2014-03-11 5:21 GMT+08:00 Eric Blake :
> On 03/10/2014 02:29 PM, Eric Blake wrote:
>
> >> +opt = qemu_opt_find(opts, name);
> >> +if (opt) {
> >> +g_free((char *)opt->str);
> >
> > ...which means the cast is pointless here.
> >
> > Hmm. This means that you are giving opt_set() the
Signed-off-by: Fam Zheng
---
block.c | 30 --
include/block/block.h | 4
2 files changed, 32 insertions(+), 2 deletions(-)
diff --git a/block.c b/block.c
index 26a43eb..199a465 100644
--- a/block.c
+++ b/block.c
@@ -52,6 +52,8 @@
struct BdrvDirt
For "dirty-bitmap" sync mode, the block job will iterate through the
given dirty bitmap to decide if a sector needs backup (backup all the
dirty clusters and skip clean ones), just as allocation conditions of
"top" sync mode.
There are two bitmap use modes for sync=dirty-bitmap:
- reset: backup
This returns the granularity (in sectors) of dirty bitmap.
Signed-off-by: Fam Zheng
---
block.c | 6 ++
include/block/block.h | 2 ++
2 files changed, 8 insertions(+)
diff --git a/block.c b/block.c
index 9de1385..26a43eb 100644
--- a/block.c
+++ b/block.c
@@ -5143,6 +5143,12 @
2014-03-11 5:44 GMT+08:00 Eric Blake :
> On 03/10/2014 01:31 AM, Chunyan Liu wrote:
> > In qemu_opt_set functions, if desc doen't exist but opts_accepts_any is
> true, it
>
> s/doen't/doesn't/
>
> I mentioned the same problem against v20. It is very depressing when
> review comments are not addre
This allows to put the dirty bitmap into a disabled state where no more
writes will be tracked.
It will be used before backup or writing to persistent file.
Signed-off-by: Fam Zheng
---
block.c | 15 +++
blockdev.c| 44 ++
This field will be set for user created dirty bitmap. Also pass in an
error pointer to bdrv_create_dirty_bitmap, so when a name is already
taken on this BDS, it can report an error message. This is not global
check, two BDSes can have dirty bitmap with a common name.
Implemented bdrv_find_dirty_bi
This makes a deep copy of an HBitmap.
Signed-off-by: Fam Zheng
---
include/qemu/hbitmap.h | 8
util/hbitmap.c | 13 +
2 files changed, 21 insertions(+)
diff --git a/include/qemu/hbitmap.h b/include/qemu/hbitmap.h
index 550d7ce..b645cfc 100644
--- a/include/qemu/hbi
This adds dirty-bitmap-add and dirty-bitmap-disable to transactions.
With this, user can stop a dirty bitmap, start backup of it, and start
another dirty bitmap atomically, so that the dirty bitmap is tracked
incrementally and we don't miss any write.
Signed-off-by: Fam Zheng
---
blockdev.c
The new command pair is added to manage user created dirty bitmap. The
dirty bitmap's name is mandatory and must be unique for the same device,
but different devices can have bitmaps with the same names.
Signed-off-by: Fam Zheng
---
blockdev.c | 60 +
Introduction
This implements incremental backup.
A convenient option "bitmap-use-mode" for drive-backup is introduced since v1.
Commands
A new sync mode for drive-backup is introduced:
drive-backup device=.. mode=.. sync=dirty-bitmap bitmap=bitmap0
bitmap-use-mode=re
bdrv_getlength could fail, check the return value before using it.
Signed-off-by: Fam Zheng
---
block.c | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/block.c b/block.c
index ce48fff..9de1385 100644
--- a/block.c
+++ b/block.c
@@ -5083,7 +5083,12 @@ BdrvDirtyBitmap
*b
2014-03-12 0:54 GMT+08:00 Eric Blake :
> On 03/10/2014 01:31 AM, Chunyan Liu wrote:
> > Change block layer to support both QemuOpts and QEMUOptionParameter.
