On 18/02/2015 06:01, David Gibson wrote:
> A number of ARM embedded boards include EHCI USB host controllers which
> appear as directly mapped devices, rather than sitting on a PCI bus.
>
> At present code to emulate such devices is included whenever EHCI support
> is included. This patch adjus
On Di, 2015-02-17 at 14:28 +0100, Markus Armbruster wrote:
> This series requires my "[PATCH 00/10] pci: Partial conversion to
> realize" and my "[PATCH 0/9] Clean up around error_get_pretty(),
> qerror_report_err()".
>
> Markus Armbruster (4):
> usb: Propagate errors through usb_register_compan
Hi
Can anybody explain what the reasons, politics and consequences of
adding RFC into patch header in this mailing list? I think, it is not
worth to add this information to
http://wiki.qemu.org/Contribute/SubmitAPatch
--
Best regards,
Vladimir
Hi,
> Thanks for this code. Is this how I should display the Serial console?
> void myDisplaySerialConsole()
> {
> QemuConsole *con;
> char *name;
> int index = 0;
>
> con = qemu_console_lookup_by_index(0);
> while(con != NULL) {
> name = qemu_console_get_label
Liu Yuan writes:
> From: Liu Yuan
>
> These functions mix up -1 and -errno in return values and would might cause
> trouble error handling in the call chain.
>
> This patch let them return -errno and add some comments.
>
> Reported-by: Markus Armbruster
> Signed-off-by: Liu Yuan
Did you revie
On 18 February 2015 at 17:10, Vladimir Sementsov-Ogievskiy
wrote:
> Can anybody explain what the reasons, politics and consequences of adding
> RFC into patch header in this mailing list? I think, it is not worth to add
> this information to http://wiki.qemu.org/Contribute/SubmitAPatch
"RFC" mean
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On Tue, 17 Feb 2015 11:03:11 -0700
Eric Blake wrote:
> On 02/17/2015 07:24 AM, Michael Mueller wrote:
> > This patch implements a new QMP request named 'query-cpu-model'.
> > It returns the cpu model of cpu 0 and its backing accelerator.
> >
> > req
Am 12.02.2015 um 18:09 schrieb Jens Freimann:
> Cornelia, Alex, Christian,
>
> this patches are a rework of guest memory access to allow access to
> guest logical address space.
>
> Frank Blaschka (1):
> s390x/pci: Rework memory access in zpci instruction
>
> Thomas Huth (24):
> s390x/mmu:
On Di, 2015-02-17 at 17:40 +0100, Michal Privoznik wrote:
> Currently, if the ticketing (password) is not set at the command
> line, there's no way how to set it afterwards. Or vice versa -
> disable previously set password. I think it's worth allowing
> users to do that. The use case may be a teac
On Tue, 17 Feb 2015 20:46:32 +0100
"Michael S. Tsirkin" wrote:
> On Mon, Feb 09, 2015 at 10:53:23AM +, Igor Mammedov wrote:
> > Adds for dynamic AML creation, which will be used
> > for piecing ASL/AML primitives together and hiding
> > from user/caller details about how nested context
> > sh
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On Tue, 17 Feb 2015 12:16:20 -0700
Eric Blake wrote:
> On 02/17/2015 07:24 AM, Michael Mueller wrote:
> > The option --probe allows to switch into probe mode also for machines
> > different from none. If one or more accelerators are specified these
>
writes:
> From: Gonglei
>
> Patch 1 check boot order arguments validation
> before vm running. Patch 2 passing &error_abort
> to restore_boot_order() because qemu_boot_set()
> can't fail in that scenario.
Reviewed-by: Markus Armbruster
Gerd Hoffmann writes:
> On Di, 2015-02-17 at 14:28 +0100, Markus Armbruster wrote:
>> This series requires my "[PATCH 00/10] pci: Partial conversion to
>> realize" and my "[PATCH 0/9] Clean up around error_get_pretty(),
>> qerror_report_err()".
