lkml.org 
[lkml]   [2017]   [Jan]   [3]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: [lkp-developer] [sched/core] 6b94780e45: unixbench.score -4.5% regression
Hi Xiaolong,

Thanks for testing, I'm going to look for another root cause
It was also mentioned a -2.9% regression with a 8 threads Intel(R)
Core(TM) i7 CPU 870 @ 2.93GHz with 6G memory. Have you checked this
platform too ?

Regards,
Vincent

On 3 January 2017 at 08:13, Ye Xiaolong <xiaolong.ye@intel.com> wrote:
> On 01/02, Vincent Guittot wrote:
>>Hi Xiaolong,
>>
>>Le Monday 19 Dec 2016 ŕ 08:14:53 (+0800), kernel test robot a écrit :
>>>
>>> Greeting,
>>>
>>> FYI, we noticed a -4.5% regression of unixbench.score due to commit:
>>
>>I have been able to restore performance on my platform with the patch below.
>>Could you test it ?
>>
>>---
>> kernel/sched/core.c | 1
>> 1 file changed, 1 insertion(+)
>>
>>diff --git a/kernel/sched/core.c b/kernel/sched/core.c
>>index 393759b..6e7d45c 100644
>>--- a/kernel/sched/core.c
>>+++ b/kernel/sched/core.c
>>@@ -2578,6 +2578,7 @@ void wake_up_new_task(struct task_struct *p)
>> __set_task_cpu(p, select_task_rq(p, task_cpu(p), SD_BALANCE_FORK, 0));
>> #endif
>> rq = __task_rq_lock(p, &rf);
>>+ update_rq_clock(rq);
>> post_init_entity_util_avg(&p->se);
>>
>> activate_task(rq, p, 0);
>>--
>>2.7.4
>>
>>Vincent
>
> Hi, Vincent,
>
> I applied your fix patch on top of 6b94780 ("sched/core: Use load_avg for selecting idlest group"),
> and here is the comparison. (60df283834fd4def3c11ad2de3 is the fix commit id).
> Seems the performance hasn't been restored back.

