On Thu, Nov 27, 2014 at 3:45 PM, Fam Zheng wrote:
On Thu, 11/27 13:59, Jason Wang wrote:
virtio_net_handle_ctrl() and other functions that process control vq
request call iov_discard_front() which will shorten the iov. This
will
lead unmapping in virtqueue_push() leaks mapping.
Fixes
Kevin Wolf writes:
> Am 26.11.2014 um 16:20 hat Max Reitz geschrieben:
>> On 2014-11-26 at 16:19, Eric Blake wrote:
>> >On 11/26/2014 02:13 AM, Max Reitz wrote:
>> >>On 2014-11-26 at 08:23, Markus Armbruster wrote:
>> >>>Max Reitz writes:
>> >>>
>> Albeit absolutely impossible right now, bdr
Cc: Gonglei
Cc: Dr. David Alan Gilbert
Signed-off-by: Gerd Hoffmann
---
hw/input/hid.c | 21 +
1 file changed, 21 insertions(+)
diff --git a/hw/input/hid.c b/hw/input/hid.c
index 8f6fbb3..6fb963f 100644
--- a/hw/input/hid.c
+++ b/hw/input/hid.c
@@ -519,6 +519,27 @@ static i
Hi Eric,
On 2014/10/31 21:53, Eric Auger wrote:
> load_dtb is renamed into arm_load_dtb and becomes non static.
> it will be used by machvirt for dynamic instantiation of
> platform devices
>
> Signed-off-by: Eric Auger
>
> ---
>
> v2 -> v3:
> load_dtb renamed into arm_load_dtb
>
> Conflicts:
Eric Blake writes:
> On 11/26/2014 05:43 AM, Max Reitz wrote:
>> On 2014-11-25 at 20:46, John Snow wrote:
>>> From: Fam Zheng
>>>
>>> Signed-off-by: Fam Zheng
>>> Signed-off-by: John Snow
>>> ---
>>> block.c | 35 +++
>>> include/block/block.h |
On Thu, 11/27 13:59, Jason Wang wrote:
> virtio_net_handle_ctrl() and other functions that process control vq
> request call iov_discard_front() which will shorten the iov. This will
> lead unmapping in virtqueue_push() leaks mapping.
>
> Fixes this by keeping the original iov untouched and using
On (Thu) 27 Nov 2014 [16:48:10], David Gibson wrote:
> VirtIO devices now remember which endianness they're operating in in order
> to support targets which may have guests of either endianness, such as
> powerpc. This endianness state is transferred in a subsection of the
> virtio device's inform
> From: Paolo Bonzini [mailto:pbonz...@redhat.com]
> On 26/11/2014 11:40, Pavel Dovgalyuk wrote:
> > + * @QEMU_CLOCK_VIRTUAL_RT: realtime clock used for icount warp
> > + *
> > + * This clock runs as a realtime clock, but is used for icount warp
> > + * and thus should be traced with record/replay
On Wed, 11/26 12:23, Vladimir Sementsov-Ogievskiy wrote:
> Mirror and migration use dirty bitmaps for their purposes, and since
> commit [block: per caller dirty bitmap] they use their own bitmaps, not
> the global one. But they use old functions bdrv_set_dirty and
> bdrv_reset_dirty, which change
John Snow writes:
> On 11/26/2014 07:19 AM, Max Reitz wrote:
>> On 2014-11-25 at 20:46, John Snow wrote:
>>> From: Fam Zheng
>>>
>>> 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 differe
On Wed, 11/26 12:41, John Snow wrote:
> From: Fam Zheng
>
> 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.
>
>
Paolo Bonzini writes:
> Use the external qemu-timer API instead.
Ignorant question: why?
On 11/27/2014 10:00 AM, Shannon Zhao wrote:
> Hi Eric,
>
> On 2014/10/31 21:53, Eric Auger wrote:
>> load_dtb is renamed into arm_load_dtb and becomes non static.
