David Abrahams wrote:
> on Mon Apr 09 2007, Avi Kivity wrote:
>
>
>> David Abrahams wrote:
>>
>>> I'm using the ondemand governor on my core duo laptop. Even when kvm
>>> is taking 99% of the CPU, somehow instead of the ondemand governor
>>> kicking it up to 2GHz, the CPU is still running
Li, Xin B wrote:
>> If the git: and svn: protocols are blocked for you, I can try to make
>>
> them
>
>> available through http.
>>
>>
>
> That would be helpful, thanks!
> -Xin
>
http://www.kernel.org/pub/scm/linux/kernel/git/avi/kvm.git, branch
avi/master
Subversion is not yet a
Gregory Haskins wrote:
> I still have not converted over to kerneldoc format as I cannot find an
> example anywhere yet, and the documentation under
> Documentation/kernel-doc-nano-HOWTO was a little vague. Pointers appreciated.
>
>
see lib/kref.c for an example.
--
Do not meddle in the
Gregory Haskins wrote:
>>> +
>>> +struct kvm_irqdevice;
>>> +
>>> +struct kvm_irqsink {
>>> + void (*raise_intr)(struct kvm_irqsink *this,
>>> + struct kvm_irqdevice *dev);
>>> +
>>> + void *private;
>>> +};
>>> +
>>> +struct kvm_irqdevice {
>>> + int (*pending)(struct
Gregory Haskins wrote:
> Hi Avi,
>
> I believe I have incorporated all of the changes requested. Please find the
> result of that patch inline.
>
> Note that I finally understand what you were getting at with the array of
> objects thing. I didn't change it yet for the same reason that you
> m
On Mon, Apr 09, 2007 at 04:38:18PM +0300, Avi Kivity ([EMAIL PROTECTED]) wrote:
> >But I don't get this "we can enhance the kernel but not userspace" vibe
> >8(
> >
>
> I've been waiting for network aio since ~2003. If it arrives in the
> next few days, I'm all for it; much more than kvm can u
On Mon, Apr 09, 2007 at 04:27:54PM -0400, Gregory Haskins wrote:
> Here is the next pass of the patch with changes based on feedback.
>
> I still have not converted over to kerneldoc format as I cannot find an
> example anywhere yet, and the documentation under
> Documentation/kernel-doc-nano-HO
On Tue, Apr 10, 2007 at 11:19:52AM +0300, Avi Kivity ([EMAIL PROTECTED]) wrote:
> I meant, network aio in the mainline kernel. I am aware of the various
> out-of-tree implementations.
If potential users do not pay attention to initial implementaion, it is
quite hard to them to get into. But actua
Hi!
Avi Kivity wrote:
> Sounds like ondemand is broken.
Maybe it only switches when the load rises in user space. With KVM, it
appears in system space. Got to cross-check that on my Core Duo.
Michael.
-
Take Surveys. Earn
Evgeniy Polyakov wrote:
>>> But it looks from this discussion, that it will not prevent from
>>> changing in-kernel driver - place a hook into skb allocation path and
>>> allocate data from opposing memory - get pages from another side and put
>>> them into fragments, then copy headers into skb->da
>>> On Tue, Apr 10, 2007 at 3:52 AM, in message <[EMAIL PROTECTED]>,
Avi Kivity <[EMAIL PROTECTED]> wrote:
>
> Hmm. The current code is synchronous in nature (the vcpu is never
> executing while we raise an interrupt, so the INTA is never needed, as
> we can ensure the cpu can process the interr
>>> On Tue, Apr 10, 2007 at 3:56 AM, in message <[EMAIL PROTECTED]>, Avi
Kivity <[EMAIL PROTECTED]> wrote:
> Gregory Haskins wrote:
>>
>> Note that I finally understand what you were getting at with the array of
> objects thing. I didn't change it yet for the same reason that you
> mentioned:
On Tue, Apr 10, 2007 at 02:21:24PM +0300, Avi Kivity ([EMAIL PROTECTED]) wrote:
> >You want to implement zero-copy network device between host and guest, if
> >I understood this thread correctly?
> >So, for sending part, device allocates pages from receiver's memory (or
> >from shared memory), rece
Gregory Haskins wrote:
>> The NIC raises an interrupt, we have to interrupt the guest to see if its
>> current IF/cr8 permit interrupt injection, and if not, we have to keep
>> the interrupt in the irqdevice and request an exit when IF/cr8 permit
>> injection.
>>
>
> Exactly. The INTR really
Gregory Haskins wrote:
>> Given that the first consumer is the local apic, it makes sense to add
>> the vcpu- local bus first, no?
>>
>
> I'm confused. I thought you didn't like the vpcu-local bus? I pulled it
> based on your feedback :) Please advise.
>
>
Sorry, my mistake.
--
erro
Evgeniy Polyakov wrote:
>> This is what Xen does. It is actually less performant than copying, IIRC.