> > After this patch, it will change backend drivers one by one. At the end,
> > QEMUOptionParameter will be removed and only QemuOpts is kep
On Tue, 2014-03-11 at 15:00 -0300, Eduardo Habkost wrote:
> On Tue, Mar 11, 2014 at 06:58:53PM +0800, chen.fan.fnst wrote:
> > From: "chen.fan.fnst"
> >
> > Signed-off-by: Chen Fan
> > ---
> > hw/i386/pc.c | 12
> > target-i386/cpu.c | 2 --
> > target-i386/cpu.h | 3 +++
> >
The patch adds a spapr-pci-vfio-host-bridge device type
which is a PCI Host Bridge with VFIO support. The new device
inherits from the spapr-pci-host-bridge device and adds
the following properties:
iommu - IOMMU group ID which represents a Partitionable
Endpoint, QEMU/ppc64
This adds SPAPR VFIO IOMMU device in order to support DMA operations
for VFIO devices.
Signed-off-by: Alexey Kardashevskiy
---
Changes:
v7:
* fixed to adjust changes to support VFIO KVM device
---
hw/ppc/spapr_iommu.c | 97 ++
include/hw/ppc/spap
This turns the sPAPR support on and enables VFIO container use
in the kernel.
Signed-off-by: Alexey Kardashevskiy
---
Changes:
v5:
* multiple returns converted to gotos
v4:
* fixed format string to use %m which is a glibc extension:
"Print output of strerror(errno). No argument is required."
---
This adds macro to extend signed 64bit value to signed 128bit value.
Signed-off-by: Alexey Kardashevskiy
---
Changes:
v2:
* (.hi = (a >> 63) ? -1 : 0) changed to (.hi = (a < 0) ? -1 : 0)
---
include/qemu/int128.h | 5 +
1 file changed, 5 insertions(+)
diff --git a/include/qemu/int128.h b/in
From: David Gibson
The only model so far supported for VFIO passthrough devices is the model
usually used on x86, where all of the guest's RAM is mapped into the
(host) IOMMU and there is no IOMMU visible in the guest.
This patch begins to relax this model, introducing the notion of a
VFIOAddres
Upcoming VFIO on SPAPR PPC64 support will initialize the IOMMU
memory region with UINT64_MAX (2^64 bytes) size so int128_get64()
will assert.
The patch takes care of this check. The existing type1 IOMMU code
is not expected to map all 64 bits of RAM so the patch does not
touch that part.
Signed-o
Yet another try with VFIO on SPAPR (server PPC64).
As the previous try was too long time ago, I did not bother with
the change log much as all of this requires review again. Also,
it depends on these 2 patchsets which I cannot get reviewed yet
(keep pinging...):
[PATCH] spapr-iommu: extend SPAPR_TC
From: David Gibson
This patch uses the new IOMMU notifiers to allow VFIO pass through devices
to work with guest side IOMMUs, as long as the host-side VFIO iommu has
sufficient capability and granularity to match the guest side. This works
by tracking all map and unmap operations on the guest IOM
From: David Gibson
At the moment, most AddressSpace objects last as long as the guest system
in practice, but that could well change in future. In addition, for VFIO
we will be introducing some private per-AdressSpace information, which must
be disposed of before the AddressSpace itself is destr
From: David Gibson
So far, VFIO has a notion of different logical DMA address spaces, but
only ever uses one (system memory). This patch extends this, creating
new VFIOAddressSpace objects as necessary, according to the AddressSpace
reported by the PCI subsystem for this device's DMAs.
This isn
This reworks vfio_connect_container() and vfio_get_group() to have
common exit path at the end of the function bodies.
Signed-off-by: Alexey Kardashevskiy
---
hw/misc/vfio.c | 60 --
1 file changed, 33 insertions(+), 27 deletions(-)
diff -
As sPAPR platform supports DMA windows on a PCI bus, the information
about their location and size should be passed into the guest via
the device tree.
The patch adds a helper to read this info from the container fd.
Signed-off-by: Alexey Kardashevskiy
---
Changes:
v6:
* added dup() to protect g
On 12/20/2013 01:47 PM, Alexey Kardashevskiy wrote:
> On 12/05/2013 08:39 PM, Alexey Kardashevskiy wrote:
>> On 11/20/2013 04:39 PM, Alexey Kardashevskiy wrote:
>>> This adds a put_tce() callback to the SPAPR TCE TABLE device class.
