On 2014/8/18 20:27, Dr. David Alan Gilbert wrote:
* zhanghailiang (zhang.zhanghaili...@huawei.com) wrote:
For all NICs(except virtio-net) emulated by qemu,
Such as e1000, rtl8139, pcnet and ne2k_pci,
Qemu can still receive packets when VM is not running.
If this happened in *migration's* last PA
On 2014/8/18 19:42, Michael Tokarev wrote:
18.08.2014 12:06, Peter Maydell wrote:
On 18 August 2014 09:00, zhanghailiang wrote:
From: Li Liu
fopen() may return NULL which will cause setbuf() segmentfault
Signed-off-by: zhanghailiang
Signed-off-by: Li Liu
---
block/vvfat.c | 1 +
1 file ch
> >> Subject: Re: [Question] Why doesn't PCIe hotplug work for Q35 machine?
> >>
> >> On Sun, 2014-08-17 at 13:00 +0200, Michael S. Tsirkin wrote:
> >>> On Fri, Aug 15, 2014 at 07:33:29AM +, Gonglei (Arei) wrote:
> Hi,
>
> I noticed that the qemu-2.1 release change log says
> >>>
If a guest reboots during a running migration, changes to the
hash page table are not necessarily updated on the destination.
Opening a new file descriptor to the HTAB forces the migration
handler to resend the entire table.
Signed-off-by: Samuel Mendoza-Jonas
---
Changes in v3: Pointed out by Da
Il 18/08/2014 22:50, Hulin, Patrick - 0559 - MITLL ha scritto:
>> >Correct. Doesn¹t work. Haven¹t fully diagnosed why, but it doesn¹t seem
>> >to ever hit the current_tb_modified passage if you invalidate beforehand.
> Yeah - mem_io_pc doesn¹t get updated until we¹re inside io_write, so
> tb_invali
On 2014-08-19 06:08, Knut Omang wrote:
>> Are you depending on interrupt remapping? If not, my patches are a bit
>> hacky and may cause their own issues if you are unlucky.
>
> It does not depend directly but interprets a NULL PciDevice pointer as
> the special bus number (0xff) for non-pci devic
commit a9f98bb5ebe6fb1869321dcc58e72041ae626ad8 vhost: multiqueue
support changed the order of stopping the device. Previously
vhost_dev_stop would disable backend and only afterwards, unset guest
notifiers. We now unset guest notifiers while vhost is still
active. This can lose interrupts causing
On Tue, 2014-08-19 at 06:08 +0200, Knut Omang wrote:
> On Mon, 2014-08-18 at 20:50 +0200, Jan Kiszka wrote:
> > On 2014-08-18 18:34, Knut Omang wrote:
> > > On Sat, 2014-08-16 at 10:47 +0200, Jan Kiszka wrote:
> > >> On 2014-08-16 10:45, Jan Kiszka wrote:
> > >>> On 2014-08-16 09:54, Knut Omang wro
On Mon, 2014-08-18 at 20:50 +0200, Jan Kiszka wrote:
> On 2014-08-18 18:34, Knut Omang wrote:
> > On Sat, 2014-08-16 at 10:47 +0200, Jan Kiszka wrote:
> >> On 2014-08-16 10:45, Jan Kiszka wrote:
> >>> On 2014-08-16 09:54, Knut Omang wrote:
> On Fri, 2014-08-15 at 19:37 +0800, Le Tan wrote:
> >
Ping http://patchwork.ozlabs.org/patch/379134/
On Mon, Aug 11, 2014 at 1:50 PM, Martin Galvan <
martin.gal...@tallertechnologies.com> wrote:
> When calling qemu_system_reset after startup on a Cortex-M CPU, the
> initial values of PC, MSP and the Thumb bit weren't set correctly. In
> particular,
On 08/19/2014 11:02 AM, Jason Wang wrote:
> commit a9f98bb5ebe6fb1869321dcc58e72041ae626ad8 vhost: multiqueue
> support changed the order of stopping the device. Previously
> vhost_dev_stop would disable backend and only afterwards, unset guest
> notifiers. We now unset guest notifiers while vhost
On 08/18/2014 07:42 PM, William Dauchy wrote:
> Hello,
