** Changed in: qemu
Status: New => Invalid
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1668556
Title:
QEMU Guest Agent service fails to start on Windows 10 RS2 preview
Status in QEMU:
On Fri, Mar 03, 2017 at 08:38:45AM +0100, Markus Armbruster wrote:
> Niels de Vos writes:
>
> > On Thu, Mar 02, 2017 at 10:44:03PM +0100, Markus Armbruster wrote:
> >> To reproduce, run
> >>
> >> $ valgrind qemu-system-x86_64 --nodefaults -S --drive
> >> driver=gluster,volume=testvol,path=/
Niels de Vos writes:
> On Fri, Mar 03, 2017 at 08:38:45AM +0100, Markus Armbruster wrote:
>> Niels de Vos writes:
>>
>> > On Thu, Mar 02, 2017 at 10:44:03PM +0100, Markus Armbruster wrote:
>> >> To reproduce, run
>> >>
>> >> $ valgrind qemu-system-x86_64 --nodefaults -S --drive
>> >> driv
Current code depends on virtio_queue_update_rings() to setup address
space cache. But this does not work for 1.0 device since we can't infer
avail and used. Fixing this by calling virtio_init_region_cache()
after subsections load which can guarantee avail and used are correct.
Fixes: 97cd965c0701
On 14.02.2017 17:02, Dr. David Alan Gilbert wrote:
Hi Pavel,
I was curious, having merged this, how you're using postcopy; do you switch
into postcopy mode immediately or wait until the first sync or what?
Do you find yourself in postcopy mode long enough that it's worth
doing the release? If
Commit c611c76417f5 ("virtio: add MemoryListener to cache ring
translations") registers a memory listener to dma_as. This may not
work when IOMMU is enabled: dma_as(bus_master_as) were correctly
initialized in pcibus_machine_done() after virtio_realize() where we
try to register listener and initia
On 2017年03月02日 18:30, Paolo Bonzini wrote:
On 02/03/2017 08:47, Jason Wang wrote:
Fixes: c611c76417f5 ("virtio: add MemoryListener to cache ring
translations")
Cc: Paolo Bonzini
Signed-off-by: Jason Wang
This is very ugly. I guess it's better than broken IOMMU ...
Paolo?
Maybe we can dela
On Fri, 3 Mar 2017 16:51:28 +0800
Jason Wang wrote:
> Current code depends on virtio_queue_update_rings() to setup address
> space cache. But this does not work for 1.0 device since we can't infer
> avail and used. Fixing this by calling virtio_init_region_cache()
> after subsections load which
On Fri, 03/03 01:02, no-re...@patchew.org wrote:
> CC ppc-softmmu/hw/misc/ivshmem.o
> In file included from
> /var/tmp/patchew-tester-tmp-krtyhjt1/src/hw/misc/ivshmem.c:19:0:
> /var/tmp/patchew-tester-tmp-krtyhjt1/src/include/qemu/osdep.h:30:25: fatal
> error: config-host.h: No such file o
Hello Juan & Dave,
We hit a bug in our test:
Network error occurs when migrating a guest, libvirt then rollback the
migration, causes qemu coredump
qemu log:
2017-03-01T12:54:33.904949+08:00|info|qemu[17672]|[33614]|monitor_qapi_event_emit[479]|:
{"timestamp": {"seconds": 1488344073, "microsecond
On 2017年03月03日 17:15, Cornelia Huck wrote:
On Fri, 3 Mar 2017 16:51:28 +0800
Jason Wang wrote:
Current code depends on virtio_queue_update_rings() to setup address
space cache. But this does not work for 1.0 device since we can't infer
avail and used. Fixing this by calling virtio_init_regi
On 2017年03月03日 17:18, Fam Zheng wrote:
On Fri, 03/03 01:02,no-re...@patchew.org wrote:
CC ppc-softmmu/hw/misc/ivshmem.o
In file included from
/var/tmp/patchew-tester-tmp-krtyhjt1/src/hw/misc/ivshmem.c:19:0:
/var/tmp/patchew-tester-tmp-krtyhjt1/src/include/qemu/osdep.h:30:25: fatal
e
> -Original Message-
> From: Stefano Stabellini [mailto:sstabell...@kernel.org]
> Sent: 02 March 2017 22:50
> To: Paul Durrant
> Cc: xen-de...@lists.xenproject.org; qemu-devel@nongnu.org; Stefano
> Stabellini ; Anthony Perard
>
> Subject: Re: [PATCH v2 5/5] xen: use libxendevicemodel when
On Fri, 03/03 09:12, Christian Borntraeger wrote:
> On 03/02/2017 05:33 PM, no-re...@patchew.org wrote:
>
> Fam, this seems to happen sometimes. Can you have a look at that why the
> qtest
> thing fails on the s390 box?
I don't know, when I run the commands manually for a few times, it is not
r
On 2 March 2017 at 03:46, Jeff Cody wrote:
> The following changes since commit c8c0a1a784cdf70ecea50e93213137c6c89337a7:
>
> Merge remote-tracking branch 'remotes/cody/tags/block-pull-request' into
> staging (2017-02-28 13:41:03 +)
>
> are available in the git repository at:
>
> g...@git
Richard Henderson writes:
> On 03/03/2017 06:53 AM, Alex Bennée wrote:
>> The translation code uses cpu_ld*_code which can trigger a tlb_fill
>> which if it fails will attempt a fault resolution. This never works
>> during translation as the TB being generated hasn't been added yet.
>> However w
On 2 March 2017 at 19:53, Alex Bennée wrote:
> While on MTTCG hosts it is very important that updates to
> cpu->interrupt_request are protected by the BQL not all guests have
> been converted to the correct locking yet. As a result we are seeing
> breaking on non-MTTCG enabled guests in production
Hi All,
Any feedback for the 4 last patches?
Thanks,
Fred
On 02/17/2017 09:17 PM, fred.kon...@greensocs.com wrote:
> From: KONRAD Frederic
>
> This series allows to execute code from mmio areas.
> The main goal of this is to be able to run code for example from an SPI
> device.
>
> The three
git repository at:
>
> git://github.com/dgibson/qemu.git tags/ppc-for-2.9-20170303
>
> for you to fetch changes up to 992d7e976c7eca2b3129cd4bae4a0d350a8065fa:
>
> target/ppc: rewrite f[n]m[add,sub] using float64_muladd (2017-03-03
> 11:38:33 +1100)
>
> ---
> -Original Message-
> From: Stefano Stabellini [mailto:sstabell...@kernel.org]
> Sent: 02 March 2017 22:50
> To: Paul Durrant
> Cc: xen-de...@lists.xenproject.org; qemu-devel@nongnu.org; Stefano
> Stabellini ; Anthony Perard
>
> Subject: Re: [PATCH v2 5/5] xen: use libxendevicemodel when
* Germano Veit Michel (germ...@redhat.com) wrote:
> qemu_announce_self() is triggered by qemu at the end of migrations
> to update the network regarding the path to the guest l2addr.
>
> however it is also useful when there is a network change such as
> an active bond slave swap. Essentially, it's
On Fri, 03/03 09:29, Gonglei (Arei) wrote:
> Hello Juan & Dave,
>
> We hit a bug in our test:
> Network error occurs when migrating a guest, libvirt then rollback the
> migration, causes qemu coredump
> qemu log:
> 2017-03-01T12:54:33.904949+08:00|info|qemu[17672]|[33614]|monitor_qapi_event_emit[4
Hi Cédric, hi David,
running "info qtree" at the HMP monitor with the pseries machine
currently results in:
[...]
