All of lore.kernel.org
 help / color / mirror / Atom feed
* [net] 11ef7a8996d: +27.3% netperf.Throughput_Mbps
       [not found] <53d1f50b.CEZoGFBjN5Kgc+zm%fengguang.wu@intel.com>
@ 2014-07-25  6:31   ` Aaron Lu
  0 siblings, 0 replies; 12+ messages in thread
From: Aaron Lu @ 2014-07-25  6:31 UTC (permalink / raw)
  To: lkp

[-- Attachment #1: Type: text/plain, Size: 3726 bytes --]

FYI, we noticed the below changes on

commit 11ef7a8996d5d433c9cd75d80651297eccbf6d42 ("net: Performance fix for process_backlog")

test case: lkp-t410/netperf/300s-200%-10K-SCTP_STREAM_MANY

68b7107b62983f2  11ef7a8996d5d433c9cd75d80  
---------------  -------------------------  
      1023 ~ 3%     +27.3%       1302 ~ 0%  TOTAL netperf.Throughput_Mbps
      0.72 ~10%     -91.4%       0.06 ~23%  TOTAL turbostat.%c3
     13385 ~12%     -92.6%        987 ~12%  TOTAL cpuidle.C6-NHM.usage
     22745 ~10%     -95.5%       1016 ~40%  TOTAL cpuidle.C3-NHM.usage
  42675736 ~12%     -84.6%    6571705 ~19%  TOTAL cpuidle.C6-NHM.time
     88342 ~11%     -82.1%      15811 ~ 9%  TOTAL softirqs.SCHED
  19148006 ~12%     -95.7%     821873 ~27%  TOTAL cpuidle.C3-NHM.time
    439.94 ~10%     -77.5%      99.05 ~ 5%  TOTAL uptime.idle
      1.35 ~ 6%     -65.8%       0.46 ~24%  TOTAL turbostat.%c6
         4 ~23%    +114.3%          9 ~ 0%  TOTAL vmstat.procs.r
      1680 ~ 3%     +40.4%       2359 ~ 3%  TOTAL proc-vmstat.nr_alloc_batch
    447921 ~ 4%     +36.2%     610047 ~ 0%  TOTAL softirqs.TIMER
      9.09 ~ 9%     +27.6%      11.60 ~ 8%  TOTAL perf-profile.cpu-cycles.memcpy.sctp_outq_flush.sctp_outq_uncork.sctp_cmd_interpreter.sctp_do_sm
   2350916 ~ 4%     +35.4%    3184088 ~ 0%  TOTAL proc-vmstat.pgalloc_dma
      7.82 ~ 9%     +24.1%       9.71 ~ 4%  TOTAL perf-profile.cpu-cycles.copy_user_generic_string.sctp_user_addto_chunk.sctp_datamsg_from_user.sctp_sendmsg.inet_sendmsg
  38837537 ~ 3%     +27.1%   49358639 ~ 0%  TOTAL proc-vmstat.numa_local
  38837537 ~ 3%     +27.1%   49358639 ~ 0%  TOTAL proc-vmstat.numa_hit
     50216 ~ 1%     +17.0%      58745 ~ 6%  TOTAL softirqs.RCU
      1.41 ~ 4%     +11.6%       1.58 ~ 1%  TOTAL perf-profile.cpu-cycles.get_page_from_freelist.__alloc_pages_nodemask.alloc_kmem_pages_node.kmalloc_large_node.__kmalloc_node_track_caller
     48539 ~ 1%     -25.5%      36171 ~ 1%  TOTAL vmstat.system.cs
     75.64 ~ 4%     +30.6%      98.82 ~ 0%  TOTAL turbostat.%c0
      3949 ~ 2%     +13.4%       4478 ~ 0%  TOTAL vmstat.system.in

Legend:
	~XX%    - stddev percent
	[+-]XX% - change percent


                                  vmstat.system.cs

  52000 ++------------------------------------------------------------------+
  50000 ++  *      .*   *.                   **        *. *  *.             |
        | .* + *.**  *. : **.* .* .**.**. * +  +    *.*  * + : * .**. *.  .**
  48000 **    *        *      *  *       * *    * .*        *   *    *  **  |
  46000 ++                                       *                          |
        |                                                                   |
  44000 ++                                                                  |
  42000 ++                                                                  |
  40000 ++                                                                  |
        |                                                                   |
  38000 +O  O                      O                                        |
  36000 O+    O  O   O  O OO OO OO  O  O   O  O  O OO  O OO O               |
        |  O   O  O O  O                  O           O                     |
  34000 ++                            O  O   O  O                           |
  32000 ++------------------------------------------------------------------+