>
> Using qemu2.1.0, a linux v3.14.X x86_64 as host and a linux v3.12.X x86_64 as
> guest
> I'm starting a VM with these network interfaces:
>
> [netdev "vifA.0"]
> type = "tap"
> vhost = "on"
> ifname = "vifA.0"
> downscript = "no"
>
commit a9f98bb5ebe6fb1869321dcc58e72041ae626ad8 vhost: multiqueue
support changed the order of stopping the device. Previously
vhost_dev_stop would disable backend and only afterwards, unset guest
notifiers. We now unset guest notifiers while vhost is still
active. This can lose interrupts causing
On 08/19/2014 03:53 AM, Michael S. Tsirkin wrote:
> On Mon, Aug 18, 2014 at 05:51:31PM +0800, Jason Wang wrote:
>> > commit a9f98bb5ebe6fb1869321dcc58e72041ae626ad8 vhost: multiqueue
>> > support changed the order of stopping the device. Previously
>> > vhost_dev_stop would disable backend and only
On 08/18/2014 09:20 PM, William Dauchy wrote:
> On Mon, Aug 18, 2014 at 11:51 AM, Jason Wang wrote:
>> err:
>> @@ -254,16 +254,16 @@ void vhost_net_stop(VirtIODevice *dev, NetClientState
>> *ncs,
>> VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(vbus);
>> int i, r;
>>
>> +for (i = 0; i <
On 08/18/2014 08:11 PM, Zhangjie (HZ) wrote:
> On 2014/8/18 17:51, Jason Wang wrote:
>> commit a9f98bb5ebe6fb1869321dcc58e72041ae626ad8 vhost: multiqueue
>> support changed the order of stopping the device. Previously
>> vhost_dev_stop would disable backend and only afterwards, unset guest
>> notif
commit 7537fe04 QMP: QMP/ -> docs/qmp/
Above commit has moved last QMP files to docs/qmp and it's not necessary
to create QMP directory. So remove it from configure.
Signed-off-by: Liming Wang
---
configure | 4
1 file changed, 4 deletions(-)
diff --git a/configure b/configure
index 283c7
If user doesn't specify numa options, nb_numa_nodes will be 0. But
PCDIMMDevice->node
is also initialized to 0. As a result, the following check will fail:
pc_dimm_realize()
{
..
if (dimm->node >= nb_numa_nodes) {
error_setg(errp, "'DIMM property " PC_DIMM_NODE_PRO
Hi Michael, Paolo
Thanks for the advices. Will send a v2 patch soon.
Thanks.
On 08/19/2014 04:04 AM, Michael S. Tsirkin wrote:
On Mon, Aug 18, 2014 at 03:58:33PM +0200, Paolo Bonzini wrote:
Il 18/08/2014 15:56, Michael S. Tsirkin ha scritto:
+/* Initialize PCDIMMDevice->node to -1 so that ev
Hi Igor, Andreas,
Could you please help to review this series? Any comment is welcome.
Regards,
Gu
On 08/07/2014 12:53 PM, Gu Zheng wrote:
> This series is based on the previous patchset from Chen Fan:
> https://lists.nongnu.org/archive/html/qemu-devel/2014-05/msg02360.html
> https://lists.nongnu
On 2014/8/18 17:58, Michael S. Tsirkin wrote:
On Mon, Aug 18, 2014 at 05:01:25PM +0800, Chen, Tiejun wrote:
On 2014/8/18 16:21, Michael S. Tsirkin wrote:
On Mon, Aug 18, 2014 at 11:06:29AM +0800, Chen, Tiejun wrote:
On 2014/8/17 18:32, Michael S. Tsirkin wrote:
On Fri, Aug 15, 2014 at 09:58
Linux should boot in EL2 or EL1. If in EL3, jump down before handing
off to Linux.
Signed-off-by: Peter Crosthwaite
---
hw/arm/boot.c | 21 +
1 file changed, 21 insertions(+)
diff --git a/hw/arm/boot.c b/hw/arm/boot.c
index 840f5da..f1f6365 100644
--- a/hw/arm/boot.c
+++ b/
Allow a fixup to be used to deposit a field in a machine-code
instruction. The option shift and length fixup fields indicate the
field being deposited by the fixup.