>>> The new callback allows to have different IOMMU types such as
On 02/12/2014 10:44 PM, Alexey Kardashevskiy wrote:
> On 01/15/2014 06:10 PM, Alexey Kardashevskiy wrote:
>> On 12/20/2013 01:47 PM, Alexey Kardashevskiy wrote:
>>> On 12/05/2013 08:39 PM, Alexey Kardashevskiy wrote:
On 11/21/2013 03:08 PM, Alexey Kardashevskiy wrote:
> Here are few rework
2014-03-11 19:59 GMT+08:00 Eric Blake :
> On 03/10/2014 11:29 PM, Chunyan Liu wrote:
> >>
> >> Why are your later callers using a common helper function, but
> >> qemu_opt_get_del() repeating a lot of the body of qemu_opt_get()?
> >> Shouldn't qemu_opt_get() and qemu_opt_get_del() call into a comm
2014-03-11 21:29 GMT+08:00 Stefan Hajnoczi :
> On Mon, Mar 10, 2014 at 03:31:37PM +0800, Chunyan Liu wrote:
> > Add def_value_str (default value) to QemuOptDesc, to replace function of
> the
> > default value in QEMUOptionParameter. And improved related functions.
> >
> > Signed-off-by: Dong Xu Wa
On 03/06/2014 12:59 PM, Alexey Kardashevskiy wrote:
> On 03/06/2014 07:42 AM, Andreas Färber wrote:
>> Am 14.02.2014 05:05, schrieb Alexey Kardashevskiy:
>>> On 02/11/2014 02:48 AM, Andreas Färber wrote:
Am 10.02.2014 04:52, schrieb Alexey Kardashevskiy:
> This adds a test if SPAPR PHB can
On Tue, Mar 11, 2014 at 09:03:29AM +0100, Paolo Bonzini wrote:
> Il 11/03/2014 04:50, Hu Tao ha scritto:
> >On Tue, Mar 04, 2014 at 03:00:45PM +0100, Paolo Bonzini wrote:
> >>Like the previous patch did in exec.c, split memory_region_init_ram and
> >>memory_region_init_ram_from_file, and push mem_p
On Wed, Mar 12, 2014 at 12:03:19AM +, Peter Maydell wrote:
> On 11 March 2014 23:58, Andreas Färber wrote:
> > Am 20.02.2014 16:58, schrieb Peter Maydell:
> >> On 16 February 2014 02:07, Edgar E. Iglesias
> >> wrote:
> >>> On Sat, Feb 15, 2014 at 03:42:56PM +, Peter Maydell wrote:
>
Am 10.03.2014 08:52, schrieb Paolo Bonzini:
> Il 10/03/2014 01:15, Andreas Färber ha scritto:
>> Note that while such functions may exist both for *-user and softmmu,
>> only *-user uses the CPUState hook, while softmmu reuses the prototype
>> for calling it directly.
>
> I'm not sure I understand
On 11 March 2014 23:58, Andreas Färber wrote:
> Am 20.02.2014 16:58, schrieb Peter Maydell:
>> On 16 February 2014 02:07, Edgar E. Iglesias
>> wrote:
>>> On Sat, Feb 15, 2014 at 03:42:56PM +, Peter Maydell wrote:
On 13 February 2014 05:07, wrote:
> From: "Edgar E. Iglesias"
>
Am 20.02.2014 16:58, schrieb Peter Maydell:
> On 16 February 2014 02:07, Edgar E. Iglesias wrote:
>> On Sat, Feb 15, 2014 at 03:42:56PM +, Peter Maydell wrote:
>>> On 13 February 2014 05:07, wrote:
From: "Edgar E. Iglesias"
cpu->exit_request is part of the execution environme
Hi Max,
Am 11.03.2014 16:02, schrieb Max Filippov:
> Hi Andreas,
>
> On Mon, Mar 10, 2014 at 4:15 AM, Andreas Färber wrote:
>> diff --git a/target-xtensa/op_helper.c b/target-xtensa/op_helper.c
>> index 4265378..8641e5d 100644
>> --- a/target-xtensa/op_helper.c
>> +++ b/target-xtensa/op_helper.c
Cc: Luiz Capitulino
Cc: Markus Armbruster
Signed-off-by: Cole Robinson
---
include/monitor/monitor.h | 1 -
monitor.c | 27 ---
stubs/Makefile.objs| 1 -
stubs/mon-print-filename.c | 6 --
4 files changed, 35 deletions(-)
delete mode 100
These errors don't seem user initiated, so forcibly printing to the
monitor doesn't seem right. Just print to stderr.
