I think I have found the reason:
There's a cache in qemu that accelerates the transform of virtual LBA to
cluster offset of qcow2 image.
The cache has a fixed size of 16x8192=128k in my configuration, which
corresponds to a 8GB (128K*64KB)
mapping size. So when the "working set" of fio exceeds 8G
Hi,
On Wed, Jun 18, 2014 at 07:19:17PM +0300, Michael S. Tsirkin wrote:
> From: Nikolay Nikolaev
>
> chardev depends on lots of external symbols that are not necessarily
> needed to be able to use, for example, 'socket chardev'. So add stubs
> for these functions:
>
> - bdrv_commit_all
> - qe
The Monday 23 Jun 2014 à 12:32:30 (+0800), Fam Zheng wrote :
> On Sat, 06/21 17:40, Benoît Canet wrote:
> > The Saturday 21 Jun 2014 à 17:39:11 (+0200), Benoît Canet wrote :
> > > We still have the issue of unlocking the bottom BDS when a subtree is
> > > detached
> > > from the graphs by a swap.
On Sat, 2014-06-14 at 20:19 +0100, Christian Burger wrote:
> Guest mouse pointer was jumpy, when moving host mouse in the vertical
> direction (see bug #1327800).
Ah, I've just done a deep dive into qemu input code to debug that
one as well :-)
It's not just "jumpy", it goes the wrong way around
On Sat, 06/21 17:40, Benoît Canet wrote:
> The Saturday 21 Jun 2014 à 17:39:11 (+0200), Benoît Canet wrote :
> > We still have the issue of unlocking the bottom BDS when a subtree is
> > detached
> > from the graphs by a swap. (It does happen in my drive-mirror arbitrary node
> > replacement serie
Hi,
Ok, your used method as the same as ivshmem,
so actually your problem is “how to write driver for ivshmem in windows guests
?”
but IMHO no one has realized this.
Best regards,
-Gonglei
From: lc [mailto:nighto...@163.com]
Sent: Monday, June 23, 2014 11:45 AM
To: Gonglei (Arei)
Cc: qemu
I want to use the memory mapping methods, so I made a memory pci devices, but I
know very little about Windows driver programming, someone has done such a
thing。
static TypeInfo mem_pci_info = {
.name = "mem_pci",
.parent = TYPE_PCI_DEVICE,
.instance_size = sizeof(PCIMEMPCIState),
>>> On 6/20/2014 at 08:08 PM, in message
, Stefano
Stabellini wrote:
> On Fri, 20 Jun 2014, Chunyan Liu wrote:
> > qemu side patch to support xen HVM direct kernel boot:
> > if -kernel exists, calls xen_load_linux(), which will read kernel/initrd
> > and add a linuxboot.bin or multiboot.bin
On Mon, 2014-06-23 at 10:45 +1000, Benjamin Herrenschmidt wrote:
>
> Another option would be to use an unused bit if the ENABLE register
> and simply force it to 1 on reads when the endian control register
> is present... though in this case it might be a better idea to use
> that bit to indicate
Cc'ing more qcow2 experts.
On Mon, 06/23 11:14, lihuiba wrote:
> >Did you prefill the image? Amplification could come from cluster allocation.
> Yes!
> I forgot to mention that I created the qcow2 image with
> 'preallocation=metadata', and I have allocated
> the data blocks with dd in VM.
>
>
Hi,
Why not configure Nic for VM to transfer those pictures ?
Best regards,
-Gonglei
From: qemu-devel-bounces+arei.gonglei=huawei@nongnu.org
[mailto:qemu-devel-bounces+arei.gonglei=huawei@nongnu.org] On Behalf Of lc
Sent: Monday, June 23, 2014 9:32 AM
To: qemu-devel@nongnu.org
Subje
>Did you prefill the image? Amplification could come from cluster allocation.
Yes!
I forgot to mention that I created the qcow2 image with
'preallocation=metadata', and I have allocated
the data blocks with dd in VM.
Creating image in host:
qemu-img create -f qcow2 -preallocation=metadata test.
On Mon, 06/23 10:06, lihuiba wrote:
> Hi, all
>
>
> I'm using a qcow2 image stored on a SSD RAID1 (2 x intel S3500), and I'm
> benchmarking the
> system using fio. Although the throughput in VM (with KVM and virtio enabled)
> is acceptable (67%
> of thoughtput in host), the IOPS performance see
Hi, all
I'm using a qcow2 image stored on a SSD RAID1 (2 x intel S3500), and I'm
benchmarking the
system using fio. Although the throughput in VM (with KVM and virtio enabled)
is acceptable (67%
of thoughtput in host), the IOPS performance seems is extremely low only
2% of IOPS in host.
