* Peter Xu (pet...@redhat.com) wrote:
> Add the latency distribution too for blocktime, using order-of-two buckets.
> It accounts for all the faults, from either vCPU or non-vCPU threads. With
> prior rework, it's very easy to achieve by adding an array to account for
> faults in each buckets.
>
在 2025/6/4 下午5:05, Alireza Sanaee 写道:
On Wed, 4 Jun 2025 14:55:00 +0800
Bibo Mao wrote:
On big endian host machine such as S390, bios-table-test fails to run.
And also linux kernel fails to boot.
This patches solves these two issues.
Bibo Mao (2):
hw/loongarch/virt: Fix big endian support
In some situations it is desirable to be able to specify the flash type
connected to a board. For example, the target operating system may not
support the default flash type, its support may be broken, or the qemu
emulation is insufficient and the default flash is not detected.
On top of that, the
If we have a server running disk requests that is for whatever reason
hanging or not able to process any more IO requests but still has some
in-flight requests previously issued by the guest OS, QEMU will still
try to drain the vring before shutting down even if it was explicitly
asked to do a "for
On Mon, Jun 09, 2025 at 03:12:54PM -0400, Peter Xu wrote:
> +static void migration_dump_blocktime(Monitor *mon, MigrationInfo *info)
> +{
> +if (info->has_postcopy_blocktime) {
> +monitor_printf(mon, "Postcopy Blocktime (ms): %" PRIu32 "\n",
> + info->postcopy_bloc
Looking up the vCPU index for each fault can be expensive when there're
hundreds of vCPUs. Provide a cache for tid->vcpu instead with a hash
table, then lookup from there.
When at it, add another counter to record how many non-vCPU faults it gets.
For example, the main thread can also access a gu
在 2025/6/5 下午5:28, Bibo Mao 写道:
Interrupt controller extioi supports 256 vectors, register EXTIOI_COREISR
records pending interrupt status with bitmap method. Size of EXTIOI_COREISR
is 256 / 8 = 0x20 bytes, EXTIOI_COREISR_END should be EXTIOI_COREISR_START
+ 0x20 rather than 0xB20.
Signed-off-by
On Thu, May 29, 2025 at 04:08:01PM +0100, Jonathan Cameron wrote:
> On Wed, 28 May 2025 12:07:23 +0100
> Jonathan Cameron wrote:
>
> > Previously these somewhat device like structures were tracked using a list
> > in the CXLState in each machine. This is proving restrictive in a few
> > cases wh
On Mon, Jun 09, 2025 at 06:05:16PM -0400, Peter Xu wrote:
> On Mon, Jun 09, 2025 at 03:12:54PM -0400, Peter Xu wrote:
> > +static void migration_dump_blocktime(Monitor *mon, MigrationInfo *info)
> > +{
> > +if (info->has_postcopy_blocktime) {
> > +monitor_printf(mon, "Postcopy Blocktime
Add the latency distribution too for blocktime, using order-of-two buckets.
It accounts for all the faults, from either vCPU or non-vCPU threads. With
prior rework, it's very easy to achieve by adding an array to account for
faults in each buckets.
Sample output for HMP (while for QMP it's simply
* Peter Xu (pet...@redhat.com) wrote:
> Blocktime so far only cares about the time one vcpu (or the whole system)
> got blocked. It would be also be helpful if it can also report the latency
> of page requests, which could be very sensitive during postcopy.
>
> Blocktime itself is sometimes not v
On Mon, Jun 9, 2025 at 11:19 PM Ben Dooks wrote:
>
> Add a (currently Genesy2 based) CVA6 machine.
