From: David Hildenbrand
STFL bit 4 and 5 are just indications to the guest, which TLB entries an
IDTE call will clear. These are performance indicators for the guest.
STFL bit 4:
INVALIDATE DAT TABLE ENTRY (IDTE) performs
the invalidation-and-clearing operation by
selectively clearin
Signed-off-by: Richard Henderson
---
target/s390x/helper.h | 6 +
target/s390x/mem_helper.c | 310 +
target/s390x/translate.c | 43 +++
target/s390x/insn-data.def | 13 ++
4 files changed, 372 insertions(+)
diff --git a/target/s390x/hel
Reviewed-by: Aurelien Jarno
Signed-off-by: Richard Henderson
---
target/s390x/cpu_models.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/target/s390x/cpu_models.c b/target/s390x/cpu_models.c
index 2c86b24..a4afdd9 100644
--- a/target/s390x/cpu_models.c
+++ b/target/s390x/cpu_models.c
@@
Signed-off-by: Richard Henderson
---
target/s390x/helper.h | 1 +
target/s390x/cpu_models.c | 2 +
target/s390x/mem_helper.c | 189 +
target/s390x/translate.c | 13 +++-
target/s390x/insn-data.def | 2 +
5 files changed, 206 insertions(+
On 07/10/2017 03:39 AM, David Gibson wrote:
On Fri, Jul 07, 2017 at 06:20:37PM -0300, Daniel Henrique Barboza wrote:
"spapr: Remove 'awaiting_allocation' DRC flag" removed the flag that
was originally was being used to prevent a race condition between
hot unplug and hotplug. The DRC code base
It appears that the recent qemu packages released for 16.04 set the
machine type to "pc-i440fx-zesty" instead of "pc-i440fx-xenial". If you
override this and make it use "pc-i440fx-xenial" instances boot again.
FYI, if you are using OpenStack, this can be done by editing nova.conf,
[libvirt] sect
(Sorry, sent to a wrong mailing list first instead of QEMU-devel)
Under certain circumstances normal xen-mapcache functioning may be broken
by guest's actions. This may lead to either QEMU performing exit() due to
a caught bad pointer (and with QEMU process gone the guest domain simply
appears hun
On 07/10/2017 09:21 AM, Alex Bennée wrote:
Previously DISAS_JUMP did ensure this but with the optimisation of
8a6b28c7 (optimize indirect branches) we might not leave the loop.
This means if any pending interrupts are cleared by changing IRQ flags
we might never get around to servicing them. You
On 07/10/2017 09:21 AM, Alex Bennée wrote:
We already have an exit condition, DISAS_UPDATE which will exit the
run-loop. Expand on the difference with DISAS_EXIT in the comments.
Signed-off-by: Alex Bennée
---
target/arm/translate.h | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
On Fri, 7 Jul 2017 06:40:58 +
"Tian, Kevin" wrote:
> > From: Alex Williamson [mailto:alex.william...@redhat.com]
> > Sent: Saturday, July 1, 2017 1:00 AM
> >
> > On Fri, 30 Jun 2017 05:14:40 +
> > "Tian, Kevin" wrote:
> >
> > > > From: Alex Williamson [mailto:alex.william...@redhat.c
On 07/10/2017 09:21 AM, Alex Bennée wrote:
-s->is_jmp = DISAS_UPDATE;
+gen_a64_set_pc_im(s->pc);
+s->is_jmp = DISAS_JUMP;
Better would be gen_goto_tb. The destination is known, so there's no need to
go through lookup_and_goto_ptr. You still get the icount_decr check a
On 07/10/2017 09:21 AM, Alex Bennée wrote:
-case DISAS_UPDATE:
-gen_set_pc_im(dc, dc->pc);
-/* fall through */
case DISAS_JUMP:
gen_goto_ptr();
break;
+case DISAS_UPDATE:
+gen_set_pc_im(dc, dc->pc);
Is there a specific reason this patch wasn't finished? If I'm not wrong
using non-scalar properties with -object is still not possible, yet
would be a very useful feature for drivers with UserCreatable objects.