The patch implements sPAPRPHBClass::format_errinjct_cmd to do the
address translation (BUID+PE number to IOMMU group ID) and then
come up with the formatted string for PCI error injection.
Signed-off-by: Gavin Shan
---
hw/ppc/spapr_pci_vfio.c | 19 +++
1 file changed, 19 insertio
The patch implements PCI error injection RTAS calls for sPAPR
platform, which are defined PAPR spec as follows. Those RTAS
calls are expected to be invoked in strict sequence of
"ibm,open-errinjct", "ibm,errinjct", "ibm,close-errinjct".
- "ibm,open-errinjct": Apply for token to do error injection
There is one guest userland utility "errinjct" used to inject various types
of errors for testing purpose. The utility works with 3 RTAS calls, which
are defined in PAPR spec as follows:
- "ibm,open-errinjct": Apply for token to do error injection
- "ibm,close-errinjct": Free the token assigned pr
The patch exports RTAS property "ibm,errinjct-tokens", which is
defined in PAPR spec and used to indicate various error types
we can inject.
Signed-off-by: Gavin Shan
---
hw/ppc/spapr.c | 19 +++
1 file changed, 19 insertions(+)
diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
index
Ping
On Tue, Jun 17, 2014 at 3:57 PM, Alistair Francis
wrote:
> Add Stefan Hajnoczi as it relies on his work
> (http://lists.gnu.org/archive/html/qemu-devel/2014-05/msg06489.html)
>
> On Tue, Jun 17, 2014 at 2:18 PM, Alistair Francis
> wrote:
>> This patch adds the Cortex-A9 ARM CPU to the A9MPC
hi,everyone
I want to copy a 3M picture from the guest to the host,which method i can use,i
think channel is too slow,because i want sent 30 pieces per second.
thanks.
Ping...
> -Original Message-
> From: Yang, Zhiyong/杨 志勇
> Sent: Wednesday, June 11, 2014 10:43 PM
> To: qemu-devel@nongnu.org
> Cc: Yang, Zhiyong/杨 志勇; peter.mayd...@linaro.org;
> mdr...@linux.vnet.ibm.com; jfor...@redhat.com
> Subject: [Qemu-devel] [PATCH v1] trace: add
> qemu_system_powe
On Sun, 2014-06-22 at 12:10 +1000, Benjamin Herrenschmidt wrote:
> On Sat, 2014-06-21 at 15:37 +1000, Benjamin Herrenschmidt wrote:
> > On Thu, 2014-06-19 at 11:36 +0200, Gerd Hoffmann wrote:
> > > If not -- then I prefer to play save and not use a vbe register to avoid
> > > ending up with two inc
On Mon, 2014-06-16 at 20:30 +1000, Benjamin Herrenschmidt wrote:
> Hi !
>
> So while trying to solve an issue we have with qemu/kvm and powerpc who
> can be both LE and BE nowadays, we thought the ideal solution would be
> to have a register in the emulated VGA to control the endian.
No reply ..
On Mon, 2014-06-16 at 20:30 +1000, Benjamin Herrenschmidt wrote:
> Hi !
>
> So while trying to solve an issue we have with qemu/kvm and powerpc who
> can be both LE and BE nowadays, we thought the ideal solution would be
> to have a register in the emulated VGA to control the endian.
No reply ...
Thanks Richard for your feedback. I am going to correct the patch and
resubmit it.
Best regards,
Jaume
On Sun, Jun 22, 2014 at 8:55 PM, Richard Henderson wrote:
> On 06/22/2014 07:55 AM, Jaume Martí wrote:
>> -cpu_x86_fsave(env, fpstate_addr, 1);
>> -fpstate->status = fpstate->sw
On 06/22/2014 07:55 AM, Jaume Martí wrote:
> -cpu_x86_fsave(env, fpstate_addr, 1);
> -fpstate->status = fpstate->sw;
> -magic = 0x;
> +cpu_x86_fsave(env, fpstate_addr);
> +fpstate->status = fpstate->sw;
> +magic = 0x;
This patch needs to be split into fo
On Fri, Jun 13, 2014 at 10:15:00AM +0200, Paolo Bonzini wrote:
> libqtest is using g_strdup_printf to format QMP commands, but
> this does not work if the argument strings need to be escaped.
> Instead, use the fancy %-formatting functionality of QObject.
> The only change required in tests is that
It can be useful to read simpletrace files that have no header. For
example, a ring buffer may not have a header record but can still be
processed if the user is sure the file format version is compatible.