>
> Has SPI and UART, the GPIO and Ethernet are currently black-holed
> as there is no hardware model for them (lowRISC ethernet and Xilinx
> GPIO)
>
> Signed-off-by: Ben Dooks
> ---
> v3:
> - fix m
On Mon, Jun 9, 2025 at 11:18 PM Ben Dooks wrote:
>
> Add TYPE_RISCV_CPU_CVA6 for the CVA6 core
>
> Signed-off-by: Ben Dooks
Acked-by: Alistair Francis
Alistair
> ---
> target/riscv/cpu-qom.h | 1 +
> target/riscv/cpu.c | 11 +++
> 2 files changed, 12 insertions(+)
>
> diff --git
On Mon, Jun 9, 2025 at 11:20 PM Ben Dooks wrote:
>
> Change to using TYPE_RISCV_CPU_CVA6 once this is merged.
You can also just change the patch order to not require this patch
>
> Signed-off-by: Ben Dooks
Reviewed-by: Alistair Francis
Alistair
> ---
> hw/riscv/cva6.c | 3 +--
> 1 file cha
This series aims to address SIGTERM/QMP quit() being a bit too graceful in
respect to devices. Both of the aforementioned ways to stop QEMU completely
bypass the guest OS so in that sense they're basically equal to pulling the
power plug on a computer, yet the device shutdown code still tries to do
This adds an ability to skip GET_VRING_BASE during device stop entirely,
and thus the expensive drain operation that this call entails as well,
which may be useful during a non-graceful shutdown in case the guest
operating system hangs or refuses to react to a previously requested
ACPI shutdown for
This can be useful for devices that might take too long to shut down
gracefully, but may have a way to shutdown quickly otherwise if needed
or explicitly requested by a force shutdown.
For now we only consider SIGTERM or the QMP quit() command a force
shutdown, since those bypass the guest entirel
From: Thomas Huth
When the testing docs were moved to a separate subfolder, the entries
in the MAINTAINERS file were missed. Update them now.
Fixes: ff41da50308 ("docs/devel: Split testing docs from the build docs and
move to separate folder")
Signed-off-by: Thomas Huth
---
MAINTAINERS | 15 +
On 07/06/2025 08.04, Haseung Bong wrote:
From: "haseung.bong"
The README file in tests/vm/ points to a non-existent file,
docs/devel/testing.rst. Update the README to point to
docs/devel/testing/main.rst, which now contains information
about VM testing.
Signed-off-by: Haseung Bong
---
tests
On 5/29/25 21:24, Steve Sistare wrote:
cpr-transfer will use the device name as a key to find the value
of the device descriptor in new QEMU. However, if the descriptor
number is specified by a command-line fd parameter, then
vfio_device_get_name creates a name that includes the fd number.
This
On 6/9/25 22:47, Steven Sistare wrote:
On 6/9/2025 4:30 PM, Cédric Le Goater wrote:
On 5/29/25 21:24, Steve Sistare wrote:
Register a vfio iommufd container and device for CPR, replacing the generic
CPR register call with a more specific iommufd register call. Add a
blocker if the kernel does
On 5/29/25 21:24, Steve Sistare wrote:
Define vfio_device_free_name to free the name created by
vfio_device_get_name. A subsequent patch will do more there.
No functional change.
Signed-off-by: Steve Sistare
---
include/hw/vfio/vfio-device.h | 1 +
hw/vfio/ap.c | 2 +-
hw/
On 5/29/25 21:24, Steve Sistare wrote:
Define a list of vfio devices in CPR state, in a subsection so that
older QEMU can be live updated to this version. However, new QEMU
will not be live updateable to old QEMU. This is acceptable because
CPR is not yet commonly used, and updates to older ver
On 6/9/25 18:44, Rorie Reyes wrote:
These functions can be invoked by the function that handles interception
of the CHSC SEI instruction for requests indicating the accessibility of
one or more adjunct processors has changed.
Signed-off-by: Rorie Reyes
Reviewed-by: Anthony Krowiak
---
hw/vfi
On 28.05.25 15:07, Andrey Zhadchenko wrote:
HBitmaps allow us to search set bits pretty fast. On the contrary,
when searching zeroes, we may be forced to fully traverse the lower
level.
