29.04.2015 09:43, Thomas Huth wrote:
> On Wed, 29 Apr 2015 09:34:51 +0300
> Michael Tokarev wrote:
[]
>> Hm. So, what's the difference? The same fields are assigned the same
>> values, why in first case we have some uninitialized data and in second
>> case everything is initialized? Does struct
General question first: through which tree should this go?
MAINTAINERS doesn't cover the QAPI generators. Closest related stanza
is QAPI (Luiz, Michael R.). Should it cover the generators? Next
closest is QAPI Schema (Luiz, you, myself).
For completeness: because we touch qmp_query_pci_bridge(
28.04.2015 12:11, arei.gong...@huawei.com wrote:
> From: Gonglei
>
> Value from xfer->packet.ep is assigned to ep here, but that
> stored value is not used before it is overwritten. Remove it.
>
> Cc: Gerd Hoffmann
> Signed-off-by: Gonglei
> ---
> hw/usb/hcd-xhci.c | 1 -
> 1 file changed, 1
On Wed, 29 Apr 2015 09:34:51 +0300
Michael Tokarev wrote:
> 27.04.2015 19:59, Thomas Huth wrote:
> > valgrind complains here about uninitialized bytes with the following
> > message:
> >
> > ==17814== Syscall param ioctl(generic) points to uninitialised byte(s)
> > ==17814==at 0x466A780: io
http://patchwork.ozlabs.org/patch/456533/
http://patchwork.ozlabs.org/patch/456535/
http://patchwork.ozlabs.org/patch/456536/
http://patchwork.ozlabs.org/patch/456534/
On Tue, Mar 31, 2015 at 11:34 AM, wrote:
> From: Itamar Tal
>
> This bundle of patches adds support for some more functionality
Applied to trivial, thank you!
/mjt
27.04.2015 19:59, Thomas Huth wrote:
> valgrind complains here about uninitialized bytes with the following message:
>
> ==17814== Syscall param ioctl(generic) points to uninitialised byte(s)
> ==17814==at 0x466A780: ioctl (in /usr/lib64/power8/libc-2.17.so)
> ==17814==by 0x100735B7: kvm_v
Applied to -trivial, thanks!
So much for removing one line :)
/mjt
27.04.2015 19:45, Emilio G. Cota wrote:
> Here is a hodge-podge of fixes for issues I found while reviewing
> QTAILQ callers, with a .gitignore patch as a bonus.
Applied to -trivial, all except the bonus. The 5/6 one (unnecessary
parentheses removal) is a bit too trivial, maybe not worh a patch,
28.04.2015 09:38, Thomas Huth wrote:
>> Date: Mon, 27 Apr 2015 20:09:50 +0200
>> From: Stefan Weil
>>
>> Am 27.04.2015 um 13:27 schrieb Michael Tokarev:
>>> 14.03.2015 09:19, Thomas Huth wrote:
Here are some more patches to remove completely unused functions
from QEMU. Please review care
Quoting David Gibson (2015-04-28 02:23:43)
> On Wed, Apr 22, 2015 at 01:28:11AM -0500, Michael Roth wrote:
> > This interface is used to fetch an OF device-tree nodes that describes a
> > newly-attached device to guest. It is called multiple times to walk the
> > device-tree node and fetch individu
Quoting David Gibson (2015-04-28 02:25:51)
> On Wed, Apr 22, 2015 at 01:28:15AM -0500, Michael Roth wrote:
> > Initially this is identifical to pseries-2.3. Subsequent patches will
> > use it to mask new 2.4 features by default.
> >
> > Signed-off-by: Michael Roth
>
> So, I've seen slightly diff
27.04.2015 17:47, Michael Tokarev wrote:
> 26.04.2015 22:10, Peter Crosthwaite wrote:
>> Hi Edgar,
>>
>> This is some code cleanup of Microblaze. Mainly unused code
>> deletion but also code share of the defconfig (P1).
>
> Applied all 6 to -trivial, as is, since the whole series
> does not perfor
27.04.2015 22:06, Peter Crosthwaite wrote:
> Does that mean a respin of this patch or should the tricore fix and
> the comment be a follow up?
Can you please resend whole thing in one change, removing single
unused function from two places and updating the comment, to fit
all the pieces together?
After removal of EXCP_NMI there's a gap in EXCP_*
numbering. Let's remove it.
Signed-off-by: Michael Tokarev
---
Since I applied Peter's series already, without renumbering,
b/c he wanted to make no code changes and that's a good plan,
I send this renumbering as a separate change. Hopefully this
Alexey Kardashevskiy writes:
> On 04/22/2015 08:35 PM, Nikunj A Dadhania wrote:
>> All the PCI enumeration and device node creation was off-loaded to
>> SLOF. With PCI hotplug support, code needed to be added to add device
>> node. This creates multiple copy of the code one in SLOF and other in
>
Signed-off-by: Fam Zheng
---
poll-glib.c | 29 ++---
1 file changed, 26 insertions(+), 3 deletions(-)
diff --git a/poll-glib.c b/poll-glib.c
index 64fde69..23d528d 100644
--- a/poll-glib.c
+++ b/poll-glib.c
@@ -14,6 +14,9 @@
#include "qemu-common.h"
#include "qemu/timer
Signed-off-by: Fam Zheng
---
tests/Makefile| 2 +
tests/test-poll.c | 272 ++
2 files changed, 274 insertions(+)
create mode 100644 tests/test-poll.c
diff --git a/tests/Makefile b/tests/Makefile
index 55aa745..d152bf5 100644
--- a/tests/
In qemu_poll_timerfd, we arm the timerfd with timeout_ns. The timerfd is
also watched by epollfd, so that when there is no other events,
epoll_wait will still return on time, even though we pass -1 (wait
infinitely).
Signed-off-by: Fam Zheng
---
poll-linux.c | 80
qemu_poll_set_fds + qemu_poll does the same, and when epoll is
available, it is faster.
Signed-off-by: Fam Zheng
---
include/qemu/timer.h | 13 -
main-loop.c | 35 ++-
qemu-timer.c | 28
3 files changed, 30
This implements qemu_poll with ppoll + epoll. The only complex part is
qemu_poll_set_fds, which will sync up epollfd with epoll_ctl by
computing the symmetric difference of previous and new fds.
The ppoll is used to retain ns precision of timeout.
Signed-off-by: Fam Zheng
---
Makefile.objs |
The AIO handler list is only modified by aio_set_fd_handler, so we can
easily add del poll fd there. Initialize a QEMUPoll and keep track of
all the fds, so we don't need to rebuild a GPollFD array for g_poll in
aio_poll.
Signed-off-by: Fam Zheng
---
aio-posix.c | 26
This is abstract of underlying poll implementation. A glib
implementation is included.
Signed-off-by: Fam Zheng
---
Makefile.objs | 2 +-
include/qemu/poll.h | 40 +++
include/qemu/typedefs.h | 4 +-
poll-glib.c | 130 +++
v4: Rebase to 2.3 and rerun tests. timerfd+epoll shows clear improvement over
current master when the virtio-console device attaches more fds to the main
loop:
syscall high # of fd low # of fd
-
qemu.git(ppoll) 24
On 2015/4/28 23:54, Michael S. Tsirkin wrote:
> On Tue, Apr 28, 2015 at 05:13:10PM +0200, Igor Mammedov wrote:
>>> Here I need to set the value of buffer to 1 or 0, we could
>>> createbitfield, but if we want to set the value to non one or zero and
>>> the BufferSize is large, how could we do? Crea
On 04/22/2015 08:35 PM, Nikunj A Dadhania wrote:
All the PCI enumeration and device node creation was off-loaded to
SLOF. With PCI hotplug support, code needed to be added to add device
node. This creates multiple copy of the code one in SLOF and other in
hotplug code. To unify this, the patch ad
On 04/27/2015 12:32 PM, Eric Blake wrote:
> On 04/27/2015 12:15 PM, Markus Armbruster wrote:
>> Eric Blake writes:
>>
>>> Previous commits demonstrated that the generator had several
>>> flaws with less-than-perfect unions:
>
>>> +++ b/tests/qapi-schema/alternate-array.json
>>> @@ -1,4 +1,4 @@
>>
On Tue, Apr 28, 2015 at 10:37:08AM +0200, Michael S. Tsirkin wrote:
> On Tue, Apr 28, 2015 at 05:31:03PM +1000, David Gibson wrote:
> > On Wed, Apr 22, 2015 at 01:28:18AM -0500, Michael Roth wrote:
> > > We need to work with PCI BARs to generate OF properties
> > > during PCI hotplug for sPAPR gues
On Tue, Apr 28, 2015 at 05:25:37PM +1000, Alexey Kardashevskiy wrote:
> On 04/10/2015 12:16 PM, David Gibson wrote:
> >On Wed, Apr 08, 2015 at 07:33:21PM +1000, Alexey Kardashevskiy wrote:
> >>Since 8dfa3a5e "target-ppc: Add "compat" CPU option" (which was a part of
> >>client-architecture-support
On Tue, Apr 28, 2015 at 01:30:29PM +0200, Thomas Huth wrote:
> On Tue, 28 Apr 2015 12:23:25 +0530
> Nikunj A Dadhania wrote:
[snip]
> Looks good to me.
>
> Reviewed-by: Thomas Huth
Same here,
Reviewed-by: David Gibson
--
David Gibson| I'll have my music baroque, and my c
On Tue, Apr 28, 2015 at 01:32:15PM +0200, Thomas Huth wrote:
> On Tue, 28 Apr 2015 12:23:26 +0530
> Nikunj A Dadhania wrote:
>
> > Signed-off-by: Nikunj A Dadhania
> > Reviewed-by: Igor Mammedov
> > ---
> > hw/ppc/spapr.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/hw/ppc/s
On Tue, 04/28 16:54, Peter Maydell wrote:
> On 28 April 2015 at 15:40, Paolo Bonzini wrote:
> > The following changes since commit e1a5476354d396773e4c555f126d752d4ae58fa9:
> >
> > Open 2.4 development tree (2015-04-25 22:05:07 +0100)
> >
> > are available in the git repository at:
> >
> > git
Test sector offset 0, 1, and the last sector(s)
in LBA28 and LBA48 modes.
Signed-off-by: John Snow
Acked-by: Stefan Hajnoczi
Message-id: 1426274523-22661-3-git-send-email-js...@redhat.com
---
tests/ahci-test.c | 68 +++--
tests/libqos/ahci.c | 1
Signed-off-by: John Snow
Message-id: 1426018503-821-7-git-send-email-js...@redhat.com
---
tests/ahci-test.c | 44
1 file changed, 44 insertions(+)
diff --git a/tests/ahci-test.c b/tests/ahci-test.c
index cd7d2ce..7c23bb2 100644
--- a/tests/ahci-test.c
Signed-off-by: John Snow
Message-id: 1426018503-821-3-git-send-email-js...@redhat.com
---
tests/ahci-test.c| 67
tests/libqos/libqos-pc.c | 5
tests/libqos/libqos-pc.h | 1 +
3 files changed, 51 insertions(+), 22 deletions(-)
diff -
From: Ed Maste
Signed-off-by: Ed Maste
Reviewed-by: John Snow
Message-id: 1427911244-22565-1-git-send-email-ema...@freebsd.org
Signed-off-by: John Snow
---
tests/libqtest.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/tests/libqtest.c b/tests/libqtest.c
index 659a3c7..a525dc5 100644
--
Allow the user to poll until a desired interrupt occurs.
Signed-off-by: John Snow
Message-id: 1426018503-821-4-git-send-email-js...@redhat.com
---
tests/ide-test.c | 11 +--
tests/libqtest.c | 16
tests/libqtest.h | 20
3 files changed, 37 insertions
Add qmp_async, which lets us send QMP commands asynchronously.
This is useful when we want to send commands that will trigger
event responses, but we don't know in what order to expect them.
Sometimes the event responses may arrive even before the command
confirmation will show up, so it is conven
This will enable the testing of high offsets without
wasting a lot of disk space, and does not impact the
previous tests.
mkimg and mkqcow2 are added to libqos for other tests.
Signed-off-by: John Snow
Acked-by: Stefan Hajnoczi
Message-id: 1426274523-22661-2-git-send-email-js...@redhat.com
---
Pull this helper out of ide-test and into libqos,
to be shared with ahci-test.
Signed-off-by: John Snow
Message-id: 1426018503-821-6-git-send-email-js...@redhat.com
---
tests/ide-test.c | 23 +--
tests/libqos/libqos.c | 22 ++
tests/libqos/libqos.h |
From: Hervé Poussineau
They were introduced in 6f7e9aec5eb5bdfa57a9e458e391b785c283a007 and
82407d1a4035e5bfefb53ffdcb270872f813b34c and lots of bug fixes were done after
that.
This fixes (at least) the detection of the floppy controller on Debian
4.0r9/SPARC,
and SS-5's OBP initialization rou
The following changes since commit a9392bc93c8615ad1983047e9f91ee3fa8aae75f:
Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging
(2015-04-28 16:55:03 +0100)
are available in the git repository at:
https://github.com/jnsnow/qemu.git tags/ide-pull-request
for you to f
Signed-off-by: John Snow
Message-id: 1426018503-821-2-git-send-email-js...@redhat.com
---
tests/ahci-test.c | 34 ++
1 file changed, 34 insertions(+)
diff --git a/tests/ahci-test.c b/tests/ahci-test.c
index 6686242..4a5c788 100644
--- a/tests/ahci-test.c
+++ b/tes
On 04/04/2015 10:07 PM, Eric Blake wrote:
> The previous commit demonstrated that the generator had several
> flaws with less-than-perfect enums:
> +++ b/tests/qapi-schema/enum-max-member.json
> @@ -1,3 +1,3 @@
> -# FIXME: we should either reject user-supplied 'max', or munge the implicit
> -# max
On Tue, Apr 28, 2015 at 01:41:13PM +0100, Leon Alrae wrote:
> Signed-off-by: Leon Alrae
I think this deserves a bit more explanation. Isn't 59 bits still the
true architectural limit?
Cheers
James
> ---
> target-mips/translate_init.c | 9 -
> 1 file changed, 9 deletions(-)
>
> diff --g
On 28 April 2015 at 22:32, Chen Gang wrote:
> The related information for cmpexch instruction:
>
> Description
>
> Compare the 8-byte contents of the CmpValue SPR with the 8-byte
> value in memory at the address held in the first source register. If
> the values are not equal, then n
On Tue, Apr 28, 2015 at 04:59:20PM +0100, Leon Alrae wrote:
> Hi James,
>
> On 28/04/2015 14:35, James Hogan wrote:
> >
> >
> > On 28/04/15 13:41, Leon Alrae wrote:
> >> Update tlb->PFN to contain PFN concatenated with PFNX. PFNX is 0 if large
> >> physical address is not supported.
> >>
> >> Si
The current documentation for -cpu merely refers the user to run
-cpu help, but whilst this lists the flags, it doesn't actually
say how to use them.
Signed-off-by: Alex Bligh
---
qemu-options.hx | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/qemu-options.hx b/qemu-optio
On 4/29/15 04:19, Chen Gang wrote:
> On 4/28/15 06:04, Richard Henderson wrote:
>> On 04/27/2015 02:12 PM, Chen Gang wrote:
>>> Now, I am blocked at SPRs: "mtspr SPR_CMPEXCH_VALUE %r12", and trying to
>>> solve it.
>>
>> SPR_CMPEXCH_VALUE should be present in your cpu structure, should be a TCGv
>>
On 28/04/15 22:07, John Snow wrote:
> On 04/28/2015 04:57 PM, Mark Cave-Ayland wrote:
>> On 17/03/15 07:23, Alexander Graf wrote:
>>
>>> On 09.03.15 23:24, Mark Cave-Ayland wrote:
This patchset attempts to separate out the IDE/ATAPI logic from the
unaligned
DMA access logic for maci
On 04/28/2015 04:57 PM, Mark Cave-Ayland wrote:
On 17/03/15 07:23, Alexander Graf wrote:
On 09.03.15 23:24, Mark Cave-Ayland wrote:
This patchset attempts to separate out the IDE/ATAPI logic from the unaligned
DMA access logic for macio which provides the following benefits:
1) Reduced code
On Tue, Jan 13, 2015 at 22:07:19 +, Peter Maydell wrote:
> On 13 January 2015 at 20:27, Paolo Bonzini wrote:
> > On 13/01/2015 21:00, Peter Maydell wrote:
> >> Hmm. That's a chunk of users who are now going to have to
> >> change the way they've been building QEMU. Does configure
> >> at least
On 17/03/15 07:23, Alexander Graf wrote:
> On 09.03.15 23:24, Mark Cave-Ayland wrote:
>> This patchset attempts to separate out the IDE/ATAPI logic from the unaligned
>> DMA access logic for macio which provides the following benefits:
>>
>> 1) Reduced code complexity
>>
>> The existing macio IDE/
On 4/28/15 06:04, Richard Henderson wrote:
> On 04/27/2015 02:12 PM, Chen Gang wrote:
>> Now, I am blocked at SPRs: "mtspr SPR_CMPEXCH_VALUE %r12", and trying to
>> solve it.
>
> SPR_CMPEXCH_VALUE should be present in your cpu structure, should be a TCGv
> like all of your other registers, and thi
On 04/28/2015 01:55 PM, Gabriel L. Somlo wrote:
> It has been reported that sometimes the .rodata section of SeaBIOS,
> containing the constant string against which the SMBIOS signature
> ends up being compared, also falls withing the guest f-segment. In
s/withing/within/
> that case, the test ob
It has been reported that sometimes the .rodata section of SeaBIOS,
containing the constant string against which the SMBIOS signature
ends up being compared, also falls withing the guest f-segment. In
that case, the test obviously fails, unless we continue searching
for the *real* smbios entry poin
On Tue, Apr 28, 2015 at 03:40:37PM -0400, Gabriel L. Somlo wrote:
> ping ?
>
> On Tue, Mar 31, 2015 at 02:32:20PM -0400, Gabriel L. Somlo wrote:
> > Rather than stopping at the first match for the SMBIOS signature
> > ("_SM_") in the f-segment (0xF-0xF), continue scanning
> > until either
On 04/04/2015 10:07 PM, Eric Blake wrote:
> Go into more details about the various types of valid expressions
> in a qapi schema, including tweaks to document fixes being done
> later in the current patch series. Also fix some stale and missing
> documentation in the QMP specification.
>
> Signed
ping ?
On Tue, Mar 31, 2015 at 02:32:20PM -0400, Gabriel L. Somlo wrote:
> Rather than stopping at the first match for the SMBIOS signature
> ("_SM_") in the f-segment (0xF-0xF), continue scanning
> until either a valid entry point table is found, or the f-segment
> has been exhausted.
>
On 04/28/2015 01:20 PM, John Snow wrote:
> Just a trivial patch to correct a QMP example in qmp-commands.hx.
>
> Signed-off-by: John Snow
> ---
> qmp-commands.hx | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com+1-919-30
On 04/28/2015 03:32 PM, John Snow wrote:
On 03/30/2015 11:10 AM, Stefan Hajnoczi wrote:
On Wed, Mar 25, 2015 at 06:57:35PM -0400, John Snow wrote:
QEMU does not compile cleanly under clang 3.5.0. These patches
eliminate the
avalanche of warnings and make the build usable.
The result is that
On 03/30/2015 11:10 AM, Stefan Hajnoczi wrote:
On Wed, Mar 25, 2015 at 06:57:35PM -0400, John Snow wrote:
QEMU does not compile cleanly under clang 3.5.0. These patches eliminate the
avalanche of warnings and make the build usable.
The result is that you *should* be able to use clang 3.5.0 *wi
Just a trivial patch to correct a QMP example in qmp-commands.hx.
Signed-off-by: John Snow
---
qmp-commands.hx | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/qmp-commands.hx b/qmp-commands.hx
index 213508f..d4a837c 100644
--- a/qmp-commands.hx
+++ b/qmp-commands.hx
@@ -2
On Tue, Apr 28, 2015 at 07:34:29PM +0100, Peter Maydell wrote:
> On 28 April 2015 at 19:14, Michael S. Tsirkin wrote:
> > On Tue, Apr 28, 2015 at 04:35:16PM +0200, Michael S. Tsirkin wrote:
> >> On Tue, Apr 28, 2015 at 03:24:19PM +0200, Cornelia Huck wrote:
> >> > Yep, see my other reply... I'm no
On Tue, Apr 28, 2015 at 08:32:51PM +0200, Michael S. Tsirkin wrote:
> On Tue, Apr 28, 2015 at 08:14:44PM +0200, Michael S. Tsirkin wrote:
> > On Tue, Apr 28, 2015 at 04:35:16PM +0200, Michael S. Tsirkin wrote:
> > > On Tue, Apr 28, 2015 at 03:24:19PM +0200, Cornelia Huck wrote:
> > > > On Tue, 28 A
On 28 April 2015 at 19:14, Michael S. Tsirkin wrote:
> On Tue, Apr 28, 2015 at 04:35:16PM +0200, Michael S. Tsirkin wrote:
>> On Tue, Apr 28, 2015 at 03:24:19PM +0200, Cornelia Huck wrote:
>> > Yep, see my other reply... I'm not quite sure what's wrong with
>> > event_idx on virtio-blk for s390-vi
On Tue, Apr 28, 2015 at 08:14:44PM +0200, Michael S. Tsirkin wrote:
> On Tue, Apr 28, 2015 at 04:35:16PM +0200, Michael S. Tsirkin wrote:
> > On Tue, Apr 28, 2015 at 03:24:19PM +0200, Cornelia Huck wrote:
> > > On Tue, 28 Apr 2015 14:16:40 +0100
> > > Peter Maydell wrote:
> > >
> > > > On 28 Apri
On 27 April 2015 at 16:00, Peter Maydell wrote:
> On 15 April 2015 at 17:02, Greg Bellows wrote:
>> From: Fabian Aggeler
>>
>> ICCICR/GICC_CTLR is banked in GICv1 implementations with Security
>> Extensions or in GICv2 in independent from Security Extensions.
>> This makes it possible to enable
On Tue, Apr 28, 2015 at 2:18 AM, Eric Auger wrote:
> The VFIO platform device gets connected to the platform bus
> on a machine init done notifier. Only at that point irqfd can be
> setup. An irq connect notifier would be helpful to do that job.
> Instead of adding a new callback at sysbus or qdev
Public bug reported:
I'm running openstack 2012.1 'icehouse' which, ultimately, calls down
into qemu-system-x86 2.0.0+dfsg-2ubuntu1.10.
I primed the process by copying all necessary base images onto the
destination host. Nonetheless, post-migration instances are much larger
than the original ima
On Tue, Apr 28, 2015 at 04:35:16PM +0200, Michael S. Tsirkin wrote:
> On Tue, Apr 28, 2015 at 03:24:19PM +0200, Cornelia Huck wrote:
> > On Tue, 28 Apr 2015 14:16:40 +0100
> > Peter Maydell wrote:
> >
> > > On 28 April 2015 at 14:13, Michael S. Tsirkin wrote:
> > > > The patches look correct to
On 28 April 2015 at 15:59, Kevin Wolf wrote:
> The following changes since commit 84cbd63f87c1d246f51ec8eee5367a5588f367fd:
>
> Merge remote-tracking branch 'remotes/ehabkost/tags/x86-pull-request' into
> staging (2015-04-28 12:22:20 +0100)
>
> are available in the git repository at:
>
>
> gi
Somehow these GPUs manage not to respond to a PCI bus reset, removing
our primary mechanism for resetting graphics cards. The result is
that these devices typically work well for a single VM boot. If the
VM is rebooted or restarted, the guest driver is not able to init the
card from the dirty sta
This is an impossible error path due to the fact that we're reading a
kernel provided, rather than user provided link, which will certainly
always fit in PATH_MAX. Currently it returns a fixed 26 char path
plus %d group number, which typically maxes out at double digits.
However, the caller of the
In an analysis by Laszlo, the resulting type of our calculation for
the end of the MSI-X table, and thus the start of memory after the
table, is uint32_t. We're therefore not correctly preventing the
corner case overflow that we intended to fix here where a BAR >=4G
could place the MSI-X table to
The following changes since commit 84cbd63f87c1d246f51ec8eee5367a5588f367fd:
Merge remote-tracking branch 'remotes/ehabkost/tags/x86-pull-request' into
staging (2015-04-28 12:22:20 +0100)
are available in the git repository at:
git://github.com/awilliam/qemu-vfio.git tags/vfio-update-20150
On 04/28/2015 08:02 AM, Markus Armbruster wrote:
>> v6 changes are noted in each patch; in particular, several new
>> patches were added (additional tests, split some patches, conversion
>> to 'struct' instead of 'type'). But most of the changes were in
>> direct response to review comments or reb
On Tue, Apr 28, 2015 at 11:06:37 +0200, Paolo Bonzini wrote:
> On 27/04/2015 19:06, Emilio G. Cota wrote:
> > Note that I'm running with -smp 1. My guess is that the iothread
> > is starved, since patch 472f4003 "Drop global lock during TCG code
> > execution"
> > removes from the iothread the abi
On 04/28/2015 02:41 AM, Markus Armbruster wrote:
> Eric Blake writes:
>
>> Previous patches have led up to the point where I create the
>> new meta-type "'alternate':'Foo'". See the previous patches
>> for documentation; I intentionally split as much work into
>> earlier patches to minimize the
On 04/28/2015 07:00 AM, Markus Armbruster wrote:
> Eric Blake writes:
>
>> A future patch will be using a 'name':{dictionary} entry in the
>> QAPI schema to specify a default value for an optional argument;
>> but existing use of inline nested structs conflicts with that goal.
>>
>> +++ b/tests/
On 04/28/2015 09:00 AM, Kevin Wolf wrote:
> From: Stefan Hajnoczi
>
> The block.c file has grown to over 6000 lines. It is time to split this
> file so there are fewer conflicts and the code is easier to maintain.
>
> Extract I/O request processing code:
> * Read
> * Write
> * Zero writes an
The arm_gic_kvm now calls kvm_irqchip_set_qemuirq_gsi to build
the hash table storing qemu_irq/gsi mappings. From that point on
irqfd can be setup directly from the qemu_irq using
kvm_irqchip_add_irqfd_notifier.
Signed-off-by: Eric Auger
---
v2 -> v3:
- kvm_irqchip_add_qemuirq_irqfd_notifier re
Check callback now takes as third argument an Object * const*. In
object_set_link_property, we pass the property child as argument.
We also assign the *child before the check call so that enhanced
check can be performed in the callback. In case the check fails,
the old value is restored and ref cou
VFIO platform device needs to setup irqfd but it does not know the
gsi corresponding to the device qemu_irq. This series proposes to
store a hash table in kvm_state using the qemu_irq as key and the gsi
as a value.
kvm_irqchip_set_qemuirq_gsi allows to insert such a pair. The interrupt
controller
Anticipating for the introduction of new add/remove functions taking
a qemu_irq parameter, let's rename existing ones with a gsi suffix.
Signed-off-by: Eric Auger
---
hw/s390x/virtio-ccw.c | 8
hw/vfio/pci.c | 6 +++---
hw/virtio/virtio-pci.c | 4 ++--
include/sysemu/kvm.h |
qdev_init_gpio_out_named takes a new argument corresponding to the
check callback passed to object_property_add_link. In qdev_init_gpio_out
and sysbus_init_irq, this callback is currently set to the dummy
object_property_allow_set_link.
This will allow qdev_init_gpio_out_named callers to specializ
Add a new callback in the SysBusDeviceClass. This callback now can
be overriden by devices inheriting from sysbus. By default the callback
is set to the dummy object_property_allow_set_link callback.
Signed-off-by: Eric Auger
---
v1 -> v2:
- use new LinkPropertySetter type
---
hw/core/sysbus.c
This patch adds the code requested to assign interrupts to
a guest. The interrupts are mediated through user handled
eventfds only.