>> it will be used by machvirt for dynamic instantiation of
>> platform devices
>>
>> Signed-off-by: Eric Auger
>>
>> ---
>>
>> v2 ->
On (Tue) 25 Nov 2014 [09:32:58], z08687 wrote:
> Public bug reported:
>
> Hi,
> I found a live migration problem and have found out the reason, but I can't
> fix it up myself. I really need help.
> When live migration vm and it's block device in save time, it will occur
> probabilistic .
>
>
On 2014/11/27 17:04, Gerd Hoffmann wrote:
> Cc: Gonglei
> Cc: Dr. David Alan Gilbert
> Signed-off-by: Gerd Hoffmann
> ---
> hw/input/hid.c | 21 +
> 1 file changed, 21 insertions(+)
>
> diff --git a/hw/input/hid.c b/hw/input/hid.c
> index 8f6fbb3..6fb963f 100644
> --- a/hw
David Gibson writes:
> VirtIO devices now remember which endianness they're operating in in order
> to support targets which may have guests of either endianness, such as
> powerpc. This endianness state is transferred in a subsection of the
> virtio device's information.
>
> With virtio-rng thi
On 2014-11-13 at 11:36, Mao Chuan Li wrote:
Currently the qemu-iotest framework is focused on the x86 platform. The suite
should be usable by other platforms as well.
To enable multi-platform usage, this patch set enhances the framework scripts
to address this. By using the default machine
type
On 31.10.2014 11:59, Stefan Hajnoczi wrote:
On Fri, Oct 31, 2014 at 09:31:53AM +0100, Peter Lieven wrote:
Am 28.10.2014 um 12:11 schrieb Stefan Hajnoczi:
On Sat, Oct 25, 2014 at 06:55:50PM +0200, Peter Lieven wrote:
The big question is whether these user-visible interfaces make sense if
write m
On 2014-11-26 at 18:41, John Snow wrote:
From: Fam Zheng
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 dir
On 2014-11-26 at 18:41, John Snow wrote:
From: Fam Zheng
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.
The granularity is an optional fi
On 2014-11-26 at 18:41, John Snow wrote:
From: Fam Zheng
This returns the granularity (in bytes) of dirty bitmap,
which matches the QMP interface and the existing query
interface.
Signed-off-by: Fam Zheng
Signed-off-by: John Snow
---
block.c | 9 +++--
include/block/bloc
Mirror and migration use dirty bitmaps for their purposes, and since
commit [block: per caller dirty bitmap] they use their own bitmaps, not
the global one. But they use old functions bdrv_set_dirty and
bdrv_reset_dirty, which change all dirty bitmaps.
Named dirty bitmaps series by Fam and Snow ar
On 2014-11-26 at 18:41, John Snow wrote:
From: Fam Zheng
This makes a deep copy of an HBitmap.
Signed-off-by: Fam Zheng
Signed-off-by: John Snow
---
include/qemu/hbitmap.h | 8
util/hbitmap.c | 16
2 files changed, 24 insertions(+)
Reviewed-by: Max Re
On 2014-11-26 at 18:41, John Snow wrote:
From: Fam Zheng
Signed-off-by: Fam Zheng
Signed-off-by: John Snow
---
block.c | 35 +++
include/block/block.h | 4
2 files changed, 35 insertions(+), 4 deletions(-)
Reviewed-by: Max Reitz
On 11/27/2014 05:08 PM, Fam Zheng wrote:
> On Thu, 11/27 13:59, Jason Wang wrote:
>> > virtio_net_handle_ctrl() and other functions that process control vq
>> > request call iov_discard_front() which will shorten the iov. This will
>> > lead unmapping in virtqueue_push() leaks mapping.
>> >
>> >
On 2014-11-27 at 10:30, Max Reitz wrote:
On 2014-11-13 at 11:36, Mao Chuan Li wrote:
Currently the qemu-iotest framework is focused on the x86 platform.
The suite should be usable by other platforms as well.
To enable multi-platform usage, this patch set enhances the framework
scripts to addres
On 26.11.2014 15:46, Kevin Wolf wrote:
This improves the performance of requests because an ACB doesn't need to
be allocated on the heap any more. It also makes the code nicer and
smaller.