Cc: Anthony Liguori
Cc: Gerd Hoffmann
Signed-off-by: Cole Robinson
---
ui/vnc.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/ui/vnc.c b/ui/vnc.c
inde
These errors don't seem user initiated, so forcibly printing to the
monitor doesn't seem right. Just print to stderr.
Drop lprint since it's now unused.
Cc: Jan Kiszka
Signed-off-by: Cole Robinson
---
checkpatch flags some pre-existing tab issues, but I didn't retab. Should I?
slirp/misc.c |
On 03/06/2014 11:09 PM, Amos Kong wrote:
> vm_config_groups[] only contains part of the options which have
> parameters, and all options which have no parameter aren't added
> to vm_config_groups[]. Current query-command-line-options only
> checks options from vm_config_groups[], so some options wi
error_printf is just a wrapper around monitor_printf, which already
drops the requested output if cur_mon is qmp.
Cc: Luiz Capitulino
Cc: Markus Armbruster
Signed-off-by: Cole Robinson
---
hw/usb/bus.c| 2 +-
hw/usb/hcd-ehci.c | 4 ++--
include/qemu/error-report.h |
Cc: Luiz Capitulino
Cc: Markus Armbruster
Signed-off-by: Cole Robinson
---
include/qemu/error-report.h | 1 -
util/qemu-error.c | 2 +-
2 files changed, 1 insertion(+), 2 deletions(-)
diff --git a/include/qemu/error-report.h b/include/qemu/error-report.h
index 3b098a9..000eae3 100644
Minor error API cleanups, and one error reporting improvement when
connected to qmp (patch #6)
Cole Robinson (6):
slirp: Remove default_mon usage
vnc: Remove default_mon usage
error: Privatize error_print_loc
monitor: Remove unused monitor_print_filename
error: Remove redundant error_pri
monitor_printf will drop the requested output if cur_mon is qmp (for
good reason). However these messages are often helpful for debugging
issues with via libvirt.
If we know the message won't hit the monitor, send it to stderr.
Cc: Luiz Capitulino
Cc: Markus Armbruster
Signed-off-by: Cole Robin
On 3/11/2014 1:48 PM, Juan Quintela wrote:
wrote:
From: ChenLiang
It is inaccuracy and complex that using the transfer speed of
migration thread to determine whether the convergence migration.
The dirty page may be compressed by XBZRLE or ZERO_PAGE.The counter
of updating dirty bitmap will be
On 03/11/2014 04:15 PM, Juan Quintela wrote:
> Eric Blake wrote:
>> On 02/18/2014 01:50 AM, mrhi...@linux.vnet.ibm.com wrote:
>>> From: "Michael R. Hines"
>
>> We're building up a LOT of migrate- tunable commands. Maybe it's time
>> to think about building a more generic migrate-set-parameter,
On 03/11/2014 04:02 PM, Juan Quintela wrote:
> mrhi...@linux.vnet.ibm.com wrote:
>> From: "Michael R. Hines"
>>
>> +# @mc-net-disable: Deactivate network buffering against outbound network
>> +# traffic while Micro-Checkpointing (@mc) is active.