When we run blockdev-backup with mode=full on top of snapshot filter
+ cbw filter, the job fills copy bitmap by
On Thu, 5 Jun 2025 at 03:26, wrote:
>
> From: Tao Tang
>
> The current definition of the SMMU_CR0_RESERVED mask is incorrect.
> It mistakenly treats bit 10 (DPT_WALK_EN) as a reserved bit
This is because our implementation pre-dates the revision
of the SMMUv3 spec which adds the DPT and this DPT
On Wed, Jun 04, 2025 at 03:28:03PM +0200, Hanna Czenczek wrote:
> The term "length" is ambiguous, use "blk_len" instead to be clear.
>
> Signed-off-by: Hanna Czenczek
> ---
> block/export/fuse.c | 12 ++--
> 1 file changed, 6 insertions(+), 6 deletions(-)
Reviewed-by: Stefan Hajnoczi
This is the trick that allows the parent-field initializer to be used
only for the object that it's meant to be initialized. This way,
the owner of a MemoryRegion must be the object that embeds it.
More information is in the comments; it's best explained with a simplified
example.
Signed-off-by:
On Mon, Jun 09, 2025 at 11:33:14AM -0400, Peter Xu wrote:
>
> Now I think I know part of what I've missed: I used to think the "config"
> of per-QMP-migrate-command can be totally temporary for a specific
> migration request, but then we need another MigrationState.parameters_2 to
> cache the old
This is a step towards safe bindings to instance_init: making the
implementation safe, though ensuring that *all* fields as initialized
is still up to the implementor. This helps making it clear what we
want from crates like pinned-init (the Linux one) or its inspiration
pin-init.
This series has
The array of BqlRefCell is not initialized yet at the
end of instance_init. In particular, the "state" field is NonNull
and therefore it is invalid to have it as zero bytes.
Note that MaybeUninit is necessary because assigning to self.timers[index]
would trigger Drop of the old value.
Signed-off
This removes undefined behavior associated to writing to uninitialized
fields, and makes it possible to remove "unsafe" from the instance_init
implementation.
However, the init function itself is still unsafe, because it must promise
(as a sort as MaybeUninit::assume_init) that all fields have bee
This is a smart pointer for MaybeUninit; it can be upcasted to the
already-initialized parent classes, or dereferenced to a MaybeUninit
for the class that is being initialized.
Signed-off-by: Paolo Bonzini
---
rust/qemu-api/src/qom.rs | 86 +++-
1 file changed
Add a macro that makes it possible to convert a MaybeUninit<> into
another MaybeUninit<> for a single field within it. Furthermore, it is
possible to use the resulting MaybeUninitField<> in APIs that take the
parent object, such as memory_region_init_io().
This allows removing some of the undefin
On 6/9/25 15:23, Andrey Zhadchenko wrote:
When hotplugging vCPUs to the Windows vms, we observed strange instance
crash on Intel(R) Xeon(R) CPU E3-1230 v6:
panic hyper-v: arg1='0x3e', arg2='0x46d359bbdff', arg3='0x56d359bbdff',
arg4='0x0', arg5='0x0'
Presumably, Windows thinks that hotplugged C
On Mon, Jun 09, 2025 at 11:51:17AM -0400, Peter Xu wrote:
> On Mon, Jun 09, 2025 at 11:37:02AM -0300, Fabiano Rosas wrote:
> > Peter Xu writes:
> >
> > > On Fri, Jun 06, 2025 at 05:23:18PM -0300, Fabiano Rosas wrote:
> > >> Peter Xu writes:
> > >>
> > >> > On Mon, Jun 02, 2025 at 10:38:08PM -03
Ping^2 on this one?
-- PMM
On Mon, 19 May 2025 at 15:32, Peter Maydell wrote:
>
> Ping? Any opinions on this?
