From: Martin Schwidefsky <schwidefsky@de.ibm.com>
To: Frederic Weisbecker <fweisbec@gmail.com>
Cc: LKML <linux-kernel@vger.kernel.org>,
Tony Luck <tony.luck@intel.com>,
Wanpeng Li <wanpeng.li@hotmail.com>,
Peter Zijlstra <peterz@infradead.org>,
Michael Ellerman <mpe@ellerman.id.au>,
Heiko Carstens <heiko.carstens@de.ibm.com>,
Benjamin Herrenschmidt <benh@kernel.crashing.org>,
Thomas Gleixner <tglx@linutronix.de>,
Paul Mackerras <paulus@samba.org>, Ingo Molnar <mingo@kernel.org>,
Fenghua Yu <fenghua.yu@intel.com>, Rik van Riel <riel@redhat.com>,
Stanislaw Gruszka <sgruszka@redhat.com>
Subject: Re: [PATCH 37/37] s390: Prevent from cputime leaks
Date: Mon, 23 Jan 2017 10:44:56 +0100 [thread overview]
Message-ID: <20170123104456.74cb2ef3@mschwideX1> (raw)
In-Reply-To: <1485109213-8561-38-git-send-email-fweisbec@gmail.com>
On Sun, 22 Jan 2017 19:20:13 +0100
Frederic Weisbecker <fweisbec@gmail.com> wrote:
> The s390 clock has a higher granularity than nanoseconds. 1 nanosec
> equals 4.096 in s390 cputime_t. Therefore we leak a remainder while
> flushing the cputime through cputime_to_nsecs().
>
> For more precision, make sure we keep that remainder on cputime
> accumulators for later accounting.
>
> Reported-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
> Cc: Paul Mackerras <paulus@samba.org>
> Cc: Michael Ellerman <mpe@ellerman.id.au>
> Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
> Cc: Tony Luck <tony.luck@intel.com>
> Cc: Fenghua Yu <fenghua.yu@intel.com>
> Cc: Peter Zijlstra <peterz@infradead.org>
> Cc: Rik van Riel <riel@redhat.com>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Ingo Molnar <mingo@kernel.org>
> Cc: Stanislaw Gruszka <sgruszka@redhat.com>
> Cc: Wanpeng Li <wanpeng.li@hotmail.com>
> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
NAK. Good intention but the patch is just broken. with 36 of the 37
patches applied all looks good but the last one completely breaks the
accounting for s390. This is from an idle system:
top - 10:39:33 up 0 min, 1 user, load average: 0,00, 0,00, 0,00
Tasks: 106 total, 1 running, 105 sleeping, 0 stopped, 0 zombie
%Cpu0 : 8,9 us, 21,6 sy, 0,0 ni, 0,0 id, 0,0 wa, 10,8 hi, 4,3 si, 54,4 st
%Cpu1 : 0,0 us, 23,5 sy, 0,0 ni, 0,0 id, 0,0 wa, 19,0 hi, 13,1 si, 44,3 st
%Cpu2 : 0,0 us, 30,3 sy, 0,0 ni, 0,0 id, 0,0 wa, 14,7 hi, 14,8 si, 40,2 st
KiB Mem : 1009304 total, 818808 free, 57284 used, 133212 buff/cache
KiB Swap: 1048556 total, 1048556 free, 0 used. 917356 avail Mem
There is another issue that affects precision, there is no s390 specific
version of cputime_to_nsecs. The generic version uses cputime_to_usecs
and mulitplies by 1000 to get nano-seconds. That already looses precision.
For now just drop that last patch please.
--
blue skies,
Martin.