	[*] 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,
Aaron

[-- Attachment #2: reproduce.ksh --]
[-- Type: text/plain, Size: 711 bytes --]

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/cpu2/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu3/cpufreq/scaling_governor
netserver
netperf -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K
netperf -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K
netperf -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K
netperf -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K
netperf -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K
netperf -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K
netperf -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K
netperf -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K


^ permalink raw reply	[flat|nested] 12+ messages in thread

* [LKP] [net] 11ef7a8996d: +27.3% netperf.Throughput_Mbps
@ 2014-07-25  6:31   ` Aaron Lu
  0 siblings, 0 replies; 12+ messages in thread
From: Aaron Lu @ 2014-07-25  6:31 UTC (permalink / raw)
  To: Tom Herbert; +Cc: David S. Miller, LKML, lkp

[-- Attachment #1: Type: text/plain, Size: 3660 bytes --]

FYI, we noticed the below changes on

commit 11ef7a8996d5d433c9cd75d80651297eccbf6d42 ("net: Performance fix for process_backlog")

test case: lkp-t410/netperf/300s-200%-10K-SCTP_STREAM_MANY

68b7107b62983f2  11ef7a8996d5d433c9cd75d80  
---------------  -------------------------  
      1023 ~ 3%     +27.3%       1302 ~ 0%  TOTAL netperf.Throughput_Mbps
      0.72 ~10%     -91.4%       0.06 ~23%  TOTAL turbostat.%c3
     13385 ~12%     -92.6%        987 ~12%  TOTAL cpuidle.C6-NHM.usage
     22745 ~10%     -95.5%       1016 ~40%  TOTAL cpuidle.C3-NHM.usage
  42675736 ~12%     -84.6%    6571705 ~19%  TOTAL cpuidle.C6-NHM.time
     88342 ~11%     -82.1%      15811 ~ 9%  TOTAL softirqs.SCHED
  19148006 ~12%     -95.7%     821873 ~27%  TOTAL cpuidle.C3-NHM.time
    439.94 ~10%     -77.5%      99.05 ~ 5%  TOTAL uptime.idle
      1.35 ~ 6%     -65.8%       0.46 ~24%  TOTAL turbostat.%c6
         4 ~23%    +114.3%          9 ~ 0%  TOTAL vmstat.procs.r
      1680 ~ 3%     +40.4%       2359 ~ 3%  TOTAL proc-vmstat.nr_alloc_batch
    447921 ~ 4%     +36.2%     610047 ~ 0%  TOTAL softirqs.TIMER
      9.09 ~ 9%     +27.6%      11.60 ~ 8%  TOTAL perf-profile.cpu-cycles.memcpy.sctp_outq_flush.sctp_outq_uncork.sctp_cmd_interpreter.sctp_do_sm
   2350916 ~ 4%     +35.4%    3184088 ~ 0%  TOTAL proc-vmstat.pgalloc_dma
      7.82 ~ 9%     +24.1%       9.71 ~ 4%  TOTAL perf-profile.cpu-cycles.copy_user_generic_string.sctp_user_addto_chunk.sctp_datamsg_from_user.sctp_sendmsg.inet_sendmsg
  38837537 ~ 3%     +27.1%   49358639 ~ 0%  TOTAL proc-vmstat.numa_local
  38837537 ~ 3%     +27.1%   49358639 ~ 0%  TOTAL proc-vmstat.numa_hit
     50216 ~ 1%     +17.0%      58745 ~ 6%  TOTAL softirqs.RCU
      1.41 ~ 4%     +11.6%       1.58 ~ 1%  TOTAL perf-profile.cpu-cycles.get_page_from_freelist.__alloc_pages_nodemask.alloc_kmem_pages_node.kmalloc_large_node.__kmalloc_node_track_caller
     48539 ~ 1%     -25.5%      36171 ~ 1%  TOTAL vmstat.system.cs
     75.64 ~ 4%     +30.6%      98.82 ~ 0%  TOTAL turbostat.%c0
      3949 ~ 2%     +13.4%       4478 ~ 0%  TOTAL vmstat.system.in