>
> In the interim we've applied commit 82707dd4f0 to drop
> the specific duplicate-label that is causing problems
> right now, so patch 2 here will need the obvious trivial
> update. Bu
There're a few things off here in that logic, rewrite it. When at it, add
rich comment to explain each of the decisions.
Since this is very sensitive path for migration, below are the list of
things changed with their reasonings.
(1) Exact pending size is only needed for precopy not postcopy
Dave suggested the HMP output for "info migrate" can not only leverage the
lines but also better grouping:
https://lore.kernel.org/r/aC4_-nMc7FwsMf9p@gallifrey
I followed Dave's suggestion, and some more modifications on top:
- Added all elements into the picture
- Use size_to_str() and d
Now after merging the precopy and postcopy version of complete() hook,
rename the precopy version from save_live_complete_precopy() to
save_complete().
Dropping the "live" when at it, because it's in most cases not live when
happening (in precopy).
No functional change intended.
Signed-off-by: P
The check over PAGE_DIRTY_FOUND isn't necessary. We could indent one less
and assert that instead.
Reviewed-by: Juraj Marcin
Signed-off-by: Peter Xu
---
migration/ram.c | 20 +++-
1 file changed, 11 insertions(+), 9 deletions(-)
diff --git a/migration/ram.c b/migration/ram.c
in
Take notes on start/end state of dirty pages for the whole system.
Reviewed-by: Juraj Marcin
Signed-off-by: Peter Xu
---
migration/ram.c| 5 +
migration/trace-events | 1 +
2 files changed, 6 insertions(+)
diff --git a/migration/ram.c b/migration/ram.c
index c66ad3cf8b..a1d0e8ada2
It's not possible to happen in bg-snapshot case.
Reviewed-by: Juraj Marcin
Signed-off-by: Peter Xu
---
migration/migration.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/migration/migration.c b/migration/migration.c
index 4098870bce..e33e39ac74 100644
--- a/migration
On Mon, Jun 09, 2025 at 11:53:47AM -0400, Peter Xu wrote:
> On Mon, Jun 09, 2025 at 04:43:40PM +0100, Daniel P. Berrangé wrote:
> > On Mon, Jun 09, 2025 at 11:33:14AM -0400, Peter Xu wrote:
> > >
> > > Now I think I know part of what I've missed: I used to think the "config"
> > > of per-QMP-migra
v2:
- Collected R-bs
- Avoid using "\b" in HMP dumps [Markus, Dave]
The series is based on a small patch from Yanfei Xu here:
Based-on: <20250514115827.3216082-1-yanfei...@bytedance.com>
https://lore.kernel.org/r/20250514115827.3216082-1-yanfei...@bytedance.com
This is a series that collected ma
Unfortunately, it was never correctly shown..
This is only found when I started to look into making the blocktime feature
more useful (so as to avoid using bpftrace, even though I'm not sure which
one will be harder to use..).
So the old dump would look like this:
Postcopy vCPU Blocktime: 0-1,
This is a follow up on the other commit "migration/ram: avoid to do log
clear in the last round" but for postcopy.
https://lore.kernel.org/r/20250514115827.3216082-1-yanfei...@bytedance.com
I can observe more than 10% reduction of average page fault latency during
postcopy phase with this optimiz
The hook is only defined in two vmstate users ("ram" and "block dirty
bitmap"), meanwhile both of them define the hook exactly the same as the
precopy version. Hence, this postcopy version isn't needed.
No functional change intended.
Signed-off-by: Peter Xu
---
include/migration/register.h |
Since we use the same save_complete() hook for both precopy and postcopy,
add a set of helpers to invoke the hook() to dedup the code.
Signed-off-by: Peter Xu
---
migration/savevm.c | 78 ++
1 file changed, 44 insertions(+), 34 deletions(-)
diff --git
Move it out of vanilla postcopy session, but instead a standalone feature.
When at it, removing the NOTE because it's incorrect now after introduction
of max-postcopy-bandwidth, which can control the throughput even for
postcopy phase.
Reviewed-by: Juraj Marcin
Signed-off-by: Peter Xu
---
docs/
On Wed, Jun 04, 2025 at 03:28:10PM +0200, Hanna Czenczek wrote:
> FUSE allows creating multiple request queues by "cloning" /dev/fuse FDs
> (via open("/dev/fuse") + ioctl(FUSE_DEV_IOC_CLONE)).
>
> We can use this to implement multi-threading.
>
> For configuration, we don't need any more informat
On Wed, Jun 04, 2025 at 03:27:52PM +0200, Hanna Czenczek wrote:
> Hi,
>
> This series:
> - Fixes some bugs/minor inconveniences,
> - Removes libfuse from the request processing path,
> - Make the FUSE export use coroutines for request handling,
> - Introduces multi-threading into the FUSE export.
On Wed, Jun 04, 2025 at 03:28:12PM +0200, Hanna Czenczek wrote:
> Run qemu-img bench on a simple multi-threaded FUSE export to test that
> it works.
>
> Signed-off-by: Hanna Czenczek
> ---
> tests/qemu-iotests/308 | 51 ++
> tests/qemu-iotests/308.out | 56 +++
Peter Xu writes:
> (I had a reply in the other thread, that might have covered most of the
> points but maybe not this one..)
>
> On Mon, Jun 09, 2025 at 05:13:00PM +0100, Daniel P. Berrangé wrote:
>> Even if only a single mgmt app is involved this is still beneficial
>> because the migration in
Moved the logic for timestamped logging (~6 lines) from
a_nocheck__trace_foo(header) into a new qemu_log_timestamp() function in
util/log.c. This avoids code duplication across binaries and enables reuse as a
standalone utility.
Encapsulation helps reduce build size significantly, particularly w
Two files are attached to show the folder-wise size (in bytes) before and
after the change. Each file lists the folder names along with their
corresponding sizes.
On Mon, Jun 9, 2025 at 11:51 PM Tanish Desai
wrote:
> Moved the logic for timestamped logging (~6 lines) from
> a_nocheck__trace_foo(
On Thu, Jun 5, 2025 at 2:49 PM Paolo Bonzini wrote:
>
> On 6/5/25 20:37, Stefan Hajnoczi wrote:
> > On Thu, Jun 5, 2025 at 9:57 AM Paolo Bonzini wrote:
> >>> It's easier to understand the code generator and the generated code when
> >>> each trace event is implemented as a single function in the
On Mon, Jun 9, 2025 at 2:21 PM Tanish Desai wrote:
>
> Moved the logic for timestamped logging (~6 lines) from
> a_nocheck__trace_foo(header) into a new qemu_log_timestamp() function in
> util/log.c. This avoids code duplication across binaries and enables reuse as
> a standalone utility.
> Enc
I am guessing it was used to be 32bits because of the atomic ops. Now all
the atomic ops are gone and we're protected by a mutex instead, it's ok we
can switch to 64 bits.
Reasons to move over:
- Allow further patches to change the unit from ms to us: with postcopy
preempt mode, we're really
Blocktime so far only cares about the time one vcpu (or the whole system)
got blocked. It would be also be helpful if it can also report the latency
of page requests, which could be very sensitive during postcopy.
Blocktime itself is sometimes not very important, especially when one
thinks about
The variable vcpu_total_blocktime isn't easy to follow. In reality, it
wants to capture the case where all vCPUs are stopped, and now there will
be some vCPUs starts running.
The name now starts to conflict with vcpu_blocktime_total[], meanwhile it's
actually not necessary to have the variable at
Now with the mutex protection it's not needed anymore.
Reviewed-by: Fabiano Rosas
Signed-off-by: Peter Xu
---
migration/postcopy-ram.c | 23 ++-
1 file changed, 10 insertions(+), 13 deletions(-)
diff --git a/migration/postcopy-ram.c b/migration/postcopy-ram.c
index 32fa06da
Now with 64bits, the offseting using start_time is not needed anymore,
because the array can always remember the whole timestamp.