Archive link since this is an old patch:
https://lists.gnu.org/archive/html/qemu-de
While an ISB will ensure any raised IRQs happen on the next
instruction it doesn't cause any to get raised by itself. We can
therefor use DISAS_JUMP for ISB instructions and rely on the
exit_request check at the top of each TB to deal with exiting if
needed.
Signed-off-by: Alex Bennée
---
target
We already have an exit condition, DISAS_UPDATE which will exit the
run-loop. Expand on the difference with DISAS_EXIT in the comments.
Signed-off-by: Alex Bennée
---
target/arm/translate.h | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/target/arm/translate.h b/target/ar
Previously DISAS_JUMP did ensure this but with the optimisation of
8a6b28c7 (optimize indirect branches) we might not leave the loop.
This means if any pending interrupts are cleared by changing IRQ flags
we might never get around to servicing them. You usually notice this
by seeing the lookup_tb_p
Hi,
This is a quick iteration from the previous series:
https://lists.nongnu.org/archive/html/qemu-devel/2017-07/msg02371.html
Which was born out of the fix:
https://lists.nongnu.org/archive/html/qemu-devel/2017-07/msg01831.html
Following feedback I reverted the attempt to use DISAS_UPDATE
DISAS_UPDATE should be used when the wider CPU state other than just
the PC has been updated and we should therefor exit the TCG runtime
and return to the main execution loop rather assuming DISAS_JUMP would
do that.
Signed-off-by: Alex Bennée
---
target/arm/translate-a64.c | 14 +++---
As a precursor to later patches attempt to come up with a more
concrete wording for what each of the common exit cases would be.
CC: Emilio G. Cota
CC: Richard Henderson
CC: Lluís Vilanova
Signed-off-by: Alex Bennée
---
include/exec/exec-all.h | 29 ++---
1 file change
Intel 82599 VFs report a PCIe capability version of 0, which is
invalid. The earliest version of the PCIe spec used version 1. This
causes Windows to fail startup on the device and it will be disabled
with error code 10. Our choices are either to drop the PCIe cap on
such devices, which has the
VFIOGroup.device_list is effectively our reference tracking mechanism
such that we can teardown a group when all of the device references
are removed. However, we also use this list from our machine reset
handler for processing resets that affect multiple devices. Generally
device removals are fu
The following changes since commit 6b06e3e49eb8c91cc286c16d6bf3181ac296f33d:
Merge remote-tracking branch 'remotes/ericb/tags/pull-nbd-2017-07-10-v2'
into staging (2017-07-10 16:12:47 +0100)
are available in the git repository at:
git://github.com/awilliam/qemu-vfio.git tags/vfio-updates-20
On 07/10/2017 08:35 AM, Alex Bennée wrote:
Well this was more in the guise of having well defined semantics across
all the translators. I agree just keeping DISAS_EXIT is cleaner w.r.t
the ARM code.
I hope we can do that within the context of (or after) LLuis' changes.