$ scripts/simpletrace.py --no-header trace-events trace-file
Signed-off-by: Stefan Hajno
SystemTap reserved words sometimes conflict with QEMU variable names.
We escape them to prevent conflicts.
Move escaping into its own function so the next patch can reuse it.
Signed-off-by: Stefan Hajnoczi
---
scripts/tracetool/format/stap.py | 11 ---
1 file changed, 8 insertions(+), 3
The simpletrace SystemTap tapset outputs simpletrace binary traces for
SystemTap probes. This is useful because SystemTap has no default way
to format or store traces. The simpletrace SystemTap tapset provides an
easy way to store traces.
The simpletrace.py tool or custom Python scripts using th
This new tracetool "format" generates a SystemTap .stp file that outputs
simpletrace binary trace data.
In contrast to simpletrace or ftrace, SystemTap does not define its own
trace format. All output from SystemTap is generated by .stp files.
This patch lets us generate a .stp file that outputs
v2:
* I realized that v1 was not complete enough after feedback from Lluís and
Frank Ch. Eigler, so here is a v2 after all.
* Add Makefile target for simpletrace .stp file [Lluís]
* Generate SystemTap probe aliases so users can select a subset of probes to
enable [Frank Ch. Eigler]
* Dele
On Sun, 2014-06-22 at 14:12 +0300, Michael S. Tsirkin wrote:
> On Sun, Jun 22, 2014 at 02:11:05PM +0300, Marcel Apfelbaum wrote:
> > On Sun, 2014-06-22 at 14:03 +0300, Michael S. Tsirkin wrote:
> > > On Sun, Jun 22, 2014 at 01:54:06PM +0300, Marcel Apfelbaum wrote:
> > > > On Thu, 2014-06-19 at 17:
On Sun, 2014-06-22 at 14:11 +0300, Michael S. Tsirkin wrote:
> On Sun, Jun 22, 2014 at 01:52:46PM +0300, Michael S. Tsirkin wrote:
> > On Sun, Jun 22, 2014 at 01:47:24PM +0300, Marcel Apfelbaum wrote:
> > > On Thu, 2014-06-19 at 17:39 +0300, Michael S. Tsirkin wrote:
> > > > On Thu, Jun 19, 2014 at
On Sun, Jun 22, 2014 at 02:11:05PM +0300, Marcel Apfelbaum wrote:
> On Sun, 2014-06-22 at 14:03 +0300, Michael S. Tsirkin wrote:
> > On Sun, Jun 22, 2014 at 01:54:06PM +0300, Marcel Apfelbaum wrote:
> > > On Thu, 2014-06-19 at 17:43 +0300, Michael S. Tsirkin wrote:
> > > > On Thu, Jun 19, 2014 at 0
On Sun, 2014-06-22 at 14:03 +0300, Michael S. Tsirkin wrote:
> On Sun, Jun 22, 2014 at 01:54:06PM +0300, Marcel Apfelbaum wrote:
> > On Thu, 2014-06-19 at 17:43 +0300, Michael S. Tsirkin wrote:
> > > On Thu, Jun 19, 2014 at 04:52:21PM +0300, Marcel Apfelbaum wrote:
> > > > Hotplug triggers both 'pr
On Sun, Jun 22, 2014 at 01:52:46PM +0300, Michael S. Tsirkin wrote:
> On Sun, Jun 22, 2014 at 01:47:24PM +0300, Marcel Apfelbaum wrote:
> > On Thu, 2014-06-19 at 17:39 +0300, Michael S. Tsirkin wrote:
> > > On Thu, Jun 19, 2014 at 04:52:20PM +0300, Marcel Apfelbaum wrote:
> > > > It is needed by ho
On Sun, Jun 22, 2014 at 01:54:06PM +0300, Marcel Apfelbaum wrote:
> On Thu, 2014-06-19 at 17:43 +0300, Michael S. Tsirkin wrote:
> > On Thu, Jun 19, 2014 at 04:52:21PM +0300, Marcel Apfelbaum wrote:
> > > Hotplug triggers both 'present detect change' and
> > > 'attention button pressed'.
> > >
> >
On Tue, Jun 17, 2014 at 07:48:36PM +0300, Michael S. Tsirkin wrote:
> On Tue, Jun 17, 2014 at 12:47:51PM -0300, Marcelo Tosatti wrote:
> >
> > It is necessary to reset RTC interrupt reinjection backlog if
> > guest time is synchronized via a different mechanism, such as
> > QGA's guest-set-time co
On Thu, 2014-06-19 at 17:43 +0300, Michael S. Tsirkin wrote:
> On Thu, Jun 19, 2014 at 04:52:21PM +0300, Marcel Apfelbaum wrote:
> > Hotplug triggers both 'present detect change' and
> > 'attention button pressed'.