Legend:
	~XX%    - stddev percent
	[+-]XX% - change percent


                                  vmstat.system.cs

  52000 ++------------------------------------------------------------------+
  50000 ++  *      .*   *.                   **        *. *  *.             |
        | .* + *.**  *. : **.* .* .**.**. * +  +    *.*  * + : * .**. *.  .**
  48000 **    *        *      *  *       * *    * .*        *   *    *  **  |
  46000 ++                                       *                          |
        |                                                                   |
  44000 ++                                                                  |
  42000 ++                                                                  |
  40000 ++                                                                  |
        |                                                                   |
  38000 +O  O                      O                                        |
  36000 O+    O  O   O  O OO OO OO  O  O   O  O  O OO  O OO O               |
        |  O   O  O O  O                  O           O                     |
  34000 ++                            O  O   O  O                           |
  32000 ++------------------------------------------------------------------+


	[*] 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,
Aaron

[-- Attachment #2: reproduce --]
[-- Type: text/plain, Size: 711 bytes --]

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/cpu2/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu3/cpufreq/scaling_governor
netserver
netperf -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K
netperf -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K
netperf -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K
netperf -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K
netperf -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K
netperf -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K
netperf -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K
netperf -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K


^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [net] 11ef7a8996d: +27.3% netperf.Throughput_Mbps
  2014-07-25  6:31   ` [LKP] " Aaron Lu
@ 2014-07-25  6:37     ` Tom Herbert
  -1 siblings, 0 replies; 12+ messages in thread
From: Tom Herbert @ 2014-07-25  6:37 UTC (permalink / raw)
  To: lkp

[-- Attachment #1: Type: text/plain, Size: 3989 bytes --]

Are you pointing out a regressions in this?

On Thu, Jul 24, 2014 at 11:31 PM, Aaron Lu <aaron.lu@intel.com> wrote:
> FYI, we noticed the below changes on
>
> commit 11ef7a8996d5d433c9cd75d80651297eccbf6d42 ("net: Performance fix for process_backlog")
>
> test case: lkp-t410/netperf/300s-200%-10K-SCTP_STREAM_MANY
>
> 68b7107b62983f2  11ef7a8996d5d433c9cd75d80
> ---------------  -------------------------
>       1023 ~ 3%     +27.3%       1302 ~ 0%  TOTAL netperf.Throughput_Mbps
>       0.72 ~10%     -91.4%       0.06 ~23%  TOTAL turbostat.%c3
>      13385 ~12%     -92.6%        987 ~12%  TOTAL cpuidle.C6-NHM.usage
