"Michael S. Tsirkin" writes:
> On Thu, Mar 21, 2013 at 02:44:50PM +0800, Amos Kong wrote:
>> The lengcy guests don't have mac programming command, we don't know when
>> it's safe to use MAC. This patch changed qemu to makes MAC change effect
>> when the last byte of MAC is written to config space.
On 03/21/2013 08:28 AM, Igor Mammedov wrote:
> ... via do_cpu_hot_add() hook called by cpu_set QMP command,
> for x86 target.
>
> * add extra check that APIC ID is in allowed range
> +if (x86_cpu_is_cpu_exist(qdev_get_machine(), &apic_id)) {
> +error_setg(errp, "Unable to add CPU with
On 03/21/2013 08:28 AM, Igor Mammedov wrote:
> Signed-off-by: Igor Mammedov
> ---
> +++ b/qapi-schema.json
> @@ -1385,6 +1385,15 @@
> { 'command': 'cpu', 'data': {'index': 'int'} }
>
> ##
> +# @cpu_set
> +#
I already mentioned naming this cpu-set on your cover letter. Additionally:
> +# Set
On 03/21/2013 02:35 AM, Hu Tao wrote:
> Signed-off-by: Hu Tao
> ---
> docs/specs/pvpanic.txt | 25 +
> 1 file changed, 25 insertions(+)
> create mode 100644 docs/specs/pvpanic.txt
Grammar nits:
>
> diff --git a/docs/specs/pvpanic.txt b/docs/specs/pvpanic.txt
> new file
On 03/22/2013 07:11 AM, Aurélien Jarno wrote:
On Thu, Mar 21, 2013 at 04:04:44PM +0900, Yeongkyoon Lee wrote:
On 03/18/2013 07:27 AM, Aurélien Jarno wrote:
On Wed, Mar 06, 2013 at 07:10:17AM +0100, Aurélien Jarno wrote:
On Wed, Mar 06, 2013 at 11:05:15AM +0900, Yeongkyoon Lee wrote:
On 03/05/
在 2013-03-21四的 10:41 +，Peter Maydell写道：
> On 21 March 2013 00:36, li guang wrote:
> > 在 2013-03-20三的 10:50 +，Peter Maydell写道：
> >> The devices should just implement appropriate signals/connections
> >> if they have a means of talking to a power controller, and the
> >> board model should w
On Thu, Mar 21, 2013 at 07:42:37PM +0200, Michael S. Tsirkin wrote:
> It doesn't actually, and our app would sometimes write to these pages.
> It simply does not care which version does the remote get in this case
> since we track writes and resend later.
Heh, somehow I thought you might say that
On Thu, Mar 21, 2013 at 08:16:33PM +0200, Michael S. Tsirkin wrote:
> This is the one I find redundant. Since the write will be done by
> the adaptor under direct control by the application, why does it
> make sense to declare this beforehand? If you don't want to allow
> local write access to me
On Thu, Mar 21, 2013 at 09:15:41PM +0200, Michael S. Tsirkin wrote:
> On Thu, Mar 21, 2013 at 12:41:35PM -0600, Jason Gunthorpe wrote:
> > On Thu, Mar 21, 2013 at 08:16:33PM +0200, Michael S. Tsirkin wrote:
> >
> > > This is the one I find redundant. Since the write will be done by
> > > the adapt
On Thu, Mar 21, 2013 at 11:39:47AM +0200, Michael S. Tsirkin wrote:
> On Thu, Mar 21, 2013 at 02:13:38AM -0700, Roland Dreier wrote:
> > On Thu, Mar 21, 2013 at 1:51 AM, Michael S. Tsirkin wrote:
> > >> In that case, no, I don't see any reason for LOCAL_WRITE, since the
> > >> only RDMA operations
On Thu, Mar 21, 2013 at 07:15:25PM +0200, Michael S. Tsirkin wrote:
> No because application does this:
> init page
>
> ...
>
> after a lot of time
>
> ..
>
> register
> send
> unregister
>
> so it can not be read only.
mprotect(READONLY)
register
send
unregister
mprotect(WRITABLE)
?
With
On 21.03.2013, at 21:01, Richard Henderson wrote:
> The first patch fixes the problem reported by agraf here:
>
> http://lists.gnu.org/archive/html/qemu-devel/2013-03/msg03747.html
>
> The subsequent patches use the macro added in the first patch
> to remove a bunch of conditional compilation.