dev: spapr-vty, id "vty@7100"
reg = 1895825408 (0x7100)
chardev = "serial0"
irq = 4098 (0x1002)
dev: spapr-rtc, id ""
dev: icp, id ""
hw/c
Peter Maydell writes:
> On 2 March 2017 at 19:53, Alex Bennée wrote:
>> While on MTTCG hosts it is very important that updates to
>> cpu->interrupt_request are protected by the BQL not all guests have
>> been converted to the correct locking yet. As a result we are seeing
>> breaking on non-MTT
On 03/03/2017 11:48 AM, Thomas Huth wrote:
> Hi Cédric, hi David,
>
> running "info qtree" at the HMP monitor with the pseries machine
> currently results in:
>
> [...]
> dev: spapr-vty, id "vty@7100"
> reg = 1895825408 (0x7100)
> chardev = "serial0"
> irq
On 02/03/2017 19:53, Alex Bennée wrote:
> We should hold the BQL before we transition to HW emulation. This is
> because all HW emulation needs to be serialised under MTTCG
> conditions. This is picked up by asserts that fire when
> cpu_mips_get_count triggers and IRQ.
>
> Reported-by: Yongbok K
On 3 March 2017 at 11:05, Alex Bennée wrote:
> According to the commit that added it
> (c552d6c038f7cf4058d1fd5987118ffd41e0e050) it is meant to be a hint to
> the compiler. Reading the GCC notes however seems to contradict that.
>
> FWIW I did test it in both builds and we do used tese for a bunc
Hold BQL when accessing timer which can cause interrupts
Signed-off-by: Yongbok Kim
---
target/mips/op_helper.c | 21 ++---
1 file changed, 18 insertions(+), 3 deletions(-)
diff --git a/target/mips/op_helper.c b/target/mips/op_helper.c
index b683fcb..e5f3ea4 100644
--- a/target/
From: Prasad J Pandit
Limit the number of arguments passed to execve(2) call from
a user program, as large number of them could lead to a bad
guest address error.
Reported-by: Jann Horn
Signed-off-by: Prasad J Pandit
---
linux-user/syscall.c | 7 +++
1 file changed, 7 insertions(+)
diff
When using a memory-backend object with prealloc turned on, QEMU
will memset() the first byte in every memory page to zero. While
this might have been acceptable for memory backends associated
with RAM, this corrupts application data for NVDIMMs.
Instead of setting every page to zero, read the cur
Hi,
This series seems to have some coding style problems. See output below for
more information:
Message-id: 20170303113255.28262-1-berra...@redhat.com
Type: series
Subject: [Qemu-devel] [PATCH v3] os: don't corrupt pre-existing memory-backend
data with prealloc
=== TEST SCRIPT BEGIN ===
#!/bin
On 02/03/2017 20:53, Alex Bennée wrote:
> IRQ modification is part of device emulation and should be done while
> the BQL is held to prevent races when MTTCG is enabled. This adds
> assertions in the hw emulation layer and wraps the calls from helpers
> in the BQL.
>
> Reported-by: Mark Cave-Ayl
My recent patches to Xen [1] introduced a new library to support
running device models for HVM guests.
This series ports QEMU onto the new library if it is available in the
build environment.
[1] Patches starting with
http://xenbits.xen.org/gitweb/?p=xen.git;a=commit;h=b108240265deea37601f1a60591
This patch modifies the wrapper functions in xen_common.h to use the
new xendevicemodel interface if it is available along with compatibility
code to use the old libxenctrl interface if it is not.
Signed-off-by: Paul Durrant
Reviewed-by: Anthony Perard
---
Cc: Stefano Stabellini
NOTE: Xen patc
This patch is a purely cosmetic change that avoids a name collision in
a subsequent patch.
Signed-off-by: Paul Durrant
Reviewed-by: Anthony Perard
---
Cc: Paolo Bonzini
Cc: Stefano Stabellini
---
include/exec/ram_addr.h | 4 ++--
include/hw/xen/xen.h| 2 +-
xen-hvm-stub.c | 2 +-
This patch adds code in configure to set CONFIG_XEN_CTRL_INTERFACE_VERSION
to a new value of 490 if libxendevicemodel is present in the build
environment.