Signed-off-by: Peter Crosthwaite
---
hw/arm/boot.c | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/hw/a
Hi Peter, Edgar,
These patches add bootloader support for Edgars upcomming EL2/3 work.
This allows for Linux boot from EL3 without a bootloader or wrapper.
Regards,
Peter
Peter Crosthwaite (2):
arm: boot: Add partial machine code fixup
arm: boot: Add EL jump-down code for Linux
hw/arm/boo
It's very useful when debugging SMP to know who disassembly or a CPU
state dump is being done on behalf of.
Signed-off-by: Peter Crosthwaite
---
target-arm/translate-a64.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/target-arm/translate-a64.c b/target-arm/translate-
On Fri, Aug 15, 2014 at 08:12:27PM +1000, Alexey Kardashevskiy wrote:
> We are going to have multiple DMA windows per PHB and we want them to
> migrate so we need a predictable way of assigning LIOBNs.
>
> This introduces a macro which makes up a LIOBN from fixed prefix,
> PHB index (unique PHB id
On Fri, Aug 15, 2014 at 08:12:24PM +1000, Alexey Kardashevskiy wrote:
> The existing KVM_CREATE_SPAPR_TCE ioctl only support 4G windows max.
> We are going to add huge DMA windows support so this will create small
> window and unexpectedly fail later.
>
> This disables KVM_CREATE_SPAPR_TCE for win
On Fri, Aug 15, 2014 at 08:12:23PM +1000, Alexey Kardashevskiy wrote:
> Current object_child_foreach() uses QTAILQ_FOREACH() to walk
> through children and that makes children removal from the callback
> impossible.
>
> This makes object_child_foreach() use QTAILQ_FOREACH_SAFE().
>
> Signed-off-b
From: Tyrel Datwyler
This uses extension of existing EPOW interrupt/event mechanism
to notify userspace tools like librtas/drmgr to handle
in-guest configuration/cleanup operations in response to
device_add/device_del.
Userspace tools that don't implement this extension will need
to be run manua
Signed-off-by: Michael Roth
---
hw/ppc/spapr.c | 23 +++
include/hw/ppc/spapr.h | 1 +
2 files changed, 24 insertions(+)
diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
index 90b25b3..39cb0bb 100644
--- a/hw/ppc/spapr.c
+++ b/hw/ppc/spapr.c
@@ -309,6 +309,29 @@ sPAPRDrc
From: Nathan Fontenot
This add entries to the root OF node to advertise our PHBs as being
DR-capable in according with PAPR specification.
Each PHB is given a name of PHB, advertised as a PHB type,
and associated with a power domain of -1 (indicating to guests that
power management is handled au
From: Nathan Fontenot
This extends the data structures currently used to report EPOW events to
gets via the check-exception RTAS interfaces to also include event types
for hotplug/unplug events.
This is currently undocumented and being finalized for inclusion in PAPR
specification, but we implem
From: Nathan Fontenot
Signed-off-by: Nathan Fontenot
Signed-off-by: Michael Roth
---
hw/ppc/spapr_pci.c | 25 +
1 file changed, 25 insertions(+)
diff --git a/hw/ppc/spapr_pci.c b/hw/ppc/spapr_pci.c
index 23a3477..f007dd6 100644
--- a/hw/ppc/spapr_pci.c
+++ b/hw/ppc/spa
Some kernels program a 0 address for io regions. PCI 3.0 spec
section 6.2.5.1 doesn't seem to disallow this.
Signed-off-by: Michael Roth
---
hw/pci/pci.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/pci/pci.c b/hw/pci/pci.c
index 351d320..9578749 100644
--- a/hw/pci/pci
From: Tyrel Datwyler
We don't actually rely on this interface to surface hotplug events, and
instead rely on the similar-but-interrupt-driven check-exception RTAS
interface used for EPOW events. However, the existence of this interface
is needed to ensure guest kernels initialize the event-report
From: Mike Day
Signed-off-by: Mike Day
Signed-off-by: Michael Roth
---
hw/ppc/spapr_pci.c | 76 ++
1 file changed, 76 insertions(+)
diff --git a/hw/ppc/spapr_pci.c b/hw/ppc/spapr_pci.c
index f007dd6..8d1351d 100644
--- a/hw/ppc/spapr_pci.c
+
This enables hotplug for PHB bridges. Upon hotplug we generate the
OF-nodes required by PAPR specification and IEEE 1275-1994
"PCI Bus Binding to Open Firmware" for the device.