On 21.03.2013, at 22:59, Scott Wood wrote:
> On 03/21/2013 04:29:02 PM, Alexander Graf wrote:
>> Am 21.03.2013 um 21:50 schrieb Scott Wood :
>> > On 03/21/2013 03:41:19 AM, Alexander Graf wrote:
>> >> Can't all the stuff above here just simply go into the qdev init function?
>> >
>> > Not if you
On Thu, Mar 21, 2013 at 05:08:34PM +0800, Hu Tao wrote:
> Signed-off-by: Hu Tao
I don't think it is a good idea to dynamically modify the DSDT. We've
been using the SSDT for that.
In any case, I think this would be a good candidate for merging after
the ACPI stuff is moved into QEMU.
-Kevin
On 03/21/2013 06:51:57 AM, Alexander Graf wrote:
On 21.03.2013, at 12:49, Alexander Graf wrote:
>
> On 21.03.2013, at 12:44, Peter Maydell wrote:
>
>> On 21 March 2013 11:38, Alexander Graf wrote:
>>>
>>> On 21.03.2013, at 12:32, Peter Maydell wrote:
>>>
On 21 March 2013 11:29, Alexander
On Thu, Mar 21, 2013 at 04:04:44PM +0900, Yeongkyoon Lee wrote:
> On 03/18/2013 07:27 AM, Aurélien Jarno wrote:
> >On Wed, Mar 06, 2013 at 07:10:17AM +0100, Aurélien Jarno wrote:
> >>On Wed, Mar 06, 2013 at 11:05:15AM +0900, Yeongkyoon Lee wrote:
> >>>On 03/05/2013 11:18 PM, Aurélien Jarno wrote:
>
On 03/21/2013 05:59 PM, Eric Blake wrote:
On 03/21/2013 12:29 PM, Stefan Berger wrote:
+++ b/tests/test-visitor-serialization.c
@@ -14,6 +14,7 @@
#include
#include
#include
+#include
Is this addition needed? I don't see any math functions added.
Indeed. Removed.
Stefan
On 03/21/2013 12:29 PM, Stefan Berger wrote:
> BER visitor tests give us some assurance that the BER visitor
> code works, and also end up by extention helping out on our
s/extention/extension/
> code coverage of the filesystem tests.
> After the output visitor invocation the resuling buffer is
> > Alon Levy writes:
> >
> > >> Alon Levy writes:
> > >>
> > >> > Note that the handler is called chr_is_guest_connected and not
> > >> > chr_is_fe_connected, consistent with other members of
> > >> > CharDriverState.
> > >>
> > >> Sorry, I don't get it.
> > >>
> > >> There isn't a notion of
On 03/21/2013 04:29:02 PM, Alexander Graf wrote:
Am 21.03.2013 um 21:50 schrieb Scott Wood :
> On 03/21/2013 03:41:19 AM, Alexander Graf wrote:
>> Can't all the stuff above here just simply go into the qdev init
function?
>
> Not if you want platform code to be able to fall back to a QEMU
On 03/21/2013 12:29 PM, Stefan Berger wrote:
> Add BER visitor hooks to test-visitor-serialization
>
> Cc: Michael Tsirkin
> Cc: Stefan Berger
> Signed-off-by: Joel Schopp
> ---
> +++ b/tests/test-visitor-serialization.c
> @@ -14,6 +14,7 @@
> #include
> #include
> #include
> +#include
> Alon Levy writes:
>
> >> Alon Levy writes:
> >>
> >> > Note that the handler is called chr_is_guest_connected and not
> >> > chr_is_fe_connected, consistent with other members of
> >> > CharDriverState.
> >>
> >> Sorry, I don't get it.
