| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <1422949668.13951.7.camel@intel.com>
Date: Tue, 03 Feb 2015 15:47:48 +0800
From: Huang Ying <ying.huang@...el.com>
To: Andy Lutomirski <luto@...capital.net>
Cc: LKML <linux-kernel@...r.kernel.org>, LKP ML <lkp@...org>
Subject: [LKP] [x86_64, entry] 103dd9885d1: +7.3%
will-it-scale.per_process_ops
FYI, we noticed the below changes on
git://git.kernel.org/pub/scm/linux/kernel/git/luto/linux.git x86/entry
commit 103dd9885d1e96be0c7493ce9a33af0732e35441 ("x86_64, entry: Use sysret to return to userspace when possible")
testbox/testcase/testparams: lkp-sb03/will-it-scale/open2
261ebee6aee55a47 103dd9885d1e96be0c7493ce9a
---------------- --------------------------
%stddev %change %stddev
\ | \
35.96 ± 6% -39.8% 21.66 ± 7% will-it-scale.time.user_time
237495 ± 1% +5.2% 249861 ± 1% will-it-scale.per_thread_ops
739981 ± 0% +7.3% 793911 ± 0% will-it-scale.per_process_ops
127 ± 20% -66.9% 42 ± 30% sched_debug.cfs_rq[14]:/.load
127 ± 20% -66.9% 42 ± 30% sched_debug.cpu#14.load
58 ± 21% -39.3% 35 ± 43% sched_debug.cfs_rq[7]:/.load
58 ± 21% -39.3% 35 ± 43% sched_debug.cpu#7.load
3442 ± 25% -42.5% 1978 ± 20% sched_debug.cpu#28.sched_goidle
3591 ± 36% +128.5% 8204 ± 17% sched_debug.cpu#10.ttwu_local
8241 ± 21% -36.9% 5202 ± 19% sched_debug.cpu#28.nr_switches
5411 ± 18% -39.0% 3301 ± 13% sched_debug.cpu#14.ttwu_local
8800 ± 15% -27.8% 6356 ± 20% sched_debug.cpu#9.ttwu_count
42 ± 22% -33.3% 28 ± 47% sched_debug.cfs_rq[14]:/.runnable_load_avg
67 ± 46% +87.1% 126 ± 31% sched_debug.cpu#12.load
573932 ± 1% +59.5% 915353 ± 1% slabinfo.kmalloc-256.active_objs
8970 ± 1% +59.5% 14305 ± 1% slabinfo.kmalloc-256.num_slabs
8970 ± 1% +59.5% 14305 ± 1% slabinfo.kmalloc-256.active_slabs
574152 ± 1% +59.5% 915575 ± 1% slabinfo.kmalloc-256.num_objs
35.96 ± 6% -39.8% 21.66 ± 7% time.user_time
1358 ± 24% -31.9% 924 ± 34% sched_debug.cpu#31.ttwu_local
2487 ± 14% -32.6% 1677 ± 9% sched_debug.cpu#31.sched_goidle
17643 ± 8% -33.6% 11719 ± 5% sched_debug.cpu#14.nr_switches
8632 ± 11% -34.9% 5617 ± 14% sched_debug.cpu#14.ttwu_count
29 ± 12% +31.9% 38 ± 15% sched_debug.cpu#12.cpu_load[1]
115453 ± 3% +48.3% 171178 ± 2% numa-meminfo.node0.SUnreclaim
28672 ± 2% +49.2% 42775 ± 2% numa-vmstat.node0.nr_slab_unreclaimable
3313 ± 15% +35.0% 4472 ± 13% sched_debug.cpu#11.curr->pid
132090 ± 3% +42.8% 188646 ± 2% numa-meminfo.node0.Slab
200693 ± 1% +42.8% 286535 ± 0% meminfo.SUnreclaim
50431 ± 1% +42.0% 71624 ± 1% proc-vmstat.nr_slab_unreclaimable
