Il 10/03/2013 16:24, Gleb Natapov ha scritto:
> On Sun, Mar 10, 2013 at 04:04:39PM +0100, Paolo Bonzini wrote:
>> Il 10/03/2013 15:55, Gleb Natapov ha scritto:
> Why not move INIT case from below as is? Vcpu is reset to correct sate
> by QEMU just like during system_reset.
APs wou
On 03/08/2013 07:03 PM, Stefan Hajnoczi wrote:
> On Thu, Mar 07, 2013 at 12:52:42PM +0200, Michael S. Tsirkin wrote:
>> On Thu, Mar 07, 2013 at 06:33:30PM +0800, Jason Wang wrote:
>>> On 03/07/2013 06:25 PM, Michael S. Tsirkin wrote:
On Thu, Mar 07, 2013 at 06:13:41PM +0800, Jason Wang wrote:
Il 11/03/2013 02:08, Anthony Liguori ha scritto:
> Juan Quintela writes:
>
>> Hi, Anthony
>>
>> please pull the migraiton queue, it includes:
>> - paolo cleanup & remove buffering series (already reviewed upstream)
>> - v2 of XBZRLE fixes from Orit
>
> After waiting for 10 minutes for repo.or.cz
On 03/08/13 17:36, Eric Blake wrote:
> On 02/28/2013 12:50 AM, Gerd Hoffmann wrote:
>> This patch adds 'udb' support to qapi.
>
> If it's not too late:
>
> s/udb/udp/
It's not, need to rebase after flow control series, I'll fix it up
along the way.
cheers,
Gerd
On Sat, Mar 9, 2013 at 4:18 PM, Kashyap Chamarthy wrote:
> Heya,
>
> After building qemu, I tried to invoke QMP by via telnet:
>
> Build:
> ==
> #./configure --target-list=x86_64-softmmu --disable-werror
> --enable-debug ; make -j5
> ==
>
> Invoke QMP:
> ==
> kashyap@qemu$
On Sun, Mar 10, 2013 at 09:01:52PM +0100, Samuel Thibault wrote:
> Stefan Hajnoczi, le Wed 06 Mar 2013 13:29:37 +0100, a écrit :
> > > What do people think about it?
> >
> > We should fix the layer that introduces the problem. Therefore I think
> > the fix needs to be net/socket.c.
> >
> > Unfor
On (Tue) 05 Mar 2013 [12:54:06], Lucas Meneghel Rodrigues wrote:
> On 03/05/2013 07:11 AM, Amit Shah wrote:
> >On (Mon) 04 Mar 2013 [16:53:55], Stefan Hajnoczi wrote:
> >>How are things looking with device emulation, migration, monitor, char, etc?
> >
> >I have a few tests coded up for virtio-seri
On 11 March 2013 01:59, Peter Crosthwaite wrote:
> On Mon, Mar 11, 2013 at 1:04 AM, Peter Maydell
> wrote:
>> Oh, and I forgot -- don't try to build system emulators with --static:
>> it's only really meant for the linux-user targets. Do specify
>
> Works for us quite nicely. Any known issues he
> > We track zero blocks at 4KB level to get small backup files. Restore
> > can handle those blocks differently, either pre-allocate or create holes.
> > That
> can even be a restore option.
>
> You are right, the behavior can be set at restore time.
>
> I'm curious how much of a win the 4 KB z
If an io_flush handler is not set, qemu_aio_wait doesn't invoke
callbacks.
Signed-off-by: MORITA Kazutaka
---
block/sheepdog.c | 11 +--
1 file changed, 9 insertions(+), 2 deletions(-)
diff --git a/block/sheepdog.c b/block/sheepdog.c
index e4ec32d..cb0eeed 100644
--- a/block/sheepdog.c
This is a fix for https://bugs.launchpad.net/qemu/+bug/1127053.
If pxe-e1000.rom has 0 size we allocate the same RAMBlock offset twice and fail
with an assertion in qemu_ram_set_idstr() later on.
Two fixes:
1. Print an error when the ROM file has zero size. Use -device ...,romfile=
instead t
find_ram_offset() does not handle size=0 gracefully. It hands out the
same RAMBlock offset multiple times, leading to obscure failures later
on.
Add an assert to warn early if something is incorrectly allocating a
zero size RAMBlock.
Signed-off-by: Stefan Hajnoczi
---
exec.c | 2 ++
1 file cha
A zero size ROM file is invalid and should produce a warning.
Attempting to use a zero size file ends up hitting an assertion
qemu_ram_set_idstr() because RAMBlocks with duplicate addresses are
allocated - due to zero size the allocator doesn't increment the next
available RAMBlock offset.
Also co
Wenchao Xia writes:
> 于 2013-3-7 15:36, Markus Armbruster 写道:
>> I intend to review at least the QMP interfaces. Unfortunately, I'm
>> struggling with getting my review queue under control. Please be
>> patient.
>>
> Thanks for your attention. I am coding for next version
> to address Eric's
Kashyap Chamarthy writes:
> Heya,
>
> After building qemu, I tried to invoke QMP by via telnet:
Exact version, please. Commit hash if you build from git.
> Build:
> ==
> #./configure --target-list=x86_64-softmmu --disable-werror
> --enable-debug ; make -j5
> ==
>
> Invoke QMP:
On Mon, Mar 11, 2013 at 10:20:19AM +0100, Stefan Hajnoczi wrote:
> This is a fix for https://bugs.launchpad.net/qemu/+bug/1127053.
>
> If pxe-e1000.rom has 0 size we allocate the same RAMBlock offset twice and
> fail
> with an assertion in qemu_ram_set_idstr() later on.
>
> Two fixes:
>
> 1. Pr
On 10/03/2013 17:28, Leonid Bloch wrote:
On Sun, Mar 10, 2013 at 4:59 PM, Peter Maydell wrote:
On 10 March 2013 14:36, Leonid Bloch wrote:
I want to cross compile QEMU itself to run on ARM, but all my searches
lead me to tutorials for cross compiling other programs for ARM and
testing them wi
Gerd Hoffmann writes:
> Hi,
>
> Little input patch series, first patch being a small cleanup, second
> patch adds a linked list of keyboard handlers, so removing a handler
> (on usb kbd unplug) will not make keyboard stop working but fallback
> to another keyboard (if present).
Reviewed-by: Ma
Gerd Hoffmann writes:
> Add a linked list of keyboard handlers. Added handlers will go
> to the head of the list. Removed handlers will be zapped from
> the list. The head of the list will be used for events.
>
> This fixes the keyboard-dead-after-usb-kbd-unplug issue, key events
> will be re-
> > > We track zero blocks at 4KB level to get small backup files. Restore
> > > can handle those blocks differently, either pre-allocate or create
> > > holes. That
> > can even be a restore option.
> >
> > You are right, the behavior can be set at restore time.
> >
> > I'm curious how much of a w
Igor Mitsyanko writes:
> Current colon position in "waiting for telnet connection" message template
> produces messages like:
> QEMU waiting for connection on: telnet::127.0.0.1,server
>
> After moving a colon to the right, we will get a correct messages like:
> QEMU waiting for connection on
On Fri, Mar 08, 2013 at 07:58:32PM +0100, Stefan Weil wrote:
> Instead of adding missing type casts which are needed by MinGW for the
> 4th argument, the patch uses qemu_setsockopt which was invented for this
> purpose.
>
> Signed-off-by: Stefan Weil
> ---
>
> I did not fix some coding style iss
brdv_truncate() is also called from readv/writev commands on self-
growing file based storage. this will result in requests waiting
for theirselves to complete.
This reverts commit 9a665b2b8640e464f0a778216fc2dca8d02acf33.
---
block.c |4
1 file changed, 4 deletions(-)
diff --git a/blo
this patch ensures that all pending IOs are completed
before a device is resized. this is especially important
if a device is shrinked as it the bdrv_check_request()
result is invalidated.
Signed-off-by: Peter Lieven
---
blockdev.c |3 +++
1 file changed, 3 insertions(+)
diff --git a/block
ensure that there are no pending I/Os before calling
the sync readcapacity commands. the block_resize monitor
command will also flush all I/O, but double check in
case iscsi_truncate() is called from elsewhere in the
future.
Signed-off-by: Peter Lieven
---
block/iscsi.c |4
1 file chan
On 03/11/13 10:25, Markus Armbruster wrote:
> Gerd Hoffmann writes:
>
>> Add a linked list of keyboard handlers. Added handlers will go
>> to the head of the list. Removed handlers will be zapped from
>> the list. The head of the list will be used for events.
>>
>> This fixes the keyboard-dead
On Mon, Mar 11, 2013 at 10:54 AM, KONRAD Frédéric
wrote:
> On 10/03/2013 17:28, Leonid Bloch wrote:
>>
>> On Sun, Mar 10, 2013 at 4:59 PM, Peter Maydell
>> wrote:
>>>
>>> On 10 March 2013 14:36, Leonid Bloch wrote:
I want to cross compile QEMU itself to run on ARM, but all my searches
Il 11/03/2013 11:05, Peter Lieven ha scritto:
> ensure that there are no pending I/Os before calling
> the sync readcapacity commands. the block_resize monitor
> command will also flush all I/O, but double check in
> case iscsi_truncate() is called from elsewhere in the
> future.
>
> Signed-off-by
Am 11.03.2013 um 11:16 schrieb Paolo Bonzini :
> Il 11/03/2013 11:05, Peter Lieven ha scritto:
>> ensure that there are no pending I/Os before calling
>> the sync readcapacity commands. the block_resize monitor
>> command will also flush all I/O, but double check in
>> case iscsi_truncate() is ca
A conflict was resolved the wrong way when merging commit 320ba5f (build:
always link device_tree.o into emulators if libfdt available, 2013-02-05).
This causes a build failure for the arm-softmmu target due to multiply
defined symbol.
Signed-off-by: Paolo Bonzini
---
hw/arm/Makefile.objs | 1 -
On 11 March 2013 10:31, Paolo Bonzini wrote:
> A conflict was resolved the wrong way when merging commit 320ba5f (build:
> always link device_tree.o into emulators if libfdt available, 2013-02-05).
> This causes a build failure for the arm-softmmu target due to multiply
> defined symbol.
>
> Signe
On 25 February 2013 16:44, Peter Maydell wrote:
> This patchset fixes a number of warnings produced by Linux 3.8
> on the vexpress boards resulting from newer kernels prodding
> more of the SYS_CFG register operations.
>
> Since some of the newly implemented functions expose voltage
> and oscillat
The general ideas are (earlier rules override the later, and exceptions
are of course possible):
- Board files go in hw/ARCH/ (already done).
- Files go in hw/ARCH/ if they reference that arch's CPUState (already done)
- Files already in a subdirectory are not moved (examples: some PPC/S390
fi
On 03/10/2013 09:06 AM, Jan Kiszka wrote:
>> @@ -2491,6 +2493,10 @@ static int gdb_handle_packet(GDBState *s, const char
>> *line_buf)
>> break;
>> }
>> #endif
>> +if (strncmp(p, "Attached", 8) == 0) {
>> +put_packet(s, gdb_attached ? "1" : "0");
>> +
In the use of snapshot a way to retrieve related info at runtime is needed,
so this serial of patches will merge some code for qemu and qemu-img, and add
following interfaces for qemu:
1) qmp: query-images, show image info for a block device
Example:
-> { "execute": "query-images" }
<- {
"
This patch adds block/snapshot.c and then moves the function
there. It also fixes small code style errors reported by check script.
Signed-off-by: Wenchao Xia
---
block/Makefile.objs |1 +
block/snapshot.c | 37 +
include/block/snapshot.h
This patch adds function bdrv_query_snapshot_info_list(), which will
retrieve snapshot info of an image in qmp object format. The implementation
is based on the code moved from qemu-img.c with modification to fit more
for qmp based block layer API.
Signed-off-by: Wenchao Xia
Reviewed-by: Eric B
This patch adds block/qapi.c and moves the functions there. To avoid
conflict and tip better, macro in header file is BLOCK_QAPI_H instead
of QAPI_H. The moving is for making review easier, those functions
will be modified and renamed later.
Signed-off-by: Wenchao Xia
---
block/Makefile.objs
This mirror function will return image info including snapshots,
and if specified backing image's info will also be returned. Now
Qemu have both query-images and query-block interfaces.
Signed-off-by: Wenchao Xia
---
block/qapi.c | 83 +
qa
Parameter *fmt was not used, so remove it.
Signed-off-by: Wenchao Xia
Reviewed-by: Eric Blake
---
qemu-img.c |5 ++---
1 files changed, 2 insertions(+), 3 deletions(-)
diff --git a/qemu-img.c b/qemu-img.c
index 471de7d..f4e5d90 100644
--- a/qemu-img.c
+++ b/qemu-img.c
@@ -1640,8 +1640,7
This patch adds function bdrv_query_image_info(), which will
retrieve image info in qmp object format. The implementation is
based on the code moved from qemu-img.c, but uses block layer
function to get snapshot info.
Signed-off-by: Wenchao Xia
Reviewed-by: Eric Blake
---
block/qapi.c
This patch using new added function in last commit which retrieve
info from qmp for snapshot info.
Signed-off-by: Wenchao Xia
---
monitor.c |2 +-
savevm.c | 64 -
2 files changed, 1 insertions(+), 65 deletions(-)
diff --git a
This function will simply call qmp interface qmp_query_snapshots()
added in last commit and then dump information in monitor console.
To get snapshot info, Now qemu and qemu-img both call block layer
function bdrv_query_snapshot_info_list() in their calling path, and
then they just translate th
This interface returns info of valid internal snapshots for whole vm.
Signed-off-by: Wenchao Xia
---
block/qapi.c | 18 +
qapi-schema.json | 14 +
qmp-commands.hx | 55 ++
3 files changed, 87 insertions(
This command will show block image's information, including
internal snapshots.
Signed-off-by: Wenchao Xia
---
hmp.c | 38 ++
hmp.h |1 +
monitor.c |7 +++
3 files changed, 46 insertions(+), 0 deletions(-)
diff --git a/hmp.c b/hmp.c
in
This function is needed later in hmp command, it is also renamed to
bdrv_image_info_dump().
Signed-off-by: Wenchao Xia
---
block/qapi.c | 67
include/block/qapi.h |1 +
qemu-img.c | 69 +-
On 11 March 2013 11:17, Paolo Bonzini wrote:
> - Files go in hw/ARCH/ if they used to be in hw/ARCH/Makefile.objs and
> they define multiple devices (example: hw/arm/tc6393xb.c)
Why are multiple devices in one file a bad thing (or at least, a
thing that should determine where a file lives)? Sho
On 2013-03-11 12:22, Fabien Chouteau wrote:
> On 03/10/2013 09:06 AM, Jan Kiszka wrote:
>>> @@ -2491,6 +2493,10 @@ static int gdb_handle_packet(GDBState *s, const char
>>> *line_buf)
>>> break;
>>> }
>>> #endif
>>> +if (strncmp(p, "Attached", 8) == 0) {
>>> +
Hi ! I am trying to emulate mips on Qemu but during the kernel loading
process ,Qemu is sending a hardware interrupt which the kernel is not able
to handle and stucks in a loop while the actual hardware and simulator
don't send interrupt at that point.I want to know that which specific
function in
This patch adds a parameter to tell whether return valid snapshots
for whole VM only.
Signed-off-by: Wenchao Xia
---
block/qapi.c | 42 --
include/block/qapi.h |1 +
qemu-img.c |3 ++-
3 files changed, 43 insertions(+), 3 dele
To make it clear about id and name in searching, the API is changed
a bit to distinguish them, and caller can choose to search by id or name.
Searching will be done with higher priority of id. This function also
returns negative value from bdrv_snapshot_list() instead of -ENOENT on
error now.
Si
Il 11/03/2013 11:19, Peter Lieven ha scritto:
>
> Am 11.03.2013 um 11:16 schrieb Paolo Bonzini :
>
>> Il 11/03/2013 11:05, Peter Lieven ha scritto:
>>> ensure that there are no pending I/Os before calling
>>> the sync readcapacity commands. the block_resize monitor
>>> command will also flush all
This allow hmp use this function, just like qemu-img.
It also returns a pointer now to make it easy to use.
Signed-off-by: Wenchao Xia
---
block/qapi.c | 67 +++--
include/block/qapi.h |2 +-
qemu-img.c |6 +++-
3 files ch
On Mon, Mar 11, 2013 at 12:33:27AM -0400,
michael.r.hines.mrhi...@linux.vnet.ibm.com wrote:
> From: "Michael R. Hines"
>
>
> Signed-off-by: Michael R. Hines
> ---
> docs/rdma.txt | 93
> +
> 1 file changed, 93 insertions(+)
> create
On 11.03.2013 12:44, Paolo Bonzini wrote:
Il 11/03/2013 11:19, Peter Lieven ha scritto:
Am 11.03.2013 um 11:16 schrieb Paolo Bonzini :
Il 11/03/2013 11:05, Peter Lieven ha scritto:
ensure that there are no pending I/Os before calling
the sync readcapacity commands. the block_resize monitor
c
On Mon, Mar 11, 2013 at 12:17:20PM +0100, Paolo Bonzini wrote:
> The general ideas are (earlier rules override the later, and exceptions
> are of course possible):
>
> - Board files go in hw/ARCH/ (already done).
>
> - Files go in hw/ARCH/ if they reference that arch's CPUState (already done)
>
Il 11/03/2013 12:31, Peter Maydell ha scritto:
> On 11 March 2013 11:17, Paolo Bonzini wrote:
>> - Files go in hw/ARCH/ if they used to be in hw/ARCH/Makefile.objs and
>> they define multiple devices (example: hw/arm/tc6393xb.c)
>
> Why are multiple devices in one file a bad thing (or at least,
On 03/11/2013 12:36 PM, Jan Kiszka wrote:
> On 2013-03-11 12:22, Fabien Chouteau wrote:
>> On 03/10/2013 09:06 AM, Jan Kiszka wrote:
diff --git a/qemu-options.hx b/qemu-options.hx
index 6f9334a..026d3eb 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -2988,6 +2988,13
On Fri, 8 Mar 2013 21:11:13 +0100
Alexander Graf wrote:
>
> On 25.02.2013, at 12:10, Christian Borntraeger wrote:
>
> > On 25/02/13 11:44, Paolo Bonzini wrote:
> >> Il 25/02/2013 09:09, Christian Borntraeger ha scritto:
> >>> Hmm, the old sequence was
> >>>
> >>> object_unparent(OBJECT(de
Il 11/03/2013 13:04, Cornelia Huck ha scritto:
> On Fri, 8 Mar 2013 21:11:13 +0100
> Alexander Graf wrote:
>
>>
>> On 25.02.2013, at 12:10, Christian Borntraeger wrote:
>>
>>> On 25/02/13 11:44, Paolo Bonzini wrote:
Il 25/02/2013 09:09, Christian Borntraeger ha scritto:
> Hmm, the old se
On 03/11/2013 01:16 PM, Paolo Bonzini wrote:
Il 11/03/2013 13:04, Cornelia Huck ha scritto:
On Fri, 8 Mar 2013 21:11:13 +0100
Alexander Graf wrote:
On 25.02.2013, at 12:10, Christian Borntraeger wrote:
On 25/02/13 11:44, Paolo Bonzini wrote:
Il 25/02/2013 09:09, Christian Borntraeger ha sc
On Mon, 11 Mar 2013 13:22:45 +0100
Alexander Graf wrote:
> On 03/11/2013 01:16 PM, Paolo Bonzini wrote:
> > Il 11/03/2013 13:04, Cornelia Huck ha scritto:
> >> On Fri, 8 Mar 2013 21:11:13 +0100
> >> Alexander Graf wrote:
> >>
> >>> On 25.02.2013, at 12:10, Christian Borntraeger wrote:
> >>>
> >>
Hi Anthony
please pull the migraiton queue, it includes:
v2:
- rebased (althought there was zero conflicts)
- moved to github (I was not having troubles with repo.or.cz, though)
v1:
- paolo cleanup & remove buffering series (already reviewed upstream)
- v2 of XBZRLE fixes from Orit
Thanks, Juan
Il 11/03/2013 13:39, Peter Maydell ha scritto:
> On 11 March 2013 11:54, Paolo Bonzini wrote:
>> Il 11/03/2013 12:31, Peter Maydell ha scritto:
>>> On 11 March 2013 11:17, Paolo Bonzini wrote:
hw/arm11mpcore.c hw/arm/arm11mpcore.c
>>>
>>> Two devices but I can spl
Le Thursday 07 Feb 2013 à 10:48:14 (+0100), Stefan Hajnoczi a écrit :
> On Wed, Feb 06, 2013 at 01:31:46PM +0100, Benoît Canet wrote:
> > diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
> > index ef91216..5b1d20d 100644
> > --- a/block/qcow2-cluster.c
> > +++ b/block/qcow2-cluster.c
> >
Il 11/03/2013 12:52, Peter Lieven ha scritto:
>>
>> For example, qcow1 (and VHDX) will already call it, but that's a bug and
>> should be fixed otherwise. Your patch will just cause an assertion
>> failure.
