Il 11/08/2014 13:39, Gonglei (Arei) ha scritto:
>> I think the VM hangs there because the VCPU thread holds the global
>> qemu metux lock and waits for IO to come back. However, in my test, the
>> IO doesn't come back (because of the multipath software). Therefore, the
>> VCPU thread never releas
Il 09/08/2014 05:33, Zhang Haoyu ha scritto:
> After receive TERMINATE signal, qemu nbd state is set to TERMINATE, then in
> the main loop,
> nbd_export_close -> nbd_export_put is performed, but sometimes exp->refcount
> still greater than zero after nbd_export_put,
> so the qemu nbd state has n
On Fri, Aug 15, 2014 at 06:33:10PM +0200, Andreas Färber wrote:
> Am 14.08.2014 18:02, schrieb Michael S. Tsirkin:
> > Patch
> > xen:hw:i386:pc_piix: introduce new machine for IGD passthrough
> > adds a new machine type.
> > Blacklist it since it forces xen accel.
> >
> > Cc: Tiejun Chen
> >
Il 13/08/2014 23:59, Marc Marí ha scritto:
> If it is not dereferenced, it can work. But all references will have to
> be casted from (void *) to (uint64_t) always which is what readl and
> writel expect. I think this is better than calculating the addresses,
> but is still a bit ugly.
It would al
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 it void,
and return -1 from m
On Fri, Aug 15, 2014 at 02:25:51PM -0400, Luiz Capitulino wrote:
> On Thu, 14 Aug 2014 12:30:10 +0200
> "Michael S. Tsirkin" wrote:
>
> > On Thu, Aug 14, 2014 at 03:29:13PM +0800, zhanghailiang wrote:
> > > The function monitor_fdset_dup_fd_find_remove() references member of
> > > 'mon_fdset'
>
Il 12/08/2014 10:02, Gal Hammer ha scritto:
> Hi,
>
> On 10/08/2014 20:22, Paolo Bonzini wrote:
>
>> Il 10/08/2014 13:32, Gal Hammer ha scritto:
>>> Based on Microsoft's sepecifications (paper can be dowloaded from
>>> http://go.microsoft.com/fwlink/?LinkId=260709), add a device
>>> description t
Il 11/08/2014 22:15, Michael R. Hines ha scritto:
> Excellent question: QEMU does have a feature called "drive-mirror"
> in block/mirror.c that was introduced a couple of years ago. I'm not
> sure what the
> adoption rate of the feature is, but I would start with that one.
block/mirror.c is asynch
On Fri, Aug 15, 2014 at 02:25:51PM -0400, Luiz Capitulino wrote:
> On Thu, 14 Aug 2014 12:30:10 +0200
> "Michael S. Tsirkin" wrote:
>
> > On Thu, Aug 14, 2014 at 03:29:13PM +0800, zhanghailiang wrote:
> > > The function monitor_fdset_dup_fd_find_remove() references member of
> > > 'mon_fdset'
>
On Fri, Aug 15, 2014 at 07:33:29AM +, Gonglei (Arei) wrote:
> Hi,
>
> I noticed that the qemu-2.1 release change log says
> " PCIe: Basic hot-plug/hot-unplug support for Q35 machine."
> And then I made a testing for the hotplugging function of Q35.
> But I'm failed, and I got the dmesg log in
On Sat, Aug 16, 2014 at 01:55:39PM +0800, Le Tan wrote:
> Expose Intel IOMMU to the BIOS. If object of TYPE_INTEL_IOMMU_DEVICE exists,
> add DMAR table to ACPI RSDT table. For now the DMAR table indicates that there
> is only one hardware unit without INTR_REMAP capability on the platform.
>
> Sig
On Sat, Aug 16, 2014 at 01:55:36PM +0800, Le Tan wrote:
> Hi,
>
> These patches are intended to introduce Intel IOMMU (VT-d) emulation to q35
> chipset. The major job in these patches is to add support for emulating Intel
> IOMMU according to the VT-d specification, including basic responses to CS
Am 17.08.2014 13:07, schrieb Michael S. Tsirkin:
> On Sat, Aug 16, 2014 at 01:55:39PM +0800, Le Tan wrote:
>> Expose Intel IOMMU to the BIOS. If object of TYPE_INTEL_IOMMU_DEVICE exists,
>> add DMAR table to ACPI RSDT table. For now the DMAR table indicates that
>> there
>> is only one hardware un
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
> > " PCIe: Basic hot-plug/hot-unplug support for Q35 machine."