> >>
> >> There isn't a notion of "connected" for the f
Am 21.03.2013 um 21:50 schrieb Scott Wood :
> On 03/21/2013 03:41:19 AM, Alexander Graf wrote:
>> On 14.02.2013, at 07:32, Scott Wood wrote:
>> > +DeviceState *kvm_openpic_create(BusState *bus, int model)
>> > +{
>> > +KVMState *s = kvm_state;
>> > +DeviceState *dev;
>> > +struct kvm
On 03/21/2013 03:41:19 AM, Alexander Graf wrote:
On 14.02.2013, at 07:32, Scott Wood wrote:
> +DeviceState *kvm_openpic_create(BusState *bus, int model)
> +{
> +KVMState *s = kvm_state;
> +DeviceState *dev;
> +struct kvm_create_device cd = {0};
> +int ret;
> +
> +if (!kvm_ch
On 03/14/13 18:49, Markus Armbruster wrote:
> Known bugs in to_json():
> My rewrite fixes them as follows:
I'll try to review this sometime later. Patch review doesn't scale *at
all*. I've spent hours on the first 3 patches. You should just be given
pull req rights. I'd need 36 hour days.
/me th
On Thu, Mar 21, 2013 at 08:35:29PM +0100, Stefan Hajnoczi wrote:
> On Thu, Mar 21, 2013 at 2:38 PM, Richard W.M. Jones wrote:
> > From: "Richard W.M. Jones"
> >
> > qemu-system-x86_64 -drive file=ssh://hostname/some/image
> >
> > QEMU will ssh into 'hostname' and open '/some/image' which is mad
On 03/14/13 18:49, Markus Armbruster wrote:
> These are all broken, too.
What are "these"? And how are they broken? And how does the patch fix them?
>
> A few test cases use noncharacters U+ and U+10. Risks testing
> noncharacters some more instead of what they're supposed to test. Use
Removing conditional compilation in the process.
Signed-off-by: Richard Henderson
---
target-ppc/translate.c | 51 ++
1 file changed, 18 insertions(+), 33 deletions(-)
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index e2d657d..2bb
I don't understand what's going on here.
On 03/14/13 18:49, Markus Armbruster wrote:
> Signed-off-by: Markus Armbruster
> ---
> tests/check-qjson.c | 5 ++---
> 1 file changed, 2 insertions(+), 3 deletions(-)
>
> diff --git a/tests/check-qjson.c b/tests/check-qjson.c
> index ec85a0c..852124a 10
Removing conditional compilation in the process.
Signed-off-by: Richard Henderson
---
target-ppc/translate.c | 7 +++
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index 2bb28b8..af936cd 100644
--- a/target-ppc/translate.c
+++ b
The set of computations used in b5a73f8d8a57e940f9bbeb399a9e47897522ee9a
are only valid if the current word size == target_long size. This failed
to take ppc64 in 32-bit (narrow) mode into account.
Add a NARROW_MODE macro to avoid conditional compilation.
Signed-off-by: Richard Henderson
---
t
Removing conditional compilation in the process.
Signed-off-by: Richard Henderson
---
target-ppc/translate.c | 41 -
1 file changed, 16 insertions(+), 25 deletions(-)
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index 92b4f6c..e2d657d 1006
Removing conditional compilation in the process.
Signed-off-by: Richard Henderson
---
target-ppc/translate.c | 62 +++---
1 file changed, 24 insertions(+), 38 deletions(-)
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index ed6415c..92b
The first patch fixes the problem reported by agraf here:
http://lists.gnu.org/archive/html/qemu-devel/2013-03/msg03747.html
The subsequent patches use the macro added in the first patch
to remove a bunch of conditional compilation.
With this (or even just the first patch), I can boot Alex's
t
Hello,
I have been noticing host hangs when trying to boot large guests
(>=40Vcpus) with the current upstream qemu.
Host is running 3.8.2 kernel.
qemu is the latest one from qemu.git.
Example qemu command line listed below... this used to work with a
slightly older qemu (about 1.5 weeks ago
Am 21.03.2013 um 20:27 schrieb Eric Blake :
> On 03/21/2013 09:57 AM, Peter Lieven wrote:
>> avoid searching for dirty pages just increment the
>> page offset. all pages are dirty anyway.
>>
>> Signed-off-by: Peter Lieven
>> ---
>> arch_init.c | 10 --
>> 1 file changed, 8 insertions(+
Luiz Capitulino writes:
> On Thu, 21 Mar 2013 11:13:13 -0500
> mdroth wrote:
>
>> > Looks like you guys have no *practical* problems to solve. Congrats!
