* windows PIO question
@ 2012-02-02 21:16 Nikola Ciprich
2012-02-05 15:10 ` Avi Kivity
0 siblings, 1 reply; 5+ messages in thread
From: Nikola Ciprich @ 2012-02-02 21:16 UTC (permalink / raw)
To: kvm; +Cc: nikola.ciprich
[-- Attachment #1: Type: text/plain, Size: 1314 bytes --]
Hello everybody,
I'm a bit confused regarding windows guests and PIO.. some time ago, I was trying
to find out, why one of my windows guests was running terribly slow.
After checking the trace-cmd output, Avi stated that windows were using PIO instead
of DMA, after changing that, windows indeed got a lot faster. I didn't check traces
after that, but I'm now testing host with two windows 2003 guests, and checking the
trac-cmd output, I see a lot of kvm_pio calls.
While windows don't seem to be running particularly slow, they're not especially
fast either, and I'm wandering whether this could be related. But checking IDE
bus in guests, they seem to be using DMA which confuses me...
so my questions is, what are the kvm_pio operations I see in traces, and how can
I tell what is OK and what is wrong, regarding disks DMA/PIO?
If somebody could shed some light on this for me, I'd be very grateful...
Thanks a lot in advance
with best regards
nikola ciprich
--
-------------------------------------
Ing. Nikola CIPRICH
LinuxBox.cz, s.r.o.
28. rijna 168, 709 01 Ostrava
tel.: +420 596 603 142
fax: +420 596 621 273
mobil: +420 777 093 799
www.linuxbox.cz
mobil servis: +420 737 238 656
email servis: servis@linuxbox.cz
-------------------------------------
[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: windows PIO question
2012-02-02 21:16 windows PIO question Nikola Ciprich
@ 2012-02-05 15:10 ` Avi Kivity
2012-02-05 21:01 ` Nikola Ciprich
0 siblings, 1 reply; 5+ messages in thread
From: Avi Kivity @ 2012-02-05 15:10 UTC (permalink / raw)
To: Nikola Ciprich; +Cc: kvm
On 02/02/2012 11:16 PM, Nikola Ciprich wrote:
> Hello everybody,
>
> I'm a bit confused regarding windows guests and PIO.. some time ago, I was trying
> to find out, why one of my windows guests was running terribly slow.
> After checking the trace-cmd output, Avi stated that windows were using PIO instead
> of DMA, after changing that, windows indeed got a lot faster. I didn't check traces
> after that, but I'm now testing host with two windows 2003 guests, and checking the
> trac-cmd output, I see a lot of kvm_pio calls.
> While windows don't seem to be running particularly slow, they're not especially
> fast either, and I'm wandering whether this could be related. But checking IDE
> bus in guests, they seem to be using DMA which confuses me...
> so my questions is, what are the kvm_pio operations I see in traces, and how can
> I tell what is OK and what is wrong, regarding disks DMA/PIO?
> If somebody could shed some light on this for me, I'd be very grateful...
Not all PIO operations are DMA. Provide a trace and we can look.
--
error compiling committee.c: too many arguments to function
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: windows PIO question
2012-02-05 15:10 ` Avi Kivity
@ 2012-02-05 21:01 ` Nikola Ciprich
2012-02-06 9:50 ` Avi Kivity
0 siblings, 1 reply; 5+ messages in thread
From: Nikola Ciprich @ 2012-02-05 21:01 UTC (permalink / raw)
To: Avi Kivity; +Cc: kvm, Nikola Ciprich
[-- Attachment #1: Type: text/plain, Size: 3782 bytes --]
> Not all PIO operations are DMA. Provide a trace and we can look.