We associate the corresponding FDT for these nodes with the DrcEntry
corresponding to the slot, which will be fetched via
Signed-off-by: Michael Roth
---
hw/ppc/spapr_pci.c | 111 +
1 file changed, 111 insertions(+)
diff --git a/hw/ppc/spapr_pci.c b/hw/ppc/spapr_pci.c
index 8d1351d..96a57be 100644
--- a/hw/ppc/spapr_pci.c
+++ b/hw/ppc/spapr_pci.c
@@ -606,6 +606,11
From: Mike Day
Signed-off-by: Mike Day
Signed-off-by: Michael Roth
---
hw/ppc/spapr_pci.c | 119 +
include/hw/ppc/spapr.h | 3 ++
2 files changed, 122 insertions(+)
diff --git a/hw/ppc/spapr_pci.c b/hw/ppc/spapr_pci.c
index 924d488..23a347
These patches are based on ppc-next, and can also be obtained from:
https://github.com/mdroth/qemu/commits/spapr-pci-hotplug-v3-ppc-next
v3:
* dropped emulation of firmware-managed BAR allocation. this will be
introduced via a follow-up series via a -machine flag and tied to
a separate hot
Reserve 32 entries of type PCI in each PHB's initial FDT. This
advertises to guests that each PHB is DR-capable device with
physical hotpluggable slots. This is necessary for allowing
hotplugging of devices to it later via bus rescan or guest rpaphp
hotplug module.
Each entry is assigned a name of
On 08/18/2014 05:11 PM, Peter Maydell wrote:
> On 18 August 2014 22:54, Joel Schopp wrote:
>> +static void vfio_fdt_add_device_node(SysBusDevice *sbdev, void *opaque)
>> +{
>> +PlatformDevtreeData *data = opaque;
>> +void *fdt = data->fdt;
>> +const char *parent_node = data->node;
>>
On 18 August 2014 22:54, Joel Schopp wrote:
>
> +static void vfio_fdt_add_device_node(SysBusDevice *sbdev, void *opaque)
> +{
> +PlatformDevtreeData *data = opaque;
> +void *fdt = data->fdt;
> +const char *parent_node = data->node;
> +int compat_str_len;
> +char *nodename;
> +
Signed-off-by: John Snow
---
hw/i386/pc_piix.c | 2 +-
hw/ide/core.c | 11 +++
include/hw/ide.h | 3 ++-
3 files changed, 10 insertions(+), 6 deletions(-)
diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c
index 47ac1b5..9da6f0e 100644
--- a/hw/i386/pc_piix.c
+++ b/hw/i386/pc_piix
Signed-off-by: John Snow
---
blockdev.c| 9 ++---
include/sysemu/blockdev.h | 2 +-
2 files changed, 7 insertions(+), 4 deletions(-)
diff --git a/blockdev.c b/blockdev.c
index 58da77f..a9efe1f 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -57,6 +57,7 @@ static const char *const
Currently, the drive definitions created by drive_new() when using
the -drive file=...[,if=ide] or -cdrom or -hd[abcd] options are not
picked up by the Q35 initialization routine.
To fix this, we have to add hooks to search for these drives using
something like pc_piix's ide_drive_get and then add
Signed-off-by: John Snow
---
hw/i386/pc_q35.c | 4
hw/ide/ahci.c| 17 +
hw/ide/ahci.h| 2 ++
3 files changed, 23 insertions(+)
diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c
index 4b5a274..4613565 100644
--- a/hw/i386/pc_q35.c
+++ b/hw/i386/pc_q35.c
@@ -86,6 +86,
Signed-off-by: John Snow
---
blockdev.c| 9 +
include/sysemu/blockdev.h | 1 +
2 files changed, 10 insertions(+)
diff --git a/blockdev.c b/blockdev.c
index 48bd9a3..58da77f 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -110,6 +110,15 @@ void blockdev_auto_del(BlockDriverSta
Signed-off-by: Maria Kustova
---
tests/image-fuzzer/qcow2/fuzz.py | 15 ++--
tests/image-fuzzer/runner.py | 51
2 files changed, 34 insertions(+), 32 deletions(-)
diff --git a/tests/image-fuzzer/qcow2/fuzz.py b/tests/image-fuzzer/qcow2/fuzz.py
+static void vfio_fdt_add_device_node(SysBusDevice *sbdev, void *opaque)
+{
+PlatformDevtreeData *data = opaque;
+void *fdt = data->fdt;
+const char *parent_node = data->node;
+int compat_str_len;
+char *nodename;
+int i, ret;
+uint32_t *irq_attr;
+uint64_t *reg_att
Heuristic values were added to fuzzing constraints and vectors.