>
> In which case can qcow1 (and VHDX) be used in conjunction with an iSCSI
> Target?
No.
Paolo Bonzini writes:
> A conflict was resolved the wrong way when merging commit 320ba5f (build:
> always link device_tree.o into emulators if libfdt available, 2013-02-05).
> This causes a build failure for the arm-softmmu target due to multiply
> defined symbol.
>
> Signed-off-by: Paolo Bonzin
Gerd Hoffmann writes:
> On 03/11/13 10:25, Markus Armbruster wrote:
>> Gerd Hoffmann writes:
>>
>>> Add a linked list of keyboard handlers. Added handlers will go
>>> to the head of the list. Removed handlers will be zapped from
>>> the list. The head of the list will be used for events.
>>>
On 11 March 2013 12:44, Paolo Bonzini wrote:
> Il 11/03/2013 13:39, Peter Maydell ha scritto:
>> On 11 March 2013 11:54, Paolo Bonzini wrote:
>>> These are little more than SoC containers, aren't they?
>>
>> They're container devices, yes. But why should container devices
>> go under hw/$ARCH ?
>
Il 11/03/2013 14:08, Peter Maydell ha scritto:
>>> These are little more than SoC containers, aren't they?
>>> >>
>>> >> They're container devices, yes. But why should container devices
>>> >> go under hw/$ARCH ?
>> >
>> > Because they don't really implement any logic, ideally a board should b
Peter Maydell writes:
> On 11 March 2013 11:54, Paolo Bonzini wrote:
>> Il 11/03/2013 12:31, Peter Maydell ha scritto:
>>> On 11 March 2013 11:17, Paolo Bonzini wrote:
hw/arm11mpcore.c hw/arm/arm11mpcore.c
>>>
>>> Two devices but I can split them if you insist.
Il 11/03/2013 05:33, michael.r.hines.mrhi...@linux.vnet.ibm.com ha scritto:
> common-obj-y += migration.o migration-tcp.o
> +common-obj-$(CONFIG_RDMA) += migration-rdma.o
> common-obj-y += qemu-char.o #aio.o
> common-obj-y += block-migration.o
> -common-obj-y += page_cache.o xbzrle.o
> +common-o
On Mon, Mar 11, 2013 at 12:17:20PM +0100, Paolo Bonzini wrote:
> hw/dataplane/Makefile.objs hw/block/dataplane/Makefile.objs
> hw/dataplane/hostmem.c hw/block/dataplane/hostmem.c
> hw/dataplane/ioq.c hw/block/dataplane/ioq.c
> hw/dat
Il 11/03/2013 05:33, michael.r.hines.mrhi...@linux.vnet.ibm.com ha scritto:
> From: "Michael R. Hines"
>
> This is the loadvm() side of the connection which is RDMA-aware,
> so that transfer of device state can use the same abstractions
> as all of the other migration protocols.
>
> Full documen
Il 11/03/2013 05:33, michael.r.hines.mrhi...@linux.vnet.ibm.com ha scritto:
> From: "Michael R. Hines"
>
> Use 'migrate rdma:host:port' to begin the migration.