Hi,
here's snippet:
qemu-kvm-16441 [000] 12887154.698329: kvm_pio: pio_read at 0xc002 size 1 count 1
qemu-kvm-16441 [000] 12887154.698331: kvm_userspace_exit: reason KVM_EXIT_IO (2)
qemu-kvm-16441 [000] 12887154.698331: kvm_fpu: unload
qemu-kvm-16441 [000] 12887154.698337: kvm_entry: vcpu 1
qemu-kvm-16441 [000] 12887154.698338: kvm_exit: [FAILED TO PARSE] exit_reason=30 guest_rip=0xfffff88000dabca3 isa=1 info1=3221356552 info2=0
qemu-kvm-16441 [000] 12887154.698339: kvm_emulate_insn: [FAILED TO PARSE] rip=18446735827386678435 csbase=0 len=1 insn=<EC><C3><CC><CC><CC><CC><CC><CC><CC>^O<B7><D1>f<ED><C3> flags=9 failed=0
qemu-kvm-16441 [000] 12887154.698340: kvm_pio: pio_read at 0xc002 size 1 count 1
qemu-kvm-16441 [000] 12887154.698340: kvm_userspace_exit: reason KVM_EXIT_IO (2)
qemu-kvm-16441 [000] 12887154.698343: kvm_entry: vcpu 1
qemu-kvm-16441 [000] 12887154.698344: kvm_exit: [FAILED TO PARSE] exit_reason=30 guest_rip=0xfffff88000dabdbd isa=1 info1=3221225472 info2=0
qemu-kvm-16441 [000] 12887154.698345: kvm_pio: pio_write at 0xc000 size 1 count 1
qemu-kvm-16441 [000] 12887154.698346: kvm_userspace_exit: reason KVM_EXIT_IO (2)
qemu-kvm-16441 [000] 12887154.698348: kvm_entry: vcpu 1
qemu-kvm-16441 [000] 12887154.698349: kvm_exit: [FAILED TO PARSE] exit_reason=30 guest_rip=0xfffff88000dabdbd isa=1 info1=3221356544 info2=0
qemu-kvm-16441 [000] 12887154.698349: kvm_pio: pio_write at 0xc002 size 1 count 1
qemu-kvm-16441 [000] 12887154.698350: kvm_userspace_exit: reason KVM_EXIT_IO (2)
qemu-kvm-16441 [000] 12887154.698352: kvm_entry: vcpu 1
qemu-kvm-16441 [000] 12887154.698353: kvm_exit: [FAILED TO PARSE] exit_reason=30 guest_rip=0xfffff88000dabca3 isa=1 info1=32964616 info2=0
qemu-kvm-16441 [000] 12887154.698354: kvm_emulate_insn: [FAILED TO PARSE] rip=18446735827386678435 csbase=0 len=1 insn=<EC><C3><CC><CC><CC><CC><CC><CC><CC>^O<B7><D1>f<ED><C3> flags=9 failed=0
qemu-kvm-16441 [000] 12887154.698354: kvm_pio: pio_read at 0x1f7 size 1 count 1
qemu-kvm-16441 [000] 12887154.698355: kvm_userspace_exit: reason KVM_EXIT_IO (2)
qemu-kvm-16441 [000] 12887154.698357: kvm_set_irq: gsi 14 level 0 source 0
qemu-kvm-16441 [000] 12887154.698358: kvm_pic_set_irq: chip 1 pin 6 (edge|masked)
qemu-kvm-16441 [000] 12887154.698358: kvm_ioapic_set_irq: pin 14 dst 3 vec=97 (LowPrio|logical|edge)
qemu-kvm-16441 [000] 12887154.698360: kvm_entry: vcpu 1
qemu-kvm-16441 [000] 12887154.698362: kvm_exit: [FAILED TO PARSE] exit_reason=30 guest_rip=0xfffff88000dabca3 isa=1 info1=32964616 info2=0
qemu-kvm-16441 [000] 12887154.698363: kvm_emulate_insn: [FAILED TO PARSE] rip=18446735827386678435 csbase=0 len=1 insn=<EC><C3><CC><CC><CC><CC><CC><CC><CC>^O<B7><D1>f<ED><C3> flags=9 failed=0
qemu-kvm-16441 [000] 12887154.698363: kvm_pio: pio_read at 0x1f7 size 1 count 1
n.