>> > Take a vacation! Please report back no later than 275 years from now,
>> > to make sure this 64 bit fd counter overflow problem gets tak
Am 21.03.2013 um 20:26 schrieb Eric Blake :
> On 03/21/2013 09:57 AM, Peter Lieven wrote:
>> during bulk stage of ram migration if a page is a
>> zero page do not send it at all.
>> the memory at the destination reads as zero anyway.
>>
>> even if there is an madvise with QEMU_MADV_DONTNEED
>> a
On 03/21/2013 12:34 PM, Orit Wasserman wrote:
> This way we send one big buffer instead of many small ones
>
> Signed-off-by: Orit Wasserman
> ---
> savevm.c | 19 +++
> 1 file changed, 15 insertions(+), 4 deletions(-)
>
> diff --git a/savevm.c b/savevm.c
> index b024354..9e8dde
On 03/21/2013 12:34 PM, Orit Wasserman wrote:
> This allow us to add a buffer to the iovec to send without copying it
s/allow/allows/
> into the static buffer, the buffer will be send later when qemu_fflush is
> called.
s/send/sent/
>
> Signed-off-by: Orit Wasserman
> ---
> include/migratio
On Thu, Mar 21, 2013 at 2:38 PM, Richard W.M. Jones wrote:
> From: "Richard W.M. Jones"
>
> qemu-system-x86_64 -drive file=ssh://hostname/some/image
>
> QEMU will ssh into 'hostname' and open '/some/image' which is made
> available as a standard block device.
>
> You can specify a username (ssh
I've (re-)read the UTF-8 article in wikipedia now... comments below
On 03/14/13 18:49, Markus Armbruster wrote:
> diff --git a/util/unicode.c b/util/unicode.c
> new file mode 100644
> index 000..954bcf7
> +/**
> + * mod_utf8_codepoint:
> + * @s: string encoded in modified UTF-8
> + * @n: max
On 03/21/2013 12:34 PM, Orit Wasserman wrote:
> Signed-off-by: Orit Wasserman
> ---
> savevm.c | 15 +++
> 1 file changed, 15 insertions(+)
>
> diff --git a/savevm.c b/savevm.c
> index 35c8d1e..6608b6e 100644
> --- a/savevm.c
> +++ b/savevm.c
> @@ -39,6 +39,7 @@
> #include "qmp-comm
On 03/21/2013 09:57 AM, Peter Lieven wrote:
> at the beginning of migration all pages are marked dirty and
> in the first round a bulk migration of all pages is performed.
>
> currently all these pages are copied to the page cache regardless
> if there are frequently updated or not. this doesn't m
On Thu, Mar 21, 2013 at 4:39 PM, Richard W.M. Jones wrote:
> On Thu, Mar 21, 2013 at 04:26:23PM +0100, Stefan Hajnoczi wrote:
>> On Thu, Mar 21, 2013 at 01:38:58PM +, Richard W.M. Jones wrote:
>> > From: "Richard W.M. Jones"
>> >
>> > qemu-system-x86_64 -drive file=ssh://hostname/some/image
On 03/21/2013 09:57 AM, Peter Lieven wrote:
> avoid searching for dirty pages just increment the
> page offset. all pages are dirty anyway.
>
> Signed-off-by: Peter Lieven
> ---
> arch_init.c | 10 --
> 1 file changed, 8 insertions(+), 2 deletions(-)
Prior review still stands.
--
Er
On 03/21/2013 09:57 AM, Peter Lieven wrote:
> the first round of ram transfer is special since all pages
> are dirty and thus all memory pages are transferred to
> the target. this patch adds a boolean variable to track
> this stage.
>
> Signed-off-by: Peter Lieven
> ---
> arch_init.c |3 +++
On 03/21/2013 09:57 AM, Peter Lieven wrote:
> during bulk stage of ram migration if a page is a
> zero page do not send it at all.
> the memory at the destination reads as zero anyway.
>
> even if there is an madvise with QEMU_MADV_DONTNEED
> at the target upon receipt of a zero page I have observ
Alon Levy writes:
>> Alon Levy writes:
>>
>> > Note that the handler is called chr_is_guest_connected and not
>> > chr_is_fe_connected, consistent with other members of
>> > CharDriverState.
>>
>> Sorry, I don't get it.
