linux-kernel - Re: /proc/stat information incorrect

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

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Fri, 20 Apr 2012 14:06:54 +0200
From:	Martin Schwidefsky <schwidefsky@...ibm.com>
To:	Bas van der Oest <bassvdo@...il.com>
Cc:	linux-kernel@...r.kernel.org
Subject: Re: /proc/stat information incorrect

On Tue, 17 Apr 2012 10:04:35 +0200
Bas van der Oest <bassvdo@...il.com> wrote:

> > Assuming that you are on a recent kernel (>= 3.2), could you please
> > try to revoke git commit a25cac5198d4ff28 "proc: Consider NO_HZ when
> > printing idle and iowait times" and try again ?
> 
> I revoked the patch and this certainly had effect, however I still
> doubt if the result is completely correct now.
> It now seems that CPU4 (this one is handling the irqs now) spends more
> time than the other CPUs:
> 
>         user    nice    system  idle    iowait  irq   softirq sum
> cpu     4       0       412     8186    0       0       79      8681
> cpu0    0       0       1       1074    0       0       1       1076
> cpu1    0       0       2       1075    0       0       0       1077
> cpu2    0       0       1       1075    0       0       0       1076
> cpu3    0       0       1       1079    0       0       0       1080
> cpu4    1       0       115     941     0       0       78      1135
> cpu5    2       0       99      984     0       0       0       1085
> cpu6    0       0       98      977     0       0       0       1075
> cpu7    0       0       95      980     0       0       0       1075
> 
> Any suggestions where this small difference is coming from?

CONFIG_IRQ_TIME_ACCOUNTING comes to mind. Is this enabled for your
kernel ?

-- 
blue skies,
   Martin.

"Reality continues to ruin my life." - Calvin.

--
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/