>> +# Enabled by default. Disabl
Am 10.03.2014 01:15, schrieb Andreas Färber:
> Signed-off-by: Andreas Färber
Needs:
diff --git a/target-arm/op_helper.c b/target-arm/op_helper.c
index af8e1dc..f8e535e 100644
--- a/target-arm/op_helper.c
+++ b/target-arm/op_helper.c
@@ -240,7 +240,7 @@ void HELPER(wfe)(CPUARMState *env)
*
Am 10.03.2014 01:15, schrieb Andreas Färber:
> Signed-off-by: Andreas Färber
Needs:
diff --git a/target-arm/op_helper.c b/target-arm/op_helper.c
index 8434f6e..931536e 100644
--- a/target-arm/op_helper.c
+++ b/target-arm/op_helper.c
@@ -232,10 +232,12 @@ void HELPER(wfi)(CPUARMState *env)
void
Eric Blake wrote:
> On 02/18/2014 01:50 AM, mrhi...@linux.vnet.ibm.com wrote:
>> From: "Michael R. Hines"
> We're building up a LOT of migrate- tunable commands. Maybe it's time
> to think about building a more generic migrate-set-parameter, which
> takes both the name of the parameter to set a
wrote:
> From: ChenLiang
>
> Clear the death code
>
> Signed-off-by: ChenLiang
> Signed-off-by: Gonglei
Reviewed-by: Juan Quintela
mrhi...@linux.vnet.ibm.com wrote:
> From: "Michael R. Hines"
>
> New capabilities include the use of RDMA acceleration,
> use of network buffering, and keepalive support, as documented
> in patch #1.
>
> Signed-off-by: Michael R. Hines
> ---
> qapi-schema.json | 36 ++
mrhi...@linux.vnet.ibm.com wrote:
> From: "Michael R. Hines"
>
> MC provides a lot of new information, including the same RAM statistics
> that ordinary migration does, so we centralize a lot of that printing
> code into a common function so that the QMP printing statements don't
> get duplicated
On 02/18/2014 01:50 AM, mrhi...@linux.vnet.ibm.com wrote:
> From: "Michael R. Hines"
>
> New capabilities include the use of RDMA acceleration,
> use of network buffering, and keepalive support, as documented
> in patch #1.
>
> Signed-off-by: Michael R. Hines
> ---
> qapi-schema.json | 36
mrhi...@linux.vnet.ibm.com wrote:
> From: "Michael R. Hines"
>
> This patch sets up the initial changes to the migration state
> machine and prototypes to be used by the checkpointing code
> to interact with the state machine so that we can later handle
> failure and recovery scenarios.
>
> Signed
When a quorum file is totally destroyed (broken NAS or SAN) the user can start a
drive-mirror job on the quorum block backend and then replace the broken
quorum file with drive-mirror-replace given it has a node-name.
Signed-off-by: Benoit Canet
---
block.c | 8 ++--
block/mi
Tests for drive-mirror-replace whose purpose is to enable quorum file mirroring
and replacement after failure.
Signed-off-by: Benoit Canet
---
tests/qemu-iotests/041| 34 +--
tests/qemu-iotests/088| 221 ++
tests/qemu-iotests/088.out
Hello,
This series for 2.1 add the missing maintainance feature for Quorum.
The first patch allows quorum to correct corrupted reads by rewriting them.
The second add the drive-mirror-replace command to be used on a drive-mirrored
device to replace a target bs by the mirror.
The series apply on
On read operations when this parameter is set and some replicas are corrupted
while quorum can be reached quorum will proceed to rewrite the correct version
of the data to fix the corrupted replicas.
This will shine with SSD where the FTL will remap the same block at another
place on rewrite.
Sig
On 02/18/2014 01:50 AM, mrhi...@linux.vnet.ibm.com wrote:
> From: "Michael R. Hines"
>
> This exposes a QMP command that allows the management software
> or policy to control the frequency of micro-checkpointing.
>
> Signed-off-by: Michael R. Hines
> ---
> hmp-commands.hx | 16 +++
On 02/18/2014 01:50 AM, mrhi...@linux.vnet.ibm.com wrote:
> From: "Michael R. Hines"
>
> MC provides a lot of new information, including the same RAM statistics
> that ordinary migration does, so we centralize a lot of that printing
> code into a common function so that the QMP printing statement
On 02/18/2014 01:50 AM, mrhi...@linux.vnet.ibm.com wrote:
> From: "Michael R. Hines"
>
> During micro-checkpointing, the VCPUs get repeatedly paused and
> resumed. We need to not freak out when the VM begins micro-checkpointing.