>>
>> Markus Armbruster (4):
>> usb: Propagate er
On Mon, 16 Feb 2015 22:36:03 +0100
"Michael S. Tsirkin" wrote:
> Drop duplicated code. Minor codechanges were required
> as geometry is a sub-structure now.
>
> Signed-off-by: Michael S. Tsirkin
> ---
> include/hw/virtio/virtio-blk.h | 77
> +-
> hw/blo
So, if I understand well, it's "just" some kind of race condition in the
handling of writes in the VDI driver of Qemu?
Does that mean it can also affect VMs running with VDI disk(s)?
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
ht
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On Tue, 17 Feb 2015 11:09:34 -0700
Eric Blake wrote:
> On 02/17/2015 07:24 AM, Michael Mueller wrote:
> > This patch implements the QMP command 'query-cpu-definitions' in the S390
> > context. The command returns a in terms of machine release date de
On Mon, 16 Feb 2015 22:36:09 +0100
"Michael S. Tsirkin" wrote:
> Drop duplicated code.
>
> Signed-off-by: Michael S. Tsirkin
> ---
> include/hw/virtio/virtio-net.h | 151
> +
> include/net/tap.h | 24 +--
> 2 files changed, 2 insertion
On Mon, 16 Feb 2015 22:36:15 +0100
"Michael S. Tsirkin" wrote:
> Drop duplicated code.
>
> Signed-off-by: Michael S. Tsirkin
> ---
> include/hw/virtio/virtio-rng.h | 4 +---
> 1 file changed, 1 insertion(+), 3 deletions(-)
>
> diff --git a/include/hw/virtio/virtio-rng.h b/include/hw/virtio/vi
On Tue, 17 Feb 2015 20:53:52 +0100
"Michael S. Tsirkin" wrote:
> On Mon, Feb 09, 2015 at 10:53:26AM +, Igor Mammedov wrote:
> > it will be reused for adding a plain integer value into AML.
> >
> > Signed-off-by: Igor Mammedov
> > ---
> > hw/acpi/aml-build.c | 19 +++
> > h
On Tue, 17 Feb 2015 20:15:42 +0100
"Michael S. Tsirkin" wrote:
> On Tue, Feb 17, 2015 at 05:47:26PM +0100, Igor Mammedov wrote:
> > On Tue, 17 Feb 2015 17:35:55 +0100
> > "Michael S. Tsirkin" wrote:
> >
> > > On Mon, Feb 09, 2015 at 10:53:27AM +, Igor Mammedov wrote:
> > > > Signed-off-by:
On Tue, 17 Feb 2015 17:42:00 +0100
"Michael S. Tsirkin" wrote:
> On Mon, Feb 09, 2015 at 10:53:28AM +, Igor Mammedov wrote:
> > it replaces prebuilt SSDT table header template copying/patching
> > with AML API
> >
> > Signed-off-by: Igor Mammedov
>
> I note that this changed the SSDT signa
On Tue, 17 Feb 2015 20:39:03 +0100
"Michael S. Tsirkin" wrote:
> On Mon, Feb 09, 2015 at 10:53:43AM +, Igor Mammedov wrote:
> > Replace AML template patching with direct composing
> > of PCI device entries in C. It allows to simplify
> > PCI tree generation further and saves us about 400LOC
>
On Tue, 17 Feb 2015 17:41:03 +0100
"Michael S. Tsirkin" wrote:
> On Mon, Feb 09, 2015 at 10:53:45AM +, Igor Mammedov wrote:
> > QEMU doesn't implement/advertize PM1b_CNT_BLK
> > register block so do not set/patch its \_Sx
> > values to avoid confusion.
> >
> > Signed-off-by: Igor Mammedov
>
From: David Gibson
A number of ARM embedded boards include EHCI USB host controllers which
appear as directly mapped devices, rather than sitting on a PCI bus.
At present code to emulate such devices is included whenever EHCI support
is included. This patch adjusts teh config options to only in
From: Markus Armbruster
It's tempting, because usbdevice_create() is so simple to use. But
there's a lot of unwanted complexity behind the simple interface.
Switch to usb_create_simple().
Cc: Alexander Graf
Cc: qemu-...@nongnu.org
Signed-off-by: Markus Armbruster
Reviewed-by: Alexander Graf
From: Markus Armbruster
USBDevice's realize method usb_qdev_realize() automatically creates a
usb-hub when only one port is left. Creating devices in realize
methods is questionable, but works.
If usb-hub creation fails, an error is reported to stderr, but the
failure is otherwise ignored. We
From: Markus Armbruster
Most LegacyUSBFactory usbdevice_init() methods realize with
qdev_init_nofail(), even though their caller usbdevice_create() can
handle failure. Okay if it really can't fail (I didn't check), but
somewhat brittle.
usb_msd_init() and usb_bt_init() call qdev_init(). The la
0212' into
staging (2015-02-13 11:44:50 +)
are available in the git repository at:
git://git.kraxel.org/qemu tags/pull-usb-20150218-1
for you to fetch changes up to c3cf77cb63b71618224129df41f114488e0f74e4:
Make sysbus EHCI devices ARM only by default (2015-02-18
From: Markus Armbruster
Signed-off-by: Markus Armbruster
Signed-off-by: Gerd Hoffmann
---
hw/usb/bus.c | 4
hw/usb/dev-bluetooth.c | 6 +-
hw/usb/dev-network.c | 3 ---
hw/usb/dev-serial.c| 3 ---
hw/usb/dev-storage.c | 3 ---
5 files changed, 1 insertion(+), 18 dele
From: Markus Armbruster
Instead of returning null pointer. Matches pci_create_simple(),
isa_create_simple(), sysbus_create_simple(). It's unused since the
previous commit, but I'll put it to use again shortly.
Signed-off-by: Markus Armbruster
Signed-off-by: Gerd Hoffmann
---
hw/usb/bus.c |
From: Markus Armbruster
Because it produces beauties like
(qemu) usb_add mouse
Failed to initialize USB device 'usb-mouse': Error: tried to attach usb
device QEMU USB Mouse to a bus with no free ports
Signed-off-by: Markus Armbruster
Signed-off-by: Gerd Hoffmann
---
hw/usb/bus.c | 4
On Tue, 17 Feb 2015 17:44:31 +0100
"Michael S. Tsirkin" wrote:
> On Mon, Feb 09, 2015 at 10:54:09AM +, Igor Mammedov wrote:
> > It completes dynamic SSDT generation and makes it
> > independed of IASL binary blobs. It also hides
> > from user all pointer arithmetic when building
> > SSDT whic
From: Markus Armbruster
It's tempting, because usbdevice_create() is so simple to use. But
there's a lot of unwanted complexity behind the simple interface.
Switch to usb_create_simple().
Cc: Magnus Damm
Signed-off-by: Markus Armbruster
Signed-off-by: Gerd Hoffmann
---
hw/sh4/r2d.c | 2 +-
On Tue, Feb 17, 2015 at 09:12:45AM -0700, Eric Blake wrote:
> On 02/17/2015 07:39 AM, Stefan Hajnoczi wrote:
> > -void bdrv_drain_all(void)
> > -{
> > -/* Always run first iteration so any pending completion BHs run */
> > -bool busy = true;
> > -BlockDriverState *bs;
> > -
> > -whi
On Di, 2015-02-17 at 10:03 +0100, Gerd Hoffmann wrote:
> On Di, 2015-02-17 at 09:42 +0100, Michael S. Tsirkin wrote:
> > On Tue, Feb 17, 2015 at 09:23:35AM +0100, Gerd Hoffmann wrote:
> > > Hi,
> > >
> > > > Should we forbid setting both legacy and modern at the same time?