"Reality continues to ruin my life." - Calvin.
next prev parent reply other threads:[~2017-01-23 9:45 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-01-22 18:19 [PATCH 00/37] cputime: Convert core use of cputime_t to nsecs v3 Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 01/37] jiffies: Reuse TICK_NSEC instead of NSEC_PER_JIFFY Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 02/37] time: Introduce jiffies64_to_nsecs() Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 03/37] sched: Remove unused INIT_CPUTIME macro Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 04/37] cputime: Convert kcpustat to nsecs Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 05/37] macintosh/rack-meter: Remove cputime_t internal use Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 06/37] cputime: Convert guest time accounting to nsecs Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 07/37] cputime: Special API to return old-typed cputime Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 08/37] cputime: Convert task/group cputime to nsecs Frederic Weisbecker
2017-01-28 11:57 ` Stanislaw Gruszka
2017-01-28 15:28 ` Frederic Weisbecker
2017-01-30 13:54 ` Stanislaw Gruszka
2017-01-30 15:38 ` Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 09/37] alpha: Convert obsolete cputime_t " Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 10/37] x86: Convert obsolete cputime type " Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 11/37] isdn: " Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 12/37] binfmt: " Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 13/37] acct: " Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 14/37] delaycct: " Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 15/37] tsacct: " Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 16/37] signal: " Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 17/37] cputime: Increment kcpustat directly on irqtime account Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 18/37] posix-timers: Use TICK_NSEC instead of a dynamically ad-hoc calculated version Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 19/37] posix-timers: Convert internals to use nsecs Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 20/37] itimer: Convert internal cputime_t units to nsec Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 21/37] sched: Remove temporary cputime_t accessors Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 22/37] cputime: Push time to account_user_time() in nsecs Frederic Weisbecker
2017-01-22 18:19 ` [PATCH 23/37] cputime: Push time to account_steal_time() " Frederic Weisbecker
2017-01-22 18:20 ` [PATCH 24/37] cputime: Push time to account_idle_time() " Frederic Weisbecker
2017-01-22 18:20 ` [PATCH 25/37] cputime: Push time to account_system_time() " Frederic Weisbecker
2017-01-22 18:20 ` [PATCH 26/37] cputime: Complete nsec conversion of tick based accounting Frederic Weisbecker
2017-01-22 18:20 ` [PATCH 27/37] vtime: Return nsecs instead of cputime_t to account Frederic Weisbecker
2017-01-22 18:20 ` [PATCH 28/37] cputime: Remove jiffies based cputime Frederic Weisbecker
2017-01-22 18:20 ` [PATCH 29/37] ia64: Move nsecs based cputime headers to the last arch using it Frederic Weisbecker
2017-01-22 18:20 ` [PATCH 30/37] ia64: Convert vtime to use nsec units directly Frederic Weisbecker
2017-01-22 18:20 ` [PATCH 31/37] ia64: Remove unused cputime definitions Frederic Weisbecker
2017-01-22 18:20 ` [PATCH 32/37] s390: Make arch_cpu_idle_time() to return nsecs Frederic Weisbecker
2017-01-22 18:20 ` [PATCH 33/37] powerpc: Remove unused cputime definitions Frederic Weisbecker
2017-01-22 18:20 ` [PATCH 34/37] s390: " Frederic Weisbecker
2017-01-22 18:20 ` [PATCH 35/37] cputime: Remove unused nsec_to_cputime Frederic Weisbecker
2017-01-22 18:20 ` [PATCH 36/37] cputime: Remove asm generic headers Frederic Weisbecker
2017-01-22 18:20 ` [PATCH 37/37] s390: Prevent from cputime leaks Frederic Weisbecker
2017-01-23 9:44 ` Martin Schwidefsky [this message]
2017-01-25 15:25 ` Frederic Weisbecker
2017-01-25 15:40 ` Martin Schwidefsky
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20170123104456.74cb2ef3@mschwideX1 \
--to=schwidefsky@de.ibm.com \
--cc=benh@kernel.crashing.org \
--cc=fenghua.yu@intel.com \
--cc=fweisbec@gmail.com \
--cc=heiko.carstens@de.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=mpe@ellerman.id.au \
--cc=paulus@samba.org \
--cc=peterz@infradead.org \
--cc=riel@redhat.com \
--cc=sgruszka@redhat.com \
--cc=tglx@linutronix.de \
--cc=tony.luck@intel.com \
--cc=wanpeng.li@hotmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox