Hi everyone,
Could you please tell me if it is possible to use a block driver that
completely avoids the guest kernel and copies block data directly
to/from the given buffer in the guest userspace?
If yes, how to activate it? If not... why not? :)
Thanks!
Darko
--
To unsubscribe from this lis
On 01/28/2011 02:52 PM, Glauber Costa wrote:
Register steal time within KVM. Everytime we sample the steal time
information, we update a local variable that tells what was the
last time read. We then account the difference.
Signed-off-by: Glauber Costa
CC: Rik van Riel
CC: Jeremy Fitzhardinge
CC
On Fri, 2011-01-28 at 22:09 +0100, Jan Kiszka wrote:
> On 2011-01-28 20:48, Glauber Costa wrote:
> > Up to know, we were relying on guest cooperation to turn off kvmclock.
> > I just realized that even though this is fine and nice, a more robust
> > method is to (also) turn it off on vcpu_reset on
On 01/28/2011 02:52 PM, Glauber Costa wrote:
This patch accounts steal time time in kernel/sched.
I kept it from last proposal, because I still see advantages
in it: Doing it here will give us easier access from scheduler
variables such as the cpu rq. The next patch shows an example of
usage for
On 01/28/2011 02:52 PM, Glauber Costa wrote:
To implement steal time, we need the hypervisor to pass the guest information
about how much time was spent running other processes outside the VM.
This is per-vcpu, and using the kvmclock structure for that is an abuse
we decided not to make.
In this
On 01/28/2011 02:52 PM, Glauber Costa wrote:
To implement steal time, we need the hypervisor to pass the guest information
about how much time was spent running other processes outside the VM.
This is per-vcpu, and using the kvmclock structure for that is an abuse
we decided not to make.
In this
On Fri, 2011-01-28 at 17:16 -0800, Jeremy Fitzhardinge wrote:
> On 01/28/2011 11:52 AM, Glauber Costa wrote:
> > This patch accounts steal time time in kernel/sched.
> > I kept it from last proposal, because I still see advantages
> > in it: Doing it here will give us easier access from scheduler
>
On 01/28/2011 11:52 AM, Glauber Costa wrote:
> This patch accounts steal time time in kernel/sched.
> I kept it from last proposal, because I still see advantages
> in it: Doing it here will give us easier access from scheduler
> variables such as the cpu rq. The next patch shows an example of
> us
From: Jan Stancek
If autotest client is run with custom harness, its name is stored
in state file. This allows autotest to continue (e.g. after reboot)
with harness that was specified on first run.
If "--harness" option is present on command line, this
option is used regardless of what is stored
ste...@us.ibm.com wrote on 01/28/2011 12:29:37 PM:
> > On Thu, 2011-01-27 at 22:05 +0200, Michael S. Tsirkin wrote:
> > > One simple theory is that guest net stack became faster
> > > and so the host can't keep up.
> >
> > Yes, that's what I think here. Some qdisc code has been changed
> > recentl
On 2011-01-28 20:48, Glauber Costa wrote:
> Up to know, we were relying on guest cooperation to turn off kvmclock.
> I just realized that even though this is fine and nice, a more robust
> method is to (also) turn it off on vcpu_reset on the hypervisor side.
> This will protect us against reboots,
Hi there,
GSoC 2011 has been announced[1]. As we were pretty successful last year,
I think we should participate again. I've already created a wiki page:
http://wiki.qemu.org/Google_Summer_of_Code_2011
We should now populate it with projects and people willing to be mentors
should say so (or ju
Register steal time within KVM. Everytime we sample the steal time
information, we update a local variable that tells what was the
last time read. We then account the difference.
Signed-off-by: Glauber Costa
CC: Rik van Riel
CC: Jeremy Fitzhardinge
CC: Peter Zijlstra
CC: Avi Kivity
---
arch/
This is a first proposal for using steal time information
to influence the scheduler. There are a lot of optimizations
and fine grained adjustments to be done, but it is working reasonably
so far for me (mostly)
With this patch (and some host pinnings to demonstrate the situation),
two vcpus with
To implement steal time, we need the hypervisor to pass the guest information
about how much time was spent running other processes outside the VM.
This is per-vcpu, and using the kvmclock structure for that is an abuse
we decided not to make.
In this patchset, I am introducing a new msr, KVM_MSR_
This patch adds documentation about usage of the newly
introduced KVM_MSR_STEAL_TIME.
Signed-off-by: Glauber Costa
---
Documentation/kvm/msr.txt | 32
1 files changed, 32 insertions(+), 0 deletions(-)
diff --git a/Documentation/kvm/msr.txt b/Documentation/kvm/
To implement steal time, we need the hypervisor to pass the guest information
about how much time was spent running other processes outside the VM.
This is per-vcpu, and using the kvmclock structure for that is an abuse
we decided not to make.
In this patchset, I am introducing a new msr, KVM_MSR_
Gentlemen,
If I may please steal a little bit of your time, for the revision of
this patchset, this is appreciated. This version only includes the steal
time part, using a new MSR+cpuid to register it.
It also includes some documentation, describing the ABI.
Mostly inspired by the kvmclock one.
This patch accounts steal time time in kernel/sched.
I kept it from last proposal, because I still see advantages
in it: Doing it here will give us easier access from scheduler
variables such as the cpu rq. The next patch shows an example of
usage for it.
Since functions like account_idle_time() c
Up to know, we were relying on guest cooperation to turn off kvmclock.
I just realized that even though this is fine and nice, a more robust
method is to (also) turn it off on vcpu_reset on the hypervisor side.
This will protect us against reboots, and we don't expect the guest
to reset its cpu dur
mashi...@linux.vnet.ibm.com wrote on 01/27/2011 02:15:05 PM:
> On Thu, 2011-01-27 at 22:05 +0200, Michael S. Tsirkin wrote:
> > One simple theory is that guest net stack became faster
> > and so the host can't keep up.
>
> Yes, that's what I think here. Some qdisc code has been changed
> recently.
Gleb Natapov writes:
> On Fri, Jan 28, 2011 at 02:29:50PM +0100, Markus Armbruster wrote:
>> Gleb Natapov writes:
>>
>> > If bootindex is specified on command line a string that describes device
>> > in firmware readable way is added into sorted list. Later this list will
>> > be passed into fi
On Fri, Jan 28, 2011, Juerg Haefliger wrote about "Re: [PATCH 0/29] nVMX:
Nested VMX, v8":
> This branch doesn't even compile:
>...
> drivers/staging/smbfs/dir.c:286: error: static declaration of
> גsmbfs_dentry_operationsג follows non-static declaration
>...
Thanks, I'll look as soon as pos
On Fri, 2011-01-28 at 18:29 +0100, André Weidemann wrote:
> Hi Alex,
>
> On 28.01.2011 01:45, Alex Williamson wrote:
>
> >> Do you mind sharing these patches?
> >
> > Attached.
>
> Thank you for attaching the patch. Unfortunately it does not apply to
> current clone of the qemu-kvm git reposito
Hi Alex,
On 28.01.2011 01:45, Alex Williamson wrote:
Do you mind sharing these patches?
Attached.
Thank you for attaching the patch. Unfortunately it does not apply to
current clone of the qemu-kvm git repository. The file hw/vfio.c does
not exist in the public repository, but your patch
On Fri, Jan 28, 2011 at 1:13 PM, Himanshu Chauhan
wrote:
> I just cloned qemu-kvm, built and installed it. But the qemu-img fails
> to create any disk image above 1G. The problem as I see is use of
> ssize_t for image size. When size is >2G, the check "if (sval < 0)"
> succeeds and I get the error
On Fri, Jan 28, 2011 at 02:29:50PM +0100, Markus Armbruster wrote:
> Gleb Natapov writes:
>
> > If bootindex is specified on command line a string that describes device
> > in firmware readable way is added into sorted list. Later this list will
> > be passed into firmware to control boot order.