Signed-off-by: Maria Kustova
---
tests/image-fuzzer/qcow2/fuzz.py | 71 +---
1 file changed, 37 insertions(+), 34 deletions(-)
diff --git a/tests/image-fuzzer/qcow2/fuzz.py b/tests/image-fuzzer/q
Additional commands were added to the default runner list to cover all qcow2
related code. This qcow2 specificity is selected to reduce number of
non-relevant tests. After implementation of a fuzzer for a new format the
default list should be updated.
Signed-off-by: Maria Kustova
---
tests/image
This patch series contains changes improving test coverage.
Maria Kustova (2):
runner: Expand the list of default test commands
fuzz: Make fuzzing functions and values relevant to the qemu
implementation
tests/image-fuzzer/qcow2/fuzz.py | 71 +++--
tests/i
Il 18/08/2014 19:47, Hulin, Patrick - 0559 - MITLL ha scritto:
>> We'll have done the page for the first byte at the top of
>> helper_{le,be}_{ld,st}_name. When we discover it's an unaligned
>> access, we should load and check the pte for the second page. We
>> might have to shuffle those two te
On 18 August 2014 22:04, Gaurav Sharma wrote:
> Hi Peter,
> I cross checked it with a AFM model, and the results are indeed different.
> The problem I think lies in how we treat de-normalized numbers which are too
> small to represent in half precision.
> In case of qemu
>>> if(exp < -10)
>>> retu
Hi Peter,
I cross checked it with a AFM model, and the results are indeed different.
The problem I think lies in how we treat de-normalized numbers which are
too small to represent in half precision.
In case of qemu
>> if(exp < -10)
>> return signed/unsigned zero.
However, in case rounding is set,
On 18 August 2014 20:36, Michael S. Tsirkin wrote:
> Does test fail if this path is triggered?
If our test harness doesn't report failure when a test
binary returns with a non-zero exit code then the
harness is broken, because there are other test
binaries that rely on that.
thanks
-- PMM
On 8/18/14, 1:47 PM, "Hulin, Patrick - 0559 - MITLL"
wrote:
>On Aug 18, 2014, at 1:37 PM, Richard Henderson wrote:
>
>>On 08/16/2014 10:21 PM, Paolo Bonzini wrote:
>Would it work to just call tb_invalidate_phys_page_range before the
>helper_ret_stb loop?
>>I doubt it.
>
>Correct. Doesn¹
On 18.08.2014 22:24, Max Reitz wrote:
On 18.08.2014 22:18, Eric Blake wrote:
On 08/18/2014 02:00 PM, Max Reitz wrote:
Add options for specifying the size of the metadata caches. This can
either be done directly for each cache (if only one is given, the other
will be derived according to a defau
On 18.08.2014 22:18, Eric Blake wrote:
On 08/18/2014 02:00 PM, Max Reitz wrote:
Add options for specifying the size of the metadata caches. This can
either be done directly for each cache (if only one is given, the other
will be derived according to a default ratio) or combined for both.
Signed
On Mon, Aug 18, 2014 at 03:32:21PM +0400, Michael Tokarev wrote:
> 18.08.2014 11:51, zhanghailiang пишет:
> > Here we don't check the return value of malloc() which may fail.
> > Use the g_malloc() instead, which will abort the program when
> > there is not enough memory.
> >
> > Signed-off-by: zh
On 08/18/2014 02:00 PM, Max Reitz wrote:
> Add options for specifying the size of the metadata caches. This can
> either be done directly for each cache (if only one is given, the other
> will be derived according to a default ratio) or combined for both.