>>
>> There isn't a notion of "connected" for the front-ends in the char
On 03/21/2013 09:57 AM, Peter Lieven wrote:
> virtually all dup pages are zero pages. remove
> the special is_dup_page() function and use the
> optimized buffer_find_nonzero_offset() function
> instead.
>
> here buffer_find_nonzero_offset() is used directly
> to avoid the unnecssary additional che
On 03/21/2013 09:57 AM, Peter Lieven wrote:
> this patch adds the usage of buffer_find_nonzero_offset()
> to skip large areas of zeroes.
>
> compared to loop unrolling presented in an earlier
> patch this adds another 50% performance benefit for
> skipping large areas of zeroes. loop unrolling alo
On Thu, Mar 21, 2013 at 12:41:35PM -0600, Jason Gunthorpe wrote:
> On Thu, Mar 21, 2013 at 08:16:33PM +0200, Michael S. Tsirkin wrote:
>
> > This is the one I find redundant. Since the write will be done by
> > the adaptor under direct control by the application, why does it
> > make sense to decl
Add BER visitor hooks to test-visitor-serialization
Cc: Michael Tsirkin
Cc: Stefan Berger
Signed-off-by: Joel Schopp
---
tests/Makefile | 2 +-
tests/test-visitor-serialization.c | 73 ++
2 files changed, 74 insertions(+), 1 deletion(-)
Am 21.03.2013 um 19:12 schrieb Eric Blake :
> On 03/21/2013 09:57 AM, Peter Lieven wrote:
>> this adds buffer_find_nonzero_offset() which is a SSE2/Altives
>
> s/Altives/Altivec/
>
>> optimized function that searches for non-zero content in a
>> buffer.
>>
>> due to the optimizations used in t
Hi Jan,
Thank you for reviewing.
On 03/21/2013 05:08 AM, Jan Kiszka wrote:
> On 2013-02-14 21:21, Satoru Moriya wrote:
>> @@ -31,6 +31,7 @@ void os_set_proc_name(const char *s); void
>> os_setup_signal_handling(void); void os_daemonize(void); void
>> os_setup_post(void);
>> +void os_mlock(vo
> Alon Levy writes:
>
> > Note that the handler is called chr_is_guest_connected and not
> > chr_is_fe_connected, consistent with other members of
> > CharDriverState.
>
> Sorry, I don't get it.
>
> There isn't a notion of "connected" for the front-ends in the char
> layer. The closest thing i
On 03/21/2013 12:05 PM, Alexandre Kandalintsev wrote:
> Hi!
>
>
> Here is the patch that allows us to specify the name of tap interface
> when -netdev bridge is used. It's like -netdev tap,ifname=xxx, but for
> bridges.
>
> +++ b/qapi-schema.json
> @@ -2676,6 +2676,7 @@
> { 'type': 'NetdevBrid
Implement an input visitor for ASN.1 BER and CER encoding.
Cc: Michael Tsirkin
Signed-off-by: Stefan Berger
Signed-off-by: Joel Schopp
---
include/qapi/ber-input-visitor.h | 30 +
qapi/Makefile.objs |2 +-
qapi/ber-input-visitor.c | 1141
Implement an output visitor for ASN.1 BER and CER encoding.
Cc: Michael Tsirkin
Signed-off-by: Stefan Berger
Signed-off-by: Joel Schopp
---
configure | 2 +-
include/qapi/ber-output-visitor.h | 28 ++
include/qapi/ber.h| 107 ++
include/qemu-commo
BER visitor tests give us some assurance that the BER visitor
code works, and also end up by extention helping out on our
code coverage of the filesystem tests.
After the output visitor invocation the resuling buffer is
compared against a known byte stream -- this will lock the
implementation into
Using the QEMUFile interface, this patch adds support functions for operating
on in-memory sized buffers that can be written to or read from.
Signed-off-by: Stefan Berger
Signed-off-by: Joel Schopp
---
include/migration/qemu-file.h | 13 ++
include/qemu-common.h | 12 ++
util/qemu-fil
On 21 March 2013 17:57, Petar Jovanovic wrote:
> From: Petar Jovanovic
>
> do_semop() is called from two places, and one of these fails to convert
> return error to target errno when semop fails. This patch changes the
> function to always return target errno in case of an unsuccessful call.
