All of lore.kernel.org
 help / color / mirror / Atom feed
From: Harri Olin <harri.olin@gmail.com>
To: Gleb Natapov <gleb@redhat.com>
Cc: Christoph Adomeit <Christoph.Adomeit@gatworks.de>,
	"kvm@vger.kernel.org" <kvm@vger.kernel.org>
Subject: Re: Freezing Windows 2008 x64bit guest
Date: Wed, 21 Jul 2010 17:05:24 +0300	[thread overview]
Message-ID: <4C46FEA4.1050709@gmail.com> (raw)
In-Reply-To: <20100721134516.GA27177@redhat.com>

Gleb Natapov wrote:
 > On Wed, Jul 21, 2010 at 04:03:28PM +0300, Harri Olin wrote:
 >> Gleb Natapov wrote:
 >>> On Wed, Jul 21, 2010 at 01:09:06PM +0300, Harri Olin wrote:
 >>>> Gleb Natapov wrote:
 >>>> On both times irq_exits go to 10000 but irq_injections stay at
 >>>> around 4600. On both times qemu process was at 100% cpu load.
 >>> irq_exits are exits due to host interrupt. What "cat /proc/interrupts"
 >>> show? Which counter grows at this rate there (or 1/4 of the rate 
but for
 >>> each host cpu).
 >> System froze again just a moment ago and there are 6100 "local timer
 >> interrupts" and around 3600 "Rescheduling interrupts" every second
 >> happening.
 >>
 >> #cat /proc/interrupts; sleep 1; cat /proc/interrupts; sleep 1; ....
 >> http://mizar.remote.agasha.com/k/kvm/interrupts.txt
 >> http://mizar.remote.agasha.com/k/kvm/interrupts1.txt
 >>
 >> quick calculations
 >>                                              total
 >> RES: 110847929 115251972 108671867 114338799 449110567 	per second:
 >> RES: 110848996 115252094 108673031 114340080 449114201 	3634
 >> RES: 110850059 115252174 108674157 114341446 449117836 	3635
 >> RES: 110851132 115252276 108675332 114342801 449121541 	3705
 >> 						
 >>                                                total		
 >> LOC: 1531251837 618157953 1727791012 522296117 4399496919 per second:
 >> LOC: 1531252861 618161024 1727792036 522297141 4399503062 6143
 >> LOC: 1531253884 618164095 1727793060 522298165 4399509204 6142
 >> LOC: 1531254907 618167166 1727794084 522299189 4399515346 6142
 >
 > Same qemu and command line as before? Without -no-hpet?

Yes those interrupt numbers area still running qemu-kvm-0.11.1 without 
-no-hpet.

I now started the quest with -no-hpet, kvm_stat, everything working:

  efer_reload                  0       0
  exits                283535303   41437
  fpu_reload            94754003   16231
  halt_exits            24716918    4696
  halt_wakeup           24416218    4651
  host_state_reload    112571913   18552
  hypercalls                   0       0
  insn_emulation        75833651   11377
  insn_emulation_fail         83       0
  invlpg                 5327501     124
  io_exits              84146648   13718
  irq_exits             10182390    2295
  irq_injections        31379031    5979
  irq_window              941352     113
  largepages                   0       0
  mmio_exits               29794       0
  mmu_cache_miss          380265      11
  mmu_flooded             167547       2
  mmu_pde_zapped          288251       2
  mmu_pte_updated         586646       2
  mmu_pte_write           505897       3
  mmu_recycled             40892       0
  mmu_shadow_zapped       365079       4
  mmu_unsync               19338      -1
  nmi_injections               0       0
  nmi_window                   0       0
  pf_fixed              20331924    1570
  pf_guest               5332473     214
  remote_tlb_flush       3213779     211
  request_irq                  0       0
  signal_exits                 0       0
  tlb_flush              8205865     335


interrupt count calculations, system working, -no-hpet:

                                                total
LOC: 1537337119 621778834 1732851735 525376978 4417344666 per sec
LOC: 1537339334 621779137 1732851925 525377284 4417347680 3014
LOC: 1537339987 621779260 1732853664 525377824 4417350735 3055
LOC: 1537340126 621779388 1732855889 525378420 4417353823 3088

RES: 112093536 116293086 109808558 115510943 453706123
RES: 112093737 116293164 109808699 115511006 453706606 483
RES: 112093921 116293182 109808731 115511040 453706874 268
RES: 112093976 116293225 109808733 115511061 453706995 121

full /proc/interrupts at: 
http://mizar.remote.agasha.com/k/kvm/interrupts-working-no-hpet.txt

command line:
/usr/local/qemu-kvm-0.11.1/bin/qemu-system-x86_64 -drive 
file=/dev/rigelvg/w2k8system,cache=none,boot=on -drive 
file=/dev/rigelvg/w2k8data,cache=none -m 6144 -vnc :1 -net 
nic,macaddr=C0:FF:12:FB:AA:01 -net tap -smp 4 -localtime -no-hpet

Still couple of hours before I can restart the host to new kernel.

-- 
Harri.

  reply	other threads:[~2010-07-21 14:05 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-07-15 13:19 Freezing Windows 2008 x64bit guest Christoph Adomeit
2010-07-15 13:44 ` Gleb Natapov
2010-07-19  7:17   ` Harri Olin
2010-07-19  7:42     ` Gleb Natapov
2010-07-21  6:25       ` Harri Olin
2010-07-21  8:28         ` Christoph Adomeit
2010-07-21  8:37         ` Gleb Natapov
2010-07-21  9:22           ` Harri Olin
2010-07-21  9:48             ` Gleb Natapov
2010-07-21 10:09               ` Harri Olin
2010-07-21 10:30                 ` Gleb Natapov
2010-07-21 10:43                   ` Harri Olin
2010-07-21 13:03                   ` Harri Olin
2010-07-21 13:45                     ` Gleb Natapov
2010-07-21 14:05                       ` Harri Olin [this message]
2010-07-27 21:53           ` Harri Olin
2010-07-28 15:18             ` Gleb Natapov
2010-12-13 19:42               ` Manfred Heubach
2010-12-13 20:12                 ` Dor Laor
2010-12-13 20:44                   ` Vadim Rozenfeld
2010-12-14 23:57                     ` AW: " Manfred Heubach
2010-12-15 10:48                       ` Vadim Rozenfeld

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=4C46FEA4.1050709@gmail.com \
    --to=harri.olin@gmail.com \
    --cc=Christoph.Adomeit@gatworks.de \
    --cc=gleb@redhat.com \
    --cc=kvm@vger.kernel.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.