Thanks for testings.
>
>
> =========================================================================================
> compiler/cpufreq_governor/kconfig/nr_task/rootfs/runtime/tbox_group/test/testcase:
> gcc-6/performance/x86_64-rhel-7.2/100%/debian-x86_64-2016-08-31.cgz/300s/lkp-wsm-ep1/shell1/unixbench
>
> commit:
> f519a3f1c6b7a990e5aed37a8f853c6ecfdee945
> 6b94780e45c17b83e3e75f8aaca5a328db583c74
> 60df283834fd4def3c11ad2de3e6fc9e81b7dff1
>
> f519a3f1c6b7a990 6b94780e45c17b83e3e75f8aac 60df283834fd4def3c11ad2de3
> ---------------- -------------------------- --------------------------
> %stddev %change %stddev %change %stddev
> \ | \ | \
> 25565 ą 0% -4.5% 24414 ą 0% -4.5% 24421 ą 0% unixbench.score
> 13223805 ą 2% -19.6% 10628072 ą 0% -21.3% 10412818 ą 1% unixbench.time.involuntary_context_switches
> 9.232e+08 ą 0% -4.3% 8.831e+08 ą 0% -4.3% 8.838e+08 ą 0% unixbench.time.minor_page_faults
> 1807 ą 0% -5.4% 1709 ą 0% -5.6% 1705 ą 0% unixbench.time.percent_of_cpu_this_job_got
> 5656 ą 0% -6.8% 5271 ą 0% -7.3% 5243 ą 0% unixbench.time.system_time
> 5743 ą 0% -4.0% 5514 ą 0% -3.9% 5516 ą 0% unixbench.time.user_time
> 29557557 ą 0% -2.6% 28781098 ą 0% -2.2% 28919280 ą 0% unixbench.time.voluntary_context_switches
> 741766 ą 2% -62.4% 279054 ą 1% -61.8% 283034 ą 1% interrupts.CAL:Function_call_interrupts
> 2912823 ą 0% -9.7% 2630010 ą 0% -8.7% 2660077 ą 0% softirqs.RCU
> 13223805 ą 2% -19.6% 10628072 ą 0% -21.3% 10412818 ą 1% time.involuntary_context_switches
> 126250 ą 0% -12.2% 110890 ą 0% -11.5% 111739 ą 0% vmstat.system.cs
> 31060 ą 1% -9.2% 28214 ą 0% -9.6% 28078 ą 0% vmstat.system.in
> 454.50 ą150% +164.7% 1203 ą166% +792.3% 4055 ą 18% numa-numastat.node0.numa_foreign
> 454.50 ą150% +164.7% 1203 ą166% +792.3% 4055 ą 18% numa-numastat.node0.numa_miss
> 4297 ą 15% -18.1% 3520 ą 57% -84.5% 666.40 ą113% numa-numastat.node1.numa_foreign
> 4297 ą 15% -18.1% 3520 ą 57% -84.5% 666.40 ą113% numa-numastat.node1.numa_miss
> 78.58 ą 0% -5.6% 74.20 ą 0% -6.0% 73.90 ą 0% turbostat.%Busy
> 2507 ą 0% -5.6% 2366 ą 0% -6.0% 2356 ą 0% turbostat.Avg_MHz
> 3.01 ą 2% +100.4% 6.03 ą 2% +100.1% 6.02 ą 0% turbostat.CPU%c3
> 2.35 ą 1% +6.8% 2.51 ą 4% +12.1% 2.64 ą 1% turbostat.CPU%c6
> 1.25 ą 5% -17.1% 1.04 ą 22% -32.3% 0.85 ą 5% perf-profile.children.cycles-pp.__irqentry_text_start
>
> Thanks,
> Xiaolong
>
>>
>>>
>>>
>>> commit: 6b94780e45c17b83e3e75f8aaca5a328db583c74 ("sched/core: Use load_avg for selecting idlest group")
>>> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
>>>
>>> in testcase: unixbench
>>> on test machine: 24 threads Nehalem-EP with 24G memory
>>> with following parameters:
>>>
>>> runtime: 300s
>>> nr_task: 100%
>>> test: shell1
>>> cpufreq_governor: performance
>>>
>>> test-description: UnixBench is the original BYTE UNIX benchmark suite aims to test performance of Unix-like system.
>>> test-url: https://github.com/kdlucas/byte-unixbench
>>>
>>> In addition to that, the commit also has significant impact on the following tests:
>>>
>>> +------------------+-----------------------------------------------------------------------+
>>> | testcase: change | unixbench: unixbench.score -2.9% regression |
>>> | test machine | 8 threads Intel(R) Core(TM) i7 CPU 870 @ 2.93GHz with 6G memory |
>>> | test parameters | nr_task=1 |
>>> | | runtime=300s |
>>> | | test=shell8 |
>>> +------------------+-----------------------------------------------------------------------+
>>>
>>>
>>> Details are as below:
>>> -------------------------------------------------------------------------------------------------->
>>>
>>>
>>> To reproduce:
>>>
>>> git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/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: unixbench/300s-100%-shell1-performance/lkp-wsm-ep1
>>>
>>> f519a3f1c6b7a990 6b94780e45c17b83e3e75f8aac
>>> ---------------- --------------------------
>>> 25565 -5% 24414 unixbench.score
>>> 29557557 28781098 unixbench.time.voluntary_context_switches
>>> 5743 -4% 5514 unixbench.time.user_time
>>> 9.232e+08 -4% 8.831e+08 unixbench.time.minor_page_faults
>>> 1807 -5% 1709 unixbench.time.percent_of_cpu_this_job_got
>>> 5656 -7% 5271 unixbench.time.system_time
>>> 13223805 -20% 10628072 unixbench.time.involuntary_context_switches
>>> 741766 -62% 279054 interrupts.CAL:Function_call_interrupts
>>> 31060 -9% 28214 vmstat.system.in
>>> 126250 -12% 110890 vmstat.system.cs
>>> 78.58 -6% 74.20 turbostat.%Busy