As a side effect, the codepath taken by aio=threads is changed to use
paio_submit_co(). This doesn't change
On Wed, Nov 26, 2014 at 11:47 AM, Pavel Dovgaluk
wrote:
> That covermail was wrong. Here is the correct one:
>
> This set of patches is related to the reverse execution and deterministic
> replay of qemu execution This implementation of deterministic replay can
> be used for deterministic debuggi
Markus Armbruster writes:
> "Michael S. Tsirkin" writes:
>
>> On Mon, Nov 03, 2014 at 05:13:15PM +0100, Markus Armbruster wrote:
>>> "Michael S. Tsirkin" writes:
>>>
>>> > On Mon, Nov 03, 2014 at 04:09:36PM +0100, Markus Armbruster wrote:
>>> >> "Michael S. Tsirkin" writes:
>>> >>
>>> >> > t
On 2014-11-26 at 18:41, John Snow wrote:
From: Fam Zheng
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
Signed-off-by: John Snow
---
block.c
virtio_net_handle_ctrl() and other functions that process control vq
request call iov_discard_front() which will shorten the iov. This will
lead unmapping in virtqueue_push() leaks mapping.
Fixes this by keeping the original iov untouched and using a temp variable
in those functions.
Cc: Wen Cong
Hi All,
Recently, I'm writing an interface of wrapper class for QCOW2 in order to
manage QCOW2 img files conveniently based on our requirements in my current
project , this wrapper includes functions such as QCOW2 creating, read/write
and snapshot relatives. Actually, these functions would f
On 26/11/14 14:01, Paolo Bonzini wrote:
> Use the external qemu-timer API instead.
>
> Signed-off-by: Paolo Bonzini
Acked-by: James Hogan
Thanks
James
> ---
> target-mips/kvm.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/target-mips/kvm.c b/target-mips/kvm.c
> i
vring_map causes a huge overhead by calling memory_region_find everytime.
the vring_map is executed already on vring_setup and there is also the memory
region referenced.
Signed-off-by: Peter Lieven
---
hw/virtio/dataplane/vring.c | 51 +--
1 file change
On 2014-11-26 at 18:41, John Snow wrote:
From: Fam Zheng
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 t
On 2014/11/27 17:19, Eric Auger wrote:
> On 11/27/2014 10:00 AM, Shannon Zhao wrote:
>> Hi Eric,
>>
>> On 2014/10/31 21:53, Eric Auger wrote:
>>> load_dtb is renamed into arm_load_dtb and becomes non static.
>>> it will be used by machvirt for dynamic instantiation of
>>> platform devices
>>>
>>> S
On 2014-11-26 at 18:41, John Snow wrote:
From: Fam Zheng
This adds three qmp commands to transactions.
Users 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: Fa
On 2014-11-26 at 18:41, John Snow wrote:
From: Fam Zheng
Signed-off-by: Fam Zheng
Signed-off-by: John Snow
---
block.c | 1 +
qapi/block-core.json | 5 -
2 files changed, 5 insertions(+), 1 deletion(-)
Reviewed-by: Max Reitz
On 2014-11-26 at 18:41, John Snow wrote:
From: Fam Zheng
Signed-off-by: Fam Zheng
Signed-off-by: John Snow
---
tests/qemu-iotests/056| 33 ++---
tests/qemu-iotests/056.out| 4 ++--
tests/qemu-iotests/iotests.py | 8
3 files changed, 40 i
Signed-off-by: Peter Lieven
---
block/block-backend.c |6 --
1 file changed, 6 deletions(-)
diff --git a/block/block-backend.c b/block/block-backend.c
index d0692b1..cf36ffb 100644
--- a/block/block-backend.c
+++ b/block/block-backend.c
@@ -15,9 +15,6 @@
#include "sysemu/blockdev.h"
#i
On Thu, 27 Nov 2014, Jason Wang wrote:
> virtio_net_handle_ctrl() and other functions that process control vq
> request call iov_discard_front() which will shorten the iov. This will
> lead unmapping in virtqueue_push() leaks mapping.
>
> Fixes this by keeping the original iov untouched and using
This patch creates a ring structure for the coroutine pool instead
of a linked list. The implementation of the list has the issue
that it always throws aways the latest coroutines instead of the
oldest ones. This is a drawback since the latest used coroutines
are more likely cached than old ones.
On 2014/11/13 20:17, Gonglei (Arei) wrote:
> From: Gonglei
>
> Gonglei (2):
> virtfs-proxy-helper: Fix possible socket leak.
> virtfs-proxy-helper: Fix handle leak to make Coverity happy
>
> fsdev/virtfs-proxy-helper.c | 16
> 1 file changed, 12 insertions(+), 4 deletions(
On Thu, 11/27 18:04, Jason Wang wrote:
> virtio_net_handle_ctrl() and other functions that process control vq
> request call iov_discard_front() which will shorten the iov. This will
> lead unmapping in virtqueue_push() leaks mapping.
>
> Fixes this by keeping the original iov untouched and using
This series removes the need for locking the coroutine pool. During
collection of perf data it showed up that there is serious time spend
in locking the coroutine pool during heavy I/O from the vServer.
This also needs serious benchmarking. Using Kevins qemu-img bench there is a
3-4%
speedup. Ins
On 11/27/2014 11:17 AM, Shannon Zhao wrote:
> On 2014/11/27 17:19, Eric Auger wrote:
>> On 11/27/2014 10:00 AM, Shannon Zhao wrote:
>>> Hi Eric,
>>>
>>> On 2014/10/31 21:53, Eric Auger wrote:
load_dtb is renamed into arm_load_dtb and becomes non static.
it will be used by machvirt for dyn
This reverts commit ac2662a913ee5854b1269256adbdc14e57ba480a.
---
include/block/coroutine.h | 11 ---
qemu-coroutine.c | 26 +++---
2 files changed, 3 insertions(+), 34 deletions(-)
diff --git a/include/block/coroutine.h b/include/block/coroutine.h
index 7
On Wed, 26 Nov 2014 20:50:05 +0200
"Michael S. Tsirkin" wrote:
> On Wed, Nov 26, 2014 at 06:28:42PM +0100, Cornelia Huck wrote:
> > virtio-net (non-vhost) and virtio-blk have everything in place to support
> > virtio 1.0: let's enable the feature bit for them.
>
> Hmm I doubt that.
> At least no
On Wed, 26 Nov 2014, Don Slutz wrote:
> On 11/26/14 13:17, Stefano Stabellini wrote:
> > On Tue, 25 Nov 2014, Andrew Cooper wrote:
> > > On 25/11/14 17:45, Stefano Stabellini wrote:
> > > > Increase maxmem before calling xc_domain_populate_physmap_exact to avoid
> > > > the risk of running out of g
Hi,
> > +evt = s->ptr.queue[(s->head+s->n) & QUEUE_MASK];
>
> s->n is QUEUE_LENGTH, can we directly delete it?
> evt = s->ptr.queue[s->head & QUEUE_MASK]
I prefer to make clear in the code that we want the last ring element
not the first and leave in the "+n", even if we could take i
On 26/11/14 14:16, Michael S. Tsirkin wrote:
On Wed, Nov 26, 2014 at 01:11:25PM +0300, Denis V. Lunev wrote:
From: Raushaniya Maksudova
Excessive virtio_balloon inflation can cause invocation of OOM-killer,
when Linux is under severe memory pressure. Various mechanisms are
responsible for corr
On Thu, 27 Nov 2014 10:30:33 +0100
Max Reitz wrote:
> On 2014-11-13 at 11:36, Mao Chuan Li wrote:
> > Currently the qemu-iotest framework is focused on the x86 platform. The
> > suite should be
> > usable by other platforms as well. To enable multi-platform usage, this
> > patch set enhances
>
(actually adding Greg)
On (Thu) 27 Nov 2014 [14:38:42], Amit Shah wrote:
> On (Thu) 27 Nov 2014 [16:48:10], David Gibson wrote:
> > VirtIO devices now remember which endianness they're operating in in order
> > to support targets which may have guests of either endianness, such as
> > powerpc. Th
On Wed, 2014-11-26 at 13:05 -0500, Luiz Capitulino wrote:
> On Wed, 26 Nov 2014 13:50:01 +0200
> Marcel Apfelbaum wrote:
>
> > The commits:
> > - 6a1fa9f5 (monitor: add del completion for peripheral device)
> > - 66e56b13 (qdev: add qdev_build_hotpluggable_device_list helper)
> >
> > cause a Q
On Thu, Nov 27, 2014 at 02:04:55PM +0300, Denis V. Lunev wrote:
> On 26/11/14 14:16, Michael S. Tsirkin wrote:
> >On Wed, Nov 26, 2014 at 01:11:25PM +0300, Denis V. Lunev wrote:
> >>From: Raushaniya Maksudova
> >>
> >>Excessive virtio_balloon inflation can cause invocation of OOM-killer,
> >>when
On 2014/11/27 19:04, Gerd Hoffmann wrote:
> Hi,
>
>>> +evt = s->ptr.queue[(s->head+s->n) & QUEUE_MASK];
>>
>> s->n is QUEUE_LENGTH, can we directly delete it?