Doing it beforehand is ju
Richard Henderson writes:
> On 07/10/2017 06:13 AM, Peter Maydell wrote:
>> On 10 July 2017 at 16:47, Alex Bennée wrote:
>>> DISAS_UPDATE should be used when the wider CPU state other than just
>>> the PC has been updated and we should therefor exit the TCG runtime
>>> and return to the main ex
Richard Henderson writes:
> On 07/10/2017 04:28 AM, Alex Bennée wrote:
>> While the SoftMMU is not emulating the target MMU of a system there is
>> a relationship between its page size and that of the target. If the
>> target MMU is full featured the functions called to re-fill the
>> entries in
On Fri, Jul 07, 2017 at 06:30:50PM -0300, Eduardo Habkost wrote:
> object_resolve_path*() ambiguous path detection breaks when
> ambiguous==NULL and the object tree have 3 objects of the same type and
> only 2 of them are under the same parent. e.g.:
>
> /container/obj1 (TYPE_FOO)
> /container/
(CCing libvir-list)
On Mon, Jul 10, 2017 at 07:45:54PM +0300, Michael S. Tsirkin wrote:
> On Mon, Jul 10, 2017 at 10:59:43AM -0300, Eduardo Habkost wrote:
> > On Mon, Jul 10, 2017 at 10:42:26AM +0300, Marcel Apfelbaum wrote:
> > > On 07/07/2017 21:03, Eduardo Habkost wrote:
> > > > On Fri, Jul 07,
On Mon, Jul 10, 2017 at 05:23:36PM +0200, Igor Mammedov wrote:
> On Mon, 10 Jul 2017 11:53:31 -0300
> Eduardo Habkost wrote:
>
> > On Mon, Jul 10, 2017 at 10:01:47AM +0200, Igor Mammedov wrote:
> > > On Fri, 7 Jul 2017 17:20:25 +0100
> > > Mark Cave-Ayland wrote:
> > >
> > > > On 07/07/17 16:
On 10.07.2017 17:13, Laurent Vivier wrote:
On 07/07/2017 16:21, Pavel Butsykin wrote:
We should guarantee that RAM will not be modified while VM has a stopped
state, otherwise it can lead to negative consequences during post-copy
migration. In RUN_STATE_FINISH_MIGRATE step, it's expected that RA
From: Juan Quintela
Once there, be consistent and use
compress_thread_{save,load}_{setup,cleanup}.
Signed-off-by: Juan Quintela
Reviewed-by: Dr. David Alan Gilbert
Message-Id: <20170628095228.4661-6-quint...@redhat.com>
Signed-off-by: Dr. David Alan Gilbert
---
migration/migration.c | 5 ---
From: Juan Quintela
We are going to use it now for more than save live regions.
Once there rename qemu_savevm_state_begin() to qemu_savevm_state_setup().
Signed-off-by: Juan Quintela
Reviewed-by: Dr. David Alan Gilbert
Message-Id: <20170628095228.4661-2-quint...@redhat.com>
Signed-off-by: Dr.
From: Juan Quintela
Once there, I rename ram_migration_cleanup() to ram_save_cleanup().
Notice that this is the first pass, and I only passed XBZRLE to the
new scheme. Moved decoded_buf to inside XBZRLE struct.
As a bonus, I don't have to export xbzrle functions from ram.c.
Signed-off-by: Juan
From: Juan Quintela
We need to do things at load time and at cleanup time.
Signed-off-by: Juan Quintela
--
Move the printing of the error message so we can print the device
giving the error.
Add call to postcopy stuff
Message-Id: <20170628095228.4661-4-quint...@redhat.com>
Reviewed-by: Dr. Da
From: Peter Xu
MigrateState object is not ready at that time, so we'll get an
assertion. Use qemu_global_option() instead.
Reported-by: Eduardo Habkost
Suggested-by: Eduardo Habkost
Reviewed-by: Eduardo Habkost
Reviewed-by: Juan Quintela
Fixes: 3df663e ("migration: move only_migratable to Mi
From: Peter Xu
[Peter collected Eduardo's patch comment and formatted into patch]
Suggested-by: Eduardo Habkost
Signed-off-by: Peter Xu
Message-Id: <1499242883-2184-5-git-send-email-pet...@redhat.com>
Reviewed-by: Eduardo Habkost
Signed-off-by: Dr. David Alan Gilbert
---
migration/migration
From: Peter Xu
It's never documented, and now we have one more parameter for it (which
obsoletes this one). Document it properly.
Suggested-by: Eduardo Habkost
Signed-off-by: Peter Xu
Message-Id: <1499396048-21657-1-git-send-email-pet...@redhat.com>
Reviewed-by: Greg Kurz
Reviewed-by: Eduardo
From: Juan Quintela
We need a cleanup for loads, so we rename here to be consistent.
Signed-off-by: Juan Quintela
Reviewed-by: Dr. David Alan Gilbert
--
Rename htab_cleanup to htap_save_cleanup as dave suggestion
Message-Id: <20170628095228.4661-3-quint...@redhat.com>
Signed-off-by: Dr. Davi
On 23 May 2017 at 12:12, Diana Craciun wrote:
> The NXP DPAA2 is a hardware architecture designed for high-speeed network
> packet processing. The DPAA2 hardware components are managed by a hardware
> component called the Management Complex (or MC) which provides an
> object-base abstraction for s
From: "Dr. David Alan Gilbert"
The following changes since commit 6b06e3e49eb8c91cc286c16d6bf3181ac296f33d:
Merge remote-tracking branch 'remotes/ericb/tags/pull-nbd-2017-07-10-v2' into
staging (2017-07-10 16:12:47 +0100)
are available in the git repository at:
git://github.com/dagrh/qemu
From: Peter Xu
Currently drive_init_func() may call migrate_get_current() while the
migrate object is still not ready yet at that time. Move the migration
object init earlier, along with the global properties, right after
acceleration init.
This fixes a breakage for iotest 055, which caused an a
Apparently none of the 32bit x86 modes are supported in 2.9 version of
qemu-system-x86_64. I realize the desire to simplify the code, and
separate i386 from x86_64, but x86_64 really does need to support all
the modes in which the processor can operate. True that for major
operating systems the p
It is called from qcow2_invalidate_cache in coroutine context, so always
load metadata from a coroutine.
---
block/qcow2.c | 45 -
1 file changed, 40 insertions(+), 5 deletions(-)
diff --git a/block/qcow2.c b/block/qcow2.c
index cb081ea47f..b5de67d113 1
On 07/09/2017 11:50 PM, Alex Bennée wrote:
Side issue. Have we considered the impact on codegen buffer utilisation
by doing an "off-code_gen_buffer" no cache translation the first time we
ever see a TB?
No we haven't. Possibly because we'd need additional infrastructure to do even
that -- we'
Suggested-by: Kevin Wolf
Signed-off-by: Paolo Bonzini
---
block.c | 43 ---
block/parallels.c | 17 +++--
block/qcow2.c | 23 +++
block/qed-check.c | 1 +
block/qed-table.c
bdrv_invalidate_cache and bdrv_check are reading/processing metadata,
which is usually protected by a lock (CoMutex). However, they are
not called from coroutine context, so they are not using the same
protection. This series moves these functions into coroutine context,
as suggested by Kevin for
Avoid recursively taking s->lock during bdrv_check.
Signed-off-by: Paolo Bonzini
---
block/qcow2-refcount.c | 28
block/qcow2.c | 20
block/qcow2.h | 2 ++
3 files changed, 34 insertions(+), 16 deletions(-)
diff --git a/block/
It is called from bdrv_invalidate_cache in coroutine context, so always
load metadata from a coroutine.
---
block/qed.c | 40 +---
1 file changed, 37 insertions(+), 3 deletions(-)
diff --git a/block/qed.c b/block/qed.c
index d43cb55923..3c2867c946 100644
--- a/
On 4 July 2017 at 15:48, Alex Bennée wrote:
> Hi Peter,
>
> A bit of a mixed set of patches here for you to pick from as you
> will. The first 2 are additional tracing fixes including a fix for
> segfaulting when generating a trace.
>
> The next two are documentation patches as requested.
>
> Then
QED's bdrv_invalidate_cache implementation would like to reuse functions
that acquire/release the metadata locks. Call it from coroutine context
to simplify the logic.