Signed-off-by: Eric Auger
---
v12 -> v13:
- start user-side eventfd handling at realize time
- remove start_irq_fn
v10 -> v11:
- use block declaration when possibl
This patch aims at optimizing IRQ handling using irqfd framework.
Instead of handling the eventfds on user-side they are handled on
kernel side using
- the KVM irqfd framework,
- the VFIO driver virqfd framework.
the virtual IRQ completion is trapped at interrupt controller
This removes the need
The platform device class has become abstract. This patch introduces
a calxeda xgmac device that derives from it.
Signed-off-by: Eric Auger
Reviewed-by: Alex Bennee
---
v10 -> v11:
- add Alex Reviewed-by
- move virt modifications in a separate patch
v8 -> v9:
- renamed calxeda_xgmac.c into calx
This patch allows the instantiation of the vfio-calxeda-xgmac device
from the QEMU command line (-device vfio-calxeda-xgmac,host="").
A specialized device tree node is created for the guest, containing
compat, dma-coherent, reg and interrupts properties.
Signed-off-by: Eric Auger
---
v12 -> v13
This includes, among other things, VFIO platform driver and
irqfd/arm.
Signed-off-by: Eric Auger
---
v12 -> v13:
- update for 4.1-rc0 headers
v10 -> v11:
- only includes header modifications related to vfio platform
driver v14 and not those related to
"vfio: type1: support for ARM SMMUS wi
This series aims at enabling KVM platform device passthrough.
Kernel dependencies were pulled for 4.1-rc0. This series now only
relies on the following QEMU series
[1] [PATCH v12 0/4] machvirt dynamic sysbus device instantiation
http://comments.gmane.org/gmane.comp.emulators.kvm.arm.devel/886
Bo
Minimal VFIO platform implementation supporting register space
user mapping but not IRQ assignment.
Signed-off-by: Kim Phillips
Signed-off-by: Eric Auger
---
v12 -> v13:
- check device name does not contain any /
- handle case where readlink fully fills the buffer
- in vfio_map_region declare s
On 04/28/2015 12:17 PM, Kashyap Chamarthy wrote:
On Thu, Apr 23, 2015 at 06:23:31PM +0200, Kashyap Chamarthy wrote:
On Thu, Apr 23, 2015 at 10:34:57AM -0400, John Snow wrote:
The qmp-shell is a little rudimentary, but it can be hacked
to give us some transactional support without too much dif
On Thu, Apr 23, 2015 at 06:23:31PM +0200, Kashyap Chamarthy wrote:
> On Thu, Apr 23, 2015 at 10:34:57AM -0400, John Snow wrote:
> > The qmp-shell is a little rudimentary, but it can be hacked
> > to give us some transactional support without too much difficulty.
> >
> > (1) Prep.
> > (2) Add suppo
Hi James,
On 28/04/2015 14:35, James Hogan wrote:
>
>
> On 28/04/15 13:41, Leon Alrae wrote:
>> Update tlb->PFN to contain PFN concatenated with PFNX. PFNX is 0 if large
>> physical address is not supported.
>>
>> Signed-off-by: Leon Alrae
>> ---
>> target-mips/op_helper.c | 32 +++
On 28 April 2015 at 15:40, Paolo Bonzini wrote:
> The following changes since commit e1a5476354d396773e4c555f126d752d4ae58fa9:
>
> Open 2.4 development tree (2015-04-25 22:05:07 +0100)
>
> are available in the git repository at:
>
> git://github.com/bonzini/qemu.git tags/for-upstream
>
> for y
On Tue, Apr 28, 2015 at 12:50:07PM +0200, Thomas Huth wrote:
> The error reporting in pci_nic_init() is quite erratic: Some errors
> are printed directly with error_report(), and some are passed back
> to the caller pci_nic_init_nofail() via an Error pointer.
> Since pci_nic_init() is only used by
On Tue, Apr 28, 2015 at 05:13:10PM +0200, Igor Mammedov wrote:
> > Here I need to set the value of buffer to 1 or 0, we could
> > createbitfield, but if we want to set the value to non one or zero and
> > the BufferSize is large, how could we do? CreateByteField? It's a little
> > complex for user.
1 - 100 of 373 matches
Mail list logo