>      22745 ~10%     -95.5%       1016 ~40%  TOTAL cpuidle.C3-NHM.usage
>   42675736 ~12%     -84.6%    6571705 ~19%  TOTAL cpuidle.C6-NHM.time
>      88342 ~11%     -82.1%      15811 ~ 9%  TOTAL softirqs.SCHED
>   19148006 ~12%     -95.7%     821873 ~27%  TOTAL cpuidle.C3-NHM.time
>     439.94 ~10%     -77.5%      99.05 ~ 5%  TOTAL uptime.idle
>       1.35 ~ 6%     -65.8%       0.46 ~24%  TOTAL turbostat.%c6
>          4 ~23%    +114.3%          9 ~ 0%  TOTAL vmstat.procs.r
>       1680 ~ 3%     +40.4%       2359 ~ 3%  TOTAL proc-vmstat.nr_alloc_batch
>     447921 ~ 4%     +36.2%     610047 ~ 0%  TOTAL softirqs.TIMER
>       9.09 ~ 9%     +27.6%      11.60 ~ 8%  TOTAL perf-profile.cpu-cycles.memcpy.sctp_outq_flush.sctp_outq_uncork.sctp_cmd_interpreter.sctp_do_sm
>    2350916 ~ 4%     +35.4%    3184088 ~ 0%  TOTAL proc-vmstat.pgalloc_dma
>       7.82 ~ 9%     +24.1%       9.71 ~ 4%  TOTAL perf-profile.cpu-cycles.copy_user_generic_string.sctp_user_addto_chunk.sctp_datamsg_from_user.sctp_sendmsg.inet_sendmsg
>   38837537 ~ 3%     +27.1%   49358639 ~ 0%  TOTAL proc-vmstat.numa_local
>   38837537 ~ 3%     +27.1%   49358639 ~ 0%  TOTAL proc-vmstat.numa_hit
>      50216 ~ 1%     +17.0%      58745 ~ 6%  TOTAL softirqs.RCU
>       1.41 ~ 4%     +11.6%       1.58 ~ 1%  TOTAL perf-profile.cpu-cycles.get_page_from_freelist.__alloc_pages_nodemask.alloc_kmem_pages_node.kmalloc_large_node.__kmalloc_node_track_caller
>      48539 ~ 1%     -25.5%      36171 ~ 1%  TOTAL vmstat.system.cs
>      75.64 ~ 4%     +30.6%      98.82 ~ 0%  TOTAL turbostat.%c0
>       3949 ~ 2%     +13.4%       4478 ~ 0%  TOTAL vmstat.system.in
>
> Legend:
>         ~XX%    - stddev percent
>         [+-]XX% - change percent
>
>
>                                   vmstat.system.cs
>
>   52000 ++------------------------------------------------------------------+
>   50000 ++  *      .*   *.                   **        *. *  *.             |
>         | .* + *.**  *. : **.* .* .**.**. * +  +    *.*  * + : * .**. *.  .**
>   48000 **    *        *      *  *       * *    * .*        *   *    *  **  |
>   46000 ++                                       *                          |
>         |                                                                   |
>   44000 ++                                                                  |
>   42000 ++                                                                  |
>   40000 ++                                                                  |
>         |                                                                   |
>   38000 +O  O                      O                                        |
>   36000 O+    O  O   O  O OO OO OO  O  O   O  O  O OO  O OO O               |
>         |  O   O  O O  O                  O           O                     |
>   34000 ++                            O  O   O  O                           |
>   32000 ++------------------------------------------------------------------+
>
>
>         [*] 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,
> Aaron