>> evt = s->ptr.queue[s->head & QUEUE_MASK]
>
> I prefer to make clear in the code that we want the last ring element
> not t
On Do, 2014-11-27 at 19:16 +0800, Gonglei wrote:
> On 2014/11/27 19:04, Gerd Hoffmann wrote:
>
> > Hi,
> >
> >>> +evt = s->ptr.queue[(s->head+s->n) & QUEUE_MASK];
> >>
> >> s->n is QUEUE_LENGTH, can we directly delete it?
> >> evt = s->ptr.queue[s->head & QUEUE_MASK]
> >
> > I prefer
On 2014/11/27 19:19, Gerd Hoffmann wrote:
> On Do, 2014-11-27 at 19:16 +0800, Gonglei wrote:
>> On 2014/11/27 19:04, Gerd Hoffmann wrote:
>>
>>> Hi,
>>>
> +evt = s->ptr.queue[(s->head+s->n) & QUEUE_MASK];
s->n is QUEUE_LENGTH, can we directly delete it?
evt = s->ptr.
On Thu, 2014-11-27 at 13:11 +0200, Marcel Apfelbaum wrote:
> On Wed, 2014-11-26 at 13:05 -0500, Luiz Capitulino wrote:
> > On Wed, 26 Nov 2014 13:50:01 +0200
> > Marcel Apfelbaum wrote:
> >
> > > The commits:
> > > - 6a1fa9f5 (monitor: add del completion for peripheral device)
> > > - 66e56b13
Am 27.11.2014 um 12:08 hat Michael Mueller geschrieben:
> On Thu, 27 Nov 2014 10:30:33 +0100
> Max Reitz wrote:
>
> > On 2014-11-13 at 11:36, Mao Chuan Li wrote:
> > > Currently the qemu-iotest framework is focused on the x86 platform. The
> > > suite should be
> > > usable by other platforms as
On 26 November 2014 at 03:09, Fam Zheng wrote:
> On Tue, 11/25 18:21, Peter Maydell wrote:
>> In qemu_poll_ns(), when we convert an int64_t nanosecond timeout into
>> a struct timespec, we may accidentally run into overflow problems if
>> the timeout is very long. This happens because the tv_sec f
On Thu, 2014-11-27 at 13:11 +0200, Marcel Apfelbaum wrote:
> On Wed, 2014-11-26 at 13:05 -0500, Luiz Capitulino wrote:
> > On Wed, 26 Nov 2014 13:50:01 +0200
> > Marcel Apfelbaum wrote:
> >
> > > The commits:
> > > - 6a1fa9f5 (monitor: add del completion for peripheral device)
> > > - 66e56b13
On Thu, 2014-11-27 at 19:35 +0800, Zhu Guihua wrote:
> On Thu, 2014-11-27 at 13:11 +0200, Marcel Apfelbaum wrote:
> > On Wed, 2014-11-26 at 13:05 -0500, Luiz Capitulino wrote:
> > > On Wed, 26 Nov 2014 13:50:01 +0200
> > > Marcel Apfelbaum wrote:
> > >
> > > > The commits:
> > > > - 6a1fa9f5 (mo
The idea is that all other virtio devices are calling this helper
to merge properties of the proxy device. This is the only difference
in between this helper and code in inside virtio_instance_init_common.