On Fri, Jan 28, 2011 at 03:31:17PM +, Prasad Joshi wrote:
>On Fri, Jan 28, 2011 at 3:15 PM, Joerg Roedel wrote:
>
> On Thu, Jan 27, 2011 at 06:43:19PM +, Prasad Joshi wrote:
> >AMD-Vi: Event logged [IO_PAGE_FAULT device=02:00.0 domain=0x0002
> >address=0xbb447000 flags=0
On 01/28/2011 04:31 PM, Yoshiaki Tamura wrote:
That's the hack I was imaging:)
So your original patch is also a hack? :)
Maybe this is just an issue of preference, but I'm not sure
adding "kemari:" to be intuitive. If there were similar
extensions having the same problem, I would have agreed
2011/1/29 Paolo Bonzini :
> On 01/28/2011 04:05 PM, Yoshiaki Tamura wrote:
>>
>> Having a scheme like "kemari:tcp:host:port" looks quite
>> challenging to me. We can of course add some quick hacks for it,
>> but adding a nice layered architecture should be more
>> appropriate. Similar to protocol
On Fri, 28 Jan 2011, KAMEZAWA Hiroyuki wrote:
> > > I see it as a tradeoff of when to check? add_to_page_cache or when we
> > > are want more free memory (due to allocation). It is OK to wakeup
> > > kswapd while allocating memory, somehow for this purpose (global page
> > > cache), add_to_page_ca
On 01/28/2011 04:05 PM, Yoshiaki Tamura wrote:
Having a scheme like "kemari:tcp:host:port" looks quite
challenging to me. We can of course add some quick hacks for it,
but adding a nice layered architecture should be more
appropriate. Similar to protocols and formats in block layer?
At the same
2011/1/28 Paolo Bonzini :
> On 01/28/2011 02:53 PM, Yoshiaki Tamura wrote:
>>>
>>> > 1) I am not sure what would happen with -incoming exec;
>>
>> Nothing happens if used with other protocols, but I assume you're
>> mentioning that it's not clear from the code, which makes sense.
>
> I assume noth
On 01/28/2011 02:53 PM, Yoshiaki Tamura wrote:
> 1) I am not sure what would happen with -incoming exec;
Nothing happens if used with other protocols, but I assume you're
mentioning that it's not clear from the code, which makes sense.
I assume nothing just because the code for other protoco
2011/1/28 Paolo Bonzini :
> On 01/28/2011 08:21 AM, Yoshiaki Tamura wrote:
>>
>> +int qemu_savevm_trans_begin(Monitor *mon, QEMUFile *f, int init)
>> +{
>> + SaveStateEntry *se;
>> + int skipped = 0;
>> +
>> + QTAILQ_FOREACH(se,&savevm_handlers, entry) {
>> + int len, stage, ret;
>>
2011/1/28 Paolo Bonzini :
> On 01/28/2011 08:21 AM, Yoshiaki Tamura wrote:
>>
>> + /* check ft_mode option */
>> + p = strstr(uri, "ft_mode");
>> + if (p&& !strcmp(p, "ft_mode")) {
>> + ft_mode = FT_INIT;
>> + }
>> +
>
> This works for TCP mode, but:
>
> 1) I am not sure what w
Gleb Natapov writes:
> If bootindex is specified on command line a string that describes device
> in firmware readable way is added into sorted list. Later this list will
> be passed into firmware to control boot order.
>
> Signed-off-by: Gleb Natapov
Out of curiosity: what about qdev "scsi-gen
On 01/28/2011 08:21 AM, Yoshiaki Tamura wrote:
+int qemu_savevm_trans_begin(Monitor *mon, QEMUFile *f, int init)
+{
+SaveStateEntry *se;
+int skipped = 0;
+
+QTAILQ_FOREACH(se,&savevm_handlers, entry) {
+int len, stage, ret;
+
+if (se->save_live_state == NULL) {
+
Hi,
I just cloned qemu-kvm, built and installed it. But the qemu-img fails
to create any disk image above 1G. The problem as I see is use of
ssize_t for image size. When size is >2G, the check "if (sval < 0)"
succeeds and I get the error:
qemu-img: Invalid image size specified! You may use k, M,
Am 27.01.2011 17:33, schrieb MORITA Kazutaka:
> This patch parses the input filename in sd_create(), and enables us
> specifying a target server to create sheepdog images.
>
> Signed-off-by: MORITA Kazutaka
Thanks, applied to the block branch.
Kevin
--
To unsubscribe from this list: send the li
On 01/28/2011 08:21 AM, Yoshiaki Tamura wrote:
+/* check ft_mode option */
+p = strstr(uri, "ft_mode");
+if (p&& !strcmp(p, "ft_mode")) {
+ft_mode = FT_INIT;
+}
+
This works for TCP mode, but:
1) I am not sure what would happen with -incoming exec;
2) it is tricky!