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [LKP] [net] 11ef7a8996d: +27.3% netperf.Throughput_Mbps
@ 2014-07-25  6:37     ` Tom Herbert
  0 siblings, 0 replies; 12+ messages in thread
From: Tom Herbert @ 2014-07-25  6:37 UTC (permalink / raw)
  To: Aaron Lu; +Cc: David S. Miller, LKML, lkp

Are you pointing out a regressions in this?

On Thu, Jul 24, 2014 at 11:31 PM, Aaron Lu <aaron.lu@intel.com> wrote:
> FYI, we noticed the below changes on
>
> commit 11ef7a8996d5d433c9cd75d80651297eccbf6d42 ("net: Performance fix for process_backlog")
>
> test case: lkp-t410/netperf/300s-200%-10K-SCTP_STREAM_MANY
>
> 68b7107b62983f2  11ef7a8996d5d433c9cd75d80
> ---------------  -------------------------
>       1023 ~ 3%     +27.3%       1302 ~ 0%  TOTAL netperf.Throughput_Mbps
>       0.72 ~10%     -91.4%       0.06 ~23%  TOTAL turbostat.%c3
>      13385 ~12%     -92.6%        987 ~12%  TOTAL cpuidle.C6-NHM.usage
>      22745 ~10%     -95.5%       1016 ~40%  TOTAL cpuidle.C3-NHM.usage
>   42675736 ~12%     -84.6%    6571705 ~19%  TOTAL cpuidle.C6-NHM.time
>      88342 ~11%     -82.1%      15811 ~ 9%  TOTAL softirqs.SCHED
>   19148006 ~12%     -95.7%     821873 ~27%  TOTAL cpuidle.C3-NHM.time
>     439.94 ~10%     -77.5%      99.05 ~ 5%  TOTAL uptime.idle
>       1.35 ~ 6%     -65.8%       0.46 ~24%  TOTAL turbostat.%c6
>          4 ~23%    +114.3%          9 ~ 0%  TOTAL vmstat.procs.r
>       1680 ~ 3%     +40.4%       2359 ~ 3%  TOTAL proc-vmstat.nr_alloc_batch
>     447921 ~ 4%     +36.2%     610047 ~ 0%  TOTAL softirqs.TIMER
>       9.09 ~ 9%     +27.6%      11.60 ~ 8%  TOTAL perf-profile.cpu-cycles.memcpy.sctp_outq_flush.sctp_outq_uncork.sctp_cmd_interpreter.sctp_do_sm
>    2350916 ~ 4%     +35.4%    3184088 ~ 0%  TOTAL proc-vmstat.pgalloc_dma
>       7.82 ~ 9%     +24.1%       9.71 ~ 4%  TOTAL perf-profile.cpu-cycles.copy_user_generic_string.sctp_user_addto_chunk.sctp_datamsg_from_user.sctp_sendmsg.inet_sendmsg
>   38837537 ~ 3%     +27.1%   49358639 ~ 0%  TOTAL proc-vmstat.numa_local
>   38837537 ~ 3%     +27.1%   49358639 ~ 0%  TOTAL proc-vmstat.numa_hit
>      50216 ~ 1%     +17.0%      58745 ~ 6%  TOTAL softirqs.RCU
>       1.41 ~ 4%     +11.6%       1.58 ~ 1%  TOTAL perf-profile.cpu-cycles.get_page_from_freelist.__alloc_pages_nodemask.alloc_kmem_pages_node.kmalloc_large_node.__kmalloc_node_track_caller
>      48539 ~ 1%     -25.5%      36171 ~ 1%  TOTAL vmstat.system.cs
>      75.64 ~ 4%     +30.6%      98.82 ~ 0%  TOTAL turbostat.%c0
>       3949 ~ 2%     +13.4%       4478 ~ 0%  TOTAL vmstat.system.in
>
> Legend:
>         ~XX%    - stddev percent
>         [+-]XX% - change percent
>
>
>                                   vmstat.system.cs
>
>   52000 ++------------------------------------------------------------------+
>   50000 ++  *      .*   *.                   **        *. *  *.             |
>         | .* + *.**  *. : **.* .* .**.**. * +  +    *.*  * + : * .**. *.  .**
>   48000 **    *        *      *  *       * *    * .*        *   *    *  **  |
>   46000 ++                                       *                          |
>         |                                                                   |
>   44000 ++                                                                  |
>   42000 ++                                                                  |
>   40000 ++                                                                  |
>         |                                                                   |
>   38000 +O  O                      O                                        |
>   36000 O+    O  O   O  O OO OO OO  O  O   O  O  O OO  O OO O               |
>         |  O   O  O O  O                  O           O                     |
>   34000 ++                            O  O   O  O                           |
>   32000 ++------------------------------------------------------------------+
>
>
>         [*] 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,
> Aaron

