On 2014/10/17 14:38, Daniel P. Berrange wrote:
> On Fri, Oct 17, 2014 at 02:34:07PM +0800, Gonglei wrote:
>> On 2014/10/16 18:46, Gerd Hoffmann wrote:
>>
>>> Hi,
>>>
> I try to prevent that by dropping the *oldest* connection, so you have a
> chance to connect even if a unprivileged atta
On Fri, Oct 17, 2014 at 02:34:07PM +0800, Gonglei wrote:
> On 2014/10/16 18:46, Gerd Hoffmann wrote:
>
> > Hi,
> >
> >>> I try to prevent that by dropping the *oldest* connection, so you have a
> >>> chance to connect even if a unprivileged attacker tries to use up all
> >>> connection slots.
>
On 2014/10/16 18:46, Gerd Hoffmann wrote:
> Hi,
>
>>> I try to prevent that by dropping the *oldest* connection, so you have a
>>> chance to connect even if a unprivileged attacker tries to use up all
>>> connection slots.
>>
>> Lets say the limit is 5. The bad guy has 5 open idle connections.
On Thu, 2014-10-16 at 13:50 +0200, Igor Mammedov wrote:
> On Mon, 6 Oct 2014 19:38:44 +0800
> Zhu Guihua wrote:
>
> > Add peripheral_device_del_completion() to let peripheral device del
> > completion be possible.
> >
> > Signed-off-by: Zhu Guihua
> > ---
> > monitor.c | 24 +++
Hi,
Please let me know if I attach a drive to KVM guest using PCI pass-through,
then will BDRV block layer receives IO?
Thanks & Regards,
Sanjay
***Legal Disclaimer***
"This communication may contain confidential and privileged material for the
On 2014/10/17 9:06, Gonglei wrote:
On 2014/10/17 8:45, Zhanghailiang wrote:
On 2014/10/16 20:50, Gonglei wrote:
On 2014/10/16 20:02, zhanghailiang wrote:
When do memory hotplug, if there is numa node, we should add
the memory size to the corresponding node memory size.
For now, it mainly af
On Fri, Oct 10, 2014 at 11:03:22AM -0500, Greg Bellows wrote:
> From: Fabian Aggeler
>
> Prepare ARMCPRegInfo to support specifying two fieldoffsets per
> register definition. This will allow us to keep one register
> definition for banked registers (different offsets for secure/
> non-secure wor
On Wed, Oct 15, 2014 at 09:30:52PM +0100, Peter Maydell wrote:
> Signed-off-by: Peter Maydell
> ---
> Update the 'general project admin' section to include me as
> well as Anthony, given that I've been in practice doing a
> bunch of said admin, and the general consensus at the QEMU
> Summit yester
On 2014/10/17 8:45, Zhanghailiang wrote:
> On 2014/10/16 20:50, Gonglei wrote:
>> On 2014/10/16 20:02, zhanghailiang wrote:
>>
>>> When do memory hotplug, if there is numa node, we should add
>>> the memory size to the corresponding node memory size.
>>>
>>> For now, it mainly affects the result o
On 2014/10/16 20:50, Gonglei wrote:
On 2014/10/16 20:02, zhanghailiang wrote:
When do memory hotplug, if there is numa node, we should add
the memory size to the corresponding node memory size.
For now, it mainly affects the result of hmp command "info numa".
Reviewed-by: Igor Mammedov
Signe
Yes, good catch. Not sure why it did not come up in my search.
Will fix in v7.
Greg
On 14 October 2014 22:17, Edgar E. Iglesias
wrote:
> On Fri, Oct 10, 2014 at 11:03:41AM -0500, Greg Bellows wrote:
> > From: Fabian Aggeler
> >
> > When EL3 is running in AArch32 (or ARMv7 with Security Exten
On Wed, Oct 15, 2014 at 10:12 PM, Max Filippov wrote:
> On Thu, Oct 16, 2014 at 6:00 AM, Pranith Kumar wrote:
+tcg_gen_add_i32(cpu_fence_count, cpu_fence_count, 1);
>>>
>>> tcg_gen_addi_i32, as you're adding an immediate.