Signed-off-by: Paolo Bonzini
---
block.c | 41 +
block/iscsi.c
On 07/10/2017 04:28 AM, Alex Bennée wrote:
While the SoftMMU is not emulating the target MMU of a system there is
a relationship between its page size and that of the target. If the
target MMU is full featured the functions called to re-fill the
entries in the SoftMMU entries start moving up the
On Mon, Jul 10, 2017 at 10:59:43AM -0300, Eduardo Habkost wrote:
> On Mon, Jul 10, 2017 at 10:42:26AM +0300, Marcel Apfelbaum wrote:
> > On 07/07/2017 21:03, Eduardo Habkost wrote:
> > > On Fri, Jul 07, 2017 at 06:17:57PM +0300, Michael S. Tsirkin wrote:
> > > > On Fri, Jul 07, 2017 at 10:39:49AM -
On Mon, Jul 10, 2017 at 04:13:54PM +0200, Laurent Vivier wrote:
> > static void virtconsole_realize(DeviceState *dev, Error **errp)
> > {
> > VirtIOSerialPort *port = VIRTIO_SERIAL_PORT(dev);
> > @@ -233,6 +253,7 @@ static void virtserialport_class_init(ObjectClass
> > *klass, void *data)
>
On 10 July 2017 at 16:04, Eric Blake wrote:
> The following changes since commit 94c56652b9079cfb9d560a6dde7ecb15eb9ef9c7:
>
> Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging
> (2017-07-10 14:06:49 +0100)
>
> are available in the git repository at:
>
> git://repo.o
On 07/10/2017 06:13 AM, Peter Maydell wrote:
On 10 July 2017 at 16:47, Alex Bennée wrote:
DISAS_UPDATE should be used when the wider CPU state other than just
the PC has been updated and we should therefor exit the TCG runtime
and return to the main execution loop rather assuming DISAS_JUMP wou
Postcopy migration of dirty bitmaps. Only named dirty bitmaps,
associated with root nodes and non-root named nodes are migrated.
If destination qemu is already containing a dirty bitmap with the same name
as a migrated bitmap (for the same node), then, if their granularities are
the same the migra
Hi all!
There is a new version of dirty bitmap postcopy migration series.
v7
clone: tag postcopy-v7 from https://src.openvz.org/scm/~vsementsov/qemu.git
online: https://src.openvz.org/users/vsementsov/repos/qemu/browse?at=postcopy-v7
- rebased on dirty-bitmap byte-based interfaces
(based on
Split common postcopy staff from ram postcopy staff.
Signed-off-by: Vladimir Sementsov-Ogievskiy
---
migration/migration.c | 39 ++-
migration/migration.h | 2 ++
migration/savevm.c| 48 +++-
3 files changed, 67
The test starts two vms (vm_a, vm_b), create dirty bitmap in
the first one, do several writes to corresponding device and
then migrate vm_a to vm_b with dirty bitmaps.
Signed-off-by: Vladimir Sementsov-Ogievskiy
Reviewed-by: Max Reitz
---
tests/qemu-iotests/169 | 86
Make it possible to set bitmap 'frozen' without a successor.
This is needed to protect the bitmap during outgoing bitmap postcopy
migration.
Signed-off-by: Vladimir Sementsov-Ogievskiy
---
block/dirty-bitmap.c | 22 --
include/block/dirty-bitmap.h | 1 +
2 files chan
On 07/10/2017 05:47 AM, Alex Bennée wrote:
We already have an exit condition that declares we should exit to the
run-loop because wider CPU state changes have been made. Use
DISAS_UPDATE and kill the architecture specific DISAS_EXIT.
Signed-off-by: Alex Bennée
---
target/arm/translate-a64.c |
Signed-off-by: Vladimir Sementsov-Ogievskiy
Reviewed-by: John Snow
Reviewed-by: Eric Blake
Reviewed-by: Juan Quintela
---
migration/migration.c | 9 +
migration/migration.h | 1 +
qapi-schema.json | 4 +++-
3 files changed, 13 insertions(+), 1 deletion(-)
diff --git a/migration/m
Enabling bitmap successor is necessary to enable successors of bitmaps
being migrated before target vm start.
Signed-off-by: Vladimir Sementsov-Ogievskiy
---
block/dirty-bitmap.c | 8
include/block/dirty-bitmap.h | 1 +
2 files changed, 9 insertions(+)
diff --git a/block/dirty-
There would be savevm states (dirty-bitmap) which can migrate only in
postcopy stage. The corresponding pending is introduced here.
Signed-off-by: Vladimir Sementsov-Ogievskiy
Reviewed-by: Juan Quintela
---
include/migration/register.h | 17 +++--
migration/block.c| 7 +
To just release successor and unfreeze bitmap without any additional
work.