>
>
On Thu, Mar 21, 2013 at 02:24:56PM -0400, Luiz Capitulino wrote:
> On Thu, 21 Mar 2013 11:13:13 -0500
> mdroth wrote:
>
> > > Looks like you guys have no *practical* problems to solve. Congrats!
> > > Take a vacation! Please report back no later than 275 years from now,
> > > to make sure this
On Thu, Mar 21, 2013 at 08:08:58PM +0200, Orit Wasserman wrote:
> On 03/21/2013 07:37 PM, Juan Quintela wrote:
> > Orit Wasserman wrote:
> >> This will remove an unneeded copy of guest memory pages.
> >> For the page header and device state we still copy the data to the
> >> static buffer the othe
This way we send one big buffer instead of many small ones
Signed-off-by: Orit Wasserman
---
savevm.c | 19 +++
1 file changed, 15 insertions(+), 4 deletions(-)
diff --git a/savevm.c b/savevm.c
index b024354..9e8ddee 100644
--- a/savevm.c
+++ b/savevm.c
@@ -622,6 +622,18 @@ int
This allow us to add a buffer to the iovec to send without copying it
into the static buffer, the buffer will be send later when qemu_fflush is
called.
Signed-off-by: Orit Wasserman
---
include/migration/qemu-file.h | 5 +
savevm.c | 36 +++-
All data is still copied into the static buffer.
Signed-off-by: Orit Wasserman
Reviewed-by: Juan Quintela
---
savevm.c | 13 +++--
1 file changed, 11 insertions(+), 2 deletions(-)
diff --git a/savevm.c b/savevm.c
index bd60169..fb20f13 100644
--- a/savevm.c
+++ b/savevm.c
@@ -114,6 +11
Update qemu_fflush and stdio_close to use writev ops if they are available
Use the buffers stored in the iovec.
Signed-off-by: Orit Wasserman
Reviewed-by: Juan Quintela
---
savevm.c | 30 +++---
1 file changed, 23 insertions(+), 7 deletions(-)
diff --git a/savevm.c b/sa
Signed-off-by: Orit Wasserman
Reviewed-by: Juan Quintela
---
savevm.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/savevm.c b/savevm.c
index 6608b6e..bd60169 100644
--- a/savevm.c
+++ b/savevm.c
@@ -648,6 +648,8 @@ void qemu_put_byte(QEMUFile *f, int v)
f->buf[f->buf_index++] = v
This will remove an unneeded copy of guest memory pages.
For the page header and device state we still copy the data to the
static buffer the other option is to allocate the memory on demand
which is more expensive.
Signed-off-by: Orit Wasserman
---
arch_init.c | 2 +-
1 file changed, 1 insertio
Signed-off-by: Orit Wasserman
---
savevm.c | 15 +++
1 file changed, 15 insertions(+)
diff --git a/savevm.c b/savevm.c
index 35c8d1e..6608b6e 100644
--- a/savevm.c
+++ b/savevm.c
@@ -39,6 +39,7 @@
#include "qmp-commands.h"
#include "trace.h"
#include "qemu/bitops.h"
+#include "qem
This will allow us to write an iovec
Signed-off-by: Orit Wasserman
Reviewed-by: Juan Quintela
---
include/migration/qemu-file.h | 7 +++
1 file changed, 7 insertions(+)
diff --git a/include/migration/qemu-file.h b/include/migration/qemu-file.h
index df81261..8d3da9b 100644
--- a/include/mi
In migration all data is copied to a static buffer in QEMUFile,
this hurts our network bandwidth and CPU usage especially with large guests.
We switched to iovec for storing different buffers to send (even a byte field is
considered as a buffer) and use writev to send the iovec.
writev was chosen (
On Thu, Mar 21, 2013 at 07:22:01PM +0100, Juan Quintela wrote:
> "Michael S. Tsirkin" wrote:
>
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >> We can optimize at some pount to write a bigger/different header and
> >> sent a bunch of pages together, but just now we don't have that code.
> >>
> >>
Since I'm throwing all this code out there I'm also signing up to maintain it.
Send me your bug reports.
Cc: Michael Tsirkin
Cc: Stefan Berger
Signed-off-by: Joel Schopp
---
MAINTAINERS | 8
1 file changed, 8 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 0ca7e1d..3e32ecb
This patch reorganizes qemu file operations to be in their own source file
instead of being lumped in savevm.c. Besides being more logical for maintenance
it also makes it easier for future users of the file functions to add tests.