> > And then I made a testing for the h
On 13/08/14 17:38, Richard Henderson wrote:
On 08/12/2014 10:02 AM, Artyom Tarasenko wrote:
Implement Short Floating-Point Store Instructions as described
in the chapter 13.5.2 of UltraSPARC-IIi User's Manual.
Particularly this instructions are used by NetBSD 4.0.1+ /sparc64
Signed-off-by: Ar
Hi Peter,
I've updated my qemu-sparc branch with the latest set of SPARC64 fixes. Please
pull.
ATB,
Mark.
The following changes since commit f2c85a2f36f57f155cda7bc9f7c42b44f1a2439e:
Merge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging
(2014-08-15 17:43:51 +0100)
ar
The SP810, which is present in the Versatile Express motherboards,
allows to set the timing reference to either REFCLK or TIMCLK.
QEMU currently sets the SP804 timer to 1MHz by default. To reflect
this, we set the TimerEn0Sel, TimerEn1Sel, TimerEn2Sel, and
TimerEn3Sel of the system control register
Hi,
The Versatile Express emulation in QEMU currently does not
have a model of the SP810 used in real hardware. The registers
provided by this System Controller can be used to set the
frequency of the SP804 timers. On newer releases of the board
the SP804 is set to 32kHz by default and has to be i
This adds a device model for the PrimeXsys System Controller (SP810)
which is present in the Versatile Express motherboards. It is
so far read-only but allows to read the SCCTRL register.
Signed-off-by: Fabian Aggeler
---
default-configs/arm-softmmu.mak | 1 +
hw/misc/Makefile.objs |
Il 15/08/2014 09:17, Peter Crosthwaite ha scritto:
> In a5e1cbc80e88ed7d73b3fcb46053a3ba167293fc the enforcement of Memory
> collisions was disabled. This means that the MemoryRegion map_overlap
> state is unused. Remove it completely.
>
> The commit mentions that it should be fixed, but we have b
Il 15/08/2014 08:53, Peter Crosthwaite ha scritto:
> Hi Paolo,
>
> Another piece to the Memory Region QOMification puzzle. This series gets
> rid of the name field of MemoryRegion and converts all users to the QOM
> managed Object name.
>
> Regards,
> Peter
>
>
> Peter Crosthwaite (5):
> load
On 15 August 2014 08:17, Peter Crosthwaite wrote:
> In a5e1cbc80e88ed7d73b3fcb46053a3ba167293fc the enforcement of Memory
> collisions was disabled. This means that the MemoryRegion map_overlap
> state is unused. Remove it completely.
>
> The commit mentions that it should be fixed, but we have be
Il 18/08/2014 00:23, Peter Maydell ha scritto:
> Um. I think in general colliding memory regions imply a bug in the
> machine model which is creating them. We have the collisions
> disabled because nobody's got round to fixing this bug in the
> PC model yet and it happens to be harmless there. I'm
On Mon, Aug 18, 2014 at 8:23 AM, Peter Maydell wrote:
> On 15 August 2014 08:17, Peter Crosthwaite
> wrote:
>> In a5e1cbc80e88ed7d73b3fcb46053a3ba167293fc the enforcement of Memory
>> collisions was disabled. This means that the MemoryRegion map_overlap
>> state is unused. Remove it completely.
On 2014/8/14 18:31, Michael S. Tsirkin wrote:
On Thu, Aug 14, 2014 at 03:29:17PM +0800, zhanghailiang wrote:
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: zhanghailiang
Re
On 2014/8/14 18:32, Michael S. Tsirkin wrote:
On Thu, Aug 14, 2014 at 03:29:19PM +0800, zhanghailiang wrote:
The function fopen() may fail, so check its return value.
Signed-off-by: zhanghailiang
Signed-off-by: Li Liu
Reviewed-by: Alex Bennée
---
tests/bios-tables-test.c | 5 +
1 file ch
On 2014/8/14 17:56, Michael S. Tsirkin wrote:
On Thu, Aug 14, 2014 at 05:09:00PM +0800, zhanghailiang wrote:
On 2014/8/14 15:15, Hu Tao wrote:
On Thu, Aug 14, 2014 at 02:31:41PM +0800, zhanghailiang wrote:
On 2014/8/13 19:50, Michael S. Tsirkin wrote:
On Wed, Aug 13, 2014 at 07:21:57PM +0800,
On 2014/8/14 18:05, Michael S. Tsirkin wrote:
On Thu, Aug 14, 2014 at 02:13:57PM +0800, zhanghailiang wrote:
When the runstate changes back to running, we definitely need to flush
queues to get packets flowing again.