Signed-off-by: Vladimir Sementsov-Ogievskiy
Signed-off-by: Denis V. Lunev
---
block/dirty-bitmap.c | 13 +
include/block/dirty-bitmap.h | 2 ++
2 files changed, 15 insertions(+)
diff --git a/block/d
Test
- start two vms (vm_a, vm_b)
- in a
- do writes from set A
- do writes from set B
- fix bitmap sha256
- clear bitmap
- do writes from set A
- start migration
- than, in b
- wait vm start (postcopy should start)
- do writes from set B
- check bitmap sha256
Only-postcopy savevm states (dirty-bitmap) don't need live iteration, so
to disable them and stop transporting empty sections there is a new
savevm handler.
Signed-off-by: Vladimir Sementsov-Ogievskiy
Reviewed-by: Juan Quintela
---
include/migration/register.h | 1 +
migration/savevm.c
Now postcopy-able states are recognized by not NULL
save_live_complete_postcopy handler. But when we have several different
postcopy-able states, it is not convenient. Ram postcopy may be
disabled, while some other postcopy enabled, in this case Ram state
should behave as it is not postcopy-able.
Signed-off-by: Vladimir Sementsov-Ogievskiy
Reviewed-by: John Snow
---
tests/qemu-iotests/iotests.py | 6 ++
1 file changed, 6 insertions(+)
diff --git a/tests/qemu-iotests/iotests.py b/tests/qemu-iotests/iotests.py
index abcf3c10e2..f64277d0b4 100644
--- a/tests/qemu-iotests/iotests.py
+++
When testing migration, auto-generated by qemu node-names differs in
source and destination qemu and migration fails. After this patch,
auto-generated by iotest nodenames will be the same.
Signed-off-by: Vladimir Sementsov-Ogievskiy
Reviewed-by: Max Reitz
---
tests/qemu-iotests/iotests.py | 2 +
Fill postcopy-able pending only if ram postcopy is enabled.
It is necessary because of there will be other postcopy-able states and
when ram postcopy is disabled, it should not spoil common postcopy
related pending.
Signed-off-by: Vladimir Sementsov-Ogievskiy
Reviewed-by: Dr. David Alan Gilbert
Add function opposite to qemu_get_counted_string.
qemu_put_counted_string puts one-byte length of the string (string
should not be longer than 255 characters), and then it puts the string,
without last zero byte.
Signed-off-by: Vladimir Sementsov-Ogievskiy
Reviewed-by: John Snow
Reviewed-by: Jua
On Mon, Jul 10, 2017 at 04:49:18PM +0100, Peter Maydell wrote:
> On 5 July 2017 at 08:14, Paolo Bonzini wrote:
> > This will be useful when the functions are called, early in the configure
> > process, to filter out targets that do not support hardware acceleration.
> >
> > Signed-off-by: Paolo Bo
Enable postcopy if dirty bitmap migration is endabled.
Signed-off-by: Vladimir Sementsov-Ogievskiy
Reviewed-by: Juan Quintela
---
migration/migration.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/migration/migration.c b/migration/migration.c
index dac67eab69..48adc812bc
On 07/10/2017 02:49 AM, Markus Armbruster wrote:
> John Snow writes:
>
>> On 07/07/2017 09:53 AM, Markus Armbruster wrote:
>>> Vladimir Sementsov-Ogievskiy writes:
>>>
07.07.2017 12:00, Markus Armbruster wrote:
> "Daniel P. Berrange" writes:
>
>> On Fri, Jul 07, 2017 at 10:05
On 10/07/2017 18:24, Peter Maydell wrote:
> On 10 July 2017 at 17:14, Paolo Bonzini wrote:
>> On 10/07/2017 17:49, Peter Maydell wrote:
>>> On 5 July 2017 at 08:14, Paolo Bonzini wrote:
This will be useful when the functions are called, early in the configure
process, to filter out targ
On 10/07/2017 18:24, Stefan Hajnoczi wrote:
> On Thu, Jul 06, 2017 at 06:38:28PM +0200, Paolo Bonzini wrote:
>> Snapshots are only created/destroyed/loaded under the BQL, while no
>> other I/O is happening. Snapshot information could be accessed while
>> other I/O is happening, but also under the
On Mon, 10 Jul 2017 12:33:31 -0300
Eduardo Habkost wrote:
> On Mon, Jul 10, 2017 at 11:45:57AM -0300, Eduardo Habkost wrote:
> > On Mon, Jul 10, 2017 at 10:10:41AM +0200, Igor Mammedov wrote:
> > > On Fri, 7 Jul 2017 18:30:51 -0300
> > > Eduardo Habkost wrote:
> [...]