Then drop the unused parameter in get_low_time_offset() altogether.
Reviewed-by: Fabiano Rosas
Signed-off-by: Peter Xu
---
migration/postcopy-ram.c | 10 --
With 64-bit fields, it is trivial. The caution is when exposing any values
in QMP, it was still declared with milliseconds (ms). Hence it's needed to
do the convertion when exporting the values to existing QMP queries.
Reviewed-by: Fabiano Rosas
Signed-off-by: Peter Xu
---
migration/postcopy-
From: novafacing
This patch adds a function to the plugins API to allow plugins to write
register contents. It also moves the qemu_plugin_read_register function
so all the register-related functions are grouped together in the file.
Reviewed-by: Pierrick Bouvier
Signed-off-by: Rowan Hart
---
From: novafacing
This patch adds functions to the plugins API to allow reading and
writing memory via virtual addresses. These functions only permit doing
so on the current CPU, because there is no way to ensure consistency if
plugins are allowed to read or write to other CPUs that aren't current
From: novafacing
This patch updates the plugin version to gate new APIs and adds notes
describing what has been added.
Signed-off-by: Rowan Hart
---
include/qemu/qemu-plugin.h | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/include/qemu/qemu-plugin.h b/include/qemu/
From: novafacing
This patch adds a plugin that implements a simple form of hypercalls
from guest code to the plugin by using the register read API. It accepts
only one hypercall, which writes a magic value to guest memory.
Signed-off-by: Rowan Hart
---
tests/tcg/Makefile.target
From: novafacing
This patch adds functions to the plugins API to allow plugins to read
and write memory via hardware addresses. The functions use the current
address space of the current CPU in order to avoid exposing address
space information to users. A later patch may want to add a function to
The postcopy blocktime feature was tricky that it used quite some atomic
operations over quite a few arrays and vars, without explaining how that
would be thread safe. The thread safety here is about concurrency between
the fault thread and the fault resolution threads, possible to access the
same
When used to report page fault latencies, the blocktime feature can be
almost useless when KVM async page fault is enabled, because in most cases
such remote fault will kickoff async page faults, then it's not trackable
from blocktime layer.
After all these recent rewrites to blocktime layer, it's
From: novafacing
This patch adds a plugin that implements a simple form of hypercalls
from guest code to the plugin by using the register read API. It accepts
only one hypercall, which writes a magic value to guest memory.
Signed-off-by: Rowan Hart
---
tests/tcg/Makefile.target
From: novafacing
This patch adds functions to the plugins API to allow plugins to read
and write memory via hardware addresses. The functions use the current
address space of the current CPU in order to avoid exposing address
space information to users. A later patch may want to add a function to
This patch series adds several new API functions focused on enabling use
cases around reading and writing guest memory from QEMU plugins. To support
these new APIs, some utility functionality around retrieving information about
address spaces is added as well.
The new qemu_plugin_write_register ut
From: novafacing
This patch updates the plugin version to gate new APIs and adds notes
describing what has been added.
Signed-off-by: Rowan Hart
---
include/qemu/qemu-plugin.h | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/include/qemu/qemu-plugin.h b/include/qemu/
Peter Xu writes:
> On Mon, Jun 09, 2025 at 03:02:06PM -0300, Fabiano Rosas wrote:
>> Peter Xu writes:
>>
>> > On Mon, Jun 09, 2025 at 11:37:02AM -0300, Fabiano Rosas wrote:
>> >> Peter Xu writes:
>> >>
>> >> > On Fri, Jun 06, 2025 at 05:23:18PM -0300, Fabiano Rosas wrote:
>> >> >> Peter Xu w
I've addressed comments by Alex and Julian WRT the new restrictions on
flags for calling qemu_plugin_read/write_register by relaxing those
restrictions using the same system for setting and getting the current
flag state, and adding a set/clear before each callback invocation.