>
> Signed-off-by: Max Reitz
> ---
> blo
On Thu, Aug 14, 2014 at 04:31:35PM +0300, Riku Voipio wrote:
> On Thu, Aug 14, 2014 at 03:29:18PM +0800, zhanghailiang wrote:
> > Signed-off-by: zhanghailiang
> > Acked-by: Riku Voipio
>
> Applied to linux-user as Michael seemed wary of passing these via
> trivial.
>
> Riku
Pls remember to add
On Mon, Aug 18, 2014 at 03:49:22PM +0400, Michael Tokarev wrote:
> 14.08.2014 11:29, zhanghailiang wrote:
> > In function virtio_blk_handle_request, it may freed memory pointed by req,
> > So do not access member of req after calling this function.
> >
> > Reviewed-by: Stefan Hajnoczi
> > Signed-
Add options for specifying the size of the metadata caches. This can
either be done directly for each cache (if only one is given, the other
will be derived according to a default ratio) or combined for both.
Signed-off-by: Max Reitz
---
block/qcow2.c | 112 ++
Add a test which tests various combinations of qcow2's cache options
(some of which are valid, some of which are not).
Signed-off-by: Max Reitz
---
tests/qemu-iotests/103 | 99 ++
tests/qemu-iotests/103.out | 29 ++
tests/qemu-iotests/g
Specifying the metadata cache sizes in clusters results in less clusters
(and much less bytes) covered for small cluster sizes and vice versa.
Using a constant byte size reduces this difference, and makes it
possible to manually specify the cache size in an easily comprehensible
unit.
Signed-off-b
With a variable cache size, the number given to qcow2_cache_create() may
be huge. Therefore, use g_try_new0().
While at it, use g_new0() instead of g_malloc0() for allocating the
Qcow2Cache object.
Signed-off-by: Max Reitz
---
block/qcow2-cache.c | 13 +
1 file changed, 9 insertions
Currently, the metadata cache size is only tunable on compile time
through macros. However, some users may want to use the minimal cache
size (for whatever reason) and others may want to increase the cache
size because they have enough memory and want to increase performance.
This series adds runt
On Mon, Aug 18, 2014 at 06:10:43PM +0200, Markus Armbruster wrote:
> They clutter the code. Unfortunately, I can't figure out how to make
> Coccinelle drop all of them, so I have to settle for common special
> cases:
>
> @@
> type T;
> T *pt;
> void *pv;
> @@
> - pt = (T *
On Mon, Aug 18, 2014 at 03:58:33PM +0200, Paolo Bonzini wrote:
> Il 18/08/2014 15:56, Michael S. Tsirkin ha scritto:
> > > +/* Initialize PCDIMMDevice->node to -1 so that even if user doesn't
> > > specify
> > > + * any numa option, PCDIMMDevice->node won't be 0, which indicates node0.
> > > + * I
The first patch adds the '--duration SECONDS' argument. After the specified
duration the runner allows to end the current test and then exits.
The second patch adds forced termination of a program under test, if the test
execution takes more than 10 minutes to indicate program freezes.
If a progr
On Mon, Aug 18, 2014 at 02:05:46PM -0400, Luiz Capitulino wrote:
> On Sun, 17 Aug 2014 11:45:17 +0200
> "Michael S. Tsirkin" wrote:
>
> > The function monitor_fdset_dup_fd_find_remove() references member of
> > 'mon_fdset' which - when remove flag is set - may be freed in function
> > monitor_fds
After the specified duration the runner stops executing new tests, but it
doesn't interrupt running ones.
Reviewed-by: Fam Zheng
Signed-off-by: Maria Kustova
---
tests/image-fuzzer/runner.py | 21 +
1 file changed, 17 insertions(+), 4 deletions(-)
diff --git a/tests/image-f
If a program under test get frozen, the test should finish and report about its
failure.
In such cases the runner waits for 10 minutes until the program ends its
execution. After this time-out the program will be terminated and the test will
be marked as failed.
For current limitation of test imag
Add a test which tests various combinations of qcow2's cache options
(some of which are valid, some of which are not).