Cc: pbonz...@redhat.com
Cc: Michael Tsirkin
Signed-off-by: Stefa
Add a sized buffer interface to qapi for serializing and
deserializing of u8[], u16[], u32[] and u64[] with proper
handling of endianess.
Cc: Michael Roth
Cc: Michael Tsirkin
Signed-off-by: Stefan Berger
Signed-off-by: Joel Schopp
---
include/qapi/visitor-impl.h | 3 +++
include/qapi/visitor.
Add 3 very short file wrapper functions to make code that follows more
readable.
Cc: Michael Tsirkin
Signed-off-by: Stefan Berger
Signed-off-by: Joel Schopp
---
include/migration/qemu-file.h | 3 +++
util/qemu-file.c | 30 ++
2 files changed, 33 insert
This patch series implements ASN.1 BER visitors for encoding and decoding
of data into byte streams.
Stefan Berger (9):
Move some contents of savevm.c to qemu-file.c
3 new file wrappers
QEMUSizedBuffer
QAPI: add type_sized_buffer
ASN.1 output visitor
ASN.1 input visitor
Extend test-v
On 03/21/2013 07:37 PM, Juan Quintela wrote:
> Orit Wasserman wrote:
>> This will remove an unneeded copy of guest memory pages.
>> For the page header and device state we still copy the data to the
>> static buffer the other option is to allocate the memory on demand
>> which is more expensive.
>
Gerd Hoffmann writes:
> Hi,
>
> If so, we're going to need to model what the hardware actually
> does, which is that there's a single connection on the back
> of the box for a monitor, and it's guest software controllable
> which of the two display devices is routed to the conne
On Thu, 21 Mar 2013 11:13:13 -0500
mdroth wrote:
> > Looks like you guys have no *practical* problems to solve. Congrats!
> > Take a vacation! Please report back no later than 275 years from now,
> > to make sure this 64 bit fd counter overflow problem gets taken care of
> > in time. ;-P
> >
"Michael S. Tsirkin" wrote:
>>
>>
>>
>>
>>
>>
>>
>> We can optimize at some pount to write a bigger/different header and
>> sent a bunch of pages together, but just now we don't have that code.
>>
>> Later, Juan.
>
> Sending the page can do vmsplice, can't it?
> Multipage is likely a good
Alon Levy writes:
> Note that the handler is called chr_is_guest_connected and not
> chr_is_fe_connected, consistent with other members of CharDriverState.
Sorry, I don't get it.
There isn't a notion of "connected" for the front-ends in the char
layer. The closest thing is whether add_handlers
On Thu, Mar 21, 2013 at 11:11:15AM -0600, Jason Gunthorpe wrote:
> On Thu, Mar 21, 2013 at 11:39:47AM +0200, Michael S. Tsirkin wrote:
> > On Thu, Mar 21, 2013 at 02:13:38AM -0700, Roland Dreier wrote:
> > > On Thu, Mar 21, 2013 at 1:51 AM, Michael S. Tsirkin
> > > wrote:
> > > >> In that case, n
On 03/21/2013 09:57 AM, Peter Lieven wrote:
> performance gain on SSE2 is approx. 20-25%. altivec
> is not tested. performance for unsigned long arithmetic
> is unchanged.
>
> Signed-off-by: Peter Lieven
> ---
> util/cutils.c |5 +
> 1 file changed, 5 insertions(+)
>
> diff --git a/util
On 03/21/2013 09:57 AM, Peter Lieven wrote:
> this adds buffer_find_nonzero_offset() which is a SSE2/Altives
s/Altives/Altivec/
> optimized function that searches for non-zero content in a
> buffer.
>
> due to the optimizations used in the function there are restrictions
> on buffer address and
Hi!
Here is the patch that allows us to specify the name of tap interface
when -netdev bridge is used. It's like -netdev tap,ifname=xxx, but for
bridges.
** Motivation **
We've got zillions of VMs and would like to see meaningful names of tap
interfaces. This is really useful for for, e.g., sy
On Thu, Mar 21, 2013 at 11:57:32AM -0600, Jason Gunthorpe wrote:
> On Thu, Mar 21, 2013 at 07:42:37PM +0200, Michael S. Tsirkin wrote:
>
> > It doesn't actually, and our app would sometimes write to these pages.
