| 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: <4354d3270705181407t6e2d6der396ea3d0fd06f790@mail.gmail.com>
Date: Sat, 19 May 2007 00:07:28 +0300
From: "Török Edvin" <edwintorok@...il.com>
To: "Ingo Molnar" <mingo@...e.hu>
Cc: mingo@...hat.com, linux-kernel@...r.kernel.org
Subject: Re: CFS Scheduler and real-time tasks
On 5/17/07, Ingo Molnar <mingo@...e.hu> wrote:
> hey, cool! Also try -v13 - it should be even more tighter.
I tried -v13. However the scheduling "error" is now 10% (vs 2% with -v12).
I also noticed strange behaviour with CPU hotplug. I offlined cpu1
(echo 0 >/sys/devices/system/cpu/cpu1/online), and the typing speed on
my terminal decreased noticably. I could hardly type something there.
The system was otherwise idle (running in Xorg, and gnome-terminal).
As soon as I started massive_intr.c workload, the keyboard typing
speed became normal again. If I have a workload running, while I
offline a CPU the keyboard is fine after cpu1 is offlined.
Also the entire system froze once, when I offlined cpu1. I couldn't
reproduce this.
I don't know how -v12 behaved wrt to CPU hotplug.
> > ./massiv_new 9 300
> > 020860 00017782
> > 020858 00017774
> >
> > RT JOB:020865 00003785
> >
> > 020859 00017048
> > 020862 00017387
> > 020855 00017494
> > 020861 00017245
> > 020863 00017492
> > 020864 00017122
Results with -v13:
edwin@...alhost ~ $ ./mo 9 300
006203 00006071
006198 00005712
006201 00006088
006196 00006208
006200 00006166
006195 00006306
006199 00006403
006197 00006288
006202 00006237
./massiv_new 9 300
013803 00010073
013801 00009889
013804 00010303
013806 00009953
013800 00010010
013802 00010171
013807 00009956
013805 00010037
RT JOB:013808 00002230
Interesting, massiv_new has better precision than massiv_orig.
Just in case it helps, here's me /proc/sched_debug (when system is
idle, and later, when running the test):
Sched Debug Version: v0.02
now at 1187357967816 nsecs
cpu: 0
.nr_running : 0
.raw_weighted_load : 0
.nr_switches : 555712
.nr_load_updates : 221754
.nr_uninterruptible : 4294967190
.jiffies : 221839
.next_balance : 221835
.curr->pid : 0
.clock : 755701854416
.prev_clock_raw : 871997978419
.clock_warps : 0
.clock_unstable_events : 97254
.clock_max_delta : 4749235
.fair_clock : 184031607224
.prev_fair_clock : 0
.exec_clock : 749010316890
.prev_exec_clock : 0
.wait_runtime : 59753136705
.wait_runtime_overruns : 2689
.wait_runtime_underruns: 1141
.cpu_load[0] : 0
.cpu_load[1] : 0
.cpu_load[2] : 0
.cpu_load[3] : 0
.cpu_load[4] : 0
.wait_runtime_rq_sum : 0
runnable tasks:
task PID tree-key delta waiting
switches prio sum-exec sum-wait sum-sleep
wait-overrun wait-underrun
------------------------------------------------------------------------------------------------------------------------------------------------------------------
cpu: 1
.nr_running : 1
.raw_weighted_load : 1024
.nr_switches : 378755
.nr_load_updates : 202398
.nr_uninterruptible : 106
.jiffies : 221839
.next_balance : 221830
.curr->pid : 20421
.clock : 794219468967
.prev_clock_raw : 874002443396
.clock_warps : 0
.clock_unstable_events : 116368
.clock_max_delta : 4531475
.fair_clock : 175306014212
.prev_fair_clock : 0
.exec_clock : 718501418488
.prev_exec_clock : 0
.wait_runtime : -59753140134
.wait_runtime_overruns : 1245