> > >
> > > No, this
qerror_report_err() is a transitional interface to help with
converting existing monitor commands to QMP. It should not be used
elsewhere. Replace by error_report_err() in initial startup helpers
machine_set_property() and object_create().
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blak
qerror_report_err() is a transitional interface to help with
converting existing monitor commands to QMP. It should not be used
elsewhere. Replace by error_report_err() in initial startup helper
configure_tpm().
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blake
---
tpm.c | 6 ++
1
Tree-wide cleanup, but the tree-wide changes are mechanical.
Includes separately posted "[PATCH] vhost-scsi: Improve error
reporting for invalid vhostfd".
Markus Armbruster (11):
vhost-scsi: Improve error reporting for invalid vhostfd
error: New convenience function error_report_err()
error
qerror_report_err() is a transitional interface to help with
converting existing monitor commands to QMP. It should not be used
elsewhere. Replace by error_report_err() in HMP command handler
hmp_trace_event().
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blake
---
monitor.c | 3 +--
1
qerror_report_err() is a transitional interface to help with
converting existing monitor commands to QMP. It should not be used
elsewhere. Replace by error_report_err().
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blake
---
qemu-img.c | 3 +--
1 file changed, 1 insertion(+), 2 deletion
qerror_report_err() is a transitional interface to help with
converting existing monitor commands to QMP. It should not be used
elsewhere. Replace by error_report_err() in HMP command handler
hmp_host_net_add() and initial startup helpers net_init_client(),
net_init_netdev(). Keep it in QMP comm
Coccinelle semantic patch:
@@
expression E;
@@
-error_report("%s", error_get_pretty(E));
-error_free(E);
+error_report_err(E);
@@
expression E, S;
@@
-error_report("%s", error_get_pretty(E));
+error_report_err(E);
(
exit(
We get two error messages: one from monitor_handle_fd_param2(), and
another one from vhost_scsi_realize(). The second one gets suppressed
in QMP context.
That's because monitor_handle_fd_param() calls qerror_report_err().
Calling qerror_report_err() is always inappropriate in realize
methods, bec
On Tue, 17 Feb 2015 20:14:06 +0100
"Michael S. Tsirkin" wrote:
> On Tue, Feb 17, 2015 at 06:51:16PM +0100, Igor Mammedov wrote:
> > On Tue, 17 Feb 2015 17:47:03 +0100
> > "Michael S. Tsirkin" wrote:
> >
> > > This seems to change the contents of the tables in too many ways. So of
> > > it need
qerror_report_err() is a transitional interface to help with
converting existing monitor commands to QMP. It should not be used
elsewhere. Replace by error_report_err() in initial startup helper
numa_init_func() and board setup helper
memory_region_allocate_system_memory().
Signed-off-by: Markus
I've typed error_report("%s", error_get_pretty(ERR)) too many times
already, and I've fixed too many instances of qerror_report_err(ERR)
to error_report("%s", error_get_pretty(ERR)) as well. Capture the
pattern in a convenience function.
Since it's almost invariably followed by error_free(), stuf
monitor_handle_fd_param() is a wrapper around
monitor_handle_fd_param2() that feeds errors to qerror_report_err()
instead of returning them. qerror_report_err() is inappropriate in
many contexts. monitor_handle_fd_param() looks simpler than
monitor_handle_fd_param2(), which tempts use. Remove th
On Tue, 17 Feb 2015 20:39:03 +0100
"Michael S. Tsirkin" wrote:
> On Mon, Feb 09, 2015 at 10:53:43AM +, Igor Mammedov wrote:
> > Replace AML template patching with direct composing
> > of PCI device entries in C. It allows to simplify
> > PCI tree generation further and saves us about 400LOC
>
qerror_report_err() is a transitional interface to help with
converting existing monitor commands to QMP. It should not be used
elsewhere. Replace by error_report_err() in legacy chardev parser
qemu_chr_parse_compat(). Legacy chardev syntax is not to be used in
QMP.
Signed-off-by: Markus Armbru
On Tue, 17 Feb 2015 20:03:52 +0100
"Michael S. Tsirkin" wrote:
> On Tue, Feb 17, 2015 at 05:47:26PM +0100, Igor Mammedov wrote:
> > On Tue, 17 Feb 2015 17:35:55 +0100
> > "Michael S. Tsirkin" wrote:
> >
> > > On Mon, Feb 09, 2015 at 10:53:27AM +, Igor Mammedov wrote:
> > > > Signed-off-by:
On Mon, Feb 16, 2015 at 04:58:05PM +, Dr. David Alan Gilbert (git) wrote:
> From: "Dr. David Alan Gilbert"
>
> It looks like the dtrace trace code gets upset if you have trace names
> with __ in, which the migration/rdma.c code does.
>
> Rename the functions and the associated traces.
>
> F
Screwed up, v2 coming, sorry!
Am 10.02.2015 um 21:29 hat Max Reitz geschrieben:
> Add a test for conversion between different refcount widths and errors
> specific to certain widths (i.e. snapshots with refcount_bits=1).
>
> Signed-off-by: Max Reitz
> Reviewed-by: Eric Blake
Can you please split this patch, so that the part
Am 10.02.2015 um 21:28 hat Max Reitz geschrieben:
> Add a refcount_order parameter to qcow2_create2(), use that value for
> the image header and for calculating the size required for
> preallocation.
>
> For now, always pass 4.
>
> This addition requires changes to the calculation of the file siz
When I can't guess a handler's name, I have to look it up in
hmp-commands.hx. Happens often enough to annoy me. Let's make them
more predictable.
Touches several subsystems because HMP command handlers are still
spread far and wide, but the changes are trivial. Speak up if you
want your subsyst
Some are called do_COMMAND() (old ones, usually), some hmp_COMMAND(),
and sometimes COMMAND pointlessly differs in spelling.
Normalize to hmp_COMMAND(), where COMMAND is exactly the command name
with '-' replaced by '_'.
Exceptions:
* do_device_add() and client_migrate_info() *not* renamed to
Some are called do_info_SUBCOMMAND() (old ones, usually), some
hmp_info_SUBCOMMAND(), some SUBCOMMAND_info(), sometimes SUBCOMMAND
pointlessly differs in spelling.
Normalize to hmp_info_SUBCOMMAND(), where SUBCOMMAND is exactly the
subcommand name with '-' replaced by '_'.
Exceptions:
* sun4m_ir
I've typed error_report("%s", error_get_pretty(ERR)) too many times
already, and I've fixed too many instances of qerror_report_err(ERR)
to error_report("%s", error_get_pretty(ERR)) as well. Capture the
pattern in a convenience function.
Since it's almost invariably followed by error_free(), stuf
We get two error messages: one from monitor_handle_fd_param2(), and
another one from vhost_scsi_realize(). The second one gets suppressed
in QMP context.
That's because monitor_handle_fd_param() calls qerror_report_err().
Calling qerror_report_err() is always inappropriate in realize
methods, bec
qerror_report_err() is a transitional interface to help with
converting existing monitor commands to QMP. It should not be used
elsewhere. Replace by error_report_err() in legacy chardev parser
qemu_chr_parse_compat(). Legacy chardev syntax is not to be used in
QMP.
Signed-off-by: Markus Armbru
Tree-wide cleanup, but the tree-wide changes are mechanical.
Includes separately posted "[PATCH] vhost-scsi: Improve error
reporting for invalid vhostfd".