>>> Enabling TCG debug would catch such errors.
>>> Not sure i
Hi Peter,
On Thu, Oct 16, 2014 at 4:05 AM, Peter Maydell wrote:
> On 16 October 2014 03:45, Pranith Kumar wrote:
>> The problem I am facing is that this seems to be crashing when run with a
>> multi-threaded executable.
>
> This is nothing to do with your changes -- user-mode QEMU does not
> sup
On 10/16/2014 09:27 AM, Max Reitz wrote:
>>> +memset(*refcount_table + old_nb_clusters, 0,
>>> + (*nb_clusters - old_nb_clusters) * sizeof(uint16_t));
>> Is this calculation unnecessarily hard-coded to refcount_order==4?
>
> While now in the process of editing this patch, no
On Thu, 16 Oct 2014, Peter Maydell wrote:
> On 16 October 2014 13:29, Stefano Stabellini
> wrote:
> > Unfortunately even though it is pretty x86 specific, it is still
> > compiled on ARM, even though it is never actually used (it is used in
> > i386 emulation with Xen acceleration support, while o
Am 09.10.2014 um 01:09 schrieb Eric Blake:
On 08/29/2014 03:41 PM, Max Reitz wrote:
+ * cluster_count clusters; therefore, we have to allocate
+ * cluster_count - contiguous_free_clusters new clusters at the end of
+ * the image (which is the current value of cluster; not
On Wed, 15 Oct 2014 12:10:39 +0200
Gerd Hoffmann wrote:
> Add verification to vmsvga_fill_rect, re-enable HW_FILL_ACCEL.
>
> Cc: qemu-sta...@nongnu.org
> Signed-off-by: Gerd Hoffmann
> ---
> hw/display/vmware_vga.c | 6 --
> 1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a
On Wed, 15 Oct 2014 12:10:36 +0200
Gerd Hoffmann wrote:
> Add verification function for rectangles, returning
> true if verification passes and false otherwise.
>
> Cc: qemu-sta...@nongnu.org
> Signed-off-by: Gerd Hoffmann
Reviewed-by: Don Koch
> ---
> hw/display/vmware_vga.c | 53
> +++
On Thu, Oct 16, 2014 at 12:38:07PM +0100, Leon Alrae wrote:
> Add myself to the maintainer list for MIPS guest cores and update the status
> from "Odd Fixes" to "Maintained".
>
> Signed-off-by: Leon Alrae
> ---
> MAINTAINERS | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --g
On 10/16/2014 07:25 AM, Max Reitz wrote:
> qcow2_write_l1_entry() may read L1 entries from beyond the end of the
> in-memory L1 table when updating a sector. Fix this and add a
> qemu-iotest.
>
>
> Max Reitz (2):
> qcow2: Do not overflow when writing an L1 sector
> iotests: Add test for qcow2
On 10/16/2014 07:13 AM, Michal Privoznik wrote:
> Currently, when the preallocating guest memory process fails, an not
s/an/a/
> so helpful error message is printed out:
>
> # virsh start migt10
> error: Failed to start domain migt10
> error: internal error: process exited while conn
Please update message body as it also adds dm{f,t}c0 for config{4,5}.
See the comment below.
Reviewed-by: Yongbok Kim
Regards,
Yongbok
On 14/07/2014 17:19, Leon Alrae wrote:
SDBBP instruction Reserved Instruction control. The purpose of this field is
to restrict availability of SDBBP to kerne
On Wed, 15 Oct 2014 12:10:38 +0200
Gerd Hoffmann wrote:
> Add verification to vmsvga_copy_rect, re-enable HW_RECT_ACCEL.
>
> Cc: qemu-sta...@nongnu.org
> Signed-off-by: Gerd Hoffmann
> ---
> hw/display/vmware_vga.c | 9 -
> 1 file changed, 8 insertions(+), 1 deletion(-)
>
> diff --git
On Wed, 15 Oct 2014 12:10:37 +0200
Gerd Hoffmann wrote:
> Switch vmsvga_update_rect over to use vmsvga_verify_rect. Slight change
> in behavior: We don't try to automatically fixup rectangles any more.