Signed-off-by: Paul Durrant
Reviewed-by: Anthony Perard
---
Cc: Stefano Stabellini
---
configure | 19 +++
1 file changed,
This patch creates inline wrapper functions in xen_common.h for all open
coded calls to xc_hvm_XXX() functions outside of xen_common.h so that use
of xen_xc can be made implicit. This again is in preparation for the move
to using libxendevicemodel.
Signed-off-by: Paul Durrant
Reviewed-by: Anthony
Doing this will make the transition to using the new libxendevicemodel
interface less intrusive on the callers of these functions, since using
the new library will require a change of handle.
NOTE: The patch also moves the 'externs' for xen_xc and xen_fmem from
xen_backend.h to xen_common.h,
On 03/03/2017 11:08, Peter Maydell wrote:
> On 2 March 2017 at 19:53, Alex Bennée wrote:
>> While on MTTCG hosts it is very important that updates to
>> cpu->interrupt_request are protected by the BQL not all guests have
>> been converted to the correct locking yet. As a result we are seeing
>>
Commit 4881658a4b introduced a call to arm_get_cpu_by_id(),
and Coverity noticed that we weren't checking that it didn't
return NULL (CID 1371652).
Normally this won't happen (because all 4 CPUs are expected
to exist), but it's possible the user requested fewer CPUs
on the command line. Handle thi
* Gonglei (Arei) (arei.gong...@huawei.com) wrote:
> Hello Juan & Dave,
cc'ing in pbonzini since it's magic involving cpu_synrhonize_all_states()
> We hit a bug in our test:
> Network error occurs when migrating a guest, libvirt then rollback the
> migration, causes qemu coredump
> qemu log:
> 201
The interrupt controller does not have its own locking.
Signed-off-by: Paolo Bonzini
---
target/nios2/op_helper.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/target/nios2/op_helper.c b/target/nios2/op_helper.c
index 538853c..efb1c48 100644
--- a/target/nios2/op_helper.c
+++ b/target/n
"Dr. David Alan Gilbert" writes:
> * Germano Veit Michel (germ...@redhat.com) wrote:
>> qemu_announce_self() is triggered by qemu at the end of migrations
>> to update the network regarding the path to the guest l2addr.
>>
>> however it is also useful when there is a network change such as
>> an
On 17 February 2017 at 03:04, Jason Wang wrote:
>
>
> On 2017年02月16日 20:29, Dmitry Fleytman wrote:
>>
>> This series fix a few issues related
>> to processing of RX packets with VLAN headers.
>>
>> See commit messages of specific patches
>> for information regarding affected devices.
>>
>> Dmitry
On Thu, 2 Mar 2017 19:59:42 +0100
Halil Pasic wrote:
> The function virtio_notify_irqfd used to ignore the return code of
> event_notifier_set. Let's fail the device should this occur.
I'm wondering if there are reasons for event_notifier_set() to fail
beyond "we've hit an internal race and sho
Can break with multiple monitors race against each other:
NAK
The power state spec section 5.1.5 AFFINITY_INFO defines the
affinity info return values as
0 ON
1 OFF
2 ON_PENDING
I grepped QEMU for power_state to ensure that no assumptions
of OFF=0 were being made.