> >
> > Hotunplug starts by triggering 'attention button pressed',
> > then waits f
On Sun, Jun 22, 2014 at 10:38:35AM +0800, Stefan Hajnoczi wrote:
> The recent changes to extend qemu-char get_msgfds to support multiple file
> descriptors caused a qemu-iotests regression and also introduced a file
> descriptor leak. This patch series fixes the bugs.
Applied, thanks!
> Stefan H
On Sun, Jun 22, 2014 at 01:47:24PM +0300, Marcel Apfelbaum wrote:
> On Thu, 2014-06-19 at 17:39 +0300, Michael S. Tsirkin wrote:
> > On Thu, Jun 19, 2014 at 04:52:20PM +0300, Marcel Apfelbaum wrote:
> > > It is needed by hot-unplug in order to get an indication
> > > from the OS when the device can
MIPS requires the pad field to 64b-align the data field just as ARM
does.
Signed-off-by: Paul Burton
---
Changes in v3:
- None.
Changes in v2:
- Apply the same padding for the mips64 target.
---
linux-user/syscall_defs.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/li
On Thu, 2014-06-19 at 17:39 +0300, Michael S. Tsirkin wrote:
> On Thu, Jun 19, 2014 at 04:52:20PM +0300, Marcel Apfelbaum wrote:
> > It is needed by hot-unplug in order to get an indication
> > from the OS when the device can be physically detached.
> >
> > Signed-off-by: Marcel Apfelbaum
> > ---
Add a definition of the KDSIGACCEPT ioctl & allow its use by target
programs.
Signed-off-by: Paul Burton
---
Changes in v3:
- Translate signal number to host value.
Changes in v2:
- None.
---
linux-user/ioctls.h | 1 +
linux-user/syscall.c | 7 +++
linux-user/syscall_defs.h |
Add support for the unshare syscall, trivially passed through to the
host.
Signed-off-by: Paul Burton
---
Changes in v3:
- None.
Changes in v2:
- None.
---
linux-user/syscall.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/linux-user/syscall.c b/linux-user/syscall.c
index b630ef
Adds support for the timerfd_create, timerfd_gettime & timerfd_settime
syscalls, allowing use of timerfds by target programs.
Signed-off-by: Paul Burton
---
Changes in v3:
- Fix coding style, checkpatch clean.
Changes in v2:
- None.
---
linux-user/strace.list | 9 +
linux-user/sysc
Implement support for the name_to_handle_at and open_by_handle_at
syscalls, allowing their use by the target program.
Signed-off-by: Paul Burton
---
Changes in v3:
- Fix coding style, checkpatch clean.
Changes in v2:
- None.
---
linux-user/strace.c| 30 ++
linux-
The settimeofday syscall accepts a tz argument indicating the desired
timezone to the kernel. QEMU previously ignored any argument provided
by the target program & always passed NULL to the kernel. Instead,
translate the argument & pass along the data userland provided.
Although this argument is d
Add the epoll_create1 syscall to strace.list in order to display that
syscall when it occurs, rather than a message about the syscall being
unknown despite QEMU already implementing support for it.
Signed-off-by: Paul Burton
---
Changes in v3:
- None.
Changes in v2:
- None.
---
linux-user/s
Calls to the mount syscall can legitimately provide NULL as the value
for the source of filesystemtype arguments, which QEMU would previously
reject & return -EFAULT to the target program. An example of this is
remounting an already mounted filesystem with different properties.
Instead of rejectin
Add support for the setns syscall, trivially passed through to the host.
Signed-off-by: Paul Burton
---
Changes in v3:
- None.
Changes in v2:
- None.
---
linux-user/strace.list | 3 +++
linux-user/syscall.c | 6 ++
2 files changed, 9 insertions(+)
diff --git a/linux-user/strace.list
Add a definition of the SIOCGIFINDEX ioctl, allowing its use by target
programs.
Signed-off-by: Paul Burton
---
Changes in v3:
- None.
Changes in v2:
- None.
---
linux-user/ioctls.h | 1 +
linux-user/syscall_defs.h | 1 +
2 files changed, 2 insertions(+)
diff --git a/linux-user/ioctl
Add support for the ioprio_get & ioprio_set syscalls, allowing their
use by target programs.
Signed-off-by: Paul Burton
---
Changes in v3:
- None.