The patch should not cause any harm as property list in generic balloon
code is empty.
This
Excessive virtio_balloon inflation can cause invocation of OOM-killer,
when Linux is under severe memory pressure. Various mechanisms are
responsible for correct virtio_balloon memory management. Nevertheless it
is often the case that these control tools does not have enough time to
react on fast c
On Thu, Nov 27, 2014 at 2:45 PM, Denis V. Lunev wrote:
> Excessive virtio_balloon inflation can cause invocation of OOM-killer,
> when Linux is under severe memory pressure. Various mechanisms are
> responsible for correct virtio_balloon memory management. Nevertheless it
> is often the case that
Excessive virtio_balloon inflation can cause invocation of OOM-killer,
when Linux is under severe memory pressure. Various mechanisms are
responsible for correct virtio_balloon memory management. Nevertheless it
is often the case that these control tools does not have enough time to
react on fast c
On Thu, 2014-11-27 at 13:41 +0200, Marcel Apfelbaum wrote:
> On Thu, 2014-11-27 at 19:35 +0800, Zhu Guihua wrote:
> > On Thu, 2014-11-27 at 13:11 +0200, Marcel Apfelbaum wrote:
> > > On Wed, 2014-11-26 at 13:05 -0500, Luiz Capitulino wrote:
> > > > On Wed, 26 Nov 2014 13:50:01 +0200
> > > > Marcel
On 2014/10/31 21:53, Eric Auger wrote:
> This new C module will be used by ARM machine files to generate
> platform bus node and their dynamic sysbus device tree nodes.
>
> Dynamic sysbus device node addition is done in a machine init
> done notifier. arm_register_platform_bus_fdt_creator does the
As per arm specs, if the src and dest register are same, write back
operation is suppressed.
[Specs]
if memop == MemOp_LOAD && wback && n == t && n != 31 then
c = ConstrainUnpredictable();
assert c IN {Constraint_WBSUPPRESS, Constraint_UNKNOWN, Constraint_UNDEF,
Constraint_NOP};
case c of
when Cons
On Thu, 2014-11-27 at 20:08 +0800, Zhu Guihua wrote:
> On Thu, 2014-11-27 at 13:41 +0200, Marcel Apfelbaum wrote:
> > On Thu, 2014-11-27 at 19:35 +0800, Zhu Guihua wrote:
> > > On Thu, 2014-11-27 at 13:11 +0200, Marcel Apfelbaum wrote:
> > > > On Wed, 2014-11-26 at 13:05 -0500, Luiz Capitulino wrot
On 27/11/14 14:50, Andrey Korolyov wrote:
On Thu, Nov 27, 2014 at 2:45 PM, Denis V. Lunev wrote:
Excessive virtio_balloon inflation can cause invocation of OOM-killer,
when Linux is under severe memory pressure. Various mechanisms are
responsible for correct virtio_balloon memory management. Ne
On 11/27/2014 01:07 PM, Shannon Zhao wrote:
> On 2014/10/31 21:53, Eric Auger wrote:
>> This new C module will be used by ARM machine files to generate
>> platform bus node and their dynamic sysbus device tree nodes.
>>
>> Dynamic sysbus device node addition is done in a machine init
>> done notifi
On Thu, Nov 27, 2014 at 03:50:11PM +0400, Andrey Korolyov wrote:
> On Thu, Nov 27, 2014 at 2:45 PM, Denis V. Lunev wrote:
> > Excessive virtio_balloon inflation can cause invocation of OOM-killer,
> > when Linux is under severe memory pressure. Various mechanisms are
> > responsible for correct vi
On 2014/11/27 19:22, Gonglei wrote:
> On 2014/11/27 19:19, Gerd Hoffmann wrote:
>
>> On Do, 2014-11-27 at 19:16 +0800, Gonglei wrote:
>>> On 2014/11/27 19:04, Gerd Hoffmann wrote:
>>>
Hi,
>> +evt = s->ptr.queue[(s->head+s->n) & QUEUE_MASK];
>
> s->n is QUEUE_LENGTH
The dirty_memory[] bitmap is used for live migration, TCG self-modifying code
detection, and VGA emulation. Up until now the bitmap was always accessed
under the QEMU global mutex. This series makes all dirty_memory[] accesses
atomic to prepare the way for threads writing to guest memory without
The new bitmap_test_and_clear_atomic() function clears a range and
returns whether or not the bits were set.