> In case we find invalid update requests we'll do a full-screen update
> instead.
This is g
On Wed, 15 Oct 2014 12:10:35 +0200
Gerd Hoffmann wrote:
> Quick & easy stopgap for CVE-2014-3689: We just compile out the
> hardware acceleration functions which lack sanity checks. Thankfully
> we have capability bits for them (SVGA_CAP_RECT_COPY and
> SVGA_CAP_RECT_FILL), so guests should dea
On 10/15/14 18:54, Ikbal Cam wrote:
Hi All,
I need to run ESXI as nested HV in KVM.
Do you know if this patch has been ported to any qemu release and
maybe available as RPM? ( Fedora, Redhat, Centos etc. )
https://lists.gnu.org/archive/html/qemu-devel/2014-05/msg03806.html
This is not i
Am 16.10.2014 um 15:34 schrieb Max Reitz:
Am 16.10.2014 um 14:10 schrieb Igor Mammedov:
change type of variable to expected IoOperationType which fixes compile
warning:
block.c:3655:20: warning: implicit conversion from enumeration
type enum IoOperationType to different enumeration type
Bloc
Am 16.10.2014 um 14:10 schrieb Igor Mammedov:
change type of variable to expected IoOperationType which fixes compile
warning:
block.c:3655:20: warning: implicit conversion from enumeration
type enum IoOperationType to different enumeration type BlockErrorAction
Signed-off-by: Igor Mammedov
While writing an L1 table sector, qcow2_write_l1_entry() copies the
respective range from s->l1_table to the local "buf" array. The size of
s->l1_table does not have to be a multiple of L1_ENTRIES_PER_SECTOR;
thus, limit the index which is used for copying all entries to the L1
size.
Cc: qemu-sta.
qcow2_write_l1_entry() may read L1 entries from beyond the end of the
in-memory L1 table when updating a sector. Fix this and add a
qemu-iotest.
Max Reitz (2):
qcow2: Do not overflow when writing an L1 sector
iotests: Add test for qcow2 L1 table update
block/qcow2-cluster.c | 6 +++--
Updating the L1 table should not result in random data being written.
This adds a test for that.
Signed-off-by: Max Reitz
---
tests/qemu-iotests/107 | 61 ++
tests/qemu-iotests/107.out | 10
tests/qemu-iotests/group | 1 +
3 files chang
Currently, when the preallocating guest memory process fails, an not
so helpful error message is printed out:
# virsh start migt10
error: Failed to start domain migt10
error: internal error: process exited while connecting to monitor:
os_mem_prealloc: failed to preallocate pages
>
EHINV bit from TLB also required to be saved/ restored.
Regards,
Yongbok
On 08/07/2014 08:57, Leon Alrae wrote:
Signed-off-by: Leon Alrae
---
target-mips/cpu.h |2 +-
target-mips/machine.c | 14 ++
2 files changed, 15 insertions(+), 1 deletions(-)
diff --git a/targe
On 15 October 2014 19:00, dcb <1381...@bugs.launchpad.net> wrote:
> Public bug reported:
>
> [qemu-2.1.2/roms/u-boot/board/renesas/ecovec/ecovec.c:66]: (error)
> Buffer is accessed out of bounds.
This is in the u-boot code which we just carry a copy of
to produce certain boot ROMs. You should repo
On 16/10/2014 11:52, Yongbok Kim wrote:
>> +ctx.ie = (env->CP0_Config4 >> CP0C4_IE) & 1;
>
> & 3
>
Yeah, initially I tested TLBINV using mips64r6-generic cpu where
Config4.IE is set to 3, thus it worked...
Thanks,
Leon
This patch adds the stm32f205 SoC. This will be used by the
Netduino 2 to create a machine.
Signed-off-by: Alistair Francis
---
default-configs/arm-softmmu.mak | 1 +
hw/arm/Makefile.objs| 1 +
hw/arm/stm32f205_soc.c | 157
includ
This patch moves the memory region init code from the
armv7m_init function to the stellaris_init function
Signed-off-by: Alistair Francis
---
hw/arm/armv7m.c | 33 +++--
hw/arm/stellaris.c | 24
include/hw/arm/arm.h | 3 +--
3 files ch
This patch allows the board to specifiy the number of NVIC interrupt
lines when using armv7m_init.