Changes in v2:
- Declare the host syscalls conditionally upon whether both host &
target define their numbers, to avoid compile time warnings
Translate the SO_SNDBUFFORCE & SO_RCVBUFFORCE options to setsockopt to
the host values & perform the syscall as expected, allowing use of those
options by target programs.
Signed-off-by: Paul Burton
---
Changes in v3:
- Fix coding style, checkpatch clean.
Changes in v2:
- None.
---
linux-us
This series fixes a number of bugs in QEMUs linux-user support, some
specific to targetting the MIPS architecture but mostly generic. It also
adds support for some previously unsupported syscalls & {g,s}etsockopt
options.
Paul Burton (16):
linux-user: translate the result of getsockopt SO_TYPE
The tv argument to the settimeofday syscall is allowed to be NULL, if
the program only wishes to provide the timezone. QEMU previously
returned -EFAULT when tv was NULL. Instead, execute the syscall &
provide NULL to the kernel as the target program expected.
Signed-off-by: Paul Burton
---
Change
Translate the SO_PASSSEC option to setsockopt to the host value &
perform the syscall as expected, allowing use of the option by target
programs.
Signed-off-by: Paul Burton
---
Changes in v3:
- None.
Changes in v2:
- Fix the value of TARGET_SO_PASSSEC for the sparc target.
---
linux-user/so
Translate the SO_ACCEPTCONN option to the host value & execute the
syscall as expected.
Signed-off-by: Paul Burton
---
Changes in v3:
- None.
Changes in v2:
- None.
---
linux-user/syscall.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/linux-user/syscall.c b/linux-user/syscall.c
in
QEMU previously passed the result of the host syscall directly to the
target program. This is a problem if the host & target have different
representations of socket types, as is the case when running a MIPS
target program on an x86 host. Introduce a host_to_target_sock_type
helper function mirrori
On Sun, Jun 22, 2014 at 12:18:02AM +0100, Peter Maydell wrote:
> > +if (!lock_user_struct(VERIFY_READ, target_tz, target_tz_addr, 1))
> > +return -TARGET_EFAULT;
>
> Coding style mandates braces even on single-line if()s; checkpatch.pl
> will catch this usually.
I copied that style fr
On Fri, Jun 20, 2014 at 09:40:31AM -0600, Eric Blake wrote:
> On 06/19/2014 11:55 PM, Hu Tao wrote:
> > Signed-off-by: Hu Tao
> > ---
> > tests/test-string-output-visitor.c | 109
> > ++---
> > 1 file changed, 90 insertions(+), 19 deletions(-)
> >
>
> >
> > +
On Sun, Jun 22, 2014 at 12:13:27AM +0100, Peter Maydell wrote:
> > diff --git a/linux-user/ioctls.h b/linux-user/ioctls.h
> > index 309fb21..cd21e64 100644
> > --- a/linux-user/ioctls.h
> > +++ b/linux-user/ioctls.h
> > @@ -64,6 +64,7 @@
> > IOCTL(KDSKBLED, 0, TYPE_INT)
> > IOCTL(KDGETL
Ping!
On Tue, 17 Jun 2014, BALATON Zoltan wrote:
On Tue, 17 Jun 2014, Alexander Graf wrote:
On 17.06.14 11:36, BALATON Zoltan wrote:
On Tue, 17 Jun 2014, Alexander Graf wrote:
On 12.04.14 11:20, BALATON Zoltan wrote:
Bring the memory map closer to a PowerMac3,1 model by removing unused
area
On 2014/6/20 20:48, Paolo Bonzini wrote:
Il 19/06/2014 11:53, Tiejun Chen ha scritto:
so this mean that isa bridge is still represented with Dev31:Func0
like the native OS. Furthermore, currently we're pushing VGA
passthrough support into qemu upstream, and with some discussion,
we wouldn't set
On 2014/6/20 20:32, Daniel Vetter wrote:
Well I have no clue about forwarding the intel gpu to virtualized
hosts and also no idea who could review this really. There's been a
bit a discussion around the iommu mapping forwarding and similar
No, this doesn't affect IOMMU mapping.
topics though.
On Fri, Jun 20, 2014 at 01:55:40PM +0800, Hu Tao wrote:
> Hi Michael,
>
> This series is on top of your pci tree.
>
> This series includes two fixups and one patch for adding test of human
> format of string output visitor, please review. thanks!
Please note we can't generally send fixups to pat
Use common rule (macro) to install and strip binaries, and use
it in all places where we install binaries, instead of fixing
bugs like 1319493 in every place.
(This fixes https://bugs.launchpad.net/bugs/1319493)
Signed-off-by: Michael Tokarev
---
Makefile| 12 ++--
Makefile.tar
68 matches
Mail list logo
