[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20170424071444.GC15491@yexl-desktop>
Date: Mon, 24 Apr 2017 15:14:44 +0800
From: kernel test robot <xiaolong.ye@...el.com>
To: "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>
Cc: Michal Hocko <mhocko@...nel.org>,
Peter Zijlstra <peterz@...radead.org>,
LKML <linux-kernel@...r.kernel.org>,
"Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>, lkp@...org
Subject: [lkp-robot] [sched,rcu] 137d662a84: will-it-scale.per_process_ops
-11.4% regression
Greeting,
FYI, we noticed a -11.4% regression of will-it-scale.per_process_ops due to commit:
commit: 137d662a84c286d28d63c9f0e593b01b61df45f1 ("sched,rcu: Make cond_resched() provide RCU quiescent state")
https://git.kernel.org/cgit/linux/kernel/git/paulmck/linux-rcu.git dev.2017.04.19c
in testcase: will-it-scale
on test machine: 88 threads Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz with 48G memory
with following parameters:
test: mmap2
cpufreq_governor: performance
test-description: Will It Scale takes a testcase and runs it from 1 through to n parallel copies to see if the testcase will scale. It builds both a process and threads based test in order to see any differences between the two.
test-url: https://github.com/antonblanchard/will-it-scale
Details are as below:
-------------------------------------------------------------------------------------------------->
To reproduce:
git clone https://github.com/01org/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
testcase/path_params/tbox_group/run: will-it-scale/mmap2-performance/lkp-bdw-ep3d
26c6ab39c97486a5 137d662a84c286d28d63c9f0e5
---------------- --------------------------
0.02 16% 0.02 will-it-scale.scalability
194310 -11% 172129 will-it-scale.per_process_ops
163240 -12% 143383 will-it-scale.per_thread_ops
17316 ± 7% 22% 21129 ± 8% perf-stat.cpu-migrations
0.02 -7% 0.02 perf-stat.dTLB-store-miss-rate%
619722 ±183% 1e+06 1819547 ±216% latency_stats.avg.max
10349 ± 60% -1e+04 0 latency_stats.avg.perf_event_alloc.SYSC_perf_event_open.SyS_perf_event_open.entry_SYSCALL_64_fastpath
46339 ±238% -5e+04 0 latency_stats.avg.rpc_wait_bit_killable.__rpc_wait_for_completion_task._nfs4_proc_open_confirm.[nfsv4].nfs4_do_open.[nfsv4].nfs4_atomic_open.[nfsv4].nfs_atomic_open.path_openat.do_filp_open.do_sys_open.SyS_open.entry_SYSCALL_64_fastpath
185365 ± 89% -2e+05 15106 ±126% latency_stats.avg.expand_files.__alloc_fd.get_unused_fd_flags.do_sys_open.SyS_open.entry_SYSCALL_64_fastpath
1165346 ±176% 3e+06 4111306 ±244% latency_stats.max.max
10349 ± 60% -1e+04 0 latency_stats.max.perf_event_alloc.SYSC_perf_event_open.SyS_perf_event_open.entry_SYSCALL_64_fastpath
46339 ±238% -5e+04 0 latency_stats.max.rpc_wait_bit_killable.__rpc_wait_for_completion_task._nfs4_proc_open_confirm.[nfsv4].nfs4_do_open.[nfsv4].nfs4_atomic_open.[nfsv4].nfs_atomic_open.path_openat.do_filp_open.do_sys_open.SyS_open.entry_SYSCALL_64_fastpath
298205 ± 97% -3e+05 23850 ±130% latency_stats.max.expand_files.__alloc_fd.get_unused_fd_flags.do_sys_open.SyS_open.entry_SYSCALL_64_fastpath
2413 ±117% 4e+04 47014 ± 66% latency_stats.sum.devkmsg_read.__vfs_read.vfs_read.SyS_read.entry_SYSCALL_64_fastpath
10349 ± 60% -1e+04 0 latency_stats.sum.perf_event_alloc.SYSC_perf_event_open.SyS_perf_event_open.entry_SYSCALL_64_fastpath
46339 ±238% -5e+04 0 latency_stats.sum.rpc_wait_bit_killable.__rpc_wait_for_completion_task._nfs4_proc_open_confirm.[nfsv4].nfs4_do_open.[nfsv4].nfs4_atomic_open.[nfsv4].nfs_atomic_open.path_openat.do_filp_open.do_sys_open.SyS_open.entry_SYSCALL_64_fastpath
1069095 ±140% -1e+06 29060 ±134% latency_stats.sum.expand_files.__alloc_fd.get_unused_fd_flags.do_sys_open.SyS_open.entry_SYSCALL_64_fastpath
90000 ++------------------------------------------------------------------O
| O O O O OO OO O O O |
80000 ++ OO O O O O O |
70000 O+ O O O O |
| O O |
60000 ++ O O |
50000 ++ O O|
| |
40000 ++ OO O O |
30000 ++ |
| O O |
20000 ++ O |
10000 ++ |
|O.O * .**O O .* .* O * * *. O** .***.*O |
0 **--*-O-OO*-*OOO--**--O**-O-*-**-*-***--***-*--*-*O*-***O-----------+
will-it-scale.per_process_ops
200000 ++-----------------------------------------------------------------+
| .****. ***.***.****.****.****.***.****.****.****.** |
190000 **.**** ****.* |
| |
180000 O+ O O OOO |
| O O OO OOO O OO OO OO OO O O O O O |
170000 ++ OO OOO O OO O OO O O O O O O O O O O OO
| |
160000 ++ |
| |
150000 ++ O |
|O O |
140000 ++ O |
| |
130000 ++-----------------------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
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,
Xiaolong
View attachment "config-4.11.0-rc2-00070-g137d662" of type "text/plain" (157984 bytes)
View attachment "job-script" of type "text/plain" (6716 bytes)
View attachment "job.yaml" of type "text/plain" (4322 bytes)
View attachment "reproduce" of type "text/plain" (144 bytes)
Powered by blists - more mailing lists