Andreas Färber writes:
> Am 08.08.2013 15:31, schrieb Anthony Liguori:
>> Rusty Russell writes:
>> We have a mechanism to do weak functions via stubs/. I think it would
>> be better to do cpu_get_byteswap() as a stub function and then overload
>> it in the ppc64 code.
>
> If this as your name in
Am 08.08.2013 um 15:37 hat Benoît Canet geschrieben:
> > Kevin's series renamed these to have a dash in the name, and also moved
> > all the throttling parameters into a sub-struct. Does it make more
> > sense to have just '*throttling' with that sub-struct containing 12
> > parameters, 6 for limi
On Fri, Aug 9, 2013 at 12:29 AM, Michael Roth wrote:
> Quoting Liu Ping Fan (2013-08-08 01:26:07)
>> Introduce struct EventsGSource. It will ease the usage of GSource
>> associated with a group of files, which are dynamically allocated
>> and release, ex, slirp.
>>
>> Signed-off-by: Liu Ping Fan
于 2013-8-6 23:07, Stefan Hajnoczi 写道:
On Thu, Jul 25, 2013 at 05:18:07PM +0200, Stefan Hajnoczi wrote:
v6:
* Fix block/stream.c:close_unused_images() dangling pointer in Patch 2
* Rebase onto qemu.git/master
v5:
* Split out bdrv_delete() drain fix [bonzini]
* Fix commit message [bonzin
Andreas Färber writes:
> Am 08.08.2013 17:40, schrieb Anthony Liguori:
>> Andreas Färber writes:
>>> Am 08.08.2013 15:31, schrieb Anthony Liguori:
We have a mechanism to do weak functions via stubs/. I think it would
be better to do cpu_get_byteswap() as a stub function and then overlo
Il 08/08/2013 17:43, Jan Kiszka ha scritto:
> On 2013-08-08 17:33, Peter Maydell wrote:
>> On 3 August 2013 09:31, Jan Kiszka wrote:
>>> --- a/ioport.c
>>> +++ b/ioport.c
>>> @@ -44,6 +44,22 @@ typedef struct MemoryRegionPortioList {
>>> MemoryRegionPortio ports[];
>>> } MemoryRegionPortioLi
On 9 August 2013 08:35, Rusty Russell wrote:
> That's a lot of replumbing and indirect function calls for a fairly
> obscure case. We certainly don't have a nice CPUState lying around in
> virtio at the moment, for example.
Actually if you're in an IO routine you do: it will be in the
global var
On Fri, 2013-08-09 at 17:05 +0930, Rusty Russell wrote:
> > Exactly, just read it as "is in ... Endian mode". On the CPUs I am more
> > familiar with (e.g., 970), this used to be controlled via an MSR bit,
970 didn't have an LE mode :-)
> > which as CPUPPCState::msr exists per CPUState. I did no
On 9 August 2013 03:58, Rusty Russell wrote:
> Anthony Liguori writes:
>> The distinction is important in QEMU. ppc64 is still
>> TARGET_WORDS_BIGENDIAN. We still want most stl_phys to treat integers
>> as big endian. There's just this extra concept that CPU loads/stores
>> are sometimes byte
Il 08/08/2013 08:26, Liu Ping Fan ha scritto:
> Each slirp has its own time to caculate timeout.
>
> Signed-off-by: Liu Ping Fan
> ---
> slirp/slirp.c | 22 ++
> slirp/slirp.h | 3 +++
> 2 files changed, 13 insertions(+), 12 deletions(-)
>
> diff --git a/slirp/slirp.c b/sli
On 2013-08-08 23:41, Alex Bligh wrote:
> This patch series adds support for timers attached to an AioContext clock
> which get called within aio_poll.
>
> In doing so it removes alarm timers and moves to use ppoll where possible.
>
> This patch set 'sort of' passes make check (see below for cavea
v7:
* Rebase onto Kevin's block-next branch to resolve conflicts [Wenchao]
v6:
* Fix block/stream.c:close_unused_images() dangling pointer in Patch 2
* Rebase onto qemu.git/master
v5:
* Split out bdrv_delete() drain fix [bonzini]
* Fix commit message [bonzini]
v4:
* Ensure pending BHs are
If a block driver has no file descriptors to monitor but there are still
active requests, it can return 1 from .io_flush(). This is used to spin
during synchronous I/O.
