Re: CONFIG_PREEMPT_RCU in next/mmotm

From: Hugh Dickins
Date: Sun Aug 09 2009 - 07:25:27 EST


On Sat, 8 Aug 2009, Paul E. McKenney wrote:
> On Sun, Aug 09, 2009 at 04:32:28AM +0100, Hugh Dickins wrote:
> >
> > Is there some particular attribute of a cpu, most relevant to its RCU
> > grace status, that I can check on with a printk?
>
> The debugfs files enabled by CONFIG_RCU_TRACE, except rcu/rcugp, which,
> since it will attempt to wait for a grace period, will probably wait

Here's ten second samples from just before and after it "goes":
the number of filp objects (my own guide to it going), rcuctrs and
rctstats. With ggp stuck on 35124, as you predicted, rcugp just hangs.

filp 5616 objects:

CPU last cur F M
0 0 0 0 1
1 0 0 0 0
2 0 -1 0 0
3 0 1 0 0
ggp = 33821, state = waitmb

na=2751281 nl=64 wa=2751217 wl=176 da=2751041 dl=0 dr=2751041 di=2751023
1=353642 e1=193343 i1=33939 ie1=118 g1=33821 a1=38041 ae1=4220 a2=33821
z1=36834 ze1=3013 z2=33821 m1=51485 me1=17665 m2=33820

filp 5780 objects:

CPU last cur F M
0 0 0 0 0
1 0 0 0 0
2 -1 0 0 0
3 1 0 0 0
ggp = 34518, state = waitack

na=2826191 nl=18 wa=2826173 wl=71 da=2826102 dl=0 dr=2826102 di=2826058
1=362636 e1=198972 i1=34640 ie1=122 g1=34518 a1=38847 ae1=4330 a2=34517
z1=37702 ze1=3185 z2=34517 m1=52475 me1=17958 m2=34517

filp 13343 objects:

CPU last cur F M
0 1 0 0 0
1 0 0 0 0
2 -1 0 0 0
3 0 0 0 0
ggp = 35124, state = waitzero

na=2898045 nl=15380 wa=2882665 wl=150 da=2882515 dl=0 dr=2882515 di=2882465
1=371717 e1=204532 i1=35248 ie1=124 g1=35124 a1=39545 ae1=4421 a2=35124
z1=39077 ze1=3954 z2=35123 m1=53315 me1=18192 m2=35123

filp 67190 objects:

CPU last cur F M
0 1 0 0 0
1 0 1 0 0
2 -1 0 0 0
3 0 0 0 0
ggp = 35124, state = waitzero

na=2961758 nl=79093 wa=2882665 wl=150 da=2882515 dl=0 dr=2882515 di=2882465
1=383301 e1=212065 i1=35248 ie1=124 g1=35124 a1=39545 ae1=4421 a2=35124
z1=43128 ze1=8005 z2=35123 m1=53315 me1=18192 m2=35123

filp 137160 objects:

CPU last cur F M
0 1 1 0 0
1 0 0 0 0
2 -1 0 0 0
3 0 0 0 0
ggp = 35124, state = waitzero

na=3041696 nl=159031 wa=2882665 wl=150 da=2882515 dl=0 dr=2882515 di=2882465
1=394717 e1=219335 i1=35248 ie1=124 g1=35124 a1=39545 ae1=4421 a2=35124
z1=47274 ze1=12151 z2=35123 m1=53315 me1=18192 m2=35123

Hugh
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/