>
> --
> error compiling committee.c: too many arguments to function
>
--
-------------------------------------
Ing. Nikola CIPRICH
LinuxBox.cz, s.r.o.
28. rijna 168, 709 01 Ostrava
tel.: +420 596 603 142
fax: +420 596 621 273
mobil: +420 777 093 799
www.linuxbox.cz
mobil servis: +420 737 238 656
email servis: servis@linuxbox.cz
-------------------------------------
[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: windows PIO question
2012-02-05 21:01 ` Nikola Ciprich
@ 2012-02-06 9:50 ` Avi Kivity
2012-02-07 8:25 ` Nikola Ciprich
0 siblings, 1 reply; 5+ messages in thread
From: Avi Kivity @ 2012-02-06 9:50 UTC (permalink / raw)
To: Nikola Ciprich; +Cc: kvm
On 02/05/2012 11:01 PM, Nikola Ciprich wrote:
> > Not all PIO operations are DMA. Provide a trace and we can look.
>
> Hi,
> here's snippet:
>
> qemu-kvm-16441 [000] 12887154.698329: kvm_pio: pio_read at 0xc002 size 1 count 1
> qemu-kvm-16441 [000] 12887154.698331: kvm_userspace_exit: reason KVM_EXIT_IO (2)
> qemu-kvm-16441 [000] 12887154.698331: kvm_fpu: unload
> qemu-kvm-16441 [000] 12887154.698337: kvm_entry: vcpu 1
> qemu-kvm-16441 [000] 12887154.698338: kvm_exit: [FAILED TO PARSE] exit_reason=30 guest_rip=0xfffff88000dabca3 isa=1 info1=3221356552 info2=0
> qemu-kvm-16441 [000] 12887154.698339: kvm_emulate_insn: [FAILED TO PARSE] rip=18446735827386678435 csbase=0 len=1 insn=<EC><C3><CC><CC><CC><CC><CC><CC><CC>^O<B7><D1>f<ED><C3> flags=9 failed=0
> qemu-kvm-16441 [000] 12887154.698340: kvm_pio: pio_read at 0xc002 size 1 count 1
>
That's virtio, not IDE.
--
error compiling committee.c: too many arguments to function
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: windows PIO question
2012-02-06 9:50 ` Avi Kivity
@ 2012-02-07 8:25 ` Nikola Ciprich
0 siblings, 0 replies; 5+ messages in thread
From: Nikola Ciprich @ 2012-02-07 8:25 UTC (permalink / raw)
To: Avi Kivity; +Cc: kvm
[-- Attachment #1: Type: text/plain, Size: 918 bytes --]
> That's virtio, not IDE.
well, this seems a bit strange to me, the only virtio device I can
see is balloon dev, can this be it?
how can I find out which operations are IDE PIO etc? sorry
for asking silly questions, but I couldn't find this anywhere
in the documentation...
thanks in advance
nik
>
> --
> error compiling committee.c: too many arguments to function
>
> --
> To unsubscribe from this list: send the line "unsubscribe kvm" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
--
-------------------------------------
Ing. Nikola CIPRICH
LinuxBox.cz, s.r.o.
28. rijna 168, 709 01 Ostrava
tel.: +420 596 603 142
fax: +420 596 621 273
mobil: +420 777 093 799
www.linuxbox.cz
mobil servis: +420 737 238 656
email servis: servis@linuxbox.cz
-------------------------------------
[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2012-02-07 8:26 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-02-02 21:16 windows PIO question Nikola Ciprich
2012-02-05 15:10 ` Avi Kivity
2012-02-05 21:01 ` Nikola Ciprich
2012-02-06 9:50 ` Avi Kivity
2012-02-07 8:25 ` Nikola Ciprich
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).