The following changes since commit cd2d5541271f1934345d8ca42f5fafff1744eee7:
Merge remote-tracking branch 'remotes/rth/tags/pull-tcg-201502
monitor_handle_fd_param() is a wrapper around
monitor_handle_fd_param2() that feeds errors to qerror_report_err()
instead of returning them. qerror_report_err() is inappropriate in
many contexts. monitor_handle_fd_param() looks simpler than
monitor_handle_fd_param2(), which tempts use. Remove th
qerror_report_err() is a transitional interface to help with
converting existing monitor commands to QMP. It should not be used
elsewhere. Replace by error_report_err().
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blake
---
qemu-img.c | 3 +--
1 file changed, 1 insertion(+), 2 deletion
qerror_report_err() is a transitional interface to help with
converting existing monitor commands to QMP. It should not be used
elsewhere. Replace by error_report_err() in HMP command handler
hmp_host_net_add() and initial startup helpers net_init_client(),
net_init_netdev(). Keep it in QMP comm
qerror_report_err() is a transitional interface to help with
converting existing monitor commands to QMP. It should not be used
elsewhere. Replace by error_report_err() in HMP command handler
hmp_trace_event().
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blake
---
monitor.c | 3 +--
1
Coccinelle semantic patch:
@@
expression E;
@@
-error_report("%s", error_get_pretty(E));
-error_free(E);
+error_report_err(E);
@@
expression E, S;
@@
-error_report("%s", error_get_pretty(E));
+error_report_err(E);
(
exit(
qerror_report_err() is a transitional interface to help with
converting existing monitor commands to QMP. It should not be used
elsewhere. Replace by error_report_err() in initial startup helpers
machine_set_property() and object_create().
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blak
Leftovers from
d1f2964 qapi: Convert query-spice
791e7c8 qapi: Convert query-migrate
0fe6a7f slirp: Drop statistic code
Signed-off-by: Markus Armbruster
---
include/migration/migration.h | 4
include/sysemu/sysemu.h | 3 ---
include/ui/qemu-spice.h | 3 ---
3 files changed, 10 d
qerror_report_err() is a transitional interface to help with
converting existing monitor commands to QMP. It should not be used
elsewhere. Replace by error_report_err() in initial startup helper
numa_init_func() and board setup helper
memory_region_allocate_system_memory().
Signed-off-by: Markus
qerror_report_err() is a transitional interface to help with
converting existing monitor commands to QMP. It should not be used
elsewhere. Replace by error_report_err() in initial startup helper
configure_tpm().
Signed-off-by: Markus Armbruster
Reviewed-by: Eric Blake
---
tpm.c | 6 ++
1
Am 10.02.2015 um 21:28 hat Max Reitz geschrieben:
> Add a creation option to qcow2 for setting the refcount order of images
> to be created, and respect that option's value.
>
> This breaks some test outputs, fix them.
>
> Signed-off-by: Max Reitz
> Reviewed-by: Eric Blake
> Reviewed-by: Stefan
On Wed, Feb 18, 2015 at 11:32:55AM +0100, Gerd Hoffmann wrote:
> On Di, 2015-02-17 at 10:03 +0100, Gerd Hoffmann wrote:
> > On Di, 2015-02-17 at 09:42 +0100, Michael S. Tsirkin wrote:
> > > On Tue, Feb 17, 2015 at 09:23:35AM +0100, Gerd Hoffmann wrote:
> > > > Hi,
> > > >
> > > > > Should we for
On Wed, Feb 18, 2015 at 10:57:18AM +0100, Igor Mammedov wrote:
> On Tue, 17 Feb 2015 17:42:00 +0100
> "Michael S. Tsirkin" wrote:
>
> > On Mon, Feb 09, 2015 at 10:53:28AM +, Igor Mammedov wrote:
> > > it replaces prebuilt SSDT table header template copying/patching
> > > with AML API
> > >
>
On Wed, Feb 18, 2015 at 11:47:08AM +0100, Igor Mammedov wrote:
> On Tue, 17 Feb 2015 20:03:52 +0100
> "Michael S. Tsirkin" wrote:
>
> > On Tue, Feb 17, 2015 at 05:47:26PM +0100, Igor Mammedov wrote:
> > > On Tue, 17 Feb 2015 17:35:55 +0100
> > > "Michael S. Tsirkin" wrote:
> > >
> > > > On Mon,
This patch adds functions to perform read and write operations
with replay log.