>
> The TCP control channel has finally been eliminated
> when RDMA is used. Documentation of the use of SEND message
> for transferring
Hi,
I ever since had a few VMs which are very hard to migrate because of a lot of
memory I/O. I found that finding the next dirty bit
seemed to be one of the culprits (apart from removing locking which Paolo is
working on).
I have to following proposal which seems to help a lot in my case. Jus
On 8 March 2013 14:22, Gerd Hoffmann wrote:
> Add a linked list of keyboard handlers. Added handlers will go
> to the head of the list. Removed handlers will be zapped from
> the list. The head of the list will be used for events.
>
> This fixes the keyboard-dead-after-usb-kbd-unplug issue, key
Il 11/03/2013 05:33, michael.r.hines.mrhi...@linux.vnet.ibm.com ha scritto:
> From: "Michael R. Hines"
>
> Very little changes here except for halting the migration after a maximum
> number of iterations is reached.
>
> When comparing against TCP, the migration never ends if we don't cap
> the m
Anthony Liguori writes:
> Gerd Hoffmann writes:
>
>> On 03/11/13 10:25, Markus Armbruster wrote:
>>> Gerd Hoffmann writes:
>>>
Add a linked list of keyboard handlers. Added handlers will go
to the head of the list. Removed handlers will be zapped from
the list. The head of th
Il 11/03/2013 05:33, michael.r.hines.mrhi...@linux.vnet.ibm.com ha scritto:
> From: "Michael R. Hines"
>
> For performance reasons, dup_page() and xbzrle() is skipped because
> they are too expensive for zero-copy RDMA.
>
> Signed-off-by: Michael R. Hines
> ---
> arch_init.c | 57 +++
Il 11/03/2013 05:33, michael.r.hines.mrhi...@linux.vnet.ibm.com ha scritto:
> From: "Michael R. Hines"
>
> ... want to use existing functions to do this.
Remember that each of the 10 steps should compile and link, both with
and without librdmacm installed. So this should be quite earlier.
> Si
On 11 March 2013 13:44, Peter Lieven wrote:
> @@ -24,12 +24,13 @@ unsigned long find_next_bit(const unsigned long *addr,
> unsigned long size,
> const unsigned long *p = addr + BITOP_WORD(offset);
> unsigned long result = offset & ~(BITS_PER_LONG-1);
> unsigned long tmp;
> +uns
Il 11/03/2013 05:33, michael.r.hines.mrhi...@linux.vnet.ibm.com ha scritto:
> From: "Michael R. Hines"
>
> RDMA needs access to this structure by including cpu-common.h
> (Including cpu-all.h causes things to throw up on me).
>
> Signed-off-by: Michael R. Hines
> ---
> include/exec/cpu-all.h
On Wed, Mar 06, 2013 at 06:02:01PM +0100, Paolo Bonzini wrote:
> Otherwise, live migration of the top layer will miss zero clusters and
> let the backing file show through. This also matches what is done in qed.
>
> QCOW2_CLUSTER_ZERO clusters are invalid in v2 image files. Check this
> directly
Il 11/03/2013 14:44, Peter Lieven ha scritto:
> Hi,
>
> I ever since had a few VMs which are very hard to migrate because of a
> lot of memory I/O. I found that finding the next dirty bit
> seemed to be one of the culprits (apart from removing locking which
> Paolo is working on).
>
> I have to f
On Fri, Mar 08, 2013 at 03:52:15PM +0100, Stefan Hajnoczi wrote:
> Make sure to rebuild generated-events.o when ./configure options change.
> This prevents linker errors when a stale generated-events.o gets linked
> with code compiled against fresh headers. For example, try building
> with ./confi
On Fri, Mar 08, 2013 at 04:10:10PM +0100, Stefan Hajnoczi wrote:
> The .gitignore file no longer covers all autogenerated tracing file. These
> patches update .gitignore.
>
> Stefan Hajnoczi (2):
> .gitignore: add trace/generated-events.[ch]
> .gitignore: rename trace/generated-tracers.dtrace
On Tue, Mar 05, 2013 at 02:47:21PM +0100, Lluís Vilanova wrote:
> Provides a generic event state description structure (TraceEvent) and a more
> detailed event control and query interface.