>
> Signed-off-by: Michael R. Hines
> ---
> cpus.c
mrhi...@linux.vnet.ibm.com wrote:
> From: "Michael R. Hines"
>
> We also later export these statistics over QMP for better
> monitoring of micro-checkpointing as the workload changes.
>
> Signed-off-by: Michael R. Hines
> ---
> arch_init.c | 34 --
> 1 file change
mrhi...@linux.vnet.ibm.com wrote:
> From: "Michael R. Hines"
>
> During micro-checkpointing, the VCPUs get repeatedly paused and
> resumed. We need to not freak out when the VM begins micro-checkpointing.
>
> Signed-off-by: Michael R. Hines
> diff --git a/include/migration/migration.h b/include/m
The Tuesday 11 Mar 2014 à 22:00:29 (+0100), Max Reitz wrote :
> On 11.03.2014 17:36, Benoît Canet wrote:
> >When a quorum file is totally destroyed (broken filer) the user can start a
>
> *file
I really meant filer as in file server: I will write NAS instead.
>
> >drive-mirror job on the quorum
Recursively walk all properties under /machine and try to retrieve their
value. This is a regression test for link<> properties and the
DeviceState::hotpluggable property.
Cf. be2f78b6b062eec5170e2612299fb8953046993f and
1a37eca107cece3ed454bae29eef0bd1fac4a244
Signed-off-by: Andreas Färber
Am 11.03.2014 11:50, schrieb Paolo Bonzini:
> From: "Liu, Jinsong"
>
> The correct size of cpuid 0x0d sub-leaf 4 is 0x40, not 0x10.
> This is confirmed by Anvin H Peter and Mallick Asit K.
>
> Signed-off-by: Liu Jinsong
> Cc: H. Peter Anvin
> Cc: Asit K Mallick
> Signed-off-by: Paolo Bonzini
wrote:
> From: ChenLiang
>
> expose the counter that log the times of updating the dirty bitmap to
> end user.
>
> Signed-off-by: ChenLiang
> Signed-off-by: Gonglei
Once Eric problems are fixed, rest of the patch looks ok.
Later, Juan.
wrote:
> From: ChenLiang
>
> Signed-off-by: ChenLiang
> Signed-off-by: Gonglei
> ---
> docs/xbzrle.txt | 7 +++
> 1 file changed, 7 insertions(+)
>
> diff --git a/docs/xbzrle.txt b/docs/xbzrle.txt
> index cc3a26a..cdf1e3e 100644
> --- a/docs/xbzrle.txt
> +++ b/docs/xbzrle.txt
> @@ -71,6 +7
wrote:
> From: ChenLiang
>
> Signed-off-by: ChenLiang
> Signed-off-by: Gonglei
You are changing teh types introduced in patch2, please fix them there?
Anyways, why are you changing the age to int64_t? Not that I expect it
to be so big, but I would expect it not to be negative?
Later, Juan.
Hello everyone,
This patch fixes the issue when -cpu specification is partially ignored
when -enable-kvm is specified. For example, |when running:
qemu-system-x86_64 -cpu core2duo
on AMD machine, the CPU vendor is properly identified as GenuineIntel.
However when running:
|
|qemu-system-x86
On 03/11/2014 06:53 AM, arei.gong...@huawei.com wrote:
> From: ChenLiang
>
The subject line says what, but there is no WHY. The commit message
should explain why you want a signed type.
> Signed-off-by: ChenLiang
> Signed-off-by: Gonglei
> ---
> arch_init.c| 4 ++--
> in
On 11.03.2014 17:36, Benoît Canet wrote:
When a quorum file is totally destroyed (broken filer) the user can start a
*file
drive-mirror job on the quorum block backend and then replace the broken
quorum file with drive-mirror-replace given it has a node-name.