Signed-off-by: Stefan Hajnoczi
---
include/qemu/bitmap.h | 2 ++
util/bitmap.c | 26 ++
2 files changed, 28 insertions(+)
diff --git a/include/qemu/bitma
The dirty memory bitmap is managed by ram_addr.h and copied to
migration_bitmap[] periodically during live migration.
Move the code to sync the bitmap to ram_addr.h where related code lives.
Signed-off-by: Stefan Hajnoczi
---
arch_init.c | 46 ++--
Use atomic_or() for atomic bitmaps where several threads may set bits at
the same time.
This avoids the race condition between threads loading an element,
bitwise ORing, and then storing the element.
Most bitmap users don't need atomicity so introduce new functions.
Signed-off-by: Stefan Hajnocz
The cpu_physical_memory_reset_dirty() function is sometimes used
together with cpu_physical_memory_get_dirty(). This is not atomic since
two separate accesses to the dirty memory bitmap are made.
Turn cpu_physical_memory_reset_dirty() into the atomic
cpu_physical_memory_test_and_clear_dirty().
T
Use set_bit_atomic() and bitmap_set_atomic() so that multiple threads
can dirty memory without race conditions.
Signed-off-by: Stefan Hajnoczi
---
I had to get creative to stay under 80 characters per line. I'm open to
suggestions if you prefer me to format it another way.
---
include/exec/ram_
The fast path of cpu_physical_memory_sync_dirty_bitmap() directly
manipulates the dirty bitmap. Use atomic_xchg() to make the
test-and-clear atomic.
Signed-off-by: Stefan Hajnoczi
---
include/exec/ram_addr.h | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/include/exe
On Thu, Nov 27, 2014 at 02:45:42PM +0300, Denis V. Lunev wrote:
> Excessive virtio_balloon inflation can cause invocation of OOM-killer,
> when Linux is under severe memory pressure. Various mechanisms are
> responsible for correct virtio_balloon memory management. Nevertheless it
> is often the ca
On Thu, Nov 27, 2014 at 06:04:03PM +0800, Jason Wang wrote:
> virtio_net_handle_ctrl() and other functions that process control vq
> request call iov_discard_front() which will shorten the iov. This will
> lead unmapping in virtqueue_push() leaks mapping.
>
> Fixes this by keeping the original iov
On 27/11/14 15:31, Michael S. Tsirkin wrote:
On Thu, Nov 27, 2014 at 02:45:42PM +0300, Denis V. Lunev wrote:
Excessive virtio_balloon inflation can cause invocation of OOM-killer,
when Linux is under severe memory pressure. Various mechanisms are
responsible for correct virtio_balloon memory man
On Thu, 27 Nov 2014 13:41:07 +0200
Marcel Apfelbaum wrote:
> On Thu, 2014-11-27 at 19:35 +0800, Zhu Guihua wrote:
> > On Thu, 2014-11-27 at 13:11 +0200, Marcel Apfelbaum wrote:
> > > On Wed, 2014-11-26 at 13:05 -0500, Luiz Capitulino wrote:
> > > > On Wed, 26 Nov 2014 13:50:01 +0200
> > > > Marce
On 27 November 2014 at 12:15, Gaurav Sharma wrote:
> As per arm specs, if the src and dest register are same, write back
> operation is suppressed.