>
> This is achieved by creating a new "non-public" tracing backend (i.e., not
> selectable by the user at co
Am 11.03.2013 um 15:14 schrieb Paolo Bonzini :
> Il 11/03/2013 14:44, Peter Lieven ha scritto:
>> Hi,
>>
>> I ever since had a few VMs which are very hard to migrate because of a
>> lot of memory I/O. I found that finding the next dirty bit
>> seemed to be one of the culprits (apart from removin
On Mon, Mar 11, 2013 at 09:26:26AM +, Dietmar Maurer wrote:
> > > > We track zero blocks at 4KB level to get small backup files. Restore
> > > > can handle those blocks differently, either pre-allocate or create
> > > > holes. That
> > > can even be a restore option.
> > >
> > > You are right,
Am 11.03.2013 um 15:22 schrieb Peter Lieven :
>
> Am 11.03.2013 um 15:14 schrieb Paolo Bonzini :
>
>> Il 11/03/2013 14:44, Peter Lieven ha scritto:
>>> Hi,
>>>
>>> I ever since had a few VMs which are very hard to migrate because of a
>>> lot of memory I/O. I found that finding the next dirty
On Thu, Mar 07, 2013 at 08:37:17 -0700, Eric Blake wrote:
> On 03/07/2013 01:23 AM, Jason Wang wrote:
> > Sometimes, we need track the state when guest is just about to start after
> > migration. There's not a accurate state available which do this accurately
> > (consider qemu may started with -S
Il 11/03/2013 15:22, Peter Lieven ha scritto:
>
> Am 11.03.2013 um 15:14 schrieb Paolo Bonzini :
>
>> Il 11/03/2013 14:44, Peter Lieven ha scritto:
>>> Hi,
>>>
>>> I ever since had a few VMs which are very hard to migrate because of a
>>> lot of memory I/O. I found that finding the next dirty bit
On Mon, Mar 11, 2013 at 06:01:02PM +0900, MORITA Kazutaka wrote:
> If an io_flush handler is not set, qemu_aio_wait doesn't invoke
> callbacks.
>
> Signed-off-by: MORITA Kazutaka
> ---
> block/sheepdog.c | 11 +--
> 1 file changed, 9 insertions(+), 2 deletions(-)
>
> diff --git a/block/
Hi,
> The major difference is that where this code sends key events
> to the first handler only, the Nokia patch sends events to
> all keys, because the use case it's trying to deal with is
> "some keys should go to one device and some to another"
> (for example, most keys go to a usb keyboard b
>>> #if defined(TARGET_ARM) || defined(TARGET_M68K) ||
>>> defined(TARGET_UNICORE32)
>>> /* Extra fields for semihosted binaries. */
>>> -uint32_t stack_base;
>>> uint32_t heap_base;
>>> uint32_t heap_limit;
>>> #endif
>>> +uint32_t stack_base;
>> Shouldn't this be abi_ulon
> > I can see space reduction up to 4% using this feature. Considering the
> > fact that it comes at no cost, it would be stupid to remove it.
>
> Okay, looks like it's useful but not a huge win.
>
> In the NBD approach pipelining writes (or discards) ought to make 4 KB zero
> blocks usable witho
Hi Peter,
On 9 March 2013 13:14, Peter Maydell wrote:
> linux-user: Fix layout of usage table to account for option text
> http://patchwork.ozlabs.org/patch/220481/
> linux-user: make bogus negative iovec lengths fail EINVAL
> http://patchwork.ozlabs.org/patch/219240/
> linux-user: Implement
Il 07/03/2013 10:33, Stefan Hajnoczi ha scritto:
>> > I have made some fix for this series, apart from the concert about
>> > GSource/ AioContext. Hope to discuss it clearly in next version and
>> > fix it too. BTW what can we benefit from AioContext besides those from
>> > GSource
> One thing I f
> > +if (dedup) {
> > +version = 3;
> > +}
>
> Lazy refcounts don't force the version. It would be consistent to
> refrain from forcing the version too.
I don't understand. Aren't the incompatible features bits implying version = 3 ?
When compat=1.1 is set the code force version
1 - 100 of 239 matches
Mail list logo