Signed-off-by: Benoit Canet
---
Hi,
On Tue, Mar 11, 2014 at 03:26:51PM +0800, Chunyan Liu wrote:
> 2014-03-11 5:21 GMT+08:00 Eric Blake :
>
> > On 03/10/2014 02:29 PM, Eric Blake wrote:
> >
> > >> +opt = qemu_opt_find(opts, name);
> > >> +if (opt) {
> > >> +g_free((char *)opt->str);
> > >
> > > ...which means th
On 03/11/2014 06:53 AM, arei.gong...@huawei.com wrote:
> From: ChenLiang
>
> Clear the death code
s/death/dead/ (twice)
>
> Signed-off-by: ChenLiang
> Signed-off-by: Gonglei
> ---
> arch_init.c | 13 -
> page_cache.c | 58
On 03/11/2014 06:53 AM, arei.gong...@huawei.com wrote:
> From: ChenLiang
s/migraion/migration/ in the subject (here and in 7/10)
s/optimiztion/optimize/ in the subject
> +++ b/arch_init.c
> @@ -381,11 +381,8 @@ static int save_xbzrle_page(QEMUFile *f, uint8_t
> **current_data,
>
> prev_
wrote:
> From: ChenLiang
>
> Reducing data copy can reduce cpu overhead.
>
> Signed-off-by: ChenLiang
> Signed-off-by: Gonglei
Reviewed-by: Juan Quintela
wrote:
> From: ChenLiang
>
> Reducing data copy can reduce cpu overhead.
>
> Signed-off-by: ChenLiang
> Signed-off-by: Gonglei
Reviewed-by: Juan Quintela
On 03/11/2014 02:48 PM, Juan Quintela wrote:
> wrote:
>> From: ChenLiang
>>
>> It is inaccuracy and complex that using the transfer speed of
>> migration thread to determine whether the convergence migration.
>> The dirty page may be compressed by XBZRLE or ZERO_PAGE.The counter
>> of updating di
On 03/11/2014 06:53 AM, arei.gong...@huawei.com wrote:
> From: ChenLiang
>
> Avoid the hot pages being replaced by others to remarkable decrease cache
s/the hot/hot/
s/remarkable/remarkably/
> missing. The counter of updating dirty bitmap is used to indicate the cached
s/missing/misses/
> pag
wrote:
> From: ChenLiang
>
> It is inaccuracy and complex that using the transfer speed of
> migration thread to determine whether the convergence migration.
> The dirty page may be compressed by XBZRLE or ZERO_PAGE.The counter
> of updating dirty bitmap will be increasing continuously if the
> m
On 03/11/2014 02:35 PM, Michael S. Tsirkin wrote:
>> If we put a command line
>> argument on it then people will still be able to incorrectly
>> rely on thread naming by using the command line option.
>
>
> Where would you document this?
>
> The idea was that "debug" in name is a hint.
Or even
wrote:
> From: ChenLiang
>
> Rebuild the cache_is_cached function.
Why?
> +bool cache_is_cached(const PageCache *cache, uint64_t addr,
> + uint64_t current_age)
> +{
> +CacheItem *it = NULL;
s/= NULL//
Thanks, Juan.
> +
> +it = cache_get_by_addr(cache, addr);
> +
On Tue, Mar 11, 2014 at 08:27:58PM +, Peter Maydell wrote:
> On 11 March 2014 20:03, Michael S. Tsirkin wrote:
> > Because we don't want people to start relying on thread naming to
> > manage the threads outside qemu.
>
> Then document that it isn't stable; anybody who relies
> on it deserves
wrote:
> From: ChenLiang
>
> Avoid the hot pages being replaced by others to remarkable decrease cache
> missing. The counter of updating dirty bitmap is used to indicate the cached
> page age.
>
> Signed-off-by: ChenLiang
> Signed-off-by: Gonglei
> /**
> * get_cached_data: Get the data ca
On 11 March 2014 20:03, Michael S. Tsirkin wrote:
> Because we don't want people to start relying on thread naming to
> manage the threads outside qemu.
Then document that it isn't stable; anybody who relies
on it deserves what they get. If we put a command line
argument on it then people will st
"Gonglei (Arei)" wrote:
>> On 03/11/2014 06:53 AM, arei.gong...@huawei.com wrote:
>> > From: ChenLiang
>> >
>> > Add counters to log the times of updating the dirty bitmap.