Signed-off-by: Andrew Jones
---
I found this with the yet to be committed kvm-unit-t
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blake
---
qapi/qmp-dispatch.c | 14 --
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/qapi/qmp-dispatch.c b/qapi/qmp-dispatch.c
index 23b0528..578c6d8 100644
--- a/qapi/qmp-dispatch.c
+++ b/qapi/qmp-dispatch.c
@@ -30,7
v4:
* PATCH 05+06: New
* PATCH 07: Messing with the list of commands breaks when multiple
monitors race against each other, use two separate lists instead
(R-by dropped)
v3:
* PATCH 12+18+19: Commit message tidied up [Eric]
* PATCH 17: Spurious whitespace change dropped [Eric]
* PATCH 22: Missi
Signed-off-by: Markus Armbruster
---
scripts/qapi-introspect.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/scripts/qapi-introspect.py b/scripts/qapi-introspect.py
index 541644e..fb72c61 100644
--- a/scripts/qapi-introspect.py
+++ b/scripts/qapi-introspect.py
@@ -64,7 +64,
qobject_input_start_struct() sets *list, except when it fails because
qobject_input_get_object() fails, i.e. the input object doesn't exist.
All the other input visitor start_struct(), start_list(),
start_alternate() always set *obj / *list.
Change qobject_input_start_struct() to match.
Signed-o
The value of key 'arguments' must be a JSON object. qemu-ga neglects
to check, and crashes. To reproduce, send
{ 'execute': 'guest-sync', 'arguments': [] }
to qemu-ga.
do_qmp_dispatch() uses qdict_get_qdict() to get the arguments. When
not a JSON object, this gets a null pointer, which fl
Signed-off-by: Markus Armbruster
---
tests/test-qobject-input-visitor.c | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/tests/test-qobject-input-visitor.c
b/tests/test-qobject-input-visitor.c
index 10c15c4..9f3a826 100644
--- a/tests/test-qobject-input-visitor.c
+++ b/
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blake
---
MAINTAINERS| 1 +
tests/Makefile.include | 5 +-
tests/libqtest.c | 17 --
tests/libqtest.h | 8 +++
tests/qmp-test.c | 139 +
5 files changed, 16
The QObject input visitor has three error message formats:
* Parameter '%s' is missing
* "Invalid parameter type for '%s', expected: %s"
* "QMP input object member '%s' is unexpected"
The '%s' are member names (or "null", but I'll fix that later).
The last error message calls the thing "QMP inpu
The qobject input visitor comes in a strict and a non-strict variant.
This test is the non-strict variant's last user. Turns out it relies
on non-strict only in test_visitor_in_null(), and just out of
laziness. We don't actually test the non-strict behavior.
Clean up test_visitor_in_null(), and
The way we get QMP commands registered is high tech:
* qapi-commands.py generates qmp_init_marshal() that does the actual work
* it also generates the magic to register it as a MODULE_INIT_QAPI
function, so it runs when someone calls
module_call_init(MODULE_INIT_QAPI)
* main() calls module_c
Drop unused QIV_STACK_SIZE and unused qobject_input_start_struct()
parameter errp.
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blake
---
qapi/qobject-input-visitor.c | 14 +++---
1 file changed, 3 insertions(+), 11 deletions(-)
diff --git a/qapi/qobject-input-visitor.c b/qapi/qo
The QERR_ macros are leftovers from the days of "rich" error objects.
QERR_QMP_BAD_INPUT_OBJECT, QERR_QMP_BAD_INPUT_OBJECT_MEMBER,
QERR_QMP_EXTRA_MEMBER are used in just one place now, except for one
use that has crept into qobject-input-visitor.c.
Drop these macros, to make the (bad) error messa
qmp_check_input_obj() duplicates qmp_dispatch_check_obj(), except the
latter screws up an error message. handle_qmp_command() runs first
the former, then the latter via qmp_dispatch(), masking the screwup.
qemu-ga also masks the screwup, because it also duplicates checks,
just differently.
qmp_c
Commit 240f64b made all qobject input visitors created outside tests
strict, except for the one in object_property_set_qobject(). That one
was left behind only because Eric couldn't spare the time to figure
out whether making it strict would break anything, with a TODO
comment. Time to resolve it
Much of test-qobject-input-strict.c duplicates
test-qobject-input-strict.c, but with less assertions on expected
output:
* test_validate_struct() duplicates test_visitor_in_struct()
* test_validate_struct_nested() duplicates
test_visitor_in_struct_nested()
* test_validate_list() duplicates the
visit_optional() is to be called only between visit_start_struct() and
visit_end_struct(). Visitors that don't support struct visits,
i.e. don't implement start_struct(), end_struct(), have no use for it.