Signed-off-by: Alistair Francis
---
hw/arm/armv7m.c | 7 ---
hw/arm/stellaris.c | 5 -
include/hw/arm/arm.h | 2 +-
3 files changed, 9 insertions(+), 5 deletions(-)
diff --git a/hw/
This patch adds the stm32f205 System Configuration
Controller. This is used to configure what memory is mapped
at address 0 (although that is not supported) as well
as configure how the EXTI interrupts work (also not
supported at the moment).
This device is not required for basic examples, but mor
This patch adds the stm32f205 USART controller
(UART also uses the same controller).
Signed-off-by: Alistair Francis
---
default-configs/arm-softmmu.mak | 1 +
hw/char/Makefile.objs | 1 +
hw/char/stm32f205_usart.c | 218 ++
include/h
This patch adds the Netduino 2 Machine.
This is a Cortex-M3 based machine. Information can be found at:
http://www.netduino.com/netduino2/specs.htm
Signed-off-by: Alistair Francis
---
hw/arm/Makefile.objs | 1 +
hw/arm/netduino2.c | 54
2
> -Original Message-
> From: Igor Mammedov [mailto:imamm...@redhat.com]
> Sent: 16 October 2014 13:41
> To: Paul Durrant
> Cc: qemu-devel@nongnu.org; xen-de...@lists.xenproject.org; Michael S.
> Tsirkin; Andreas Faerber
> Subject: Re: [PATCH v3 1/2] Add device listener interface
>
> On Wed
This patch adds the stm32f205 timers: TIM2, TIM3, TIM4 and TIM5
to QEMU.
Signed-off-by: Alistair Francis
---
V4:
- Update timer units again
- Thanks to Peter C
V3:
- Update debug statements
- Correct the units for timer_mod
- Correctly set timer_offset from resets
V2:
- Reorder the Makef
This patch series adds the Netduino 2 Machine to QEMU
Information on the board is avalible at:
http://www.netduino.com/netduino2/specs.htm
The git tree can be found at:
https://github.com/alistair23/qemu/tree/netduino2.5
This patch series makes some changes to the armv7m_init function
that allow
On 2014/10/16 20:02, zhanghailiang wrote:
> When do memory hotplug, if there is numa node, we should add
> the memory size to the corresponding node memory size.
>
> For now, it mainly affects the result of hmp command "info numa".
>
> Reviewed-by: Igor Mammedov
> Signed-off-by: zhanghailiang
On Wed, 15 Oct 2014 10:05:32 +
Paul Durrant wrote:
>
>
> > -Original Message-
> > From: Igor Mammedov [mailto:imamm...@redhat.com]
> > Sent: 15 October 2014 10:54
> > To: Paul Durrant
> > Cc: qemu-devel@nongnu.org; xen-de...@lists.xenproject.org; Michael
> > S. Tsirkin; Andreas Faer
On 16 October 2014 13:29, Stefano Stabellini
wrote:
> Unfortunately even though it is pretty x86 specific, it is still
> compiled on ARM, even though it is never actually used (it is used in
> i386 emulation with Xen acceleration support, while on ARM we only use
> the PV machine).
Really? CONFIG
Hi,
> > How stable are these numbers?
>
> In theory the mesa/gallium numbers aren't stable, though I've never
> seen them change yet,
>
> If they diverge in the future I'll just provide a remapping table
> inside the guest driver.
>
> So it should be fine to expose these formats for 2D use.
This patch series is v4 of what was originally the single patch
"Xen: Use the ioreq-server API when available".
v2 of the series moved the code that added the PCI bus listener
to patch #1 and the remainder of the changes to patch #2. Patch #2
was then re-worked to constrain the #ifdefing to xen_co
The ioreq-server API added to Xen 4.5 offers better security than
the existing Xen/QEMU interface because the shared pages that are
used to pass emulation request/results back and forth are removed
from the guest's memory space before any requests are serviced.
