.cycles-pp.sched_ttwu_pending.__flush_smp_call_function_queue.flush_smp_call_function_queue.do_idle.cpu_startup_entry
1.44 ± 9% +0.2 1.68 ± 7% perf-profile.calltrace.cycles-pp.bprm_execve.do_execveat_common.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
6.58 ± 7% -0.5 6.10 ± 4% perf-profile.children.cycles-pp.process_one_work
6.74 ± 8% -0.5 6.26 ± 4% perf-profile.children.cycles-pp.worker_thread
2.37 ± 9% -0.4 1.92 ± 11% perf-profile.children.cycles-pp.getxattr
5.35 ± 3% -0.3 5.08 ± 2% perf-profile.children.cycles-pp.console_flush_all
5.35 ± 3% -0.3 5.08 ± 2% perf-profile.children.cycles-pp.console_unlock
5.35 ± 3% -0.3 5.09 ± 2% perf-profile.children.cycles-pp.devkmsg_emit
5.35 ± 3% -0.3 5.09 ± 2% perf-profile.children.cycles-pp.vprintk_emit
5.35 ± 3% -0.3 5.10 ± 2% perf-profile.children.cycles-pp.devkmsg_write
0.54 ± 17% -0.1 0.39 ± 16% perf-profile.children.cycles-pp.cifs_d_revalidate
0.57 ± 15% -0.1 0.44 ± 14% perf-profile.children.cycles-pp.lookup_one_qstr_excl
0.57 ± 14% -0.1 0.43 ± 15% perf-profile.children.cycles-pp.lookup_dcache
0.21 ± 45% -0.1 0.08 ± 51% perf-profile.children.cycles-pp._copy_from_iter
0.03 ±141% +0.1 0.13 ± 38% perf-profile.children.cycles-pp.__run_timers
0.09 ± 56% +0.1 0.21 ± 34% perf-profile.children.cycles-pp.__x64_sys_munmap
0.04 ±147% +0.1 0.16 ± 47% perf-profile.children.cycles-pp.tmigr_inactive_up
0.17 ± 25% +0.1 0.29 ± 22% perf-profile.children.cycles-pp.xas_find
0.23 ± 22% +0.1 0.36 ± 10% perf-profile.children.cycles-pp.native_irq_return_iret
0.19 ± 24% +0.1 0.34 ± 9% perf-profile.children.cycles-pp.entry_SYSRETQ_unsafe_stack
0.04 ±147% +0.2 0.18 ± 46% perf-profile.children.cycles-pp.tmigr_cpu_deactivate
2.17 ± 3% +0.2 2.33 ± 2% perf-profile.children.cycles-pp.execve
2.19 ± 3% +0.2 2.36 perf-profile.children.cycles-pp.__x64_sys_execve
2.18 ± 2% +0.2 2.34 perf-profile.children.cycles-pp.do_execveat_common
0.50 ± 12% +0.2 0.69 ± 27% perf-profile.children.cycles-pp.enqueue_entity
0.15 ± 57% +0.2 0.36 ± 16% perf-profile.children.cycles-pp.perf_read
0.49 ± 25% +0.3 0.74 ± 23% perf-profile.children.cycles-pp.tick_nohz_restart_sched_tick
0.35 ± 29% +0.3 0.63 ± 15% perf-profile.children.cycles-pp.readn
0.77 ± 18% +0.3 1.09 ± 19% perf-profile.children.cycles-pp.tick_nohz_idle_exit
0.19 ± 24% +0.1 0.31 ± 10% perf-profile.self.cycles-pp.entry_SYSRETQ_unsafe_stack
0.23 ± 22% +0.1 0.36 ± 10% perf-profile.self.cycles-pp.native_irq_return_iret
Disclaimer:
Results have been estimated based on internal Intel analysis and are provided
for informational purposes only. Any difference in system hardware or software
design or configuration may affect actual performance.
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