Clarify documentation.
The string input visitor doesn't support struct visits. Its
parse_op
When you try to visit beyond the end of a list, the qobject input
visitor crashes, and the string visitor screws returns garbage. The
generated list visits never go beyond the list end, but manual visits
could.
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blake
---
tests/test-opts-visito
The string input visitor tries to cope with null input. Null input
isn't used anywhere, and isn't covered by tests. Unsurprisingly, it
doesn't fully work: start_list() crashes because it passes the input
via parse_str() to strtoll() unchecked.
Make string_input_visitor_new() assert its argument
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blake
---
include/qapi/qobject-input-visitor.h | 37 ++-
include/qapi/qobject-output-visitor.h | 35 +
2 files changed, 67 insertions(+), 5 deletions(-)
diff --git a/include/qapi/
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blake
---
qapi/qobject-input-visitor.c | 11 ---
tests/test-qobject-input-visitor.c | 2 --
2 files changed, 8 insertions(+), 5 deletions(-)
diff --git a/qapi/qobject-input-visitor.c b/qapi/qobject-input-visitor.c
index 34065ba..
On 2 March 2017 at 06:20, Michael S. Tsirkin wrote:
> The following changes since commit 1e0addb682c3c552fd97480037d4f8ff18e2b87e:
>
> Merge remote-tracking branch 'remotes/sstabellini/tags/xen-20170228-tag'
> into staging (2017-03-01 20:33:47 +)
>
> are available in the git repository at:
Error messages refer to nodes of the QObject being visited by name.
Trouble is the names are sometimes less than helpful:
* The name of the root QObject is whatever @name argument got passed
to the visitor, except NULL gets mapped to "null". We commonly pass
NULL. Not good.
Avoiding error
Call visitor_input_teardown() from visitor_input_test_init(), so you
don't have to call it from the actual tests.
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blake
---
tests/test-string-input-visitor.c | 19 ++-
1 file changed, 2 insertions(+), 17 deletions(-)
diff --git
Demonstrates a design flaw: there is no way to for input visitors to
report that a list visit didn't visit the complete input list. The
generated list visits always do, but manual visits needn't.
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blake
---
tests/test-opts-visitor.c |
The command registry encapsulates a single command list. Give the
functions using it a parameter instead. Define suitable command lists
in monitor, guest agent and test-qmp-commands.
Signed-off-by: Markus Armbruster
---
include/qapi/qmp/dispatch.h | 22 ++
monitor.c
Fix the design flaw demonstrated in the previous commit: new method
check_list() lets input visitors report that unvisited input remains
for a list, exactly like check_struct() lets them report that
unvisited input remains for a struct or union.
Implement the method for the qobject input visitor (
On 03/03/2017 10:53 AM, Fam Zheng wrote:
> On Fri, 03/03 09:12, Christian Borntraeger wrote:
>> On 03/02/2017 05:33 PM, no-re...@patchew.org wrote:
>>
>> Fam, this seems to happen sometimes. Can you have a look at that why the
>> qtest
>> thing fails on the s390 box?
>
> I don't know, when I run
On 03/03/2017 13:00, Dr. David Alan Gilbert wrote:
> Ouch that's pretty nasty; I remember Paolo explaining to me a while ago that
> their were times when run_on_cpu would have to drop the BQL and I worried
> about it,
> but this is the 1st time I've seen an error due to it.
>
> Do you know what
The split between tests/test-qobject-input-visitor.c and
tests/test-qobject-input-strict.c now makes less sense than ever. The
next commit will take care of that.