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [net] 11ef7a8996d: +27.3% netperf.Throughput_Mbps
  2014-07-25  6:37     ` [LKP] " Tom Herbert
@ 2014-07-25  6:44       ` Aaron Lu
  -1 siblings, 0 replies; 12+ messages in thread
From: Aaron Lu @ 2014-07-25  6:44 UTC (permalink / raw)
  To: lkp

[-- Attachment #1: Type: text/plain, Size: 4265 bytes --]

On 07/25/2014 02:37 PM, Tom Herbert wrote:
> Are you pointing out a regressions in this?

No, the report is an indication your commit makes the Throughput_Mbps of
the netperf benchamrk increased 27.3%, a good thing I think.

Thanks,
Aaron

> 
> On Thu, Jul 24, 2014 at 11:31 PM, Aaron Lu <aaron.lu@intel.com> wrote:
>> FYI, we noticed the below changes on
>>
>> commit 11ef7a8996d5d433c9cd75d80651297eccbf6d42 ("net: Performance fix for process_backlog")
>>
>> test case: lkp-t410/netperf/300s-200%-10K-SCTP_STREAM_MANY
>>
>> 68b7107b62983f2  11ef7a8996d5d433c9cd75d80
>> ---------------  -------------------------
>>       1023 ~ 3%     +27.3%       1302 ~ 0%  TOTAL netperf.Throughput_Mbps
>>       0.72 ~10%     -91.4%       0.06 ~23%  TOTAL turbostat.%c3
>>      13385 ~12%     -92.6%        987 ~12%  TOTAL cpuidle.C6-NHM.usage
>>      22745 ~10%     -95.5%       1016 ~40%  TOTAL cpuidle.C3-NHM.usage
>>   42675736 ~12%     -84.6%    6571705 ~19%  TOTAL cpuidle.C6-NHM.time
>>      88342 ~11%     -82.1%      15811 ~ 9%  TOTAL softirqs.SCHED
>>   19148006 ~12%     -95.7%     821873 ~27%  TOTAL cpuidle.C3-NHM.time
>>     439.94 ~10%     -77.5%      99.05 ~ 5%  TOTAL uptime.idle
>>       1.35 ~ 6%     -65.8%       0.46 ~24%  TOTAL turbostat.%c6
>>          4 ~23%    +114.3%          9 ~ 0%  TOTAL vmstat.procs.r
>>       1680 ~ 3%     +40.4%       2359 ~ 3%  TOTAL proc-vmstat.nr_alloc_batch
>>     447921 ~ 4%     +36.2%     610047 ~ 0%  TOTAL softirqs.TIMER
>>       9.09 ~ 9%     +27.6%      11.60 ~ 8%  TOTAL perf-profile.cpu-cycles.memcpy.sctp_outq_flush.sctp_outq_uncork.sctp_cmd_interpreter.sctp_do_sm
>>    2350916 ~ 4%     +35.4%    3184088 ~ 0%  TOTAL proc-vmstat.pgalloc_dma
>>       7.82 ~ 9%     +24.1%       9.71 ~ 4%  TOTAL perf-profile.cpu-cycles.copy_user_generic_string.sctp_user_addto_chunk.sctp_datamsg_from_user.sctp_sendmsg.inet_sendmsg
>>   38837537 ~ 3%     +27.1%   49358639 ~ 0%  TOTAL proc-vmstat.numa_local
>>   38837537 ~ 3%     +27.1%   49358639 ~ 0%  TOTAL proc-vmstat.numa_hit
>>      50216 ~ 1%     +17.0%      58745 ~ 6%  TOTAL softirqs.RCU
>>       1.41 ~ 4%     +11.6%       1.58 ~ 1%  TOTAL perf-profile.cpu-cycles.get_page_from_freelist.__alloc_pages_nodemask.alloc_kmem_pages_node.kmalloc_large_node.__kmalloc_node_track_caller
>>      48539 ~ 1%     -25.5%      36171 ~ 1%  TOTAL vmstat.system.cs
>>      75.64 ~ 4%     +30.6%      98.82 ~ 0%  TOTAL turbostat.%c0
>>       3949 ~ 2%     +13.4%       4478 ~ 0%  TOTAL vmstat.system.in
>>
>> Legend:
>>         ~XX%    - stddev percent
>>         [+-]XX% - change percent
>>
>>
>>                                   vmstat.system.cs
>>
>>   52000 ++------------------------------------------------------------------+
>>   50000 ++  *      .*   *.                   **        *. *  *.             |
>>         | .* + *.**  *. : **.* .* .**.**. * +  +    *.*  * + : * .**. *.  .**
>>   48000 **    *        *      *  *       * *    * .*        *   *    *  **  |
>>   46000 ++                                       *                          |
>>         |                                                                   |
>>   44000 ++                                                                  |
>>   42000 ++                                                                  |
>>   40000 ++                                                                  |
>>         |                                                                   |
>>   38000 +O  O                      O                                        |
>>   36000 O+    O  O   O  O OO OO OO  O  O   O  O  O OO  O OO O               |
>>         |  O   O  O O  O                  O           O                     |
>>   34000 ++                            O  O   O  O                           |
>>   32000 ++------------------------------------------------------------------+
>>
>>
>>         [*] 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,
>> Aaron


^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [LKP] [net] 11ef7a8996d: +27.3% netperf.Throughput_Mbps
@ 2014-07-25  6:44       ` Aaron Lu
  0 siblings, 0 replies; 12+ messages in thread
From: Aaron Lu @ 2014-07-25  6:44 UTC (permalink / raw)
  To: Tom Herbert; +Cc: David S. Miller, LKML, lkp

On 07/25/2014 02:37 PM, Tom Herbert wrote:
> Are you pointing out a regressions in this?

No, the report is an indication your commit makes the Throughput_Mbps of
the netperf benchamrk increased 27.3%, a good thing I think.

Thanks,
Aaron

> 
> On Thu, Jul 24, 2014 at 11:31 PM, Aaron Lu <aaron.lu@intel.com> wrote:
>> FYI, we noticed the below changes on
>>
>> commit 11ef7a8996d5d433c9cd75d80651297eccbf6d42 ("net: Performance fix for process_backlog")
>>
>> test case: lkp-t410/netperf/300s-200%-10K-SCTP_STREAM_MANY
>>
>> 68b7107b62983f2  11ef7a8996d5d433c9cd75d80
>> ---------------  -------------------------
>>       1023 ~ 3%     +27.3%       1302 ~ 0%  TOTAL netperf.Throughput_Mbps
>>       0.72 ~10%     -91.4%       0.06 ~23%  TOTAL turbostat.%c3
>>      13385 ~12%     -92.6%        987 ~12%  TOTAL cpuidle.C6-NHM.usage
>>      22745 ~10%     -95.5%       1016 ~40%  TOTAL cpuidle.C3-NHM.usage
>>   42675736 ~12%     -84.6%    6571705 ~19%  TOTAL cpuidle.C6-NHM.time
>>      88342 ~11%     -82.1%      15811 ~ 9%  TOTAL softirqs.SCHED
>>   19148006 ~12%     -95.7%     821873 ~27%  TOTAL cpuidle.C3-NHM.time
>>     439.94 ~10%     -77.5%      99.05 ~ 5%  TOTAL uptime.idle
>>       1.35 ~ 6%     -65.8%       0.46 ~24%  TOTAL turbostat.%c6
>>          4 ~23%    +114.3%          9 ~ 0%  TOTAL vmstat.procs.r
>>       1680 ~ 3%     +40.4%       2359 ~ 3%  TOTAL proc-vmstat.nr_alloc_batch
>>     447921 ~ 4%     +36.2%     610047 ~ 0%  TOTAL softirqs.TIMER
>>       9.09 ~ 9%     +27.6%      11.60 ~ 8%  TOTAL perf-profile.cpu-cycles.memcpy.sctp_outq_flush.sctp_outq_uncork.sctp_cmd_interpreter.sctp_do_sm
>>    2350916 ~ 4%     +35.4%    3184088 ~ 0%  TOTAL proc-vmstat.pgalloc_dma
>>       7.82 ~ 9%     +24.1%       9.71 ~ 4%  TOTAL perf-profile.cpu-cycles.copy_user_generic_string.sctp_user_addto_chunk.sctp_datamsg_from_user.sctp_sendmsg.inet_sendmsg
>>   38837537 ~ 3%     +27.1%   49358639 ~ 0%  TOTAL proc-vmstat.numa_local
>>   38837537 ~ 3%     +27.1%   49358639 ~ 0%  TOTAL proc-vmstat.numa_hit
>>      50216 ~ 1%     +17.0%      58745 ~ 6%  TOTAL softirqs.RCU
>>       1.41 ~ 4%     +11.6%       1.58 ~ 1%  TOTAL perf-profile.cpu-cycles.get_page_from_freelist.__alloc_pages_nodemask.alloc_kmem_pages_node.kmalloc_large_node.__kmalloc_node_track_caller
>>      48539 ~ 1%     -25.5%      36171 ~ 1%  TOTAL vmstat.system.cs
>>      75.64 ~ 4%     +30.6%      98.82 ~ 0%  TOTAL turbostat.%c0
>>       3949 ~ 2%     +13.4%       4478 ~ 0%  TOTAL vmstat.system.in
>>
>> Legend:
>>         ~XX%    - stddev percent
>>         [+-]XX% - change percent
>>
>>
>>                                   vmstat.system.cs
>>
>>   52000 ++------------------------------------------------------------------+
>>   50000 ++  *      .*   *.                   **        *. *  *.             |
>>         | .* + *.**  *. : **.* .* .**.**. * +  +    *.*  * + : * .**. *.  .**
>>   48000 **    *        *      *  *       * *    * .*        *   *    *  **  |
>>   46000 ++                                       *                          |
>>         |                                                                   |
>>   44000 ++                                                                  |
>>   42000 ++                                                                  |
>>   40000 ++                                                                  |
>>         |                                                                   |
>>   38000 +O  O                      O                                        |
>>   36000 O+    O  O   O  O OO OO OO  O  O   O  O  O OO  O OO O               |
>>         |  O   O  O O  O                  O           O                     |
>>   34000 ++                            O  O   O  O                           |
>>   32000 ++------------------------------------------------------------------+
>>
>>
>>         [*] 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,
>> Aaron


^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [net] 11ef7a8996d: +27.3% netperf.Throughput_Mbps
  2014-07-25  6:37     ` [LKP] " Tom Herbert
@ 2014-07-25  6:48       ` David Miller
  -1 siblings, 0 replies; 12+ messages in thread
From: David Miller @ 2014-07-25  6:48 UTC (permalink / raw)
  To: lkp

[-- Attachment #1: Type: text/plain, Size: 204 bytes --]

From: Tom Herbert <therbert@google.com>
Date: Thu, 24 Jul 2014 23:37:07 -0700

> Are you pointing out a regressions in this?

I think he's letting you know that performance has improved, in fact :)

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [LKP] [net] 11ef7a8996d: +27.3% netperf.Throughput_Mbps
@ 2014-07-25  6:48       ` David Miller
  0 siblings, 0 replies; 12+ messages in thread
From: David Miller @ 2014-07-25  6:48 UTC (permalink / raw)
  To: therbert; +Cc: aaron.lu, linux-kernel, lkp

From: Tom Herbert <therbert@google.com>
Date: Thu, 24 Jul 2014 23:37:07 -0700

> Are you pointing out a regressions in this?