>>
>> The problem with flipping pages around is that physical addresses are
>> cached both in the kvm mmu and in the on-chip tlbs, necessitating
>> expensive page table walks and tlb invalidation
Eric Sesterhenn / Snakebyte wrote:
> * Avi Kivity ([EMAIL PROTECTED]) wrote:
>
>> Eric Sesterhenn / Snakebyte wrote:
>>
>>> i was testing the gcc 4.3 against the latest git kernel, and got a
>>> warning in your code (using -Wstrict-overflow=1)
>>>
>>> drivers/kvm/vmx.c: In function 'inject_
On Tue, Apr 10, 2007 at 03:17:45PM +0300, Avi Kivity ([EMAIL PROTECTED]) wrote:
> >Check a link please in case we are talking about different ideas:
> >http://marc.info/?l=linux-netdev&m=112262743505711&w=2
> >
> >
>
> I don't really understand what you're testing there. in particular, how
> c
Evgeniy Polyakov wrote:
> On Tue, Apr 10, 2007 at 03:17:45PM +0300, Avi Kivity ([EMAIL PROTECTED])
> wrote:
>
>>> Check a link please in case we are talking about different ideas:
>>> http://marc.info/?l=linux-netdev&m=112262743505711&w=2
>>>
>>>
>>>
>> I don't really understand what y
Evgeniy Polyakov wrote:
> On Mon, Apr 09, 2007 at 04:38:18PM +0300, Avi Kivity ([EMAIL PROTECTED])
> wrote:
>
>>> But I don't get this "we can enhance the kernel but not userspace" vibe
>>> 8(
>>>
>>>
>> I've been waiting for network aio since ~2003. If it arrives in the
>> next few
>>> On Tue, Apr 10, 2007 at 8:01 AM, in message <[EMAIL PROTECTED]>,
Avi Kivity <[EMAIL PROTECTED]> wrote:
> Gregory Haskins wrote:
>>
>> And if we replace read_vector() with ack(), how do we learn of the vector in
> the first place? I am guessing that you are thinking that both pending and
>
Gregory Haskins wrote:
On Tue, Apr 10, 2007 at 8:01 AM, in message <[EMAIL PROTECTED]>,
> Avi Kivity <[EMAIL PROTECTED]> wrote:
>
>> Gregory Haskins wrote:
>>
>>> And if we replace read_vector() with ack(), how do we learn of the vector
>>> in
>>>
>> the first
>>> On Tue, Apr 10, 2007 at 10:46 AM, in message <[EMAIL PROTECTED]>,
Avi Kivity <[EMAIL PROTECTED]> wrote:
>
> Both the vcpu and the irqdevice are involved in the decision:
>
> - the vcpu holds the IF (or rather, the "interrupt window")
> - the lapic holds the tpr
> - but the vcpu contains t
>>> On Tue, Apr 10, 2007 at 11:39 AM, in message <[EMAIL PROTECTED]>,
Gregory Haskins wrote:
> We could change the API to handle the pending condition as a tri-level
I just thought of a better way. Leave the return value as a boolean, but use
the concept of the mask_reasons. It simplifies the
Gregory Haskins wrote:
>
>> where - >pending() has to program a tpr exit if current tpr masks an
>> interrupt.
>>
>
> I wasn't planning on the irqdevice having this type of control because I
> assumed we would always have these exit types set, but you have highlighted
> an interesting po
Gregory Haskins wrote:
On Tue, Apr 10, 2007 at 11:39 AM, in message <[EMAIL PROTECTED]>,
> Gregory Haskins wrote:
>
>> We could change the API to handle the pending condition as a tri-level
>>
>
> I just thought of a better way. Leave the return value as a boolean, but
Li, Xin B wrote:
> I can't get the source code through http or https, can anyone help?
>
>
A snapshot (kvm-18) is available through the sourceforge download page
(http://sourceforge.net/project/showfiles.php?group_id=180599). If the
git: and svn: protocols are blocked for you, I can try to
>>> On Tue, Apr 10, 2007 at 12:11 PM, in message <[EMAIL PROTECTED]>,
Avi Kivity <[EMAIL PROTECTED]> wrote:
>
>
> It's more complex than that;
Isn't it always ;)
> you need to program the hardware to exit on
> a specific threshold level of tpr values (e.g. exit when tpr <= n).
>
Agreed, but
Avi Kivity <[EMAIL PROTECTED]> writes:
> That's a good indication the problem is with handling I/O instructions.
> I have an image that exhibits similar behavior (but on Windows chkdsk on
> boot) so I'll try debugging that. If that does not work I'll try to
> send you debug patches.
>
Would be
On Mon, 2007-04-09 at 16:38 +0300, Avi Kivity wrote:
> Moreover, some things just don't lend themselves to a userspace
> abstraction. If we want to expose tso (tcp segmentation offload), we
> can easily do so with a kernel driver since the kernel interfaces are
> all tso aware. Tacking on tso
Rusty Russell wrote:
> On Mon, 2007-04-09 at 16:38 +0300, Avi Kivity wrote:
>
>> Moreover, some things just don't lend themselves to a userspace
>> abstraction. If we want to expose tso (tcp segmentation offload), we
>> can easily do so with a kernel driver since the kernel interfaces are
>>
31 matches
Mail list logo