> > It simply does not care which version does the remote get in this case
> > since w
From: Petar Jovanovic
do_semop() is called from two places, and one of these fails to convert
return error to target errno when semop fails. This patch changes the
function to always return target errno in case of an unsuccessful call.
Signed-off-by: Petar Jovanovic
---
linux-user/syscall.c |
On Thu, Mar 21, 2013 at 11:21:50AM -0600, Jason Gunthorpe wrote:
> On Thu, Mar 21, 2013 at 07:15:25PM +0200, Michael S. Tsirkin wrote:
>
> > No because application does this:
> > init page
> >
> > ...
> >
> > after a lot of time
> >
> > ..
> >
> > register
> > send
> > unregister
> >
> > so i
On Thu, Mar 21, 2013 at 06:44:14PM +0100, Juan Quintela wrote:
> "Michael S. Tsirkin" wrote:
> > On Thu, Mar 21, 2013 at 06:27:42PM +0200, Orit Wasserman wrote:
> >> On 03/21/2013 06:16 PM, Michael S. Tsirkin wrote:
> >> > On Thu, Mar 21, 2013 at 06:05:40PM +0200, Orit Wasserman wrote:
> >> >> Thi
"Michael S. Tsirkin" wrote:
> On Thu, Mar 21, 2013 at 06:27:42PM +0200, Orit Wasserman wrote:
>> On 03/21/2013 06:16 PM, Michael S. Tsirkin wrote:
>> > On Thu, Mar 21, 2013 at 06:05:40PM +0200, Orit Wasserman wrote:
>> >> This way we send one big buffer instead of many small ones
>> >>
>> >> Signe
Orit Wasserman wrote:
> In migration all data is copied to a static buffer in QEMUFile,
> this hurts our network bandwidth and CPU usage especially with large guests.
> We switched to iovec for storing different buffers to send (even a byte field
> is
> considered as a buffer) and use writev to s
Orit Wasserman wrote:
> This way we send one big buffer instead of many small ones
>
> Signed-off-by: Orit Wasserman
> ---
> savevm.c | 21 +
> 1 file changed, 17 insertions(+), 4 deletions(-)
>
> diff --git a/savevm.c b/savevm.c
> index 50e8fb2..13a533b 100644
> --- a/savevm
On 03/21/2013 07:38 PM, Eric Blake wrote:
> On 03/21/2013 10:05 AM, Orit Wasserman wrote:
>> Signed-off-by: Orit Wasserman
>> ---
>> savevm.c | 2 ++
>> 1 file changed, 2 insertions(+)
>>
>> diff --git a/savevm.c b/savevm.c
>> index baa45ae..686c8c8 100644
>> --- a/savevm.c
>> +++ b/savevm.c
>> @
On 03/21/2013 10:05 AM, Orit Wasserman wrote:
> Signed-off-by: Orit Wasserman
> ---
> savevm.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/savevm.c b/savevm.c
> index baa45ae..686c8c8 100644
> --- a/savevm.c
> +++ b/savevm.c
> @@ -647,6 +647,8 @@ void qemu_put_byte(QEMUFile *f, in
On 03/21/2013 07:34 PM, Juan Quintela wrote:
> Orit Wasserman wrote:
>> This allow us to add a buffer to the iovec to send without copying it
>> into the static buffer.
>>
>> Signed-off-by: Orit Wasserman
>> ---
>> include/migration/qemu-file.h | 5 +
>> savevm.c | 37 +
Orit Wasserman wrote:
> This will remove an unneeded copy of guest memory pages.
> For the page header and device state we still copy the data to the
> static buffer the other option is to allocate the memory on demand
> which is more expensive.
>
> Signed-off-by: Orit Wasserman
> ---
> arch_ini
On 03/21/2013 06:29 PM, Michael S. Tsirkin wrote:
> On Thu, Mar 21, 2013 at 06:27:42PM +0200, Orit Wasserman wrote:
>> On 03/21/2013 06:16 PM, Michael S. Tsirkin wrote:
>>> On Thu, Mar 21, 2013 at 06:05:40PM +0200, Orit Wasserman wrote:
This way we send one big buffer instead of many small one
1 - 100 of 338 matches
Mail list logo