>> >
>> > Signed-off-by: ChenLiang
>> > Signed-off-by: Gonglei
>> > Reviewed-by: Dr. David Alan Gilbert
>> > Reviewed-by:
wrote:
> From: ChenLiang
>
> The page may not be inserted into cache after executing save_xbzrle_page.
> In case of failure to insert, the original page should be sent rather
> than the page in the cache.
>
> Signed-off-by: ChenLiang
> Signed-off-by: Gonglei
> Reviewed-by: Dr. David Alan Gilber
The Tuesday 11 Mar 2014 à 20:58:06 (+0100), Max Reitz wrote :
> On 11.03.2014 17:36, Benoît Canet wrote:
> >On read operations when this parameter is set and some replicas are corrupted
> >while quorum can be reached quorum will proceed to rewrite the correct
> >version
> >of the data to fix the c
* Peter Maydell (peter.mayd...@linaro.org) wrote:
> On 11 March 2014 19:46, Dr. David Alan Gilbert wrote:
> > * Michael S. Tsirkin (m...@redhat.com) wrote:
> >
> >> Shouldn't we prevent the unsupported option from
> >> appearing in qemu output though?
> >> Otherwise how does user know whether it w
On Tue, Mar 11, 2014 at 07:53:20PM +, Peter Maydell wrote:
> On 11 March 2014 19:46, Dr. David Alan Gilbert wrote:
> > * Michael S. Tsirkin (m...@redhat.com) wrote:
> >
> >> Shouldn't we prevent the unsupported option from
> >> appearing in qemu output though?
> >> Otherwise how does user know
Finally I tracked down the reason why my early test failed, but then
it just started working "by magic".
The reason is this:
If you:
- create an ext4 filesystem
- mount it WITHOUT -o discard
- create and remove some big files
- unmount
- mount -o discard
- fstrim
then the fstrim has no eff
On 11.03.2014 17:36, Benoît Canet wrote:
On read operations when this parameter is set and some replicas are corrupted
while quorum can be reached quorum will proceed to rewrite the correct version
of the data to fix the corrupted replicas.
This will shine with SSD where the FTL will remap the s
On 11 March 2014 10:50, Paolo Bonzini wrote:
> The following changes since commit 9fbee91a131a05e443d7108d7fbdf3ca91020290:
>
> Merge remote-tracking branch 'remotes/kvm/uq/master' into staging
> (2014-02-27 16:00:31 +)
>
> are available in the git repository at:
>
>
> git://git.kernel.or
On 11 March 2014 19:46, Dr. David Alan Gilbert wrote:
> * Michael S. Tsirkin (m...@redhat.com) wrote:
>
>> Shouldn't we prevent the unsupported option from
>> appearing in qemu output though?
>> Otherwise how does user know whether it works?
>
> I'll make it spit a warning if you try and enable it
* Michael S. Tsirkin (m...@redhat.com) wrote:
> Shouldn't we prevent the unsupported option from
> appearing in qemu output though?
> Otherwise how does user know whether it works?
I'll make it spit a warning if you try and enable it on a platform
that doesn't have a way of doing it; ifdef all th
On Tue, 2014-03-11 at 18:11 +0100, Andreas Färber wrote:
> Hi Marcel,
>
> Am 11.03.2014 16:35, schrieb Marcel Apfelbaum:
> > To be applied on top of:
> > hw/boards: converted current_machine to be an instance of MachineClass
> > Rebased for qom-next branch of Andrea's QOM tree.
> >
> > Patch
Alexey Kardashevskiy wrote:
> This replaces DPRINTF macro with tracepoints.
>
> This moves some messages from migration.c to savevm.c.
>
> This adds tracepoint to signal about fileds failed to migrate.
>
> Signed-off-by: Alexey Kardashevskiy
> ---
> Changes:
> v4:
> * made use of new qemu_ether_n
Alexey Kardashevskiy wrote:
> This adds @idstr to savevm_section_start and savevm_section_end
> tracepoints.
>
> Signed-off-by: Alexey Kardashevskiy
Already in
1 - 100 of 313 matches
Mail list logo