Stop relying on .io_flush() and instead check
QLIST_EMPTY(&bs->tracked_requests) to decide whether there are active
requests.
T
In bdrv_delete() make sure to call bdrv_make_anon() *after* bdrv_close()
so that the device is still seen by bdrv_drain_all() when iterating
bdrv_states.
Cc: qemu-sta...@nongnu.org
Signed-off-by: Stefan Hajnoczi
---
block.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/
Check exit conditions before entering blocking aio_poll(). This is
mainly for consistency since it's unlikely that we are stopping in the
first event loop iteration.
Signed-off-by: Stefan Hajnoczi
---
hw/block/dataplane/virtio-blk.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
dif
aio_poll(ctx, true) will soon block if any fd handlers have been set.
Previously it would only block when .io_flush() returned true.
This means that callers must check their wait condition *before*
aio_poll() to avoid deadlock.
Reviewed-by: Paolo Bonzini
Signed-off-by: Stefan Hajnoczi
---
test
aio_poll(ctx, true) will soon block when fd handlers have been set.
Previously aio_poll() would return early if all .io_flush() returned
false. This means we need to check the equivalent of the .io_flush()
condition *before* calling aio_poll(ctx, true) to avoid deadlock.
Reviewed-by: Paolo Bonzin
Since .io_flush() is no longer called we do not need
qemu_gluster_aio_flush_cb() anymore. It turns out that qemu_aio_count
is unused now and can be dropped.
Signed-off-by: Stefan Hajnoczi
---
block/gluster.c | 16 +---
1 file changed, 1 insertion(+), 15 deletions(-)
diff --git a/bl
Now that aio_poll() users check their termination condition themselves,
it is no longer necessary to call .io_flush() handlers.
The behavior of aio_poll() changes as follows:
1. .io_flush() is no longer invoked and file descriptors are *always*
monitored. Previously returning 0 from .io_flush()
.io_flush() is no longer called so drop have_co_req() and
aio_flush_request().
Signed-off-by: Stefan Hajnoczi
---
block/sheepdog.c | 25 +
1 file changed, 5 insertions(+), 20 deletions(-)
diff --git a/block/sheepdog.c b/block/sheepdog.c
index 7699aad..9252b1d 100644
---
.io_flush() is no longer called so drop flush_true() and flush_io().
Signed-off-by: Stefan Hajnoczi
---
hw/block/dataplane/virtio-blk.c | 17 ++---
1 file changed, 2 insertions(+), 15 deletions(-)
diff --git a/hw/block/dataplane/virtio-blk.c b/hw/block/dataplane/virtio-blk.c
index 5
.io_flush() is no longer called so drop qemu_laio_completion_cb(). It
turns out that count is now unused so drop that too.
Signed-off-by: Stefan Hajnoczi
---
block/linux-aio.c | 17 ++---
1 file changed, 2 insertions(+), 15 deletions(-)
diff --git a/block/linux-aio.c b/block/linux-
.io_flush() is no longer called so drop iscsi_process_flush().
Signed-off-by: Stefan Hajnoczi
---
block/iscsi.c | 9 +
1 file changed, 1 insertion(+), 8 deletions(-)
diff --git a/block/iscsi.c b/block/iscsi.c
index e7c1c2b..180b827 100644
--- a/block/iscsi.c
+++ b/block/iscsi.c
@@ -146,
.io_flush() is no longer called so drop return_true().
Signed-off-by: Stefan Hajnoczi
---
block/ssh.c | 10 +-
1 file changed, 1 insertion(+), 9 deletions(-)
diff --git a/block/ssh.c b/block/ssh.c
index d7e7bf8..e149da9 100644
--- a/block/ssh.c
+++ b/block/ssh.c
@@ -740,14 +740,6 @@ sta
Drop the io_flush argument to aio_set_event_notifier().
Reviewed-by: Paolo Bonzini
Signed-off-by: Stefan Hajnoczi
---
tests/test-aio.c | 22 --
1 file changed, 8 insertions(+), 14 deletions(-)
diff --git a/tests/test-aio.c b/tests/test-aio.c
index 1251952..7b2892a 100644
--
.io_flush() is no longer called so drop thread_pool_active(). The block
layer is the only thread-pool.c user and it already tracks in-flight
requests, therefore we do not need thread_pool_active().
Signed-off-by: Stefan Hajnoczi
---
thread-pool.c | 8 +---
1 file changed, 1 insertion(+), 7
.io_flush() is no longer called so drop qemu_rbd_aio_flush_cb().
qemu_aio_count is unused now so drop it too.
Signed-off-by: Stefan Hajnoczi
---
block/rbd.c | 14 +-
1 file changed, 1 insertion(+), 13 deletions(-)
diff --git a/block/rbd.c b/block/rbd.c
index cb71751..71b4a0c 100644
.io_flush() is no longer called so drop nbd_have_request(). We cannot
drop in_flight since it is still used by other block/nbd.c code.
Signed-off-by: Stefan Hajnoczi
---
block/nbd.c | 13 +++--
1 file changed, 3 insertions(+), 10 deletions(-)
diff --git a/block/nbd.c b/block/nbd.c
inde
The .io_flush() handler no longer exists and has no users. Drop the
io_flush argument to aio_set_fd_handler() and related functions.
The AioFlushEventNotifierHandler and AioFlushHandler typedefs are no
longer used and are dropped too.
Reviewed-by: Paolo Bonzini
Signed-off-by: Stefan Hajnoczi
-
On Fri, Aug 09, 2013 at 03:14:11PM +0800, Wenchao Xia wrote:
> 于 2013-8-6 23:07, Stefan Hajnoczi 写道:
> >On Thu, Jul 25, 2013 at 05:18:07PM +0200, Stefan Hajnoczi wrote:
> >Ping?
> >
> I tried to apply this series to do more work above it, but seems
> code conflict with upstream.
I sent a new rev
On Fri, Aug 9, 2013 at 4:12 PM, Jan Kiszka wrote:
> On 2013-08-08 23:41, Alex Bligh wrote:
>> This patch series adds support for timers attached to an AioContext clock
>> which get called within aio_poll.
>>
>> In doing so it removes alarm timers and moves to use ppoll where possible.
>>
>> This p
On 2013-08-09 10:24, liu ping fan wrote:
> Hi Jan, do you work on pushing hpet onto dedicated thread? If you do,
> I will cease my current work.
No, the HPET is not needed for our current use case, just the MC146818.
Jan
--
Siemens AG, Corporate Technology, CT RTC ITP SES-DE
Corporate Competenc
.io_flush() is no longer called so drop curl_aio_flush(). The acb[]
array that the function checks is still used in other parts of
block/curl.c. Therefore we cannot remove acb[], it is needed.
Signed-off-by: Stefan Hajnoczi
---
block/curl.c | 22 +++---
1 file changed, 3 insert
On Fri, Aug 9, 2013 at 4:06 PM, Paolo Bonzini wrote:
> Il 08/08/2013 08:26, Liu Ping Fan ha scritto:
>> Each slirp has its own time to caculate timeout.
>>
>> Signed-off-by: Liu Ping Fan
>> ---
>> slirp/slirp.c | 22 ++
>> slirp/slirp.h | 3 +++
>> 2 files changed, 13 insert
A PCI device's DMA address space (possibly an IOMMU) is returned by a
method on the PCIBus. At the moment that only has one caller, so the
method is simply open coded. We'll need another caller for VFIO, so
this patch introduces a helper/wrapper function.
Signed-off-by: David Gibson
[aik: added
spapr-pci config space accessors use find_dev() to find a PCI device.
However find_dev() only searched on a primary bus and did not do
recursive search through secondary buses so config space access was not
possible for devices other that on a primary bus.
This fixed find_dev() by using the PCI AP
On Thu, Aug 08, 2013 at 10:42:02PM +0100, Alex Bligh wrote:
> Where supported, called prctl(PR_SET_TIMERSLACK, 1, ...) to
> set one nanosecond timer slack to increase precision of timer
> calls.
>
> Signed-off-by: Alex Bligh
> ---
> qemu-timer.c |7 +++
> 1 file changed, 7 insertions(+)
On Thu, Aug 08, 2013 at 10:42:08PM +0100, Alex Bligh wrote:
> @@ -213,13 +214,41 @@ QEMUClock *timerlist_get_clock(QEMUTimerList
> *timer_list);
> bool timerlist_run_timers(QEMUTimerList *timer_list);
>
> /**
> + * timerlist_set_notify_cb:
> + * @timer_list: the timer list to use
> + * @cb: th
Hi Josh,
just opened
http://tracker.ceph.com/issues/5919
with all collected information incl. debug-log.
Hope it helps,
Oliver.
On 08/08/2013 07:01 PM, Josh Durgin wrote:
On 08/08/2013 05:40 AM, Oliver Francke wrote:
Hi Josh,
I have a session logged with:
debug_ms=1:debug_rbd=20:deb
On Thu, Aug 08, 2013 at 10:42:20PM +0100, Alex Bligh wrote:
> @@ -269,17 +299,17 @@ bool timerlist_expired(QEMUTimerList *timer_list);
> int64_t timerlist_deadline_ns(QEMUTimerList *timer_list);
>
> /**
> - * timerlist_getclock:
> + * timerlist_get_clock:
> * @timer_list: the timer list to op
Il 09/08/2013 10:48, liu ping fan ha scritto:
> After patch 3,4, the timeout will be an param for GSource's prepare.
> So I think I will set the right timeout value in patch4?
No, please do it before (in fact you can extract this patch and the
timeout change to a seprate series). Then patch 4 can
Il 09/08/2013 10:49, Alexey Kardashevskiy ha scritto:
> A PCI device's DMA address space (possibly an IOMMU) is returned by a
> method on the PCIBus. At the moment that only has one caller, so the
> method is simply open coded. We'll need another caller for VFIO, so
> this patch introduces a help
On Thu, Aug 08, 2013 at 10:41:57PM +0100, Alex Bligh wrote:
> This patch series adds support for timers attached to an AioContext clock
> which get called within aio_poll.
Patch 29/30 didn't make it to my inbox. It may have been bounced due to
size. Using your git repo instead.
Stefan
Hi,
> Converting src/smm.c to use a runtime value isn't hard - just change
> the assembler from: "mov $" __stringify(PORT_ACPI_PM_BASE) " + 0x04,
> %dx\n" to: "mov 4(my_acpi_base), %dx\n" and make sure to define the
> global variable my_acpi_base as VARFSEG.
The apm fix brought a ctl register v
On 08/09/2013 07:40 PM, Paolo Bonzini wrote:
> Il 09/08/2013 10:49, Alexey Kardashevskiy ha scritto:
>> A PCI device's DMA address space (possibly an IOMMU) is returned by a
>> method on the PCIBus. At the moment that only has one caller, so the
>> method is simply open coded. We'll need another
Hi Allqemu 1.6.1 has bug when starting it with -vnc, when i connect it
using vncviewer, the vncviewer will close automatically.
$uname -aDarwin Peters-MacBook-Air.local 11.4.2 Darwin Kernel Version 11.4.2:
Thu Aug 23 16:25:48 PDT 2012; root:xnu-1699.32.7~1/RELEASE_X86_64 x86_64
compile by:./
Il 09/08/2013 11:48, Alexey Kardashevskiy ha scritto:
> On 08/09/2013 07:40 PM, Paolo Bonzini wrote:
>> Il 09/08/2013 10:49, Alexey Kardashevskiy ha scritto:
>>> A PCI device's DMA address space (possibly an IOMMU) is returned by a
>>> method on the PCIBus. At the moment that only has one caller,
we need bdrv_new() to properly initialize BDS, don't allocate memory
manually.
Signed-off-by: Fam Zheng
---
block/vvfat.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/block/vvfat.c b/block/vvfat.c
index cd3b8ed..a827d91 100644
--- a/block/vvfat.c
+++ b/block/vvfat.c
@@ -29
[resend to the correct list]
BlockDriverState lifecycle management is needed by future features such as
image fleecing and blockdev-add. This series adds reference count to
BlockDriverState.
The first two patches clean up two odd BlockDriverState use cases, so all code
uses bdrv_new() to create B
Introduce bdrv_ref/bdrv_unref to manage the lifecycle of
BlockDriverState. They are unused for now but will used to replace
bdrv_delete() later.
Signed-off-by: Fam Zheng
---
block.c | 21 +
include/block/block.h | 2 ++
include/block/block_int.h | 1 +
We call bdrv_attach_dev when initializing whether or not bs is created
locally, so call bdrv_detach_dev and let the refcnt handle the
lifecycle.
Signed-off-by: Fam Zheng
---
hw/block/xen_disk.c | 11 +--
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/hw/block/xen_disk.c b/
BlockDriverState structure needs bdrv_new() to initialize refcnt, don't
allocate a local structure variable and memset to 0, becasue with coming
refcnt implementation, bdrv_unref will crash if bs->refcnt not
initialized to 1.
Signed-off-by: Fam Zheng
---
block/iscsi.c | 16 +---
1 fi
Il 08/08/2013 23:42, Alex Bligh ha scritto:
> Add QEMUTimerListGroup and helper functions, to represent
> a QEMUTimerList associated with each clock. Add a default
> QEMUTimerListGroup representing the default timer lists
> which are not associated with any other object (e.g.
> an AioContext as add
Manage BlockDriverState lifecycle with refcnt, so bdrv_delete() is no
longer public and should be called by bdrv_unref() if refcnt is
decreased to 0.
This is an identical change because effectively, there's no multiple
reference of BDS now: no caller of bdrv_ref() yet, only bdrv_new() sets
bs->ref
Il 08/08/2013 23:42, Alex Bligh ha scritto:
> Split QEMUClock into QEMUClock and QEMUTimerList so that we can
> have more than one QEMUTimerList associated with the same clock.
>
> Introduce a main_loop_timerlist concept and make existing
> qemu_clock_* calls that actually should operate on a QEMU
Previously, nbd calls drive_get_ref() on the drive of bs. A BDS doesn't
always have associated dinfo, which nbd doesn't care either. We already
have BDS ref count, so use it to make it safe for a BDS w/o blockdev.
Signed-off-by: Fam Zheng
---
blockdev-nbd.c | 10 +-
nbd.c | 5 +
Il 08/08/2013 23:42, Alex Bligh ha scritto:
> Convert mainloop to use timeout from default timerlist group
> (i.e. the current 3 static timers)
And with two AioContexts in the main loop this patch disappears
completely, since the deadline is computed by the second AioContext.
Paolo
> Signed-off-
Block jobs used drive_get_ref(drive_get_by_blockdev(bs)) to avoid BDS
being deleted. Now we have BDS reference count, and block jobs don't
care about dinfo, so replace them to get cleaner code. It is also the
safe way when BDS has no drive info.
Signed-off-by: Fam Zheng
---
blockdev.c | 53 -
Il 08/08/2013 23:42, Alex Bligh ha scritto:
> Add a notify pointer to QEMUTimerList so it knows what to notify
> on a timer change.
If we do the "two AioContexts" trick, this can simply be a back-pointer
to the AioContext.
Paolo
> Signed-off-by: Alex Bligh
> ---
> async.c |7 +
Il 08/08/2013 23:42, Alex Bligh ha scritto:
> On qemu_mod_timer_ns, ensure qemu_notify or aio_notify is called to
> end the appropriate poll(), irrespective of use_icount value.
>
> On qemu_clock_enable, ensure qemu_notify or aio_notify is called for
> all QEMUTimerLists attached to the QEMUClock.
Il 08/08/2013 23:42, Alex Bligh ha scritto:
> Calculate the timeout in aio_ctx_prepare taking into account
> the timers attached to the AioContext.
>
> Alter aio_ctx_check similarly.
>
> Signed-off-by: Alex Bligh
> ---
> async.c | 13 +++--
> 1 file changed, 11 insertions(+), 2 deleti
Il 08/08/2013 23:42, Alex Bligh ha scritto:
> Add aio_timer_new wrapper function.
>
> Signed-off-by: Alex Bligh
> ---
> include/block/aio.h | 19 +++
> 1 file changed, 19 insertions(+)
>
> diff --git a/include/block/aio.h b/include/block/aio.h
> index a13f6e8..bd6f17c 100644
>
On 08/09/2013 07:53 PM, Paolo Bonzini wrote:
> Il 09/08/2013 11:48, Alexey Kardashevskiy ha scritto:
>> On 08/09/2013 07:40 PM, Paolo Bonzini wrote:
>>> Il 09/08/2013 10:49, Alexey Kardashevskiy ha scritto:
A PCI device's DMA address space (possibly an IOMMU) is returned by a
method on th
Il 09/08/2013 12:13, Alexey Kardashevskiy ha scritto:
> On 08/09/2013 07:53 PM, Paolo Bonzini wrote:
>> Il 09/08/2013 11:48, Alexey Kardashevskiy ha scritto:
>>> On 08/09/2013 07:40 PM, Paolo Bonzini wrote:
Il 09/08/2013 10:49, Alexey Kardashevskiy ha scritto:
> A PCI device's DMA address
On Fri, 2013-08-09 at 11:40 +0200, Paolo Bonzini wrote:
> (BTW, do you need to enable bus-master DMA on PCI bridges, to do DMA
> for devices sitting on the secondary bus?)
In theory yes though I have seen bridges who ignore it...
Cheers,
Ben.
Il 08/08/2013 23:42, Alex Bligh ha scritto:
> Add scripts/switch-timer-api to programatically rewrite source
> files to use the new timer system.
>
> Signed-off-by: Alex Bligh
> ---
> scripts/switch-timer-api | 178
> ++
> 1 file changed, 178 inserti
Il 08/08/2013 23:42, Alex Bligh ha scritto:
> + # these have just changed name
> + $line =~ s/\bqemu_mod_timer\b/qemu_timer_mod/g;
> + $line =~ s/\bqemu_mod_timer_(ns|us|ms)\b/qemu_timer_mod_$1/g;
> + $line =~ s/\bqemu_free_timer\b/qemu_timer_free/g;
> + $lin
block-migration.c does not actually use DriveInfo anywhere. Hence it's
safe to drive ref code, we really only care about referencing BDS.
Signed-off-by: Fam Zheng
---
block-migration.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/block-migration.c b/block-migration.c
Il 08/08/2013 23:41, Alex Bligh ha scritto:
> This patch series adds support for timers attached to an AioContext clock
> which get called within aio_poll.
>
> In doing so it removes alarm timers and moves to use ppoll where possible.
>
> This patch set 'sort of' passes make check (see below for
Il 09/08/2013 12:20, Benjamin Herrenschmidt ha scritto:
> On Fri, 2013-08-09 at 11:40 +0200, Paolo Bonzini wrote:
>> (BTW, do you need to enable bus-master DMA on PCI bridges, to do DMA
>> for devices sitting on the secondary bus?)
>
> In theory yes though I have seen bridges who ignore it...
And
Jan,
--On 9 August 2013 10:12:45 +0200 Jan Kiszka wrote:
Do you have this in git somewhere? I'd like to port my (almost) BQL-free
threaded RTC device model on top to check how well the new API works.
But it looks promising.
Per my subsequent message:
For ease of review, the finished result
On Fri, Aug 09, 2013 at 07:48:16PM +1000, Alexey Kardashevskiy wrote:
> On 08/09/2013 07:40 PM, Paolo Bonzini wrote:
> > Il 09/08/2013 10:49, Alexey Kardashevskiy ha scritto:
> >> A PCI device's DMA address space (possibly an IOMMU) is returned by a
> >> method on the PCIBus. At the moment that on
On 08/09/2013 08:19 PM, Paolo Bonzini wrote:
> Il 09/08/2013 12:13, Alexey Kardashevskiy ha scritto:
>> On 08/09/2013 07:53 PM, Paolo Bonzini wrote:
>>> Il 09/08/2013 11:48, Alexey Kardashevskiy ha scritto:
On 08/09/2013 07:40 PM, Paolo Bonzini wrote:
> Il 09/08/2013 10:49, Alexey Kardashe
Il 09/08/2013 12:58, Alexey Kardashevskiy ha scritto:
> On 08/09/2013 08:19 PM, Paolo Bonzini wrote:
>> Il 09/08/2013 12:13, Alexey Kardashevskiy ha scritto:
>>> On 08/09/2013 07:53 PM, Paolo Bonzini wrote:
Il 09/08/2013 11:48, Alexey Kardashevskiy ha scritto:
> On 08/09/2013 07:40 PM, Pao
On Thu, Aug 08, 2013 at 10:41:57PM +0100, Alex Bligh wrote:
> This patch series adds support for timers attached to an AioContext clock
> which get called within aio_poll.
>
> In doing so it removes alarm timers and moves to use ppoll where possible.
>
> This patch set 'sort of' passes make check
On 08/09/2013 09:07 PM, Paolo Bonzini wrote:
> Il 09/08/2013 12:58, Alexey Kardashevskiy ha scritto:
>> On 08/09/2013 08:19 PM, Paolo Bonzini wrote:
>>> Il 09/08/2013 12:13, Alexey Kardashevskiy ha scritto:
On 08/09/2013 07:53 PM, Paolo Bonzini wrote:
> Il 09/08/2013 11:48, Alexey Kardashe
Il 09/08/2013 13:21, Alexey Kardashevskiy ha scritto:
> On 08/09/2013 09:07 PM, Paolo Bonzini wrote:
>> Il 09/08/2013 12:58, Alexey Kardashevskiy ha scritto:
>>> On 08/09/2013 08:19 PM, Paolo Bonzini wrote:
Il 09/08/2013 12:13, Alexey Kardashevskiy ha scritto:
> On 08/09/2013 07:53 PM, Pao
Am 08.08.2013 07:15, schrieb Prerna Saxena:
> On 08/01/2013 06:32 AM, Andreas Färber wrote:
>> By default on KVM or when user asks for it via -cpu host, cpu_model will
>> be "host" and sPAPR merely upper-cases it for the SLOF device tree.
>>
>> Change it so that we get the underlying CPU type, e.g.
This just binds both patches properly together and fixes up the return
value size of unassigned_io_read as suggested. Please merge before the
1.6 release.
Jan Kiszka (2):
memory: Provide separate handling of unassigned io ports accesses
Revert "memory: Return -1 again on reads from unsigned re
Accesses to unassigned io ports shall return -1 on read and be ignored
on write. Ensure these properties via dedicated ops, decoupling us from
the memory core's handling of unassigned accesses.
Signed-off-by: Jan Kiszka
---
exec.c|3 ++-
include/exec/ioport.h |2 ++
iopor
This reverts commit 9b8c69243585a32d14b9bb9fcd52c37b0b5a1b71.
The commit was wrong: We only return -1 on invalid accesses, not on
valid but unbacked ones. This broke various corner cases.
Signed-off-by: Jan Kiszka
---
memory.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --
> It fail with the following error message at exit and I don't know why yet.
> qemu-system-x86_64: block.c:1489: bdrv_drain_all: Assertion
> `((&bs->tracked_requests)->lh_first == ((void *)0))' failed.
I solved this issue: bdrv_drain_all was bogus.
Best regards
Benoît
>
> block.c
Gerd Hoffmann writes:
> On 08/08/13 18:38, Anthony Liguori wrote:
>> This breaks migration and is unneeded with modern SeaBIOS.
>
> No. Dropping for piix is fine. It will break q35 though.
Can you elaborate? When Michael and I discussed this I was under the
impression that latest SeaBIOS had
On 8 August 2013 13:51, Peter Maydell wrote:
> For ARM you can't get at feature info of the host from userspace
> (unless you want to get into parsing /proc/cpuinfo), so my current
> idea is to have KVM_ARM_VCPU_INIT support a target-cpu-type
> which means "whatever host CPU is".
To expand on thi
On 23 July 2013 10:33, Mian M. Hamayun wrote:
> From: "Mian M. Hamayun"
>
> The cpu init function tries to initialize with all possible cpu types, as
> KVM does not provide a means to detect the real cpu type and simply refuses
> to initialize on cpu type mis-match. By using the loop based init f
On 08/09/13 14:38, Anthony Liguori wrote:
> Gerd Hoffmann writes:
>
>> On 08/08/13 18:38, Anthony Liguori wrote:
>>> This breaks migration and is unneeded with modern SeaBIOS.
>>
>> No. Dropping for piix is fine. It will break q35 though.
>
> Can you elaborate? When Michael and I discussed th
[CC'ing qemu-devel list]
On 09.08.2013 15:17, Daniel P. Berrange wrote:
> On Fri, Aug 09, 2013 at 07:13:58AM -0600, Eric Blake wrote:
>> On 08/09/2013 06:56 AM, Michal Privoznik wrote:
>>> This function is to guess the correct limit for maximal memory
>>> usage by qemu for given domain. This can ne
Now in KVM, when RAM snapshot, vcpus needs stopped, it is Unfriendly
restrictions to users.