26 ± 12% +29.8% 33 ± 12% sched_debug.cpu#12.cpu_load[4]
20434 ± 2% +38.4% 28286 ± 2% numa-vmstat.node1.nr_slab_unreclaimable
82040 ± 2% +36.9% 112273 ± 2% numa-meminfo.node1.SUnreclaim
14623 ± 26% +55.6% 22751 ± 12% sched_debug.cpu#10.nr_switches
235487 ± 1% +36.5% 321407 ± 0% meminfo.Slab
842597 ± 20% -30.2% 588471 ± 26% sched_debug.cfs_rq[27]:/.min_vruntime
6502 ± 25% +54.2% 10028 ± 11% sched_debug.cpu#10.sched_goidle
100197 ± 2% +29.4% 129675 ± 2% numa-meminfo.node1.Slab
237178 ± 1% -23.2% 182260 ± 2% softirqs.RCU
7310 ± 4% -26.2% 5397 ± 6% sched_debug.cpu#14.sched_goidle
6148 ± 12% -22.7% 4751 ± 10% sched_debug.cpu#31.nr_switches
44905 ± 20% -28.5% 32101 ± 25% sched_debug.cfs_rq[27]:/.exec_clock
32 ± 6% +38.3% 44 ± 19% sched_debug.cpu#1.cpu_load[0]
139 ± 15% -25.4% 104 ± 18% sched_debug.cfs_rq[23]:/.tg_load_contrib
3454 ± 14% -17.7% 2844 ± 6% sched_debug.cpu#31.curr->pid
438 ± 8% -11.5% 388 ± 5% sched_debug.cfs_rq[31]:/.tg_runnable_contrib
30 ± 1% +11.6% 33 ± 6% sched_debug.cpu#23.cpu_load[4]
20115 ± 7% -11.3% 17841 ± 6% sched_debug.cfs_rq[31]:/.avg->runnable_avg_sum
35 ± 11% -19.0% 28 ± 9% sched_debug.cpu#21.cpu_load[0]
32 ± 6% +31.2% 42 ± 19% sched_debug.cfs_rq[1]:/.runnable_load_avg
32 ± 11% +27.1% 41 ± 10% sched_debug.cpu#9.cpu_load[2]
30 ± 1% +9.9% 33 ± 6% sched_debug.cpu#23.cpu_load[3]
575 ± 5% +12.4% 647 ± 6% sched_debug.cfs_rq[23]:/.tg_runnable_contrib
26412 ± 5% +12.5% 29710 ± 6% sched_debug.cfs_rq[23]:/.avg->runnable_avg_sum
29 ± 18% +33.6% 38 ± 14% sched_debug.cpu#12.cpu_load[0]
32 ± 5% +32.1% 43 ± 17% sched_debug.cpu#1.cpu_load[1]
33 ± 6% +26.3% 42 ± 16% sched_debug.cpu#1.cpu_load[2]
424232 ± 2% +13.7% 482375 ± 1% numa-meminfo.node0.MemUsed
3558 ± 6% +14.5% 4076 ± 3% sched_debug.cpu#15.curr->pid
1567 ± 0% -9.3% 1422 ± 1% vmstat.system.cs
testbox/testcase/testparams: lkp-sb03/will-it-scale/signal1
261ebee6aee55a47 103dd9885d1e96be0c7493ce9a
---------------- --------------------------
fail:runs %reproduction fail:runs
| | |
40.59 ± 3% -33.0% 27.21 ± 3% will-it-scale.time.user_time
0.11 ± 0% -10.5% 0.10 ± 0% will-it-scale.scalability
297467 ± 0% +10.2% 327822 ± 0% will-it-scale.per_thread_ops
526508 ± 0% +7.2% 564555 ± 0% will-it-scale.per_process_ops
2527 ± 43% -74.9% 633 ± 20% sched_debug.cpu#24.ttwu_local
3355 ± 33% -66.8% 1113 ± 14% sched_debug.cpu#24.ttwu_count
6776 ± 38% -61.6% 2601 ± 11% sched_debug.cpu#24.nr_switches
47 ± 44% -59.6% 19 ± 12% sched_debug.cfs_rq[10]:/.load