Signed-off-by: Pavel Dovgalyuk
---
replay/Makefile.objs |1
replay/replay-internal.c | 143 ++
replay/replay-internal.h | 44 ++
3 files changed, 18
Processing CPU_INTERRUPT_POLL requests in cpu_has_work functions
break the determinism of cpu_exec. This patch is required to make
interrupts processing deterministic.
Signed-off-by: Paolo Bonzini
Signed-off-by: Pavel Dovgalyuk
---
cpu-exec.c|6 ++
target-i386/cpu.c | 10 ++--
On Wed, Feb 18, 2015 at 11:42:15AM +0100, Igor Mammedov wrote:
> On Tue, 17 Feb 2015 20:39:03 +0100
> "Michael S. Tsirkin" wrote:
>
> > On Mon, Feb 09, 2015 at 10:53:43AM +, Igor Mammedov wrote:
> > > Replace AML template patching with direct composing
> > > of PCI device entries in C. It all
This patch adds icount event to the replay subsystem. This event corresponds
to execution of several instructions and used to synchronize input events
in the replay phase.
Reviewed-by: Paolo Bonzini
Signed-off-by: Pavel Dovgalyuk
---
replay/replay-internal.c | 21 +
repla
This patch removes "static" specifier from several qemu function to make
them visible to the replay module. It also invents several system functions
that will be used by replay.
Reviewed-by: Paolo Bonzini
Signed-off-by: Pavel Dovgalyuk
---
cpus.c |4 ++--
include/exec/exec
This set of patches is related to the reverse execution and deterministic
replay of qemu execution. This implementation of deterministic replay can
be used for deterministic debugging of guest code through gdb remote
interface.
These patches include only core function of the replay,
excluding th
Clock ticks are considered as the sources of non-deterministic data for
virtual machine. This patch implements saving the clock values when they
are acquired (virtual, host clock).
When replaying the execution corresponding values are read from log and
transfered to the module, which wants to read
This patch is required for deterministic replay to generate an exception
by trying executing an instruction without changing icount.
It adds new flag to TB for disabling icount while translating it.
Signed-off-by: Paolo Bonzini
Signed-off-by: Pavel Dovgalyuk
---
cpu-exec.c |7
This patch adds global variables, defines, functions declarations,
and function stubs for deterministic VM replay used by external modules.
Reviewed-by: Paolo Bonzini
Reviewed-by: Eric Blake
Signed-off-by: Pavel Dovgalyuk
---
Makefile.target |1
docs/replay.txt | 161 +
This patch includes modifications of common cpu files. All interrupts and
exceptions occured during recording are written into the replay log.
These events allow correct replaying the execution by kicking cpu thread
when one of these events is found in the log.
Signed-off-by: Pavel Dovgalyuk
---
This patch adds calls to replay functions into the icount setup block.
In record mode number of executed instructions is written to the log.
In replay mode number of istructions to execute is taken from the replay log.
Signed-off-by: Pavel Dovgalyuk
---
cpus.c | 38 +++
Bottom halves in AIO context are stored and removes
in LIFO order. It makes their execution non-deterministic.
This patch replaces the stack with queue to preserve the
order of bottom halves processing.
Signed-off-by: Pavel Dovgalyuk
---
async.c | 25 +
inc
This mutex will protect read/write operations for replay log.
Using mutex is necessary because most of the events consist of
several fields stored in the log. The mutex will help to avoid races.
