>
>> Second, we currently use Qemu's device emulation were the ne2k device
>> does dozens of IO accesses per packet! The rtl8139 is better and does
>> about 3 IO(MMIO) per packet. The current maximum throughput using the
>
>and this is where paravirtualized device drivers can help, like Ingo's
>net
> Second, we currently use Qemu's device emulation were the ne2k device
> does dozens of IO accesses per packet! The rtl8139 is better and does
> about 3 IO(MMIO) per packet. The current maximum throughput using the
and this is where paravirtualized device drivers can help, like Ingo's
network dr
>> > Intel is indeed slower than AMD with this one.
>> > I didn't measure such roundtrip specifically. We usually see that
the
>> > VMETNRY/VMEXIT cycle limits our performance (for io/mmio,..)
>> > On Intel core duo we cannot get more than 150k-200k VMEXITS per
second
>> > for regular guest doin
>I noticed a recent commit for disk I/O support. It's a terribly useful
>idea but unfortunately it was only done for VMDK.
>
>Any chance this could be generalized for other disk types? Also,
>wouldn't it be better to push this sort of things to QEMU upstream
>before committing to KVM?
>
You're a
> > Intel is indeed slower than AMD with this one.
> > I didn't measure such roundtrip specifically. We usually see that the
> > VMETNRY/VMEXIT cycle limits our performance (for io/mmio,..)
> > On Intel core duo we cannot get more than 150k-200k VMEXITS per second
> > for regular guest doing extens
On 2/12/07, Omar Khan <[EMAIL PROTECTED]> wrote:
> The paravirtualization announcement by Ingo Molnar said that:
>
> - it provides an ad-hoc paravirtualization hypercall API between a Linux
> guest and a Linux host. (this will be replaced with a proper
> hypercall later on.)
>
> Is it possible
Howdy,
%lu only works on 64 bit hosts for uint64_t's. The most portable thing
to do is use PRIu64. This is what is done in the rest of QEMU.
Regards,
Anthony Liguori
diff -r f2e188be2c43 qemu/monitor.c
--- a/qemu/monitor.c Mon Feb 12 22:04:38 2007 -0600
+++ b/qemu/monitor.c Mon Feb 12 22:06
I've been tracking down a rather strange SEGV in SVN. I'm on a 32bit
SVM system. I think people are seeing the same on 32bit VMX.
Basically, we get to the first EINTR returned by kvm_run(). Everything
goes as expected until we get to cpu_loop_exit(). When we get to this
point, the jmp_buf a
I noticed a recent commit for disk I/O support. It's a terribly useful
idea but unfortunately it was only done for VMDK.
Any chance this could be generalized for other disk types? Also,
wouldn't it be better to push this sort of things to QEMU upstream
before committing to KVM?
Regards,
Ant
On Sun, 2007-02-11 at 05:18 -0800, SourceForge.net wrote:
>
>
> KVM works really slowly when doing anything.
> however it works EVEN SLOWER when the kvm-intel module is loaded!!!
>
> vmware is like 10 times faster the KVM but people on ubuntu forums are saying
> that their implimentation of KV
On Sat, Feb 10, 2007 at 02:34:43PM +0100, Waba wrote:
> It took me a while, but I figured it out... nearly!
Great. We get closer to the real problem.
> Further investigating, I tricked ls(1) into using the optimised libc
> through LD_LIBRARY_PATH and obtained a core. mdb(1) told me that the
> cul
>Avi Kivity <[EMAIL PROTECTED]> wrote:
>> Waba wrote:
>>> It took me a while, but I figured it out... nearly!
>>>
>>> Everything SIGILLs after the fs- root service is started. Its start
>>> method does several things, but the problematic bit is replacing the
>>> libc with an optimised version (name
>>> On Mon, Feb 12, 2007 at 4:48 AM, in message <[EMAIL PROTECTED]>,
Avi Kivity <[EMAIL PROTECTED]> wrote:
> Waba wrote:
>> It took me a while, but I figured it out... nearly!
>>
>> Everything SIGILLs after the fs- root service is started. Its start
>> method does several things, but the problema
>>> On Mon, Feb 12, 2007 at 9:23 AM, in message
<[EMAIL PROTECTED]>,
"Dor Laor" <[EMAIL PROTECTED]> wrote:
>
> Thanks man, my mistake, I meant microsecond when wrote the 'ms'.
FYI: The SI symbol for micro is the greek lower-case letter Mu. However, since
that would be a pain in ASCII, the le
>On Mon, Feb 12, 2007 at 05:21:26AM -0800, Dor Laor wrote:
>> >>
>> >> >hi,
>> >> >Can someone briefly explain what the cr3-cache feature on
intel
>> VT
>> >> >processors is and how it is used to improve performance in
>> >> >paravirtualization.
>> >>
>> >> In short, this is a unique Intel VT f
On Mon, Feb 12, 2007 at 05:21:26AM -0800, Dor Laor wrote:
> >>
> >> >hi,
> >> >Can someone briefly explain what the cr3-cache feature on intel
> VT
> >> >processors is and how it is used to improve performance in
> >> >paravirtualization.
> >>
> >> In short, this is a unique Intel VT feature th
>>
>> >hi,
>> >Can someone briefly explain what the cr3-cache feature on intel
VT
>> >processors is and how it is used to improve performance in
>> >paravirtualization.
>>
>> In short, this is a unique Intel VT feature that supposed to save
vmexit
>> on cr3 switch. The vmexit operation is quite
On Sat, Feb 10, 2007 at 02:22:25PM -0800, Dor Laor wrote:
>
> >hi,
> >Can someone briefly explain what the cr3-cache feature on intel VT
> >processors is and how it is used to improve performance in
> >paravirtualization.
>
> In short, this is a unique Intel VT feature that supposed to save v
The paravirtualization announcement by Ingo Molnar said that:
- it provides an ad-hoc paravirtualization hypercall API between a Linux
guest and a Linux host. (this will be replaced with a proper
hypercall later on.)
Is it possible to modify a non-linux guest to take advantage of this hyperc
Thanks Dor.
Omar
-
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier.
Download IBM WebSphere Application Server v.1.0.1
Waba wrote:
It took me a while, but I figured it out... nearly!
Everything SIGILLs after the fs-root service is started. Its start
method does several things, but the problematic bit is replacing the
libc with an optimised version (namely, /usr/lib/libc/libc_hwcap1.so.1,
which makes use of the S
21 matches
Mail list logo