48 ± 40% -60.8% 19 ± 12% sched_debug.cpu#10.load
184 ± 33% -63.1% 68 ± 5% sched_debug.cfs_rq[1]:/.blocked_load_avg
80 ± 36% +97.8% 158 ± 34% sched_debug.cfs_rq[15]:/.blocked_load_avg
98 ± 30% +77.7% 175 ± 31% sched_debug.cfs_rq[15]:/.tg_load_contrib
215 ± 28% -54.3% 98 ± 3% sched_debug.cfs_rq[1]:/.tg_load_contrib
212 ± 32% -60.7% 83 ± 39% sched_debug.cfs_rq[26]:/.blocked_load_avg
238 ± 34% -56.0% 104 ± 35% sched_debug.cfs_rq[26]:/.tg_load_contrib
1175 ± 15% +82.0% 2139 ± 36% sched_debug.cpu#30.sched_goidle
40.59 ± 3% -33.0% 27.21 ± 3% time.user_time
183 ± 23% -32.1% 124 ± 27% sched_debug.cfs_rq[19]:/.tg_load_contrib
3095 ± 11% +81.1% 5606 ± 43% sched_debug.cpu#30.nr_switches
14 ± 5% +39.7% 20 ± 19% sched_debug.cpu#31.cpu_load[0]
1296 ± 11% +28.2% 1661 ± 15% sched_debug.cpu#5.ttwu_count
43 ± 9% +21.5% 52 ± 10% sched_debug.cpu#0.cpu_load[0]
59711 ± 0% -11.8% 52641 ± 7% sched_debug.cfs_rq[5]:/.exec_clock
29 ± 0% +13.8% 33 ± 7% sched_debug.cpu#21.cpu_load[0]
29 ± 6% -12.6% 26 ± 4% sched_debug.cfs_rq[5]:/.runnable_load_avg
30 ± 5% -13.2% 26 ± 6% sched_debug.cpu#5.load
29 ± 1% +12.0% 32 ± 6% sched_debug.cpu#21.cpu_load[2]
29 ± 0% +14.7% 33 ± 6% sched_debug.cpu#21.cpu_load[1]
15 ± 2% +24.6% 19 ± 18% sched_debug.cpu#31.cpu_load[1]
15 ± 2% +19.7% 18 ± 15% sched_debug.cpu#31.cpu_load[2]
30 ± 5% -12.4% 26 ± 6% sched_debug.cfs_rq[5]:/.load
29 ± 0% +13.8% 33 ± 7% sched_debug.cfs_rq[21]:/.runnable_load_avg
29 ± 0% +13.8% 33 ± 7% sched_debug.cfs_rq[21]:/.load
29 ± 0% +13.8% 33 ± 7% sched_debug.cpu#21.load
lkp-sb03: Sandy Bridge-EP
Memory: 64G
will-it-scale.time.user_time
45 ++---------------------------------------------------------------------+
| |
40 ++ .*.. .*.. .*.. |
35 *+.*..*.. .*..* *..*..*. *..*. *..*..*.. ..*..*. *.. .*
| *. : : *. *. |
30 ++ : : |
25 ++ : : |
O O O O O O: O :O O O O O O O O O O O |
20 ++ : : O |
15 ++ : : |
| : : |
10 ++ : : |
5 ++ : : |
| : |
0 ++-----------------*---------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
To reproduce:
apt-get install ruby ruby-oj
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/setup-local job.yaml # the job file attached in this email
bin/run-local job.yaml
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,
Huang, Ying
View attachment "job.yaml" of type "text/plain" (1687 bytes)
View attachment "reproduce" of type "text/plain" (40 bytes)
_______________________________________________
LKP mailing list
LKP@...ux.intel.com
Powered by blists - more mailing lists