>>> 2507 -6% 2366 turbostat.Avg_MHz
>>> 9134 ą 47% -6e+03 2973 ą 36% latency_stats.max.pipe_read.__vfs_read.vfs_read.SyS_read.entry_SYSCALL_64_fastpath
>>> 380879 ą 10% 5e+05 887692 ą 49% latency_stats.sum.wait_on_page_bit_killable.__lock_page_or_retry.filemap_fault.__do_fault.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
>>> 31710 ą 15% -2e+04 10583 ą 14% latency_stats.sum.call_rwsem_down_write_failed.__vma_adjust.__split_vma.do_munmap.vm_munmap.elf_map.load_elf_binary.search_binary_handler.do_execveat_common.SyS_execve.do_syscall_64.return_from_SYSCALL_64
>>> 51796 ą 4% -4e+04 15457 ą 10% latency_stats.sum.call_rwsem_down_write_failed.unlink_file_vma.free_pgtables.unmap_region.do_munmap.vm_munmap.elf_map.load_elf_binary.search_binary_handler.do_execveat_common.SyS_execve.do_syscall_64
>>> 111998 ą 18% -7e+04 37074 ą 14% latency_stats.sum.call_rwsem_down_write_failed.__vma_adjust.__split_vma.do_munmap.mmap_region.do_mmap.vm_mmap_pgoff.SyS_mmap_pgoff.SyS_mmap.entry_SYSCALL_64_fastpath
>>> 275087 ą 15% -2e+05 81973 ą 3% latency_stats.sum.call_rwsem_down_write_failed.unlink_file_vma.free_pgtables.unmap_region.do_munmap.mmap_region.do_mmap.vm_mmap_pgoff.SyS_mmap_pgoff.SyS_mmap.entry_SYSCALL_64_fastpath
>>> 930993 ą 12% -6e+05 320520 ą 4% latency_stats.sum.call_rwsem_down_write_failed.vma_link.mmap_region.do_mmap.vm_mmap_pgoff.vm_mmap.elf_map.load_elf_binary.search_binary_handler.do_execveat_common.SyS_execve.do_syscall_64
>>> 4755783 ą 9% -3e+06 1619348 ą 4% latency_stats.sum.call_rwsem_down_write_failed.__vma_adjust.__split_vma.split_vma.mprotect_fixup.do_mprotect_pkey.SyS_mprotect.entry_SYSCALL_64_fastpath
>>> 5536067 ą 10% -4e+06 1929338 ą 3% latency_stats.sum.call_rwsem_down_write_failed.copy_process._do_fork.SyS_clone.do_syscall_64.return_from_SYSCALL_64
>>> 9.032e+08 -4% 8.64e+08 perf-stat.page-faults
>>> 9.032e+08 -4% 8.64e+08 perf-stat.minor-faults
>>> 2.329e+09 2.269e+09 perf-stat.node-load-misses
>>> 2.2e+09 -9% 2.011e+09 ą 5% perf-stat.dTLB-store-misses
>>> 3.278e+10 -9% 2.987e+10 ą 6% perf-stat.dTLB-load-misses
>>> 19484819 13% 21974129 perf-stat.cpu-migrations
>>> 3.755e+13 -6% 3.54e+13 perf-stat.cpu-cycles
>>> 3244 4% 3379 perf-stat.instructions-per-iTLB-miss
>>> 4.536e+12 -4% 4.332e+12 perf-stat.branch-instructions
>>> 2.303e+13 -4% 2.208e+13 perf-stat.instructions
>>> 5.768e+12 -4% 5.517e+12 perf-stat.dTLB-loads
>>> 3.567e+11 -4% 3.414e+11 perf-stat.cache-references
>>> 2.97 2.93 perf-stat.branch-miss-rate%
>>> 2.768e+10 2.699e+10 perf-stat.node-stores
>>> 5.446e+10 -3% 5.275e+10 perf-stat.cache-misses
>>> 0.03 -4% 0.03 perf-stat.iTLB-load-miss-rate%
>>> 9.673e+09 -4% 9.294e+09 perf-stat.node-loads
>>> 3.596e+12 -4% 3.442e+12 perf-stat.dTLB-stores
>>> 0.61 0.62 perf-stat.ipc
>>> 1.347e+11 -6% 1.27e+11 perf-stat.branch-misses
>>> 7.098e+09 -8% 6.533e+09 perf-stat.iTLB-load-misses
>>> 2.309e+13 -4% 2.206e+13 perf-stat.iTLB-loads
>>> 79911173 -12% 70187035 perf-stat.context-switches
>>>
>>>
>>>
>>> turbostat._Busy
>>>
>>> 90 ++-------------------------------------*---*---------------------------+
>>> | .. *...*.. |
>>> 80 *+..*..*...*..*...*..*...*..*...O...* O O O O O...O..O...O O O
>>> 70 O+ O O O O O O O O |
>>> | |
>>> 60 ++ |
>>> 50 ++ |
>>> | |
>>> 40 ++ |
>>> 30 ++ |
>>> | |
>>> 20 ++ |
>>> 10 ++ |
>>> | |
>>> 0 ++----------------------------------O----------------------------------+
>>>
>>>
>>>
>>>
>>>
>>> unixbench.time.percent_of_cpu_this_job_got
>>>
>>> 2500 ++-------------------------------------------------------------------+
>>> | |
>>> | .*... |
>>> 2000 ++ .*. *..*... |
>>> *..*...*..*...*..*...*..*...*..O...*. O O O O O..O...O..O O O
>>> O O O O O O O O O |
>>> 1500 ++ |
>>> | |
>>> 1000 ++ |
>>> | |
>>> | |
>>> 500 ++ |
>>> | |
>>> | |
>>> 0 ++---------------------------------O---------------------------------+
>>>
>>>
>>> vmstat.system.in
>>>
>>> 40000 ++------------------------------------------------------------------+
>>> | .*...*.. |
>>> 35000 ++ .*...*. |
>>> 30000 *+.*...*..*...*..*..*...*..*...*..*. *..*...*..* |
>>> O O O O O O O O O O O O O O O O O O O O
>>> 25000 ++ |
>>> | |
>>> 20000 ++ |
>>> | |
>>> 15000 ++ |
>>> 10000 ++ |
>>> | |
>>> 5000 ++ |
>>> | |
>>> 0 ++--------------------------------O---------------------------------+
>>>
>>> [*] 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
>>

\
 
 \ /
  Last update: 2017-01-03 10:01    [W:0.121 / U:0.084 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site