This prevents the guest from mapping
The Xen ioreq-server API, introduced in Xen 4.5, requires that PCI device
models explicitly register with Xen for config space accesses. This patch
adds a listener interface into qdev-core which can be used by the Xen
interface code to monitor for arrival and departure of PCI devices.
Signed-off-b
change type of variable to expected IoOperationType which fixes compile
warning:
block.c:3655:20: warning: implicit conversion from enumeration
type enum IoOperationType to different enumeration type BlockErrorAction
Signed-off-by: Igor Mammedov
---
block.c | 6 +++---
1 file changed, 3 insert
When do memory hotplug, if there is numa node, we should add
the memory size to the corresponding node memory size.
For now, it mainly affects the result of hmp command "info numa".
Reviewed-by: Igor Mammedov
Signed-off-by: zhanghailiang
---
v5:
- reword the subject (Igor Mammedov)
- turn quer
Friendly ping.
I've saw a couple of scripts affected by this bug.
Thanks,
Zifei
On Sep 19, 2014 10:03 PM, "Kirill Batuzov" wrote:
> On 19.09.2014 11:12, Zifei Tong wrote:
>
>> After commit 812c1057f6175ac9a9829fa2920a2b5783814193 (Handle G_IO_HUP
>> in tcp_chr_read for tcp chardev), connections
On Mon, 6 Oct 2014 19:38:44 +0800
Zhu Guihua wrote:
> Add peripheral_device_del_completion() to let peripheral device del
> completion be possible.
>
> Signed-off-by: Zhu Guihua
> ---
> monitor.c | 24
> 1 file changed, 24 insertions(+)
>
> diff --git a/monitor.c b/mo
On Mon, 6 Oct 2014 19:38:43 +0800
Zhu Guihua wrote:
> For peripheral device del completion, add a function to build a list
> for hotpluggable devices.
>
> Signed-off-by: Zhu Guihua
> ---
> hw/core/qdev.c | 14 ++
> include/hw/qdev-core.h | 2 ++
> 2 files changed, 16 inser
Add myself to the maintainer list for MIPS guest cores and update the status
from "Odd Fixes" to "Maintained".
Signed-off-by: Leon Alrae
---
MAINTAINERS | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index 206bf7e..e07662f 100644
--- a/MAINTAINERS
On Wed, 15 Oct 2014, Peter Maydell wrote:
> On 15 October 2014 11:16, Paul Durrant wrote:
> > The ioreq-server API added to Xen 4.5 offers better security than
> > the existing Xen/QEMU interface because the shared pages that are
> > used to pass emulation request/results back and forth are remove
On 2014/10/16 18:36, Igor Mammedov wrote:
On Thu, 9 Oct 2014 20:21:57 +0800
zhanghailiang wrote:
reword subj with:
numa: make 'info numa' take into account hotplugged memory
When do memory hotplug, if there is numa node, we should add
the memory size to the corresponding node memory size.
Fo
On Thu, 16 Oct 2014, Alex Bennée wrote:
> >
> > From Valgrind experience there are enough genericism. Valgrind can translate
> > SSE, AltiVec and NEON instructions to vector opcodes. Most of the opcodes
> > are
> > reused between instruction sets.
>
> Doesn't Valgrind have the advantage of same-
From: Gonglei
This patch fixes compilation errors when building against glib < 2.16.0
due to the missing g_strcmp0() function.
Suggested-by: Peter Maydell
Signed-off-by: Gonglei
---
Because g_strcmp0() was called in three places, I provide
a back-compat implementation.
---
include/glib-com
On Mi, 2014-10-15 at 17:43 +0200, Michael Tokarev wrote:
> On 15.10.2014 12:10, Gerd Hoffmann wrote:
> >Hi,
> >
> > vmware-vga emulation lacks sanity checks in the hardware acceleration
> > (blit + fill) functions. This patch series plugs the holes.