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blake
---
block/nbd.c | 2 +-
block/nfs.c
On 03/03/2017 12:16 PM, Cédric Le Goater wrote:
> On 03/03/2017 11:48 AM, Thomas Huth wrote:
>> Hi Cédric, hi David,
>>
>> running "info qtree" at the HMP monitor with the pseries machine
>> currently results in:
>>
>> [...]
>> dev: spapr-vty, id "vty@7100"
>> reg = 1895825408
The next commit is going to add a test that calls qmp("null").
Curiously, this hangs. Here's why.
qmp_fd_sendv() doesn't send newlines. Not even when @fmt contains
some. At first glance, the QMP parser seems to be fine with that.
However, it turns out that it fails to react to input until it se
On Fri, 3 Mar 2017 13:43:32 +0100
Halil Pasic wrote:
> On 03/03/2017 01:21 PM, Cornelia Huck wrote:
> > On Thu, 2 Mar 2017 19:59:42 +0100
> > Halil Pasic wrote:
> >
> >> The function virtio_notify_irqfd used to ignore the return code of
> >> event_notifier_set. Let's fail the device should thi
To enforce capability negotiation before normal operation,
handle_qmp_command() inspects every command before it's handed off to
qmp_dispatch(). This is a bit of a layering violation, and results in
duplicated code.
Before capability negotiation (!cur_mon->in_command_mode), we fail
commands other
The recent changes on the XICS layer removed the XICSState object to
let the sPAPR machine handle the ICP and ICS directly. The reset of
these objects was previously handled by XICSState, which was a SysBus
device, and to keep the same behavior, the ICP and ICS were assigned
to SysbBus.
But that b
Lists with elements above INT64_MAX don't work (known bug). Empty
lists don't work (weird).
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blake
---
tests/test-string-input-visitor.c | 85 +--
1 file changed, 72 insertions(+), 13 deletions(-)
diff --git
On 03/03/2017 01:21 PM, Cornelia Huck wrote:
> On Thu, 2 Mar 2017 19:59:42 +0100
> Halil Pasic wrote:
>
>> The function virtio_notify_irqfd used to ignore the return code of
>> event_notifier_set. Let's fail the device should this occur.
>
> I'm wondering if there are reasons for event_notifi
On 21/02/2017 09:51, KONRAD Frederic wrote:
> Ping!
>
> Would be nice for us if we can get this into 2.9.
Sorry, I've been much busier with icount than I would have liked. :(
Paolo
> Thanks,
> Fred
>
> Le 17/02/2017 à 21:17, fred.kon...@greensocs.com a écrit :
>> From: KONRAD Frederic
>>
>>
Yongbok Kim writes:
> On 02/03/2017 19:53, Alex Bennée wrote:
>> We should hold the BQL before we transition to HW emulation. This is
>> because all HW emulation needs to be serialised under MTTCG
>> conditions. This is picked up by asserts that fire when
>> cpu_mips_get_count triggers and IRQ.
On 2 March 2017 at 06:20, Michael S. Tsirkin wrote:
> dtc | 2 +-
I just noticed this erroneous submodule update in here, unfortunately
after I pushed it to master :-(
I'll fix up the tree...