Are there plans to achieve ram live Snapshot feature?
in my mind, Snapshots can not occupy additional too much memory, So when the
memory needs to be changed, the old memory page is needed to flush to th
When individual CONFIG_ switches for the A9MPcore and A15MPcore
devices were created, they were inadvertently given incorrect names
(CONFIG_ARM9MPCORE and CONFIG_ARM15MPCORE). These CPUs are
"Cortex-A9MP" and "Cortex-A15MP", and in particular the ARM9 is
a different (rather older) CPU than the Cort
On 9 Aug 2013, at 09:53, Stefan Hajnoczi wrote:
> The ./configure change should also be in this patch. I think it crept
> into another patch by mistake.
Oops - I will fix.
--
Alex Bligh
Rusty Russell writes:
> Anthony Liguori writes:
>> I suspect this is a premature optimization. With a weak function called
>> directly in the accessors below, I suspect you would see no measurable
>> performance overhead compared to this approach.
>>
>> It's all very predictable so the CPU shou
commit 41cb383f42d0cb51d8e3e25e3ecebc954dd4196f made a guest-visible
change by adding the PCLMULQDQ bit to Westmere without adding
compatibility code to keep the ABI older machine-types. This patch fixes
it by adding the missing compat code.
Signed-off-by: Eduardo Habkost
---
Bug detected by the
Rusty Russell writes:
> Anthony Liguori writes:
>> "Daniel P. Berrange" writes:
>>
>> The distinction is important in QEMU. ppc64 is still
>> TARGET_WORDS_BIGENDIAN. We still want most stl_phys to treat integers
>> as big endian. There's just this extra concept that CPU loads/stores
>> are s
On 9 Aug 2013, at 10:02, Stefan Hajnoczi wrote:
> When looking at thread-safety I had to think about set_notify_cb() for a
> while. The issue is that we add the timerlist to the clock source's
> ->timerlists *before* notify_cb has been assigned.
>
> This could be a problem is another thread re-
On 9 Aug 2013, at 10:23, Stefan Hajnoczi wrote:
> On Thu, Aug 08, 2013 at 10:42:20PM +0100, Alex Bligh wrote:
>> @@ -269,17 +299,17 @@ bool timerlist_expired(QEMUTimerList *timer_list);
>> int64_t timerlist_deadline_ns(QEMUTimerList *timer_list);
>>
>> /**
>> - * timerlist_getclock:
>> + * timer
Stefan,
On 9 Aug 2013, at 10:41, Stefan Hajnoczi wrote:
> Patch 29/30 didn't make it to my inbox. It may have been bounced due to
> size. Using your git repo instead.
I can break this up into 1 patch per file if you prefer, which is
what I was originally going to do. The automated program does
On 9 Aug 2013, at 11:03, Paolo Bonzini wrote:
>>
>> +/* New format calling conventions for timers */
>> +
>> +/**
>> + * timer_free:
>> + * @ts: the timer
>> + *
>> + * Free a timer (it must not be on the active list)
>> + */
>> +static inline void timer_free(QEMUTimer *ts)
>> +{
>> +qemu_fr
Am 09.08.2013 09:00, schrieb Rusty Russell:
> Andreas Färber writes:
>> Am 08.08.2013 15:31, schrieb Anthony Liguori:
>>> Rusty Russell writes:
>>> We have a mechanism to do weak functions via stubs/. I think it would
>>> be better to do cpu_get_byteswap() as a stub function and then overload
>>
Il 09/08/2013 16:18, Alex Bligh ha scritto:
>
> On 9 Aug 2013, at 10:23, Stefan Hajnoczi wrote:
>
>> On Thu, Aug 08, 2013 at 10:42:20PM +0100, Alex Bligh wrote:
>>> @@ -269,17 +299,17 @@ bool timerlist_expired(QEMUTimerList *timer_list);
>>> int64_t timerlist_deadline_ns(QEMUTimerList *timer_list
Il 09/08/2013 16:19, Alex Bligh ha scritto:
> Stefan,
>
> On 9 Aug 2013, at 10:41, Stefan Hajnoczi wrote:
>
>> Patch 29/30 didn't make it to my inbox. It may have been bounced due to
>> size. Using your git repo instead.
>
> I can break this up into 1 patch per file if you prefer, which is
> w
1 - 100 of 204 matches
Mail list logo
