* Len Brown <l...@kernel.org> wrote:
> On Thu, May 7, 2015 at 6:23 AM, Ingo Molnar <mi...@kernel.org> wrote:
>
> These numbers do not look right.
>
> > Here's what the boot time looks like on a 120 CPUs system, with the
> > patch applied:
> >
> > [ 0.558947] x86: Booting SMP configuration:
> > [ 0.563375] .... node #0, CPUs: #1 #2 #3 #4 #5 #6
> > #7 #8 #9 #10 #11 #12 #13 #14
> > [ 0.644851] .... node #1, CPUs: #15 #16 #17 #18 #19 #20 #21
> > #22 #23 #24 #25 #26 #27 #28 #29
> > [ 0.830474] .... node #2, CPUs: #30 #31 #32 #33 #34 #35 #36
> > #37 #38 #39 #40 #41 #42 #43 #44
> > [ 1.016357] .... node #3, CPUs: #45 #46 #47 #48 #49 #50 #51
> > #52 #53 #54 #55 #56 #57 #58 #59
> > [ 1.202342] .... node #0, CPUs: #60 #61 #62 #63 #64 #65 #66
> > #67 #68 #69 #70 #71 #72 #73 #74
> > [ 1.283864] .... node #1, CPUs: #75 #76 #77 #78 #79 #80 #81
> > #82 #83 #84 #85 #86 #87 #88 #89
> > [ 1.397131] .... node #2, CPUs: #90 #91 #92 #93 #94 #95 #96
> > #97 #98 #99 #100 #101 #102 #103 #104
> > [ 1.510417] .... node #3, CPUs: #105 #106 #107 #108 #109 #110 #111
> > #112 #113 #114 #115 #116 #117 #118 #119
> > [ 1.620967] x86: Booted up 4 nodes, 120 CPUs
> > [ 1.625928] smpboot: Total of 120 processors activated (672866.16
> > BogoMIPS)
> >
> > 1.1 seconds to boot 120 CPUs, 10.8 seconds to hit init, that's an
> > entirely reasonable runtime I think.
> >
> > It was 20+ seconds before that, 10+ seconds for the SMP bootup
> > sequence.
>
> (1.625928-0.558947) = 1.07 seconds to online 119 additional cpus.
> /119 = .0089662268 each, lets call it 9ms.
>
> Here is my ivb-ex running stock fedora 21's Linux-3.19 (no patch applied):
>
> [ 0.404369] x86: Booting SMP configuration:
> [ 0.409492] .... node #0, CPUs: #1
> [ 0.439900] NMI watchdog: enabled on all CPUs, permanently consumes
> one hw-PMU counter.
> [ 0.450533] #2 #3 #4 #5 #6 #7 #8 #9 #10 #11 #12 #13
> #14
> [ 0.652409] .... node #1, CPUs: #15 #16 #17 #18 #19 #20
> #21 #22 #23 #24 #25 #26 #27 #28 #29
> [ 0.999645] .... node #2, CPUs: #30 #31 #32 #33 #34 #35
> #36 #37 #38 #39 #40 #41 #42 #43 #44
> [ 1.346991] .... node #3, CPUs: #45 #46 #47 #48 #49 #50
> #51 #52 #53 #54 #55 #56 #57 #58 #59
> [ 1.694171] .... node #0, CPUs: #60 #61 #62 #63 #64 #65
> #66 #67 #68 #69 #70 #71 #72 #73 #74
> [ 1.928248] .... node #1, CPUs: #75 #76 #77 #78 #79 #80
> #81 #82 #83 #84 #85 #86 #87 #88 #89
> [ 2.198370] .... node #2, CPUs: #90 #91 #92 #93 #94 #95
> #96 #97 #98 #99 #100 #101 #102 #103 #104
> [ 2.468574] .... node #3, CPUs: #105 #106 #107 #108 #109 #110
> #111 #112 #113 #114 #115 #116 #117 #118 #119
> [ 2.737884] x86: Booted up 4 nodes, 120 CPUs
> [ 2.743758] smpboot: Total of 120 processors activated (671097.18 BogoMIPS)
>
> (2.743758-0.404369) = 2.339389 for all 119 processors
> /119 = .01965873109243697478 - lets call it 19ms each
>
> so this baseline case is 19ms/processor -- which matches above, where
> we delete 10ms/processor.
> But even at 19ms each, this is totals only 2.3 seconds for all 119
> processors. So I don't understand your reference to 10+ seconds,
> above.
Hm, so what I was referring to is:
[ 0.205318] x86: Booting SMP configuration:
[ 0.206005] .... node #0, CPUs: #1 #2 #3 #4 #5 #6 #7
#8 #9 #10 #11 #12 #13 #14 #15 #16 #17 #18 #19 #20 #21 #22 #23
#24 #25 #26 #27 #28 #29 #30 #31 #32 #33 #34 #35 #36 #37 #38 #39
#40 #41 #42 #43 #44 #45 #46 #47 #48 #49 #50 #51 #52 #53 #54 #55
#56 #57 #58 #59 #60 #61 #62 #63 #64 #65 #66 #67 #68 #69 #70 #71
#72 #73 #74 #75 #76 #77 #78 #79 #80 #81 #82 #83 #84 #85 #86 #87
#88 #89 #90 #91 #92 #93 #94 #95 #96 #97 #98 #99 #100 #101 #102 #103
#104 #105 #106 #107 #108 #109 #110 #111 #112 #113 #114 #115 #116 #117 #118 #119
[ 9.021191] x86: Booted up 1 node, 120 CPUs
[ 9.021803] smpboot: Total of 120 processors activated (672102.75 BogoMIPS)
[ 9.027282] devtmpfs: initialized
So it's 8.9 seconds - but still much more than you'd expect. The delay
is evenly distributed amongst CPUs.
> ps. I wish the BIOS boot time on this box were as fast as Linux:-)
LOL! :-)
Thanks,
Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