Signed-off-by: Max Reitz
---
tests/qemu-iotests/103 | 99 ++
tests/qemu-iotests/103.out | 29 ++
tests/qemu-iotests/g
On 18.08.2014 22:00, Max Reitz wrote:
With a variable cache size, the number given to qcow2_cache_create() may
be huge. Therefore, use g_try_new0().
While at it, use g_new0() instead of g_malloc0() for allocating the
Qcow2Cache object.
Signed-off-by: Max Reitz
---
block/qcow2-cache.c | 7 +++
Add options for specifying the size of the metadata caches. This can
either be done directly for each cache (if only one is given, the other
will be derived according to a default ratio) or combined for both.
Signed-off-by: Max Reitz
---
block/qcow2.c | 112 ++
Currently, the metadata cache size is only tunable on compile time
through macros. However, some users may want to use the minimal cache
size (for whatever reason) and others may want to increase the cache
size because they have enough memory and want to increase performance.
This series adds runt
With a variable cache size, the number given to qcow2_cache_create() may
be huge. Therefore, use g_try_new0().
While at it, use g_new0() instead of g_malloc0() for allocating the
Qcow2Cache object.
Signed-off-by: Max Reitz
---
block/qcow2-cache.c | 7 +--
1 file changed, 5 insertions(+), 2
Specifying the metadata cache sizes in clusters results in less clusters
(and much less bytes) covered for small cluster sizes and vice versa.
Using a constant byte size reduces this difference, and makes it
possible to manually specify the cache size in an easily comprehensible
unit.
Signed-off-b
On Mon, Aug 18, 2014 at 06:10:42PM +0200, Markus Armbruster wrote:
> Signed-off-by: Markus Armbruster
> ---
> qemu-io-cmds.c | 11 +--
> 1 file changed, 1 insertion(+), 10 deletions(-)
>
> diff --git a/qemu-io-cmds.c b/qemu-io-cmds.c
> index afd8867..b224ede 100644
> --- a/qemu-io-cmds.c
On Mon, Aug 18, 2014 at 06:10:41PM +0200, Markus Armbruster wrote:
> g_new(T, n) is safer than g_malloc(sizeof(*v) * n) for two reasons.
> One, it catches multiplication overflowing size_t. Two, it returns
> T * rather than void *, which lets the compiler catch more type
> errors.
>
> Perhaps a c
On Mon, Aug 18, 2014 at 05:51:31PM +0800, Jason Wang wrote:
> commit a9f98bb5ebe6fb1869321dcc58e72041ae626ad8 vhost: multiqueue
> support changed the order of stopping the device. Previously
> vhost_dev_stop would disable backend and only afterwards, unset guest
> notifiers. We now unset guest noti
On 8/15/2014 3:05 PM, Richard Henderson wrote:
> On 08/11/2014 09:23 AM, Tom Musta wrote:
>> Also fix the special case of MB=31 and ME=0 to copy the entire contents
>> of the source GPR.
>
> Err, that's not what you did.
>
>> if (likely(sh == 0 && mb == 0 && me == 31)) {
>> +#if defined(TARG
On Mon, Aug 18, 2014 at 06:10:40PM +0200, Markus Armbruster wrote:
> g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
> for two reasons. One, it catches multiplication overflowing size_t.
> Two, it returns T * rather than void *, which lets the compiler catch
> more type error
On Mon, Aug 18, 2014 at 06:24:02PM +0400, Michael Tokarev wrote:
> 18.08.2014 17:44, Michael S. Tsirkin wrote:
> > On Mon, Aug 18, 2014 at 03:38:12PM +0400, Michael Tokarev wrote:
> >> 18.08.2014 11:54, zhanghailiang wrote:
> >>> The function fopen() may fail, so check its return value.
> >>>
> >>>
From: Hani Benhabiles
Signed-off-by: Hani Benhabiles
Signed-off-by: Luiz Capitulino
---
monitor.c | 11 +--
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/monitor.c b/monitor.c
index cdbaa60..48f0fdc 100644
--- a/monitor.c
+++ b/monitor.c
@@ -4521,16 +4521,15 @@ void net
From: Chen Gang
In dump_init(), when failure occurs, need notice about 'fd' and memory
mapping. So call dump_cleanup() for it (need let all initializations at
front).
Also simplify dump_cleanup(): remove redundant 'ret' and redundant 'fd'
checking.
