Hi all,
this is a patch on top of "[PATCH 4/5] linux-user: fake
/proc/self/stat" (sorry I couldn't find the git repo bound to this
patchwork) that also implements PID and binary name reading them from
the fake TSS.
The pid was just a "why not" while the binary name is required (as
well as the backe
virtio-mmio: fix features negotiation problem
virtio-mmio should invoke virtio_set_features() when it gets
VIRTIO_MMIO_GUESTFEATURES command, since some devices (at least, virtio-net)
register their own set_features method and setup their features in that.
Signed-off-by: Ying-Shiuan Pan
---
hw/
virtio-mmio: fix switch-case in virtio_mmio_write()
The switch-case in virtio_mmio_write() forgets to break when it
finishes a case statement.
Signed-off-by: Ying-Shiuan Pan
---
hw/virtio-mmio.c |3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/hw/virtio-mmio.c b/hw/vir
Hi, all
I'm not sure if it is appropriate to put patches here because
virtio-mmio is actually not in official repository yet.
In case that someone else wants to test virtio-mmio as I did, I decide
to send them out.
Currently, I've successfully tested virtio-blk, virtio-net, and virtio-balloon.
Y
2012/1/3 Stefan Hajnoczi :
> On Mon, Jan 2, 2012 at 5:19 AM, Ying-Shiuan Pan
> wrote:
>> 2011/12/30 Stefan Hajnoczi
>>>
>>> On Wed, Dec 28, 2011 at 06:16:42PM +0800, Ying-Shiuan Pan wrote:
>>> > I'm very interested in virtio-mmio Peter Maydell did for QEMU,
>>> > (http://lists.nongnu.org/archive/
From: Khansa Butt
This patch includes sigcontext for mips64 suggested by Andreas Färber and
sa_flags byte
swap fix by Richard Henderson setup_frame() and do_sigreturn need non RT
syscall numbers
which do not exist in linux 64 bit syscalls. so both function should have
different implementation
From: Khansa Butt
This patch includes sigcontext for mips64 suggested by Andreas Färber and
sa_flags byte
swap fix by Richard Henderson setup_frame() and do_sigreturn need non RT
syscall numbers
which do not exist in linux 64 bit syscalls. so both function should have
different implementation
From: Khansa Butt
MIPS_HFLAG_UX is included in env->hflags so that the address computation
for LD instruction does not get treated as 32 bit code, see gen_op_addr_add()
in translate.c.
Signed-off-by: Khansa Butt
---
target-mips/translate.c |3 +++
1 files changed, 3 insertions(+), 0 delet
From: Khansa Butt
This patch incorporates
1)default configs for mips64
2)updated target-list with mips64-linux-user
3)syscall differences: Linux 64-bit syscalls have different number range and
because of n64 ABI, 8 argument registers are suffient for do_syscall()
4)define TARGET_QEMU_ESIGRETURN
From: Khansa Butt
in translate_init.c fpu_init() initializes active_fpu for given cpu
model afterwards cpu_reset() resets all the members upto breakpoints
due to this whatever the value of cpu_model->CR1_fcr0 was , the value of
env->active_fpu.fcr0 is zero now. This patch resolve this issue so t
From: Khansa Butt
in translate_init.c fpu_init() initializes active_fpu for given cpu
model afterwards cpu_reset() resets all the members upto breakpoints
due to this whatever the value of cpu_model->CR1_fcr0 was , the value of
env->active_fpu.fcr0 is zero now. This patch resolve this issue so t
From: Khansa Butt
This patch incorporates
1)default configs for mips64
2)updated target-list with mips64-linux-user
3)syscall differences: Linux 64-bit syscalls have different number range and
because of n64 ABI, 8 argument registers are suffient for do_syscall()
4)define TARGET_QEMU_ESIGRETURN
From: Khansa Butt
[sorry version missed in previous set]
This is the team work of Ehsan-ul-Haq, Abdul Qadeer, Abdul Waheed, Khansa Butt
from HPCN Lab KICS UET Lahore.
In previous patch set we were including Cavium specific instructions along with
Cavium specifc registers in UME. Because of these
From: Khansa Butt
MIPS_HFLAG_UX is included in env->hflags so that the address computation
for LD instruction does not get treated as 32 bit code, see gen_op_addr_add()
in translate.c.
Signed-off-by: Khansa Butt
---
target-mips/translate.c |3 +++
1 files changed, 3 insertions(+), 0 delet
From: Khansa Butt
This is the team work of Ehsan-ul-Haq, Abdul Qadeer, Abdul Waheed, Khansa Butt
from HPCN Lab KICS UET Lahore.
In previous patch set we were including Cavium specific instructions along with
Cavium specifc registers in UME. Because of these register fields we had to bump
the cpu
Right now, DeviceInfo acts as the class for qdev. In order to switch to a
proper ObjectClass derivative, we need to ween all of the callers off of
interacting directly with the info pointer.
Signed-off-by: Anthony Liguori
---
hw/acpi_piix4.c|4 +-
hw/cirrus_vga.c|2 +-
h
We already have a QOM interface for this so let's use it.
Signed-off-by: Anthony Liguori
---
hw/e1000.c |2 +-
hw/eepro100.c |2 +-
hw/etraxfs_eth.c|2 +-
hw/hda-audio.c |2 +-
hw/intel-hda.c |2 +-
hw/lan9118.c|
This is a "shallow", half hearted, and untested conversion.
Signed-off-by: Anthony Liguori
---
hw/macio.c | 43 ++-
1 files changed, 30 insertions(+), 13 deletions(-)
diff --git a/hw/macio.c b/hw/macio.c
index cc6ae40..357e0ea 100644
--- a/hw/macio.c
++
Signed-off-by: Anthony Liguori
---
hw/applesmc.c| 18 --
hw/cs4231a.c | 20 +---
hw/debugcon.c| 16 +++-
hw/fdc.c | 24 +++-
hw/gus.c | 20 +---
hw/i8254.c | 22 ++
On 01/02/2012 04:41 PM, Andreas Färber wrote:
Am 20.12.2011 17:51, schrieb Anthony Liguori:
I have no idea what's going on here, but this is broken and cannot possibly
work because:
1) It's qdev name conflicts with a device that sits on SysBus
2) The PCI constructor expects a SysBus device
Signed-off-by: Anthony Liguori
---
hw/s390-virtio-bus.c | 91 +++---
1 files changed, 64 insertions(+), 27 deletions(-)
diff --git a/hw/s390-virtio-bus.c b/hw/s390-virtio-bus.c
index a047bbf..3bce5e4 100644
--- a/hw/s390-virtio-bus.c
+++ b/hw/s390-vi
This converts two types because smbus is implemented as a subclass of i2c. It's
extremely difficult to convert these two independently.
Signed-off-by: Anthony Liguori
---
hw/ds1338.c | 21 +
hw/i2c.c | 83 +---
hw/i2
This class provides the main building block for QEMU Object Model and is
extensively documented in the header file. It is largely inspired by GObject.
Signed-off-by: Anthony Liguori
---
v1 -> v2
- remove printf() in type registration
- fix typo in comment (Paolo)
- make Interface private
- m
Signed-off-by: Anthony Liguori
---
hw/ccid-card-emulated.c | 27 ++--
hw/ccid-card-passthru.c | 27 ++--
hw/ccid.h | 28 +++--
hw/usb-ccid.c | 76 +++
4 files changed, 109 insertions(+)
These are various small stylistic changes which help make things more
consistent such that the automated conversion script can be simpler.
It's not necessary to agree or disagree with these style changes because all
of this code is going to be rewritten by the patch monkey script anyway.
Signed-o
Signed-off-by: Anthony Liguori
---
hw/usb-bt.c | 33 ++
hw/usb-bus.c| 179 +++
hw/usb-ccid.c | 38 +++-
hw/usb-desc.c | 18 +++---
hw/usb-hid.c| 108 --
hw/usb-hub.c|
From: David Gibson
For forgotten historical reasons, PAPR hypercalls for specific virtual IO
devices (oh which there are quite a number) are registered via a callback
in the VIOsPAPRDeviceInfo structure.
This is kind of ugly, so this patch instead registers hypercalls from
device_init() function
This is series 2/4 of the QOM refactoring. These series are divided up based
on the major scripted code conversions.
This series makes qdev a proper Object and converts qdev's type inheritance to
QOM inheritance.
The first half of the series are manual cleanups/refactorings. The second half
is
Same problem as with grackle. This code can't possibly work.
Signed-off-by: Anthony Liguori
---
hw/unin_pci.c | 77 -
1 files changed, 0 insertions(+), 77 deletions(-)
diff --git a/hw/unin_pci.c b/hw/unin_pci.c
index 14d9914..b3365fe 10
Signed-off-by: Anthony Liguori
---
hw/intel-hda.c |4 +---
hw/lsi53c895a.c |2 +-
hw/qdev.c | 22 +-
hw/qdev.h |8
4 files changed, 27 insertions(+), 9 deletions(-)
diff --git a/hw/intel-hda.c b/hw/intel-hda.c
index 8cb92d3..6e1c5de 100644
Signed-off-by: Anthony Liguori
---
hw/scsi-bus.c | 95 +++-
hw/scsi-disk.c| 159 -
hw/scsi-generic.c | 29 ++
hw/scsi.h | 31 ++
4 files changed, 201 insertions(+), 113 deletions
Signed-off-by: Anthony Liguori
---
hw/ds1338.c | 10 +-
hw/hw.h |4 ++--
hw/i2c.c | 30 +++---
hw/i2c.h | 18 ++
hw/lm832x.c | 10 +-
hw/max7310.c | 10 +-
hw/pxa2xx.c | 10 +-
hw/smbus.
Signed-off-by: Anthony Liguori
---
hw/hda-audio.c | 58 ++-
hw/intel-hda.c | 31 +
hw/intel-hda.h | 26
3 files changed, 71 insertions(+), 44 deletions(-)
diff --git a/hw/hda-audio.c b/
Signed-off-by: Anthony Liguori
---
hw/ads7846.c | 17 -
hw/max111x.c | 34 --
hw/spitz.c | 39 +++
hw/ssd0323.c | 17 -
hw/ssi-sd.c| 17 -
hw/ssi.c |
This is a very shallow integration. We register a TYPE_DEVICE but only use
QOM as basically a memory allocator. This will make all devices show up as
QOM objects but they will all carry the TYPE_DEVICE.
Signed-off-by: Anthony Liguori
---
v1 -> v2
- update for new location of object.h
---
hw/q
Signed-off-by: Anthony Liguori
---
hw/ide/internal.h | 20 ++---
hw/ide/qdev.c | 111 +
2 files changed, 82 insertions(+), 49 deletions(-)
diff --git a/hw/ide/internal.h b/hw/ide/internal.h
index 00b28df..c808a0d 100644
--- a/hw/ide/
I have no idea what's going on here, but this is broken and cannot possibly
work because:
1) It's qdev name conflicts with a device that sits on SysBus
2) The PCI constructor expects a SysBus device... but's it's a PCIDevice
If you do qdev_create("grackle"), the code expects to get the SysBus
Signed-off-by: Anthony Liguori
---
v1 -> v2
- use QOM to check for the default console
---
hw/spapr_llan.c | 37 +++--
hw/spapr_vio.c | 47 ---
hw/spapr_vio.h | 35 ++-
hw/spapr
Since we are still dynamically creating TypeInfo, we need to chain the
class_init function in order to be able to make use of it within subclasses of
TYPE_DEVICE.
This will disappear once we register TypeInfos directly.
Signed-off-by: Anthony Liguori
---
hw/qdev.c |4
hw/qdev.h |5
Signed-off-by: Anthony Liguori
---
hw/virtio-console.c| 73 ++
hw/virtio-serial-bus.c | 66 +++---
hw/virtio-serial.h | 81 ++--
3 files changed, 129 insertions(+), 91
Untested beyond compile. But it's a very simple conversion.
Signed-off-by: Anthony Liguori
---
hw/prep_pci.c | 38 +++---
1 files changed, 27 insertions(+), 11 deletions(-)
diff --git a/hw/prep_pci.c b/hw/prep_pci.c
index ea9fb69..16c036f 100644
--- a/hw/prep_
This board never worked with TCG. It hasn't been updated since 0.13.0. I'm
fairly sure hardware doesn't exist anymore that you can run the KVM support
with.
So let's remove it. It can always be restored later if there is interest again.
Signed-off-by: Anthony Liguori
---
Makefile.target|
On 01/02/2012 11:59 AM, Paolo Bonzini wrote:
On 12/20/2011 05:51 PM, Anthony Liguori wrote:
hw/object.c | 469 +
hw/object.h | 427 +++
A couple more remarks:
1) Please put it outside hw/
D
On 01/02/2012 09:54 AM, Peter Maydell wrote:
On 2 January 2012 13:46, Andreas Färber wrote:
* when can we expect to be able to model SoCs rather than CPUs? Will
this affect command line usage - are we going to have '-device
ti-tms570' rather than '-cpu cortex-r4' then, or -cpu overriding the
co
On 01/02/2012 08:11 AM, Paolo Bonzini wrote:
On 01/02/2012 02:46 PM, Andreas Färber wrote:
QOM: If Anthony is available, I'd be interested in hearing an update on
the roadmap. In particular,
* when can we expect to be able to model SoCs rather than CPUs? Will
this affect command line usage - are
On 01/02/2012 07:46 AM, Andreas Färber wrote:
Am 02.01.2012 13:09, schrieb Juan Quintela:
First of all, Happy New Year to everybody (even for the people whose
calendar is different O:-)
+1
Please send in any agenda items you are interested in covering.
QOM: If Anthony is available, I'd be
Because now all PCI devices are converted to qdev.
Signed-off-by: Anthony Liguori
---
hw/pci.c |7 ++-
1 files changed, 2 insertions(+), 5 deletions(-)
diff --git a/hw/pci.c b/hw/pci.c
index 399227f..fd575ac 100644
--- a/hw/pci.c
+++ b/hw/pci.c
@@ -159,11 +159,8 @@ void pci_device_deass
On 01/02/2012 05:01 PM, Andreas Färber wrote:
Am 22.12.2011 20:57, schrieb Anthony Liguori:
On 12/22/2011 12:00 PM, Kevin O'Connor wrote:
On Thu, Dec 22, 2011 at 11:41:08AM -0600, Anthony Liguori wrote:
On 12/22/2011 11:25 AM, Kevin O'Connor wrote:
Why not declare types with something like th
On 01/02/2012 04:55 PM, Andreas Färber wrote:
Am 20.12.2011 17:51, schrieb Anthony Liguori:
Obviously, this isn't an acceptable approach. We need to convert these devices
to qdev.
---
hw/macio.c |5 +
hw/openpic.c|5 +
hw/ppc4xx_pci.c |5 +
hw/prep_pci.c
In order to introduce inheritance while still using the qdev registration
interfaces, we need to be able to use a parent other than TYPE_DEVICE. Add a
new interface that allows this.
Signed-off-by: Anthony Liguori
---
hw/qdev.c |9 +++--
hw/qdev.h |1 +
2 files changed, 8 insertions
bus is always NULL so the code in this if clause is dead (and therefore
untested).
Signed-off-by: Anthony Liguori
---
hw/openpic.c | 30 +++---
hw/openpic.h |2 +-
hw/ppc_newworld.c |2 +-
3 files changed, 5 insertions(+), 29 deletions(-)
diff --git a
Am 22.12.2011 20:57, schrieb Anthony Liguori:
> On 12/22/2011 12:00 PM, Kevin O'Connor wrote:
>> On Thu, Dec 22, 2011 at 11:41:08AM -0600, Anthony Liguori wrote:
>>> On 12/22/2011 11:25 AM, Kevin O'Connor wrote:
Why not declare types with something like the following:
TypeInfo my_dev
Am 20.12.2011 17:51, schrieb Anthony Liguori:
> Obviously, this isn't an acceptable approach. We need to convert these
> devices
> to qdev.
> ---
> hw/macio.c |5 +
> hw/openpic.c|5 +
> hw/ppc4xx_pci.c |5 +
> hw/prep_pci.c |5 +
> 4 files changed, 20
Am 20.12.2011 17:51, schrieb Anthony Liguori:
> Same problem as with grackle. This code can't possibly work.
Nack. Cf. grackle, it does work (-M mac99).
Solution should be renaming, not breaking.
Andreas
--
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Jeff Hawn, Jennife
Am 20.12.2011 17:51, schrieb Anthony Liguori:
> I have no idea what's going on here, but this is broken and cannot possibly
> work because:
>
> 1) It's qdev name conflicts with a device that sits on SysBus
>
> 2) The PCI constructor expects a SysBus device... but's it's a PCIDevice
>
> If you
On Mon, Jan 2, 2012 at 3:39 PM, Christoph Hellwig wrote:
> On Fri, Dec 30, 2011 at 10:35:01AM +, Stefan Hajnoczi wrote:
>> If you can reproduce this bug and suspect coroutines are involved then I
>
> It's entirely reproducable.
>
> I've played around a bit and switched from the ucontext to the
On 01/03/2012 03:33 AM, Avi Kivity wrote:
> Avi Kivity (16):
> memory: move endianness compensation to memory core
> exec: make phys_page_find() return a temporary
> memory: move mmio access to functions
> memory: remove MemoryRegion::backend_registered
> Fix wrong region_offset when over
On 01/03/2012 03:33 AM, Avi Kivity wrote:
> -if (is_ram_rom_romd(pd)) {
> +if (!is_ram_rom_romd(pd)) {
Should be merged with a previous patch.
r~
On 01/03/2012 03:33 AM, Avi Kivity wrote:
> -if ((pd & ~TARGET_PAGE_MASK) > IO_MEM_ROM && !(pd & IO_MEM_ROMD)) {
> +if (!is_ram_rom(pd) && !(pd & IO_MEM_ROMD)) {
> /* IO memory case (romd handled later) */
Perhaps (!is_ram_rom_romd(pd))?
At least that's what I see from DeMorgan's
Signed-off-by: Paolo Bonzini
---
hw/9pfs/virtio-9p-device.h |2 ++
hw/virtio.h|6 --
2 files changed, 2 insertions(+), 6 deletions(-)
diff --git a/hw/9pfs/virtio-9p-device.h b/hw/9pfs/virtio-9p-device.h
index 7f2a4eb..28b0012 100644
--- a/hw/9pfs/virtio-9p-device.h
++
These patches simplify the jungle of lists provided by qemu-queue from
4 to 3. QCIRCLEQ is dropped, since it provides no real advantage over
QTAILQ. QSIMPLEQ is simplified to no longer permit insertion at the
tail, with the advantage that it is more suited for simple free lists.
Stefan, these ar
Signed-off-by: Paolo Bonzini
---
hw/9pfs/virtio-9p-device.h |5 +
hw/virtio-pci.c|4 +---
2 files changed, 6 insertions(+), 3 deletions(-)
diff --git a/hw/9pfs/virtio-9p-device.h b/hw/9pfs/virtio-9p-device.h
index 65789db..7f2a4eb 100644
--- a/hw/9pfs/virtio-9p-device.h
+
CCID emulation needs to insert at tail. Use a QTAILQ, even though
the double links are strictly not necessary.
Signed-off-by: Paolo Bonzini
---
hw/ccid-card-emulated.c | 24
1 files changed, 12 insertions(+), 12 deletions(-)
diff --git a/hw/ccid-card-emulated.c b/hw/
Signed-off-by: Paolo Bonzini
---
hw/virtio-net.h |3 +++
hw/virtio.h |3 ---
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/hw/virtio-net.h b/hw/virtio-net.h
index ee55a85..96cd8b1 100644
--- a/hw/virtio-net.h
+++ b/hw/virtio-net.h
@@ -168,6 +168,9 @@ struct virtio_net
Signed-off-by: Paolo Bonzini
---
hw/s390-virtio-bus.c |3 +--
hw/virtio-pci.c |4 +---
hw/virtio-serial.h |5 +
3 files changed, 7 insertions(+), 5 deletions(-)
diff --git a/hw/s390-virtio-bus.c b/hw/s390-virtio-bus.c
index 80b11e5..282688a 100644
--- a/hw/s390-virtio-bus.
Signed-off-by: Paolo Bonzini
---
hw/s390-virtio-bus.c |2 +-
hw/virtio-blk.h |5 +
hw/virtio-pci.c |3 +--
3 files changed, 7 insertions(+), 3 deletions(-)
diff --git a/hw/s390-virtio-bus.c b/hw/s390-virtio-bus.c
index cf7cfef..71720b9 100644
--- a/hw/s390-virtio-bus.c
QED needs to insert at tail. Use a QTAILQ, even though
the double links are strictly not necessary.
Signed-off-by: Paolo Bonzini
---
block/qed.c | 20 ++--
block/qed.h |4 ++--
2 files changed, 12 insertions(+), 12 deletions(-)
diff --git a/block/qed.c b/block/qed.c
index
Signed-off-by: Paolo Bonzini
---
hw/virtio-serial.h |7 +--
hw/virtio.h|3 ---
2 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/hw/virtio-serial.h b/hw/virtio-serial.h
index 2e1a428..8469956 100644
--- a/hw/virtio-serial.h
+++ b/hw/virtio-serial.h
@@ -45,10 +45,1
Signed-off-by: Paolo Bonzini
---
hw/virtio-pci.h |2 +-
hw/virtio.h |4 ++--
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/hw/virtio-pci.h b/hw/virtio-pci.h
index e0df6ef..2edd249 100644
--- a/hw/virtio-pci.h
+++ b/hw/virtio-pci.h
@@ -34,7 +34,7 @@ typedef struct {
The main advantage of circular lists (the fact that the head node
has the same memory layout as any other node, so you do not need
conditionals) is completely negated by the implementation in qemu-queue.h.
Not surprisingly, nobody uses QCIRCLEQ. While this might change if RCU
is ever adopted by QE
Block migration needs to insert at tail. Use a QTAILQ, even though
the double links are strictly not necessary.
Signed-off-by: Paolo Bonzini
---
block-migration.c | 44 ++--
1 files changed, 22 insertions(+), 22 deletions(-)
diff --git a/block-migratio
Signed-off-by: Paolo Bonzini
---
hw/s390-virtio-bus.c |4 ++--
hw/s390-virtio-bus.h |3 ++-
hw/virtio-blk.c | 13 ++---
hw/virtio-blk.h |5 +
hw/virtio-pci.c |4 ++--
hw/virtio-pci.h |3 ++-
hw/virtio.h |3 ++-
7 files changed, 21
QSIMPLEQ is still relatively heavyweight when used as a free list,
compared to a simple singly-linked list. One disadvantage is that
it requires an initializer macro, unlike for example QLIST.
The patch removes the double links so that there is a "really really
simple" queue type.
Signed-off-by:
Also move the 9p.h file to 9pfs/virtio-9p-device.h, for consistency
with the corresponding .c file.
Signed-off-by: Paolo Bonzini
---
hw/9pfs/virtio-9p-device.c | 36 +
hw/{9p.h => 9pfs/virtio-9p-device.h} |4 +-
hw/9pfs/virtio-9p.c
Signed-off-by: Paolo Bonzini
---
hw/virtio-blk.h |7 +--
hw/virtio.h |7 ---
2 files changed, 5 insertions(+), 9 deletions(-)
diff --git a/hw/virtio-blk.h b/hw/virtio-blk.h
index 518e1cc..70d939a 100644
--- a/hw/virtio-blk.h
+++ b/hw/virtio-blk.h
@@ -97,11 +97,14 @@ struct vi
QSIMPLEQ is now the best candidate for a free list, use it.
Signed-off-by: Paolo Bonzini
---
coroutine-ucontext.c | 10 +-
qemu-coroutine-int.h |2 +-
2 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/coroutine-ucontext.c b/coroutine-ucontext.c
index 3d01075..65b0bb7 1
Also call blockdev_mark_auto_del from virtio_blk_exit, rather than
just virtio_blk_exit_pci.
Signed-off-by: Paolo Bonzini
---
hw/s390-virtio-bus.c |4 +---
hw/s390-virtio-bus.h |1 -
hw/virtio-blk.c | 17 +
hw/virtio-blk.h |2 ++
hw/virtio-pci.c |
QSIMPLEQ is equivalent to an open-coded free list, use it.
Signed-off-by: Paolo Bonzini
---
block.c |9 -
block_int.h |4 ++--
2 files changed, 6 insertions(+), 7 deletions(-)
diff --git a/block.c b/block.c
index 3f072f6..10e80d7 100644
--- a/block.c
+++ b/block.c
@@ -3242,9
Signed-off-by: Paolo Bonzini
---
hw/s390-virtio-bus.c |3 +--
hw/s390-virtio-bus.h |1 -
hw/virtio-net.c | 13 ++---
hw/virtio-net.h |2 ++
hw/virtio-pci.c |3 +--
hw/virtio-pci.h |1 -
hw/virtio.h |3 +--
7 files changed, 11 insertio
Notifiers do not need to access both ends of the list, and using
a QLIST also simplifies the API.
Signed-off-by: Paolo Bonzini
---
I do plan to exploit the simplified API in a future patch...
input.c |2 +-
migration.c |2 +-
notify.c | 10 +-
notify.h |
Signed-off-by: Paolo Bonzini
---
hw/s390-virtio-bus.h |8 +---
hw/virtio-pci.h | 10 ++
2 files changed, 11 insertions(+), 7 deletions(-)
diff --git a/hw/s390-virtio-bus.h b/hw/s390-virtio-bus.h
index 414fd22..b9f711d 100644
--- a/hw/s390-virtio-bus.h
+++ b/hw/s390-virtio-
Signed-off-by: Paolo Bonzini
---
hw/virtio-balloon.h |3 +++
hw/virtio-pci.h |1 +
hw/virtio.h |2 --
3 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/hw/virtio-balloon.h b/hw/virtio-balloon.h
index 73300dd..cde2e67 100644
--- a/hw/virtio-balloon.h
+++ b/hw/v
Signed-off-by: Paolo Bonzini
---
hw/s390-virtio-bus.c |6 +-
hw/virtio-net.h | 41 ++---
hw/virtio-pci.c |7 +--
3 files changed, 24 insertions(+), 30 deletions(-)
diff --git a/hw/s390-virtio-bus.c b/hw/s390-virtio-bus.c
index c4b9a
This series cleans up device configuration in virtio:
1) all properties (features and not) are moved to a single macro
DEFINE_VIRTIO_XYZ_PROPERTIES. This makes the definitions
uniform across all three virtio backends.
2) all configuration is moved to a single struct, which embeds
the NICConf/Blo
On 12/20/2011 05:51 PM, Anthony Liguori wrote:
hw/object.c | 469 +
hw/object.h | 427 +++
A couple more remarks:
1) Please put it outside hw/
2) Why do you need a list for the inte
On 01/02/2012 06:05 PM, Andrea Arcangeli wrote:
On Thu, Dec 29, 2011 at 06:01:45PM +0200, Avi Kivity wrote:
On 12/29/2011 06:00 PM, Avi Kivity wrote:
The NFS client has exactly the same issue, if you mount it with the intr
option. In fact you could use the NFS client as a trivial umem/cuse
pro
Am 01.01.2012 18:23, schrieb Avi Kivity:
> cpu_register_physical_memory_log() does not update region_offset
> if a page was previously registered for the same address. This
> could cause mmio accesses going to the wrong place, by using the
> old region_offset.
>
> Signed-off-by: Avi Kivity
Line
Instead of doing device endianness compensation in cpu_register_io_memory(),
do it in the memory core.
Signed-off-by: Avi Kivity
---
exec-obsolete.h |2 +-
exec.c | 142 ---
memory.c| 37 --
3 files changed,
Instead of returning a PhysPageDesc pointer, return a temporary.
This lets us move away from actually storing PhysPageDesc's, and
instead sythesising them when needed.
Signed-off-by: Avi Kivity
---
exec.c | 148 +---
1 files changed, 4
Signed-off-by: Avi Kivity
---
cpu-common.h |1 -
exec.c | 72 +++--
2 files changed, 24 insertions(+), 49 deletions(-)
diff --git a/cpu-common.h b/cpu-common.h
index fd787b9..98dddfe 100644
--- a/cpu-common.h
+++ b/cpu-common.h
@@ -
The code sometimes uses range comparisons on io indexes (e.g.
index =< IO_MEM_ROM). Avoid these as they make moving to objects harder.
Signed-off-by: Avi Kivity
---
exec-all.h |2 +-
exec.c | 37 -
memory.c |3 ++-
soft
On Mon, Jan 2, 2012 at 9:28 AM, Dor Laor wrote:
> At the end of the day, it's performance testing using various scenarios that
> can educate us whether post-copy worth the extra complexity over slowing
> down the guest on the source.
True. It's certainly an interesting patch series to benchmark
Now that all mmio goes through MemoryRegions, we can convert
io_mem_opaque to be a MemoryRegion pointer, and remove the thunks
that convert from old-style CPU{Read,Write}MemoryFunc to MemoryRegionOps.
Signed-off-by: Avi Kivity
---
exec-all.h |4 +-
exec-obsolete.h |5 +-
exec.c
On Mon, Jan 2, 2012 at 5:19 AM, Ying-Shiuan Pan
wrote:
> 2011/12/30 Stefan Hajnoczi
>>
>> On Wed, Dec 28, 2011 at 06:16:42PM +0800, Ying-Shiuan Pan wrote:
>> > I'm very interested in virtio-mmio Peter Maydell did for QEMU,
>> > (http://lists.nongnu.org/archive/html/qemu-devel/2011-11/msg01870.htm
Signed-off-by: Avi Kivity
---
exec.c | 70 ---
1 files changed, 18 insertions(+), 52 deletions(-)
diff --git a/exec.c b/exec.c
index c8c839a..fb8c23e 100644
--- a/exec.c
+++ b/exec.c
@@ -2472,6 +2472,7 @@ static inline void tlb_set_di
Its use of IO_MEM_ROM and friends will later cause #include loops; and it
is too large to merit inlining.
Signed-off-by: Avi Kivity
---
exec-all.h | 26 +-
exec.c | 26 ++
2 files changed, 27 insertions(+), 25 deletions(-)
diff --git a/exe
Currently mmio access goes directly to the io_mem_{read,write} arrays.
In preparation for eliminating them, add indirection via a function.
Signed-off-by: Avi Kivity
---
exec-all.h |7 -
exec.c | 54 ++--
memory.c
cpu_register_physical_memory_log() does not update region_offset
if a page was previously registered for the same address. This
could cause mmio accesses going to the wrong place, by using the
old region_offset.
Signed-off-by: Avi Kivity
---
exec.c |1 +
1 files changed, 1 insertions(+), 0
On Thu, Dec 29, 2011 at 06:01:45PM +0200, Avi Kivity wrote:
> On 12/29/2011 06:00 PM, Avi Kivity wrote:
> > The NFS client has exactly the same issue, if you mount it with the intr
> > option. In fact you could use the NFS client as a trivial umem/cuse
> > prototype.
>
> Actually, NFS can return
Still internally using ram_addr.
Signed-off-by: Avi Kivity
---
exec-obsolete.h | 24 +++-
exec.c | 24 +++-
memory.c| 44
3 files changed, 38 insertions(+), 54 deletions(-)
diff --git a/e
1 - 100 of 124 matches
Mail list logo