> >
> > v2 changes:
> > * small whitespace
Hi Leon,
As the issues below are all corrected in the PRPL already,
Reviewed-by: Yongbok Kim
Regards,
Yongbok
On 08/07/2014 08:57, Leon Alrae wrote:
For Standard TLB configuration (Config.MT=1):
TLBINV invalidates a set of TLB entries based on ASID. The virtual address is
ignored in the e
On Thu, 16 Oct 2014, Paul Durrant wrote:
> > -Original Message-
> > From: Stefano Stabellini [mailto:stefano.stabell...@eu.citrix.com]
> > Sent: 15 October 2014 15:38
> > To: Paul Durrant
> > Cc: qemu-devel@nongnu.org; xen-de...@lists.xenproject.org; Stefano
> > Stabellini; Peter Maydell; P
Hi,
> > I try to prevent that by dropping the *oldest* connection, so you have a
> > chance to connect even if a unprivileged attacker tries to use up all
> > connection slots.
>
> Lets say the limit is 5. The bad guy has 5 open idle connections.
> The good guy opens a new one and pushes off on
On Thu, 9 Oct 2014 20:21:57 +0800
zhanghailiang wrote:
reword subj with:
numa: make 'info numa' take into account hotplugged memory
> When do memory hotplug, if there is numa node, we should add
> the memory size to the corresponding node memory size.
>
> For now, it mainly affects the result o
> -Original Message-
> From: Paul Durrant
> Sent: 16 October 2014 11:17
> To: 'Paolo Bonzini'; Peter Maydell
> Cc: QEMU Developers; xen-de...@lists.xenproject.org; Stefano Stabellini;
> Michael Tokarev; Stefan Hajnoczi; Stefan Weil; Olaf Hering; Gerd Hoffmann;
> Alexey Kardashevskiy; Alexan
Il 16/10/2014 12:16, Paul Durrant ha scritto:
>> What exactly is the right semantics? Note that save _can_ fail,
>> so you need the ability to roll back to the source machine. I
>> think this is missing from your patch, and there is no post_save
>> hook that you can use.
>
> I need something tha
On 16/10/2014 11:12, Aurelien Jarno wrote:
> Hi,
>
> On Wed, Oct 01, 2014 at 04:35:46PM +0100, Leon Alrae wrote:
>> Hi All,
>>
>> I noticed that it's quite difficult to get target-mips changes
>> reviewed/accepted. There is already a queue of relatively big features
>> and bug fixes which are stuc
> -Original Message-
> From: Paolo Bonzini [mailto:pbonz...@redhat.com]
> Sent: 16 October 2014 11:10
> To: Paul Durrant; Peter Maydell
> Cc: QEMU Developers; xen-de...@lists.xenproject.org; Stefano Stabellini;
> Michael Tokarev; Stefan Hajnoczi; Stefan Weil; Olaf Hering; Gerd Hoffmann;
> A
Hi,
On Wed, Oct 01, 2014 at 04:35:46PM +0100, Leon Alrae wrote:
> Hi All,
>
> I noticed that it's quite difficult to get target-mips changes
> reviewed/accepted. There is already a queue of relatively big features
> and bug fixes which are stuck for months. Does anyone have an idea how
> to impro
On 2014/10/16 18:04, Peter Maydell wrote:
> On 15 October 2014 11:05, Gerd Hoffmann wrote:
>> From: Gonglei
>>
>> Introduce del_boot_device_path() to clean up fw_cfg content when
>> hot-unplugging a device that refers to a bootindex or update a
>> existent devcie's bootindex.
>
>> +void del_boo
Il 16/10/2014 10:25, Paul Durrant ha scritto:
>>> +static void xen_hvm_pre_save(void *opaque)
>>> +{
>>> +XenIOState *state = opaque;
>>> +
>>> +/* Stop servicing emulation requests */
>>> +xen_set_ioreq_server_state(xen_xc, xen_domid, state->ioservid, 0);
>>> +xen_destroy_ioreq_ser
ping...
On 10/10/2014 10:15 AM, Gu Zheng wrote:
> Previously we use cpu_added_notifiers to register cpu hotplug notifier
> callback
> which is not able to pass/handle errors, so we switch it to unified hotplug
> handler API which allows to pass errors and would allow to cancel device_add
> in ca
On 15 October 2014 11:05, Gerd Hoffmann wrote:
> From: Gonglei
>
> Introduce del_boot_device_path() to clean up fw_cfg content when
> hot-unplugging a device that refers to a bootindex or update a
> existent devcie's bootindex.