thanks
-- PMM
On 03/03/2017 10:06, Jason Wang wrote:
> Commit c611c76417f5 ("virtio: add MemoryListener to cache ring
> translations") registers a memory listener to dma_as. This may not
> work when IOMMU is enabled: dma_as(bus_master_as) were correctly
> initialized in pcibus_machine_done() after virtio_reali
On 03/03/2017 12:54, Alex Bennée wrote:
>
> Yongbok Kim writes:
>
>> On 02/03/2017 19:53, Alex Bennée wrote:
>>> We should hold the BQL before we transition to HW emulation. This is
>>> because all HW emulation needs to be serialised under MTTCG
>>> conditions. This is picked up by asserts tha
Hi,
This series seems to have some coding style problems. See output below for
more information:
Message-id: 1488544368-30622-1-git-send-email-arm...@redhat.com
Type: series
Subject: [Qemu-devel] [PATCH v4 00/28] qapi: QMP dispatch and input visitor work
=== TEST SCRIPT BEGIN ===
#!/bin/bash
BA
On 03.03.2017 13:51, Cédric Le Goater wrote:
> The recent changes on the XICS layer removed the XICSState object to
> let the sPAPR machine handle the ICP and ICS directly. The reset of
> these objects was previously handled by XICSState, which was a SysBus
> device, and to keep the same behavior,
Am 02.03.2017 um 22:43 hat Markus Armbruster geschrieben:
> As a bdrv_snapshot_delete() method, sd_snapshot_delete() must set an
> error and return negative errno on failure. It sometimes returns -1,
> and sometimes neglects to set an error. It also prints error messages
> with error_report(). F
On 03/03/2017 01:50 PM, Cornelia Huck wrote:
> On Fri, 3 Mar 2017 13:43:32 +0100
> Halil Pasic wrote:
>
>> On 03/03/2017 01:21 PM, Cornelia Huck wrote:
>>> On Thu, 2 Mar 2017 19:59:42 +0100
>>> Halil Pasic wrote:
>>>
The function virtio_notify_irqfd used to ignore the return code of
icount has become much slower after tcg_cpu_exec has stopped
using the BQL. There is also a latent bug that is masked by
the slowness.
The slowness happens because every occurrence of a QEMU_CLOCK_VIRTUAL
timer now has to wake up the I/O thread and wait for it. The rendez-vous
is mediated by the
This optimization is not necessary anymore, because the vCPU now drops
the I/O thread lock even with TCG. Drop it to simplify the code and
avoid the "I/O thread spun for 1000 iterations" warning.
Signed-off-by: Paolo Bonzini
---
vl.c | 5 +
1 file changed, 1 insertion(+), 4 deletions(-)
di
If the first timer is exactly at the current value of the clock, the
deadline is met and the timer should fire. This fixes itself without icount,
but with icount execution of instructions will stop exactly at the deadline.
Signed-off-by: Paolo Bonzini
---
util/qemu-timer.c | 2 +-
1 file change
This dependency is the wrong way, and we will need util/qemu-timer.h from
sysemu/cpus.h
in the next patch.
Signed-off-by: Paolo Bonzini
---
cpu-exec.c| 1 +
hw/core/ptimer.c | 1 +
include/qemu/timer.h | 1 -
include/sysemu/cpus.h | 2 ++
kvm-all.c | 1 +
monitor.c
icount has become much slower after tcg_cpu_exec has stopped
using the BQL. There is also a latent bug that is masked by
the slowness.
The slowness happens because every occurrence of a QEMU_CLOCK_VIRTUAL
timer now has to wake up the I/O thread and wait for it. The rendez-vous
is mediated by the
There is no change for now, because the callback just invokes
qemu_notify_event.
Signed-off-by: Paolo Bonzini
---
cpus.c | 5 +
include/qemu/timer.h | 4 ++--
include/sysemu/cpus.h| 1 +
stubs/cpu-get-icount.c | 6 ++
tests/test-aio-multith
Rewrite the -numa documentation to clarify what exactly it does.
Signed-off-by: Eduardo Habkost
Message-Id: <20170123180632.28942-3-ehabk...@redhat.com>
Reviewed-by: Igor Mammedov
Signed-off-by: Eduardo Habkost
---
qemu-options.hx | 38 --
1 file changed, 28
* Paolo Bonzini (pbonz...@redhat.com) wrote:
>
>
> On 03/03/2017 13:00, Dr. David Alan Gilbert wrote:
> > Ouch that's pretty nasty; I remember Paolo explaining to me a while ago that
> > their were times when run_on_cpu would have to drop the BQL and I worried
> > about it,
> > but this is the 1
1 - 100 of 320 matches
Mail list logo