Signed-off-by: Chen Gang
Reviewed-by: Laszlo
Three little birds.
The following changes since commit 08ab59770da57648bfb8fc9be37f0ef7fb50b0f9:
Merge remote-tracking branch 'remotes/mcayland/qemu-sparc' into staging
(2014-08-18 12:55:02 +0100)
are available in the git repository at:
git://repo.or.cz/qemu/qmp-unstable.git queue/qmp
fo
From: "Michael S. Tsirkin"
The function monitor_fdset_dup_fd_find_remove() references member of
'mon_fdset' which - when remove flag is set - may be freed in function
monitor_fdset_cleanup().
remove is set by monitor_fdset_dup_fd_remove which in practice
does not need the returned value, so make
On 2014-08-18 18:34, Knut Omang wrote:
> On Sat, 2014-08-16 at 10:47 +0200, Jan Kiszka wrote:
>> On 2014-08-16 10:45, Jan Kiszka wrote:
>>> On 2014-08-16 09:54, Knut Omang wrote:
On Fri, 2014-08-15 at 19:37 +0800, Le Tan wrote:
> Hi Knut,
>
> 2014-08-15 19:15 GMT+08:00 Knut Omang :
On 8/18/14, 1:47 PM, "Hulin, Patrick - 0559 - MITLL"
wrote:
>On Aug 17, 2014, at 1:21 AM, Paolo Bonzini wrote:
>
>> Il 15/08/2014 23:49, Hulin, Patrick - 0559 - MITLL ha scritto:
> In this case, the write is 8 bytes and unaligned, so it gets split
> into 8 single-byte writes. In stock QE
On Sun, 17 Aug 2014 11:45:17 +0200
"Michael S. Tsirkin" wrote:
> The function monitor_fdset_dup_fd_find_remove() references member of
> 'mon_fdset' which - when remove flag is set - may be freed in function
> monitor_fdset_cleanup().
> remove is set by monitor_fdset_dup_fd_remove which in practic
On Aug 17, 2014, at 1:21 AM, Paolo Bonzini wrote:
> Il 15/08/2014 23:49, Hulin, Patrick - 0559 - MITLL ha scritto:
In this case, the write is 8 bytes and unaligned, so it gets split
into 8 single-byte writes. In stock QEMU, these writes are done in
reverse order (see the loop in so
>El Mon, 18 Aug 2014 14:46:07 +0200
>Marc Marí escribió:
> +void qvirtio_pci_device_enable(QVirtioPCIDevice *d)
> +{
> +qpci_device_enable(d->pdev);
> +d->addr = qpci_iomap(d->pdev, 0);
> +g_assert(d->addr != NULL);
> +}
> +
qpci_iomap changed its prototype in one of the lastest pull
On Aug 18, 2014, at 1:37 PM, Richard Henderson wrote:
> On 08/16/2014 10:21 PM, Paolo Bonzini wrote:
Would it work to just call tb_invalidate_phys_page_range before the
helper_ret_stb loop?
>
> I doubt it.
Correct. Doesn’t work. Haven’t fully diagnosed why, but it doesn’t seem to eve
>El Mon, 18 Aug 2014 14:46:11 +0200
>Marc Marí escribió:
> +void qpci_msix_enable(QPCIDevice *dev)
> +{
> +uint8_t addr;
> +uint16_t val;
> +uint32_t table;
> +uint8_t bir_table;
> +uint8_t bir_pba;
> +void *offset;
> +
> +addr = qpci_find_capability(dev, PCI_CAP_ID_MSI
On Fri, 2014-08-15 at 20:12 +1000, Alexey Kardashevskiy wrote:
> Since the changes are not in upstream yet, no tag or branch is specified here.
>
> Signed-off-by: Alexey Kardashevskiy
> ---
> linux-headers/linux/vfio.h | 37 -
> 1 file changed, 36 insertions(+
On 08/16/2014 10:21 PM, Paolo Bonzini wrote:
>>> Would it work to just call tb_invalidate_phys_page_range before the
>>> helper_ret_stb loop?
I doubt it.
>> Maybe. I think there’s another issue, which is that QEMU’s ending up
>> in the I/O read/write code instead of the normal memory RW. This cou
1 - 100 of 286 matches
Mail list logo