> +void del_boot_device_path(DeviceState *dev, const char *suffix)
>
Kirill Batuzov writes:
>> (4) Consider supporting generic vector operations in the TCG?
>
> I gave it a go and was quite happy with the result. I have implemented the
> add_i32x4
> opcode which is addition of 128-bit vectors composed of four 32-bit integers
> and used it to translate NEON vadd.
On 16/10/2014 10:49, Peter Maydell wrote:
>
> One question for now:
>
>> Leon Alrae (17):
>> softfloat: add functions corresponding to IEEE-2008 min/maxNumMag
>
> Can you confirm that you're happy for your softfloat changes
> to be licensed under both of the softfloat-2a and softfloat-2b
> lic
> -Original Message-
> From: Stefano Stabellini [mailto:stefano.stabell...@eu.citrix.com]
> Sent: 15 October 2014 15:38
> To: Paul Durrant
> Cc: qemu-devel@nongnu.org; xen-de...@lists.xenproject.org; Stefano
> Stabellini; Peter Maydell; Paolo Bonzini; Michael Tokarev; Stefan Hajnoczi;
> Ste
On 15 October 2014 11:53, Leon Alrae wrote:
> Hi,
>
> This pull request has been assembled from pending target-mips patches which
> look good to me and received in my opinion sufficient review comments. They
> were tested mainly in context of MIPS. Please have a look and pull.
Thanks for putting
affects: ubuntu/qemu
importance: high
affects: qemu
importance: high
** Also affects: qemu (Ubuntu)
Importance: High
Status: New
** Also affects: qemu
Importance: High
Status: New
** No longer affects: qemu-kvm (Ubuntu)
--
You received this bug notification because you
Ping!
At KVM Forum I had a discussion with (someone, sorry!) that having some
pointers to which specifications to look at here might be helpful, since
some of the fixes were just spec-adherence fixes.
See below, in-line, for some additional notes on how to review these
patches.
On 10/02/20
On 15 October 2014 11:37, Gerd Hoffmann wrote:
> Hi,
>
> vga patch queue with two patches, one deleting some unused bits and one
> adding the qext region to stdvga mmio to allow switching vga framebuffer
> endianess (needed for ppc64le).
>
> please pull,
> Gerd
>
> The following changes since
On 2014/10/16 16:35, Dr. David Alan Gilbert wrote:
* zhanghailiang (zhang.zhanghaili...@huawei.com) wrote:
+case MIG_RPCOMM_ACK:
+tmp32 = be32_to_cpup((uint32_t *)buf);
+DPRINTF("RP: Received ACK 0x%x", tmp32);
+atomic_xchg(&ms->rp_state.latest_ack, t
Introduce INDEX_op_add_i32x4 opcode which adds two 128-bit variables as vectors
of four 32-bit integers.
Add tcg_gen_add_i32x4 wrapper function that generates this opcode. If a TCG
target
does not support it, the wrapper falls back to emulation of vector operation as
a series of scalar ones. Wrap
To be able to generate vector operations in TCG backend we need to do several
things.
1. We need to tell the register allocator about the target's vector registers.
In the case of x86 we'll use xmm0..xmm7. xmm7 is designated as a scratch
register, others can be used by register allocator.
2
> (4) Consider supporting generic vector operations in the TCG?
I gave it a go and was quite happy with the result. I have implemented the
add_i32x4
opcode which is addition of 128-bit vectors composed of four 32-bit integers
and used it to translate NEON vadd.i32 to SSE paddd instruction. I used
Signed-off-by: Kirill Batuzov
---
target-arm/translate.c | 12
1 file changed, 12 insertions(+)
diff --git a/target-arm/translate.c b/target-arm/translate.c
index 22855d8..00ea5cf 100644
--- a/target-arm/translate.c
+++ b/target-arm/translate.c
@@ -5239,6 +5239,18 @@ static int d
Introduce TCG_TYPE_V128 and corresponding TCGv_v128 for TCG temps. Add wrapper
functions that work with temps of this new type.
Signed-off-by: Kirill Batuzov
---
tcg/tcg-op.h | 23 +++
tcg/tcg.c| 24
tcg/tcg.h| 28 +++
When a TCG backend does not support some vector operation we need to emulate
this operation. Unlike arguments of the scalar operations vector values are
hard to operate on directly or to be passed as function arguments (because
a target may lack corresponding type support). To avoid this we will us
Currently every field of CPUArchState can be accessed from the TCG-generated
code
as a memory location or as a global but not both. In order to be able to mix
these two approaches we need to restore consistency between value of global
(possibly kept on register) and value in corresponding memory l
To support 128-bit guest registers as globals we need to do two things:
1) create corresponding globals,
2) add sync_temp/discard to code that access these registers as memory
locations.
Note that the second part is not complete in this RFC yet and mixing NEON with
VFP code can result in misco
* zhanghailiang (zhang.zhanghaili...@huawei.com) wrote:
> >+case MIG_RPCOMM_ACK:
> >+tmp32 = be32_to_cpup((uint32_t *)buf);
> >+DPRINTF("RP: Received ACK 0x%x", tmp32);
> >+atomic_xchg(&ms->rp_state.latest_ack, tmp32);
>
> I didn't see *ms->rp_state.lat
Am 22.09.2014 um 17:36 hat Max Reitz geschrieben:
> As its comment states, raw_co_get_block_status() should unconditionally
> return 0 and set *pnum to 0 for after EOF.
>
> An assertion after lseek(..., SEEK_HOLE) tried to catch this case by
> asserting that errno != -ENXIO (which would indicate a
On 2014/10/4 1:47, Dr. David Alan Gilbert (git) wrote:
From: "Dr. David Alan Gilbert"
Open a return path, and handle messages that are received upon it.
Signed-off-by: Dr. David Alan Gilbert
---
include/migration/migration.h | 10 +++
migration.c | 181 ++
On 15 October 2014 11:05, Gerd Hoffmann wrote:
> Hi,
>
> It's finally sorted, here comes the pull request for the bootindex patch
> series which turns bootindex into a writable qom property and thereby
> allows changing the bootorder at runtime via monitor.
>
> please pull,
> Gerd
>
> The foll
> -Original Message-
> From: Paolo Bonzini [mailto:pbonz...@redhat.com]
> Sent: 16 October 2014 08:37
> To: Peter Maydell; Paul Durrant
> Cc: QEMU Developers; xen-de...@lists.xenproject.org; Stefano Stabellini;
> Michael Tokarev; Stefan Hajnoczi; Stefan Weil; Olaf Hering; Gerd Hoffmann;
> A
> -Original Message-
> From: Peter Maydell [mailto:peter.mayd...@linaro.org]
> Sent: 15 October 2014 18:30
> To: Paul Durrant
> Cc: QEMU Developers; xen-de...@lists.xenproject.org; Stefano Stabellini;
> Paolo Bonzini; Michael Tokarev; Stefan Hajnoczi; Stefan Weil; Olaf Hering;
> Gerd Hoffma
Am 08.10.2014 um 19:55 hat Max Reitz geschrieben:
> Export names may be used with nbd+unix, too, fix nbd_refresh_filename()
> accordingly. Also, for nbd+tcp, the documented path schema is
> "nbd://host[:port]/export", so use it. Furthermore, as can be seen from
> that schema, the port is optional.
* Juan Quintela (quint...@redhat.com) wrote:
> "Dr. David Alan Gilbert (git)" wrote:
> G> From: "Dr. David Alan Gilbert"
> >
> > The migration code now occupies a fair chunk of the top level .c
> > files, it seems time to give it it's own directory.
> >
> > I've not touched:
> >arch_init.c -
"Dr. David Alan Gilbert (git)" wrote:
> From: "Dr. David Alan Gilbert"
>
> The splitting of qemu-file and addition of the buffered file landed
> at the same time; so now split the buffered file code out.
>
> Signed-off-by: Dr. David Alan Gilbert
Reviewed-by: Juan Quintela
1 - 100 of 118 matches
Mail list logo