Here we implement this in the net layer:
(1) add a member 'VMChangeStateEntry
On 2014/8/14 18:09, Michael S. Tsirkin wrote:
On Thu, Aug 14, 2014 at 02:13:57PM +0800, zhanghailiang wrote:
When the runstate changes back to running, we definitely need to flush
queues to get packets flowing again.
Here we implement this in the net layer:
(1) add a member 'VMChangeStateEntry
It's a constant based on a macro. Just use the macro in place.
Signed-off-by: Peter Crosthwaite
---
hw/microblaze/petalogix_ml605_mmu.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/hw/microblaze/petalogix_ml605_mmu.c
b/hw/microblaze/petalogix_ml605_mmu.c
index aea9c
This have 6 inline usages. Make it a bit more readable by using a local
variable.
Signed-off-by: Peter Crosthwaite
---
Theres three more usages of this in coming patches, that would push
the number of usages higher
hw/dma/xilinx_axidma.c | 10 ++
1 file changed, 6 insertions(+), 4 delet
On 2014/8/14 18:36, Michael S. Tsirkin wrote:
On Thu, Aug 14, 2014 at 03:29:21PM +0800, 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 | 5 -
1 file changed, 4 insertio
>> >> Hi, all
>> >> I'm using VFIO to assign intel 82599 VF to VM, now I encounter a problem,
>> >> 82599 PF and its VFs belong to the same iommu_group, but I only want to
>> >> assign some VFs to one VM, and some other VFs to another VM, ...,
>> >> so how to only unbind (part of) the VFs but PF?
On Mon, 2014-08-18 at 09:00 +0800, Zhang Haoyu wrote:
> >> >> Hi, all
> >> >> I'm using VFIO to assign intel 82599 VF to VM, now I encounter a
> >> >> problem,
> >> >> 82599 PF and its VFs belong to the same iommu_group, but I only want to
> >> >> assign some VFs to one VM, and some other VFs to
This argument is a MemoryRegion and not an AddressSpace.
"Address space" means something quite different to "memory region"
in QEMU parlance so rename the variable to reduce confusion.
Signed-off-by: Peter Crosthwaite
---
hw/arm/armv7m.c | 8
include/hw/arm/arm.h | 2 +-
2 files
It's a MemoryRegion and not an AddressSpace. But since it's single use,
just inline the get_system_memory() call to the only usage to remove it.
Signed-off-by: Peter Crosthwaite
---
hw/arm/stellaris.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/hw/arm/stellaris.c b/hw/
Caches are 16 bytes in A9. Self identify in CCSIDR accordingly.
QEMU doesn't emulate caches, but we should still report the correct
cache-line size to the guest. Some guests (like u-boot) complain if
the cache-line size mismatches a requested flush or invalidate
operation.
Signed-off-by: Peter Cr
On Sun, Aug 17, 2014 at 06:59:59PM -0700, Peter Crosthwaite wrote:
> Caches are 16 bytes in A9. Self identify in CCSIDR accordingly.
Hi,
This is a bit confusing, caches are not 16bytes on the a9, they are
configurable.
Cache lines are 32 bytes. The reported associativity doesn't match
either. Fo
On Mon, Aug 18, 2014 at 12:18 PM, Edgar E. Iglesias
wrote:
> On Sun, Aug 17, 2014 at 06:59:59PM -0700, Peter Crosthwaite wrote:
>> Caches are 16 bytes in A9. Self identify in CCSIDR accordingly.
>
> Hi,
>
> This is a bit confusing, caches are not 16bytes on the a9, they are
> configurable.
Sorry
On Mon, Aug 18, 2014 at 12:57:43PM +1000, Peter Crosthwaite wrote:
> On Mon, Aug 18, 2014 at 12:18 PM, Edgar E. Iglesias
> wrote:
> > On Sun, Aug 17, 2014 at 06:59:59PM -0700, Peter Crosthwaite wrote:
> >> Caches are 16 bytes in A9. Self identify in CCSIDR accordingly.
> >
> > Hi,
> >
> > This is
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
---
hw/ppc/spapr.c | 6 ++
1 fi
On Wed, Aug 13, 2014 at 09:48:35AM -0500, Greg Bellows wrote:
> Hi Edgar,
>
> I was just writing a test to verify the correct behavior of the SCR AW/FW
> bits and I think there is an issue.