> [Specs]
> if memop == MemOp_LOAD && wback && n == t && n != 31 then
> c = ConstrainUnpredictable();
> assert c IN {Constraint_WBSUPPRESS, Constraint
On 27 November 2014 at 12:33, Michael S. Tsirkin wrote:
> On Thu, Nov 27, 2014 at 06:04:03PM +0800, Jason Wang wrote:
>> virtio_net_handle_ctrl() and other functions that process control vq
>> request call iov_discard_front() which will shorten the iov. This will
>> lead unmapping in virtqueue_pus
Excessive virtio_balloon inflation can cause invocation of OOM-killer,
when Linux is under severe memory pressure. Various mechanisms are
responsible for correct virtio_balloon memory management. Nevertheless it
is often the case that these control tools does not have enough time to
react on fast c
Excessive virtio_balloon inflation can cause invocation of OOM-killer,
when Linux is under severe memory pressure. Various mechanisms are
responsible for correct virtio_balloon memory management. Nevertheless it
is often the case that these control tools does not have enough time to
react on fast c
The idea is that all other virtio devices are calling this helper
to merge properties of the proxy device. This is the only difference
in between this helper and code in inside virtio_instance_init_common.
The patch should not cause any harm as property list in generic balloon
code is empty.
This
Konrad, I think we should have this fix in 4.5: without it
vif=[ 'model=virtio-net' ] crashes QEMU.
On Thu, 27 Nov 2014, Peter Maydell wrote:
> On 27 November 2014 at 12:33, Michael S. Tsirkin wrote:
> > On Thu, Nov 27, 2014 at 06:04:03PM +0800, Jason Wang wrote:
> >> virtio_net_handle_ctrl() an
writes:
> From: Gonglei
>
> Coverity report:
> (94) Event open_fn: Returning handle opened by function "proxy_socket(char
> const *, uid_t, gid_t)". [details]
> (95) Event var_assign: Assigning: "sock" = handle returned from
> "proxy_socket(sock_name, own_u, own_g)".
> (103) Event leaked_han
I was taking into consideration the behavior of afm, which it seems
suppresses write back.
However, i do get your point on this.
Regards,
Gaurav
On Thu, Nov 27, 2014 at 6:10 PM, Peter Maydell
wrote:
> On 27 November 2014 at 12:15, Gaurav Sharma
> wrote:
> > As per arm specs, if the src and des
writes:
> From: Gonglei
>
> Signed-off-by: Gonglei
Reviewed-by: Markus Armbruster
Hi,
Running a redhat-6.4-64bit (kernel 2.6.32-358.el6.x86_64) or elder guest on
qemu-2.1, with kvm enabled and -cpu host, non default cpu-topology and guest
numa
I'm seeing a reliable kernel panic from the guest shortly after boot. It is
happening in
find_busiest_group().
We also found it happ
On 2014/11/27 20:25, Eric Auger wrote:
> On 11/27/2014 01:07 PM, Shannon Zhao wrote:
>> On 2014/10/31 21:53, Eric Auger wrote:
>>> This new C module will be used by ARM machine files to generate
>>> platform bus node and their dynamic sysbus device tree nodes.
>>>
>>> Dynamic sysbus device node add
Cc'ing Paolo and Benoît.
Best regards,
-Gonglei
> -Original Message-
> From: Gonglei (Arei)
> Sent: Thursday, November 27, 2014 8:58 PM
> To: qemu-devel@nongnu.org
> Subject: [BUG] Redhat-6.4_64bit-guest kernel panic with cpu-passthrough and
> guest numa
>
> Hi,
>
> Running a redhat-6
On 27 November 2014 at 12:49, Gaurav Sharma wrote:
> I was taking into consideration the behavior of afm, which it seems
> suppresses write back.
The Fast Models are just one implementation -- like every other
implementation, they have to choose a kind of unpredictable
behaviour in cases like the
On 27.11.14 13:56, Shannon Zhao wrote:
> On 2014/11/27 20:25, Eric Auger wrote:
>> On 11/27/2014 01:07 PM, Shannon Zhao wrote:
>>> On 2014/10/31 21:53, Eric Auger wrote:
This new C module will be used by ARM machine files to generate
platform bus node and their dynamic sysbus device tre
1 - 100 of 284 matches
Mail list logo