On Thu, Jan 27, 2011 at 01:30:38PM -0800, Shirley Ma wrote:
> On Thu, 2011-01-27 at 13:02 -0800, David Miller wrote:
> > > Interesting. Could this is be a variant of the now famuous
> > bufferbloat then?
> >
> > Sigh, bufferbloat is the new global warming... :-/
>
> Yep, some places become colde
* KAMEZAWA Hiroyuki [2011-01-28 17:17:44]:
> On Fri, 28 Jan 2011 13:49:28 +0530
> Balbir Singh wrote:
>
> > * KAMEZAWA Hiroyuki [2011-01-28 16:56:05]:
>
> > > BTW, it seems this doesn't work when some apps use huge shmem.
> > > How to handle the issue ?
> > >
> >
> > Could you elaborate furt
* MinChan Kim [2011-01-28 16:24:19]:
> >
> > But the assumption for LRU order to change happens only if the page
> > cannot be successfully freed, which means it is in some way active..
> > and needs to be moved no?
>
> 1. holded page by someone
> 2. mapped pages
> 3. active pages
>
> 1 is rare
Hi all,
I have to set up a virtual network environment, with +/- 15 virtual
hosts and synchronize some of those virtual machines with ptp (precision
time protocol). PTP was developed to provide very precise time
coordination of LAN connected computers. The aim is to be a lot more
precise than
Am 28.01.2011 08:21, schrieb Yoshiaki Tamura:
> event-tap function is called only when it is on, and requests were
> sent from device emulators.
>
> Signed-off-by: Yoshiaki Tamura
Acked-by: Kevin Wolf
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to
Hi Nadav,
> Hi,
>
> This is the eighth iteration of the nested VMX patch set. This iteration
> solves a number of bugs and issues that bothered the reviewers. Some more
> issues raised in the previous review remain open, but don't worry - I *am*
> working to resolve all of them.
>
> The biggest i
On 01/27/2011 02:09 PM, Jan Kiszka wrote:
Will be required for SIGBUS handling. For obvious reasons, this will
remain a nop on Windows hosts.
Signed-off-by: Jan Kiszka
---
Makefile.objs |2 +-
cpus.c| 117 +++--
2 files changed
On Fri, 28 Jan 2011 13:49:28 +0530
Balbir Singh wrote:
> * KAMEZAWA Hiroyuki [2011-01-28 16:56:05]:
> > BTW, it seems this doesn't work when some apps use huge shmem.
> > How to handle the issue ?
> >
>
> Could you elaborate further?
>
==
static inline unsigned long zone_unmapped_file_pages
* KAMEZAWA Hiroyuki [2011-01-28 16:56:05]:
> On Fri, 28 Jan 2011 16:24:19 +0900
> Minchan Kim wrote:
>
> > On Fri, Jan 28, 2011 at 3:48 PM, Balbir Singh
> > wrote:
> > > * MinChan Kim [2011-01-28 14:44:50]:
> > >
> > >> On Fri, Jan 28, 2011 at 11:56 AM, Balbir Singh
> > >> wrote:
> > >> > O
On 01/27/2011 02:09 PM, Jan Kiszka wrote:
Move qemu_kvm_eat_signals around and call it also when the IO-thread is
not used. Do not yet process SIGBUS, will be armed in a separate step.
Signed-off-by: Jan Kiszka
---
cpus.c | 88 ---
On 01/27/2011 02:09 PM, Jan Kiszka wrote:
Block SIG_IPI, unblock it during KVM_RUN, just like in io-thread mode.
It's unused so far, but this infrastructure will be required for
self-IPIs and to process SIGBUS plus, in KVM mode, SIGIO and SIGALRM. As
Windows doesn't support signal services, we ne
On Fri, 28 Jan 2011 16:24:19 +0900
Minchan Kim wrote:
> On Fri, Jan 28, 2011 at 3:48 PM, Balbir Singh
> wrote:
> > * MinChan Kim [2011-01-28 14:44:50]:
> >
> >> On Fri, Jan 28, 2011 at 11:56 AM, Balbir Singh
> >> wrote:
> >> > On Thu, Jan 27, 2011 at 4:42 AM, Minchan Kim
> >> > wrote:
> >>
53 matches
Mail list logo