>
> During an SCR write an initial valid mask is set from SCR_MASK which is
> defined to not include these
On Sat, Aug 2, 2014 at 1:28 AM, Peter Maydell wrote:
> On 26 June 2014 06:02, Alistair Francis wrote:
>> This patch adds support for the ARMv8 version of the PMCCNTR and
>> related registers. It also starts to implement the PMCCFILTR_EL0
>> register.
>>
>> Signed-off-by: Peter Crosthwaite
>> Sig
On Thu, Jun 26, 2014 at 9:28 PM, Peter Crosthwaite
wrote:
> On Thu, Jun 26, 2014 at 3:02 PM, Alistair Francis
> wrote:
>> Include a helper function to determine if the CCNT counter
>> is enabled as well as the constants used to mask the pmccfiltr_el0
>> and c9_pmxevtyper registers.
>>
>> Signed-o
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 v2: Forgot check on k
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 PAUSE VM stage,
The new dirty RAM related to the packets will be missed,
And this will lead serious network fau
On Fri, 08/15 17:55, Maria Kustova wrote:
> Signed-off-by: Maria Kustova
> ---
> tests/image-fuzzer/runner.py | 23 ++-
> 1 file changed, 18 insertions(+), 5 deletions(-)
>
> diff --git a/tests/image-fuzzer/runner.py b/tests/image-fuzzer/runner.py
> index 3fa7fca..688d470 100
On 2014/8/14 18:31, Michael S. Tsirkin wrote:
On Thu, Aug 14, 2014 at 03:29:17PM +0800, zhanghailiang wrote:
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: zhanghailiang
Re
On Fri, Aug 15, 2014 at 03:59:04PM +0200, Benoît Canet wrote:
> The Friday 15 Aug 2014 à 13:05:17 (+0800), Liu Yuan wrote :
> > This patch adds single read pattern to quorum driver and quorum vote is
> > default
> > pattern.
> >
> > For now we do a quorum vote on all the reads, it is designed for
On 2014/8/14 18:15, Michael S. Tsirkin wrote:
On Thu, Aug 14, 2014 at 03:29:16PM +0800, zhanghailiang wrote:
In this file, we don't check the return value of malloc/strdup/realloc which
may fail.
Instead of using these routines, we use the GLib memory APIs
g_malloc/g_strdup/g_realloc.
They wil
On 2014/8/16 0:53, Richard Henderson wrote:
On 08/15/2014 05:03 AM, Michael Tokarev wrote:
f = fopen("/tmp/op.log", "w");
Gosh. So why are we still use fixed filenames in /tmp?
Every such use is a potential security holw... :( Ughm.
Can't we get rid of this somehow, by requiring a
On Tue, 2014-08-12 at 12:00 +1000, Peter Crosthwaite wrote:
> On Mon, Aug 4, 2014 at 2:21 PM, Chen Fan wrote:
> > the memdev_list in hmp_info_memdev() is never freed.
> > so we use existent method qapi_free_MemdevList() to free it.
> > and also we can use qapi_free_MemdevList() to replace list lo
Signed-off-by: Chen Fan
Reviewed-by: Peter Crosthwaite
Reviewed-by: Hu Tao
---
numa.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/numa.c b/numa.c
index c78cec9..aa772aa 100644
--- a/numa.c
+++ b/numa.c
@@ -318,10 +318,11 @@ void memory_region_allocate_system_memory
when using valgrind to test the command "query memdev", I had
found some memory leaks. the test result:
==13802== 4 bytes in 1 blocks are definitely lost in loss record 125 of 8,508
==13802==at 0x4A08934: malloc (vg_replace_malloc.c:291)
==13802==by 0x4A08AA8: realloc (vg_replace_malloc.c:
the memdev_list in hmp_info_memdev() is never freed.
so we use existent method qapi_free_MemdevList() to free it.
and also we can use qapi_free_MemdevList() to replace list loops
to clean up the memdev list in error path.
Signed-off-by: Chen Fan
Reviewed-by: Peter Crosthwaite
Reviewed-by: Hu Tao
string_output_get_string() uses g_string_free(str, false) to
transfer the 'str' pointer to callers and never free it.
Signed-off-by: Chen Fan
Reviewed-by: Peter Crosthwaite
Reviewed-by: Hu Tao
---
hmp.c| 6 --
qom/object.c | 12 ++--
2 files changed, 14 insertions(+), 4 d
On Wed, 08/06 17:24, Milos Vyletel wrote:
> VMDK's streamOptimized format is different from regular sparse format.
> L1(GD) and L2(GT) tables are not predefined but rather generated and
> written during image creation mainly because there is no way to tell
> how much space data will occupy once the
57 matches
Mail list logo