> > > > +
> > > > +
On Thu, Jul 06, 2017 at 06:38:17PM +0200, Paolo Bonzini wrote:
> Here, patches 1-6 make the remaining part of the block layer thread safe.
> Patches 7-11 start removing aio_context_acquire/release, so the line
> count goes down instead of up.
>
> This is the penultimate series. The remaining part
On 10 July 2017 at 17:14, Paolo Bonzini wrote:
> On 10/07/2017 17:49, Peter Maydell wrote:
>> On 5 July 2017 at 08:14, Paolo Bonzini wrote:
>>> This will be useful when the functions are called, early in the configure
>>> process, to filter out targets that do not support hardware acceleration.
>
On Thu, Jul 06, 2017 at 06:38:28PM +0200, Paolo Bonzini wrote:
> Snapshots are only created/destroyed/loaded under the BQL, while no
> other I/O is happening. Snapshot information could be accessed while
> other I/O is happening, but also under the BQL so they cannot be
> modified concurrently. T
On 07/10/2017 03:53 AM, Kashyap Chamarthy wrote:
>
> They are useful suggestions. :-) I'm a stickler for English, and I like
> to get it right. Appreciate the patience!
>
Hardly any such thing as "right". It's all very subjective.
On 07/10/2017 11:16 AM, Paolo Bonzini wrote:
> On 10/07/2017 15:21, Stefan Hajnoczi wrote:
>> On Thu, Jul 06, 2017 at 06:38:18PM +0200, Paolo Bonzini wrote:
>>> Code refactoring only.
>>>
>>> Signed-off-by: Paolo Bonzini
>>> ---
>>> block/write-threshold.c | 28
>>> 1
On 10/07/2017 15:21, Stefan Hajnoczi wrote:
> On Thu, Jul 06, 2017 at 06:38:18PM +0200, Paolo Bonzini wrote:
>> Code refactoring only.
>>
>> Signed-off-by: Paolo Bonzini
>> ---
>> block/write-threshold.c | 28
>> 1 file changed, 16 insertions(+), 12 deletions(-)
>>
>>
On 10/07/2017 17:49, Peter Maydell wrote:
> On 5 July 2017 at 08:14, Paolo Bonzini wrote:
>> This will be useful when the functions are called, early in the configure
>> process, to filter out targets that do not support hardware acceleration.
>>
>> Signed-off-by: Paolo Bonzini
>
>> +supported_x
On 10 July 2017 at 16:47, Alex Bennée wrote:
> DISAS_UPDATE should be used when the wider CPU state other than just
> the PC has been updated and we should therefor exit the TCG runtime
> and return to the main execution loop rather assuming DISAS_JUMP would
> do that.
>
> As some DISAS_UPDATE use
On 10/07/2017 17:52, Stefan Hajnoczi wrote:
> On Thu, Jul 06, 2017 at 06:38:20PM +0200, Paolo Bonzini wrote:
>> void notifier_list_notify(NotifierList *list, void *data)
>> {
>> Notifier *notifier, *next;
>>
>> -QLIST_FOREACH_SAFE(notifier, &list->notifiers, node, next) {
>> +QLIST
On 07/09/2017 11:12 PM, Jiang Biao wrote:
Reserve a register for the guest_base using ppc code for reference.
By doing so, we do not have to recompute it for every memory load.
Signed-off-by: Jiang Biao
Signed-off-by: Richard Henderson
---
tcg/mips/tcg-target.inc.c | 17 +
1 f
On Thu, Jul 06, 2017 at 06:38:25PM +0200, Paolo Bonzini wrote:
> The resulting drained section is a replacement for the AioContext.