Reviewed-by: Paolo Bonzini
Signed-off-by: Pavel Dovgalyuk
---
replay/replay-internal.c | 27 +
This patch adds module for saving and replaying asynchronous events.
These events include network packets, keyboard and mouse input,
USB packets, thread pool and bottom halves callbacks.
All events are stored in the queue to be processed at synchronization points
such as beginning of TB execution,
This patch replaces time() function calls with calls to
qemu_clock_get_ns(QEMU_CLOCK_HOST). It makes such requests deterministic
in record/replay mode of icount.
Signed-off-by: Pavel Dovgalyuk
---
vl.c | 12
1 files changed, 8 insertions(+), 4 deletions(-)
diff --git a/vl.c b/vl.
Some devices are not supported by record/replay subsystem.
This patch introduces replay blocker which denies starting record/replay
if such devices are included into the configuration.
Signed-off-by: Pavel Dovgalyuk
---
hw/bt/hci.c |7 +++
include/qapi/qmp/qerror.h |3 +
On Wed, Feb 18, 2015 at 10:50:54AM +0100, Igor Mammedov wrote:
> On Tue, 17 Feb 2015 20:15:42 +0100
> "Michael S. Tsirkin" wrote:
>
> > On Tue, Feb 17, 2015 at 05:47:26PM +0100, Igor Mammedov wrote:
> > > On Tue, 17 Feb 2015 17:35:55 +0100
> > > "Michael S. Tsirkin" wrote:
> > >
> > > > On Mon,
This patch records and replays simulator shutdown event.
Reviewed-by: Paolo Bonzini
Signed-off-by: Pavel Dovgalyuk
---
replay/replay-internal.h |2 ++
replay/replay.c | 12
replay/replay.h |5 +
vl.c |1 +
4 files changed, 20
This patch introduces bottom half event for replay queue. It saves the events
into the queue and process them at the checkpoints and instructions execution.
Signed-off-by: Pavel Dovgalyuk
---
async.c | 24 +++-
dma-helpers.c|4 +++-
hw/ide/a
This records user input (keyboard and mouse events) in record mode and replays
these input events in replay mode.
Signed-off-by: Pavel Dovgalyuk
---
include/ui/input.h |2 +
replay/Makefile.objs |1
replay/replay-events.c | 31 +
replay/replay-input.c| 159 +++
This patch introduces checkpoints that synchronize cpu thread and iothread.
When checkpoint is met in the code all asynchronous events from the queue
are executed.
Signed-off-by: Pavel Dovgalyuk
---
block.c | 12
cpus.c |5 +
main-loop.c
This patch adds identifier to aio requests. ID is used for creating bottom
halves and identifying them while replaying.
The patch also introduces several functions that make possible replaying
of the aio requests.
Signed-off-by: Pavel Dovgalyuk
---
block.c| 81 +
This patch modifies thread pool to allow replaying asynchronous thread tasks
synchronously in replay mode.
Signed-off-by: Pavel Dovgalyuk
---
block/raw-posix.c |6 -
block/raw-win32.c |4 +++-
include/block/thread-pool.h |4 +++-
replay/replay-events.c |
This patch introduces the functions for enabling the record/replay and for
freeing the resources when simulator closes.
Signed-off-by: Pavel Dovgalyuk
---
block.c |2 -
exec.c |1
replay/replay-internal.h |2 +
replay/replay.c | 138 +
This patch moves typedefs for QemuOpts and related types
to qemu/typedefs.h file.
Signed-off-by: Pavel Dovgalyuk
---
include/qemu/option.h |5 +
include/qemu/typedefs.h |3 +++
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/include/qemu/option.h b/include/qemu/optio
This patch introduces command line options for enabling recording or replaying
virtual machine behavior. "-record" option starts recording of the execution
and saves it into the log, specified with "fname" parameter. "-replay" option
is intended for replaying previously saved log.
Signed-off-by: P
1 - 100 of 398 matches
Mail list logo