I think he's letting you know that performance has improved, in fact :)

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [net] 11ef7a8996d: +27.3% netperf.Throughput_Mbps
  2014-07-25  6:48       ` [LKP] " David Miller
@ 2014-07-25  6:49         ` Aaron Lu
  -1 siblings, 0 replies; 12+ messages in thread
From: Aaron Lu @ 2014-07-25  6:49 UTC (permalink / raw)
  To: lkp

[-- Attachment #1: Type: text/plain, Size: 292 bytes --]

On 07/25/2014 02:48 PM, David Miller wrote:
> From: Tom Herbert <therbert@google.com>
> Date: Thu, 24 Jul 2014 23:37:07 -0700
> 
>> Are you pointing out a regressions in this?
> 
> I think he's letting you know that performance has improved, in fact :)

Yes, exactly :-)

-Aaron


^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [LKP] [net] 11ef7a8996d: +27.3% netperf.Throughput_Mbps
@ 2014-07-25  6:49         ` Aaron Lu
  0 siblings, 0 replies; 12+ messages in thread
From: Aaron Lu @ 2014-07-25  6:49 UTC (permalink / raw)
  To: David Miller, therbert; +Cc: linux-kernel, lkp

On 07/25/2014 02:48 PM, David Miller wrote:
> From: Tom Herbert <therbert@google.com>
> Date: Thu, 24 Jul 2014 23:37:07 -0700
> 
>> Are you pointing out a regressions in this?
> 
> I think he's letting you know that performance has improved, in fact :)

Yes, exactly :-)

-Aaron


^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [net] 11ef7a8996d: +27.3% netperf.Throughput_Mbps
  2014-07-25  6:49         ` [LKP] " Aaron Lu
@ 2014-07-25  7:16           ` Tom Herbert
  -1 siblings, 0 replies; 12+ messages in thread
From: Tom Herbert @ 2014-07-25  7:16 UTC (permalink / raw)
  To: lkp

[-- Attachment #1: Type: text/plain, Size: 423 bytes --]

On Thu, Jul 24, 2014 at 11:49 PM, Aaron Lu <aaron.lu@intel.com> wrote:
> On 07/25/2014 02:48 PM, David Miller wrote:
>> From: Tom Herbert <therbert@google.com>
>> Date: Thu, 24 Jul 2014 23:37:07 -0700
>>
>>> Are you pointing out a regressions in this?
>>
>> I think he's letting you know that performance has improved, in fact :)
>
> Yes, exactly :-)
>
Very nice, thanks for posting the results!

> -Aaron
>

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [LKP] [net] 11ef7a8996d: +27.3% netperf.Throughput_Mbps
@ 2014-07-25  7:16           ` Tom Herbert
  0 siblings, 0 replies; 12+ messages in thread
From: Tom Herbert @ 2014-07-25  7:16 UTC (permalink / raw)
  To: Aaron Lu; +Cc: David Miller, LKML, lkp

On Thu, Jul 24, 2014 at 11:49 PM, Aaron Lu <aaron.lu@intel.com> wrote:
> On 07/25/2014 02:48 PM, David Miller wrote:
>> From: Tom Herbert <therbert@google.com>
>> Date: Thu, 24 Jul 2014 23:37:07 -0700
>>
>>> Are you pointing out a regressions in this?
>>
>> I think he's letting you know that performance has improved, in fact :)
>
> Yes, exactly :-)
>
Very nice, thanks for posting the results!

> -Aaron
>

^ permalink raw reply	[flat|nested] 12+ messages in thread

end of thread, other threads:[~2014-07-25  7:16 UTC | newest]

Thread overview: 12+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
     [not found] <53d1f50b.CEZoGFBjN5Kgc+zm%fengguang.wu@intel.com>
2014-07-25  6:31 ` [net] 11ef7a8996d: +27.3% netperf.Throughput_Mbps Aaron Lu
2014-07-25  6:31   ` [LKP] " Aaron Lu
2014-07-25  6:37   ` Tom Herbert
2014-07-25  6:37     ` [LKP] " Tom Herbert
2014-07-25  6:44     ` Aaron Lu
2014-07-25  6:44       ` [LKP] " Aaron Lu
2014-07-25  6:48     ` David Miller
2014-07-25  6:48       ` [LKP] " David Miller
2014-07-25  6:49       ` Aaron Lu
2014-07-25  6:49         ` [LKP] " Aaron Lu
2014-07-25  7:16         ` Tom Herbert
2014-07-25  7:16           ` [LKP] " Tom Herbert

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.