>
> Signed-off-by: Paolo Bonzini
> ---
> blockdev.c | 8 ++--
> 1 file changed, 2 insertions(+), 6 deletions(-)
Reviewed-by: Stefan Hajnoczi
signature.asc
D
On Sun, Jul 09, 2017 at 11:44:10 -1000, Richard Henderson wrote:
> On 07/09/2017 11:14 AM, Emilio G. Cota wrote:
> >On Sun, Jul 09, 2017 at 10:45:55 -1000, Richard Henderson wrote:
> >>On 07/08/2017 09:50 PM, Emilio G. Cota wrote:
> >>>+/* includes aborted translations because of exceptions */
On Thu, Jul 06, 2017 at 06:38:24PM +0200, Paolo Bonzini wrote:
> The only caller does it already.
>
> Signed-off-by: Paolo Bonzini
> ---
> block.c | 13 +++--
> 1 file changed, 3 insertions(+), 10 deletions(-)
Reviewed-by: Stefan Hajnoczi
signature.asc
Description: PGP signature
On Thu, Jul 06, 2017 at 06:38:23PM +0200, Paolo Bonzini wrote:
> Signed-off-by: Paolo Bonzini
> ---
> include/block/block_int.h | 10 +++---
> 1 file changed, 7 insertions(+), 3 deletions(-)
Reviewed-by: Stefan Hajnoczi
signature.asc
Description: PGP signature
On Thu, Jul 06, 2017 at 06:38:22PM +0200, Paolo Bonzini wrote:
> Protect the list of inflight reqs and the CoQueues for dependent
> requests.
>
> Signed-off-by: Paolo Bonzini
> ---
> block/backup.c | 20 +++-
> block/replication.c | 2 +-
> include/block/b
On Thu, Jul 06, 2017 at 06:38:20PM +0200, Paolo Bonzini wrote:
> void notifier_list_notify(NotifierList *list, void *data)
> {
> Notifier *notifier, *next;
>
> -QLIST_FOREACH_SAFE(notifier, &list->notifiers, node, next) {
> +QLIST_FOREACH_SAFE_RCU(notifier, &list->notifiers, node,
On 5 July 2017 at 08:14, Paolo Bonzini wrote:
> This will be useful when the functions are called, early in the configure
> process, to filter out targets that do not support hardware acceleration.
>
> Signed-off-by: Paolo Bonzini
> +supported_xen_target() {
> +test "$xen" = "yes" || return
On Thu, Jul 06, 2017 at 06:38:21PM +0200, Paolo Bonzini wrote:
> Reads access the list in RCU style, so be careful to avoid use-after-free
> scenarios in the backup block job. Apart from this, all that's needed
> is protecting updates with a mutex.
>
> Signed-off-by: Paolo Bonzini
> ---
> block
DISAS_UPDATE should be used when the wider CPU state other than just
the PC has been updated and we should therefor exit the TCG runtime
and return to the main execution loop rather assuming DISAS_JUMP would
do that.
As some DISAS_UPDATE users may update the PC dynamically via a helper
we also pus
DISAS_UPDATE should be used when the wider CPU state other than just
the PC has been updated and we should therefor exit the TCG runtime
and return to the main execution loop rather assuming DISAS_JUMP would
do that.
As some DISAS_UPDATE users may update the PC dynamically via a helper
we also pus
Previously DISAS_JUMP did ensure this but with the optimisation of
8a6b28c7 (optimize indirect branches) we might not leave the loop.
This means if any pending interrupts are cleared by changing IRQ flags
we might never get around to servicing them. You usually notice this
by seeing the lookup_tb_p
While an ISB will ensure any raised IRQs happen on the next
instruction it doesn't cause any to get raised by itself. We can
therefor use DISAS_JUMP for ISB instructions and rely on the
exit_request check at the top of each TB to deal with exiting if
needed.
Signed-off-by: Alex Bennée
---
target
Hi,
This is a follow-up to the eret patch I posted on Friday:
https://lists.nongnu.org/archive/html/qemu-devel/2017-07/msg01831.html
It fixes the same bug but does it in a slightly more holistic way. We
start by expanding the commentary on the common DISAS_n exit flags in
exec-all.h. I then up
101 - 200 of 411 matches
Mail list logo