.wait_runtime_underruns: 1604
.cpu_load[0] : 0
.cpu_load[1] : 0
.cpu_load[2] : 0
.cpu_load[3] : 0
.cpu_load[4] : 0
.wait_runtime_rq_sum : -34602
runnable tasks:
task PID tree-key delta waiting
switches prio sum-exec sum-wait sum-sleep
wait-overrun wait-underrun
------------------------------------------------------------------------------------------------------------------------------------------------------------------
R cat 20421 175306048814 34602 -34602
2 120 69205 -34602 853
0 0
And sched_debug output while running the test:
Sched Debug Version: v0.02
now at 1261993795754 nsecs
cpu: 0
.nr_running : 5
.raw_weighted_load : 5120
.nr_switches : 588021
.nr_load_updates : 232164
.nr_uninterruptible : 4294967190
.jiffies : 240498
.next_balance : 240508
.curr->pid : 21096
.clock : 770190363479
.prev_clock_raw : 895084968848
.clock_warps : 0
.clock_unstable_events : 125214
.clock_max_delta : 0
.fair_clock : 190695995910
.prev_fair_clock : 0
.exec_clock : 763286081256
.prev_exec_clock : 0
.wait_runtime : 68703429598
.wait_runtime_overruns : 2978
.wait_runtime_underruns: 1273
.cpu_load[0] : 5120
.cpu_load[1] : 5119
.cpu_load[2] : 5110
.cpu_load[3] : 5077
.cpu_load[4] : 5037
.wait_runtime_rq_sum : -54771343
runnable tasks:
task PID tree-key delta waiting
switches prio sum-exec sum-wait sum-sleep
wait-overrun wait-underrun
------------------------------------------------------------------------------------------------------------------------------------------------------------------
R mo 21096 190703231293 7235383 -7235383
445 120 1813432308 -8250501 211806979
0 1
mo 21093 190704201003 8205093 -8472612
450 120 1818737228 -12394371 206140729
0 6
mo 21094 190704732208 8736298 -9665761
464 120 1816895620 -12925451 204333757
0 4
mo 21092 190704971831 8975921 -15574122
456 120 1817099155 -16680247 208648471
0 2
mo 21091 190705631487 9635577 -13823465
457 120 1816983212 -16160000 204693203
0 4
cpu: 1
.nr_running : 5
.raw_weighted_load : 5120
.nr_switches : 415103
.nr_load_updates : 211674
.nr_uninterruptible : 106
.jiffies : 240498
.next_balance : 240516
.curr->pid : 21291
.clock : 806335506481
.prev_clock_raw : 897090740058
.clock_warps : 0
.clock_unstable_events : 150133
.clock_max_delta : 0
.fair_clock : 179777580248
.prev_fair_clock : 0
.exec_clock : 730473748729
.prev_exec_clock : 0
.wait_runtime : -68769955037
.wait_runtime_overruns : 1250
.wait_runtime_underruns: 1885
.cpu_load[0] : 4096
.cpu_load[1] : 4293
.cpu_load[2] : 5044
.cpu_load[3] : 7236
.cpu_load[4] : 8445
.wait_runtime_rq_sum : -11785269
runnable tasks:
task PID tree-key delta waiting
switches prio sum-exec sum-wait sum-sleep
wait-overrun wait-underrun
------------------------------------------------------------------------------------------------------------------------------------------------------------------
R cat 21291 179777631414 51166 -51166
2 120 66773 -51166 53218
0 0
mo 21095 179778195659 615411 -799325
737 120 2088825958 -799325 201402067
0 0
mo 21090 179778267640 687392 -711781
760 120 2094378519 -711781 208607795
0 0
mo 21097 179781627293 4047045 -5262328
789 120 2090034861 -5262328 202105919
0 0
mo 21089 179782321359 4741111 -4960669
757 120 2094677840 -4960669 202830299
0 0
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists