Avi, Marcelo, if there are no objections, maybe this series can be
merged through the new shiny qemu-kvm upstream branch? Thanks!
- Forwarded message from Michael S. Tsirkin m...@redhat.com -
Date: Mon, 4 Jan 2010 21:48:56 +0200
From: Michael S. Tsirkin m...@redhat.com
To: Anthony
On Fri, Dec 18, 2009 at 11:05 PM, Aurelien Jarno aurel...@aurel32.netwrote:
On Fri, Dec 18, 2009 at 10:32:37PM +0100, Marc Andre Tanner wrote:
On Tue, Dec 01, 2009 at 09:41:43PM +0100, Filip Navara wrote:
While most of the ARMv5 instructions are backward compatible with
ARMv4, there
are
Since you wanted a no, because... there's at least
one :-)
On Tue, Dec 1, 2009 at 9:41 PM, Filip Navara filip.nav...@gmail.com wrote:
While most of the ARMv5 instructions are backward compatible with ARMv4, there
are few important differences. Most notably the stack pop and load
instructions
Am 14.01.2010 00:26, schrieb Anthony Liguori:
On 01/12/2010 06:49 AM, Christoph Hellwig wrote:
The backing device is only modified from bdrv_commit. So instead of
flushing it every time bdrv_flush is called for the front-end device
only flush it after we're written data to it in bdrv_commit.
Hi,
If you're looking to pull in 32bit pcibios support, then I don't think
it would be worthwhile to rebase to a stable branch, as the 32bit
pcibios support is easily the biggest user visible change in v0.5.1
(in the sense that Linux will call 32bit pcibios if it's available).
Unless
On Wed, 13 Jan 2010, Cole Robinson wrote:
On 01/13/2010 07:11 PM, malc wrote:
On Wed, 13 Jan 2010, Cole Robinson wrote:
Current CDROM detection is hardcoded based on source file name.
Make this smarter by attempting a CDROM specific ioctl.
This makes '-cdrom /dev/sr0' succeed with
On Sun, Dec 20, 2009 at 04:52:22PM +0100, Stefan Weil wrote:
Handling of transmit commands is rather complex,
so about 80 lines of code were moved from function
action_command to the new function tx_command.
The two new values tx and cb_address in the
eepro100 status structure made this
There's nothing target-dependent in the virtio-serial code so allow it
to be compiled just once for all the targets.
Signed-off-by: Amit Shah amit.s...@redhat.com
---
Makefile.objs |2 +-
Makefile.target |2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git
The port 'id' or number is internal state between the guest kernel and
our bus implementation. This is invocation-dependent and isn't part of
the guest-host ABI.
To correcly enumerate and map ports between the host and the guest, the
'name' property is used.
Example:
-device
This commit converts the virtio-console device to create a new
virtio-serial bus that can host console and generic serial ports. The
file hosting this code is now called virtio-serial-bus.c.
The virtio console is now a very simple qdev device that sits on the
virtio-serial-bus and communicates
This patch adds a new property named 'ver' to scsi-disk which allows to
specify the version which the virtual disk/cdrom should report to the
guest. By default this is the qemu version (i.e. 0.12). usage:
-drive if=none,id=disk,file=...
-device lsi
-device
This patch adds a new property named 'ver' to ide-drive which allows to
specify the version which the virtual disk/cdrom should report to the
guest. By default this is the qemu version (i.e. 0.12). usage:
-drive if=none,id=disk,file=...
-device ide-drive,bus=ide.0,unit=0,drive=disk,ver=42
Hi,
This patch series makes the version reported by ide and scsi drives
configurable using qdev properties. Should help keeping the virtual
hardware more constant on qemu updates, so it becomes less likely
that Windows wants be re-activated.
Maybe 0.12 candidate?
cheers,
Gerd
Signed-off-by: Amit Shah amit.s...@redhat.com
---
hw/virtio-serial-bus.c |2 ++
hw/virtio-serial.h |1 +
2 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/hw/virtio-serial-bus.c b/hw/virtio-serial-bus.c
index 1ec67d2..d045ea5 100644
--- a/hw/virtio-serial-bus.c
+++
We have to buffer data from guest as ports may not consume all the data
in one go or the guest could be fast in sending data and the apps may
not consume at the same rate.
We keep caching data the guest sends us till a port accepts it. However,
this could lead to an OOM condition where a rogue
On Wed, Jan 13, 2010 at 05:58:35PM -0600, Anthony Liguori wrote:
I actually need the compiler fix to build on my laptop (F12) so I've
included that too. Care to take a look at
git://git.qemu.org/seabios.git stable-0.5.0? It survives some light
testing and I'll be doing more thorough testing
Hello people,
This iteration of the series removes the START and END flags (and
hence the header associated with each buffer). That's the major change
since the last submission.
Please review.
Obligatory disclaimer:
This series splits up the patches by functionality. Note, however,
that patches
Via control channel messages, the guest can tell us whether a port got
opened or closed. Similarly, we can also indicate to the guest of host
port open/close events.
Signed-off-by: Amit Shah amit.s...@redhat.com
---
hw/virtio-serial-bus.c | 91
VIRTIO_PCI_QUEUE_MAX is redefined in hw/virtio.c. Let's just keep it in
hw/virtio.h.
Also, bump up the value of the maximum allowed virtqueues to 64. This is
in preparation to allow multiple ports per virtio-console device.
Signed-off-by: Amit Shah amit.s...@redhat.com
---
hw/virtio.c |2 --
I pulled from master today and Can't build ppc-softmmu anymore:
% make
LINK ppcemb-softmmu/qemu-system-ppcemb
ppce500_mpc8544ds.o: In function `mpc8544_copy_soc_cell':
/home/edgar/src/c/qemu/git/qemu/hw/ppce500_mpc8544ds.c:57: undefined reference
to `kvmppc_read_host_property'
On 01/14/2010 03:32 AM, Michael S. Tsirkin wrote:
Avi, Marcelo, if there are no objections, maybe this series can be
merged through the new shiny qemu-kvm upstream branch? Thanks!
FWIW, I've been waiting for at least a comment from Avi since that was
requested in the original post.
On 01/14/2010 04:33 AM, Vadim Rozenfeld wrote:
The following patch allows to suppress virtio queue notification with
the number of requests passed as parameter.
Changes from v1:
- code styling,
- parameter x-queue-depth-suppress-notify for queue depth adjustment.
repository:
On Thu, Jan 14, 2010 at 02:50:27PM +0100, Edgar E. Iglesias wrote:
I pulled from master today and Can't build ppc-softmmu anymore:
% make
LINK ppcemb-softmmu/qemu-system-ppcemb
ppce500_mpc8544ds.o: In function `mpc8544_copy_soc_cell':
On 01/14/2010 07:44 AM, Gerd Hoffmann wrote:
Hi,
This patch series makes the version reported by ide and scsi drives
configurable using qdev properties. Should help keeping the virtual
hardware more constant on qemu updates, so it becomes less likely
that Windows wants be re-activated.
Naphtali Sprei (3):
Make CDROM a read-only drive
Switch to bit-flags based read-only drive option implementation
Disable fall-back to read-only when cannot open drive's file for
read-write
block.c | 29 +
block.h |4 ++--
Signed-off-by: Naphtali Sprei nsp...@redhat.com
---
vl.c |4
1 files changed, 4 insertions(+), 0 deletions(-)
diff --git a/vl.c b/vl.c
index 06cb40d..4f19505 100644
--- a/vl.c
+++ b/vl.c
@@ -2234,6 +2234,10 @@ DriveInfo *drive_init(QemuOpts *opts, void *opaque,
Signed-off-by: Naphtali Sprei nsp...@redhat.com
---
block.c |4
1 files changed, 0 insertions(+), 4 deletions(-)
diff --git a/block.c b/block.c
index cbd72cc..0b4b9ad 100644
--- a/block.c
+++ b/block.c
@@ -457,10 +457,6 @@ int bdrv_open2(BlockDriverState *bs, const char *filename,
int
On 01/14/2010 04:27 PM, Anthony Liguori wrote:
On 01/14/2010 03:32 AM, Michael S. Tsirkin wrote:
Avi, Marcelo, if there are no objections, maybe this series can be
merged through the new shiny qemu-kvm upstream branch? Thanks!
FWIW, I've been waiting for at least a comment from Avi since that
On 01/14/2010 07:17 AM, Amit Shah wrote:
Hello people,
This iteration of the series removes the START and END flags (and
hence the header associated with each buffer). That's the major change
since the last submission.
I think the biggest issue remaining is the buffering.
I think this is
On 01/14/10 15:31, Anthony Liguori wrote:
On 01/14/2010 07:44 AM, Gerd Hoffmann wrote:
Hi,
This patch series makes the version reported by ide and scsi drives
configurable using qdev properties. Should help keeping the virtual
hardware more constant on qemu updates, so it becomes less likely
On 01/14/2010 07:46 AM, Kevin O'Connor wrote:
On Wed, Jan 13, 2010 at 05:58:35PM -0600, Anthony Liguori wrote:
I actually need the compiler fix to build on my laptop (F12) so I've
included that too. Care to take a look at
git://git.qemu.org/seabios.git stable-0.5.0? It survives some light
On Wed, Dec 23, 2009 at 04:33:24PM -0800, Kazu Hirata wrote:
Hi,
Attached is a patch to fix a typo in 'P' packet processing for M68K.
Without this patch, QEMU fails to honor GDB's P packets from GDB
(writing to registers) for the address registers (A0 - A7).
The problem is because of an
This fixes eg. -nographic -serial mon:stdio [-serial ...].
Signed-off-by: Jan Kiszka jan.kis...@siemens.com
---
vl.c |9 +
1 files changed, 9 insertions(+), 0 deletions(-)
diff --git a/vl.c b/vl.c
index 06cb40d..fa3e8ee 100644
--- a/vl.c
+++ b/vl.c
@@ -5171,6 +5171,9 @@ int
Starting with commit 0e8c9214ba1d4128cf92442cd343bc3733478261, when
trying to run make without running configure first, the error message
has been changed from:
| $ make
| Please call configure before running make!
| make: *** [config-host.mak] Erreur 1
to this more cryptic message:
| $ make
|
This patch adds compat property entries for ide-disk.ver and
scsi-disk.ver to pc-0.10 and pc-0.11. With this patch applied
the scsi and ide disks report 0.10 and 0.11 as version when
you start qemu with -M pc-0.10 or -M pc-0.11.
Signed-off-by: Gerd Hoffmann kra...@redhat.com
---
hw/pc.c | 16
On 01/14/10 15:49, Jan Kiszka wrote:
This fixes eg. -nographic -serial mon:stdio [-serial ...].
Looks good to me.
Acked-by: Gerd Hoffmann kra...@redhat.com
cheers,
Gerd
On Tue, Dec 29, 2009 at 12:01:22AM -0500, Vince Weaver wrote:
Hello
The stat64/fstat64 syscalls are broken for alpha linux-user.
This is because Alpha, even though it is native 64-bits, has a stat64
syscall that is different than regular stat. This means that the
TARGET_LONG_BITS==64
On Tue, Dec 29, 2009 at 10:39:21PM +0100, Loïc Minier wrote:
Hey there
This patch for linux-user adapts the output of the emulated uname()
syscall to match the configured CPU. Tested with x86, x86-64 and arm
emulation.
This patch look ok, but is missing a Signed-off-by:
On 01/14/2010 08:49 AM, Jan Kiszka wrote:
This fixes eg. -nographic -serial mon:stdio [-serial ...].
Is this really an appropriate invocation though?
-nographic != mon:stdio so the semantics of how this is supposed to
behave is at best ill-defined.
Regards,
Anthony Liguori
On Fri, Jan 01, 2010 at 04:41:06PM +0100, Alexander Graf wrote:
The recent transition to always have the DCR helper functions take 32 bit
values broke the PPC64 target, as tlong became 64 bits there.
This patch moves all translate.c callers to a _tl function that simply
calls the uint32_t
On Fri, Jan 01, 2010 at 03:59:39PM +0900, Shin-ichiro KAWASAKI wrote:
This patch adds hardware cursor feature to SM501 graphics chip emulation,
to make the graphic console more useful for QEMU SH4 users.
Signed-off-by: Shin-ichiro KAWASAKI kawas...@juno.dti.ne.jp
Thanks, applied.
On 14.01.2010, at 16:13, Aurelien Jarno wrote:
On Fri, Jan 01, 2010 at 04:41:06PM +0100, Alexander Graf wrote:
The recent transition to always have the DCR helper functions take 32 bit
values broke the PPC64 target, as tlong became 64 bits there.
This patch moves all translate.c callers to
Anthony Liguori wrote:
On 01/14/2010 08:49 AM, Jan Kiszka wrote:
This fixes eg. -nographic -serial mon:stdio [-serial ...].
Is this really an appropriate invocation though?
-nographic != mon:stdio so the semantics of how this is supposed to
behave is at best ill-defined.
Original
qemu_chr_open_fd() calls qemu_chr_generic_open(),
so qemu_chr_open_tty() doesn't need to call it.
Signed-off-by: Kusanagi Kouichi sl...@ac.auone-net.jp
---
qemu-char.c |1 -
1 files changed, 0 insertions(+), 1 deletions(-)
diff --git a/qemu-char.c b/qemu-char.c
index b13f8d4..800ee6c 100644
On Tue, Jan 05, 2010 at 04:03:00PM -0800, Richard Henderson wrote:
This avoids an unnecessary REX.W prefix when dealing with AND
operands that fit into a 32-bit quantity. The most common change
actually seen is movz[wb]q - movz[wb]l.
Similarly, avoid REXW in ext{8,16}u_i64 tcg opcodes.
On 01/14/2010 07:57 AM, Aurelien Jarno wrote:
On Tue, Jan 05, 2010 at 04:03:00PM -0800, Richard Henderson wrote:
This avoids an unnecessary REX.W prefix when dealing with AND
operands that fit into a 32-bit quantity. The most common change
actually seen is movz[wb]q - movz[wb]l.
Similarly,
On Tue, Jan 05, 2010 at 04:31:11PM -0800, Richard Henderson wrote:
A while ago Laurent pointed out that the setcc opcode emitted by
the setcond patch had unnecessary REX prefixes.
The existing P_REXB internal opcode flag unconditionally emits
the REX prefix. Technically it's not needed if
On 01/14/2010 05:45 AM, malc wrote:
On Wed, 13 Jan 2010, Cole Robinson wrote:
On 01/13/2010 07:11 PM, malc wrote:
On Wed, 13 Jan 2010, Cole Robinson wrote:
Current CDROM detection is hardcoded based on source file name.
Make this smarter by attempting a CDROM specific ioctl.
This makes
Current legacy floppy detection is hardcoded based on source file
name. Make this smarter on linux by attempting a floppy specific
ioctl.
v2:
Give ioctl check higher priority than filename check
s/IDE/legacy/
v3:
Actually initialize 'prio' variable
Check for ioctl success rather
Signed-off-by: Kevin Wolf kw...@redhat.com
---
024 | 104 +++
024.out | 81 +
group |1 +
3 files changed, 186 insertions(+), 0 deletions(-)
create mode 100755 024
create
Currently the 'status' key is a string whose value can be
disabled or enabled, change it to the QMP's standard
'enabled' key, which is a bool.
Note that 'status' in being dropped and this wouldn't be
allowed if QMP were stable.
Signed-off-by: Luiz Capitulino lcapitul...@redhat.com
---
vnc.c |
There is no reason to have it as optional and the code
in the server and client gets slightly simpler if the
key is mandatory.
While there also do some cleanup on how the server info is
collected.
Signed-off-by: Luiz Capitulino lcapitul...@redhat.com
---
vnc.c | 26 ++
It's the SASL username, so it's better to call it 'sasl_username'
to be consistent.
Note that this change wouldn't be allowed if QMP were stable.
Signed-off-by: Luiz Capitulino lcapitul...@redhat.com
---
vnc.c |9 +
1 files changed, 5 insertions(+), 4 deletions(-)
diff --git
Hi there,
This series contains two VNC related changes. First a small cleanup
is done in the current 'query-vnc' command _response_, then the following
QMP events are introduced:
- VNC_CONNECTED: emitted when a VNC client establishes a connection
- VNC_INITIALIZED: emitted when the VNC session
It contains the socket adress family name, like ipv4 or
ipv6.
This is useful for clients so that they can interpret the
'host' key reliably.
Signed-off-by: Luiz Capitulino lcapitul...@redhat.com
---
vnc.c | 26 +-
1 files changed, 25 insertions(+), 1 deletions(-)
diff
When a disconnection happens the client's socket on QEMU
side may become invalid, this way it won't be possible
to query it to get client information, which is going to
be needed by the future QMP VNC_DISCONNECTED event.
To always have this information available we query the
socket at connection
It's emitted when a VNC client connects to QEMU, client's information
such as port and IP address are provided.
Note that this event is emitted right when the connection is
established. This means that it happens before authentication
procedure and session initialization.
Event example:
{
It's emitted when a VNC client disconnects from QEMU, client's
information such as port and IP address are provided.
Event example:
{ event: VNC_DISCONNECTED,
timestamp: { seconds: 1262976601, microseconds: 975795 },
data: {
server: { auth: sasl, family: ipv4,
It's emitted when a VNC client session is activated by QEMU,
client's information such as port, IP and auth ID (if the
session is authenticated) are provided.
Event example:
{ event: VNC_INITIALIZED,
timestamp: {seconds: 1263475302, microseconds: 150772},
data: {
server: { auth:
On Thu, Jan 14, 2010 at 04:19:31PM +0100, Alexander Graf wrote:
On 14.01.2010, at 16:13, Aurelien Jarno wrote:
On Fri, Jan 01, 2010 at 04:41:06PM +0100, Alexander Graf wrote:
The recent transition to always have the DCR helper functions take 32 bit
values broke the PPC64 target, as
On 14.01.2010, at 18:02, Aurelien Jarno wrote:
On Thu, Jan 14, 2010 at 04:19:31PM +0100, Alexander Graf wrote:
On 14.01.2010, at 16:13, Aurelien Jarno wrote:
On Fri, Jan 01, 2010 at 04:41:06PM +0100, Alexander Graf wrote:
The recent transition to always have the DCR helper functions take
Hi Aurelien,
I don't have a write access to the repository. Could someone apply
this patch if it's OK?
This patch looks ok, but is missing a Signed-of-by:
Thank you for a review. I'm going to resubmit the patch in the git style.
Kazu Hirata
Hi,
Attached is a patch to fix a typo in 'P' packet processing for M68K.
Without this patch, QEMU fails to honor GDB's P packets from GDB
(writing to registers) for the address registers (A0 - A7).
The problem is because of an obvious typo. Notice that the second
if condition is meant to be n
When make is called without a valid configuration,
it should tell the user what to do.
Revision 0e8c9214ba1d4128cf92442cd343bc3733478261
was a regression which resulted in a message
which was no longer user friendly
(reported by Aurelien Jarno).
This patch restores the old behaviour.
Cc:
Am 14.01.2010 um 15:50 schrieb Aurelien Jarno:
Starting with commit 0e8c9214ba1d4128cf92442cd343bc3733478261, when
trying to run make without running configure first, the error message
has been changed from:
| $ make
| Please call configure before running make!
| make: *** [config-host.mak]
On Thu, Jan 14, 2010 at 02:50:51PM -0200, Luiz Capitulino wrote:
Hi there,
This series contains two VNC related changes. First a small cleanup
is done in the current 'query-vnc' command _response_, then the following
QMP events are introduced:
- VNC_CONNECTED: emitted when a VNC client
Am 14.01.2010 um 18:11 schrieb Stefan Weil:
When make is called without a valid configuration,
it should tell the user what to do.
Revision 0e8c9214ba1d4128cf92442cd343bc3733478261
was a regression which resulted in a message
which was no longer user friendly
(reported by Aurelien Jarno).
On 01/14/2010 08:10 AM, Aurelien Jarno wrote:
On Tue, Jan 05, 2010 at 04:31:11PM -0800, Richard Henderson wrote:
A while ago Laurent pointed out that the setcc opcode emitted by
the setcond patch had unnecessary REX prefixes.
The existing P_REXB internal opcode flag unconditionally emits
the
On Thu, Jan 14, 2010 at 10:09:48AM -0800, Richard Henderson wrote:
On 01/14/2010 08:10 AM, Aurelien Jarno wrote:
On Tue, Jan 05, 2010 at 04:31:11PM -0800, Richard Henderson wrote:
A while ago Laurent pointed out that the setcc opcode emitted by
the setcond patch had unnecessary REX prefixes.
On Thu, Jan 14, 2010 at 08:05:58AM -0800, Richard Henderson wrote:
On 01/14/2010 07:57 AM, Aurelien Jarno wrote:
On Tue, Jan 05, 2010 at 04:03:00PM -0800, Richard Henderson wrote:
This avoids an unnecessary REX.W prefix when dealing with AND
operands that fit into a 32-bit quantity. The most
On Thu, 14 Jan 2010 17:32:14 +
Daniel P. Berrange berra...@redhat.com wrote:
On Thu, Jan 14, 2010 at 02:50:51PM -0200, Luiz Capitulino wrote:
Hi there,
This series contains two VNC related changes. First a small cleanup
is done in the current 'query-vnc' command _response_, then
On Thu, Jan 14, 2010 at 06:11:43PM +0100, Stefan Weil wrote:
When make is called without a valid configuration,
it should tell the user what to do.
Revision 0e8c9214ba1d4128cf92442cd343bc3733478261
was a regression which resulted in a message
which was no longer user friendly
(reported by
On Thu, Jan 14, 2010 at 09:08:00AM -0800, Kazu Hirata wrote:
Hi,
Attached is a patch to fix a typo in 'P' packet processing for M68K.
Without this patch, QEMU fails to honor GDB's P packets from GDB
(writing to registers) for the address registers (A0 - A7).
The problem is because of an
On Thu, Jan 14, 2010 at 3:09 AM, Amit Shah amit.s...@redhat.com wrote:
On (Wed) Jan 13 2010 [19:08:11], Blue Swirl wrote:
Thanks. I fixed the warnings related to Sparc32. Were there really no
new warnings for Sparc64?
Looks like it; vl.c gets reported three times at the same locations so 3
When using a control/QMP monitor in tandem with a regular monitor, asynchronous
messages can get lost depending on the order of the QEMU program arguments.
QEMU events issued by monitor_protocol_event() always go to cur_mon. If the
user monitor was specified on the command line first (or it has
The existing P_REXB internal opcode flag unconditionally emits
the REX prefix. Technically it's not needed if the register in
question is %al, %bl, %cl, %dl.
Eliding the prefix requires splitting the P_REXB flag into two,
in order to indicate whether the byte register in question is
in the REG
Previously, mmap_find_vma could return addresses not properly aligned
to the target page size. This of course led to all sorts of odd
problems down the road.
The trivial fix, to simply reject the unaligned address and continue
searching the address space by increments of one page, is not a good
Richard Henderson wrote:
On 01/14/2010 08:10 AM, Aurelien Jarno wrote:
With the above change, rex can be 0xff. Not sure it's not a good idea
to not have an explicit cast when calling tcg_out8(), even if it
technically works.
What's the reason for removing the ' 0xff' part? tcg_out8() takes
diff --git a/hw/rtl8139.c b/hw/rtl8139.c
index 1f4f585..f04dd54 100644
--- a/hw/rtl8139.c
+++ b/hw/rtl8139.c
@@ -1909,6 +1909,7 @@ static int rtl8139_cplus_transmit_one(RTL8139State *s)
cpu_physical_memory_read(cplus_tx_ring_desc,(uint8_t *)val, 4);
txdw0 = le32_to_cpu(val);
+
+static QString *get_sock_family(const struct sockaddr_storage *sa)
+{
+const char *name;
+
+switch (sa-ss_family)
+{
+case AF_INET:
+name = ipv4;
+break;
+case AF_INET6:
+name = ipv6;
+break;
+default:
+
80 matches
Mail list logo