Please check
htt
On Mon, Jun 09, 2025 at 03:02:06PM -0300, Fabiano Rosas wrote:
> Peter Xu writes:
>
> > On Mon, Jun 09, 2025 at 11:37:02AM -0300, Fabiano Rosas wrote:
> >> Peter Xu writes:
> >>
> >> > On Fri, Jun 06, 2025 at 05:23:18PM -0300, Fabiano Rosas wrote:
> >> >> Peter Xu writes:
> >> >>
> >> >> > On
On Sun, Jun 8, 2025 at 1:26 AM Akihiko Odaki
wrote:
> On 2025/06/07 5:50, John Snow wrote:
> > In advance of actually bumping the build system requirements for Sphinx,
> > bump the version used for the static analysis tests. Update the minimum
> > requirements accordingly.
> >
> > This changes th
Add a field to count how many remote faults one vCPU has taken. So far
it's still not used, but will be soon.
Reviewed-by: Fabiano Rosas
Signed-off-by: Peter Xu
---
migration/postcopy-ram.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/migration/postcopy-ram.c b/migration/postcopy-r
This patch series adds several new API functions focused on enabling use
cases around reading and writing guest memory from QEMU plugins. To support
these new APIs, some utility functionality around retrieving information about
address spaces is added as well.
The new qemu_plugin_write_register ut
From: novafacing
This patch exposes the gdb_write_register function from
gdbstub/gdbstub.c via the exec/gdbstub.h header file to support use in
plugins to write register contents.
Reviewed-by: Alex Bennée
Reviewed-by: Julian Ganz
Reviewed-by: Pierrick Bouvier
Signed-off-by: Rowan Hart
---
g
From: novafacing
This patch adds a plugin that exercises the virtual and hardware memory
read-write API functions added in a previous patch. The plugin takes a
target and patch byte sequence, and will overwrite any instruction
matching the target byte sequence with the patch.
Signed-off-by: Rowa
Add a global property to allow enabling postcopy-blocktime feature.
Reviewed-by: Fabiano Rosas
Signed-off-by: Peter Xu
---
migration/options.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/migration/options.c b/migration/options.c
index 162c72cda4..4e923a2e07 100644
--- a/migration/opti
Currently, the postcopy blocktime feature maintains vCPU fault information
using an array (vcpu_addr[]). It has two issues.
Issue 1: Performance Concern
The old algorithm was almost OK and fast on inserts, except that the lookup
is slow and won't scale if there are a
This patch adds functionality to enforce the requested QEMU_PLUGIN_CB_
flags level passed when registering a callback function using the
plugins API. Each time a callback is about to be invoked, a thread-local
variable will be updated with the level that callback requested. Then,
called API functio
This series is based on the other series I posted here:
Based-on: <20250609161855.6603-1-pet...@redhat.com>
https://lore.kernel.org/r/20250609161855.6603-1-pet...@redhat.com
v2:
- Collect tags
- Use two spaces always in qapi/ documentations for patch 8/13 [Markus]
- English error fix (s/system-wi
From: novafacing
This patch adds a function to the plugins API to allow plugins to write
register contents. It also moves the qemu_plugin_read_register function
so all the register-related functions are grouped together in the file.
Reviewed-by: Pierrick Bouvier
Signed-off-by: Rowan Hart
---
Hello Everyone,
I want to embed a QEMU window directly in my GUI application. Are Python
APIs available to embed QEMU into a GNOME GUI application?
I am the developer of Cubic (Custom Ubuntu ISO Creator), a tool which
allows users to customize Ubuntu and Debian based Live ISOs.
Screenshots & inf
From: novafacing
This patch exposes the gdb_write_register function from
gdbstub/gdbstub.c via the exec/gdbstub.h header file to support use in
plugins to write register contents.
Reviewed-by: Alex Bennée
Reviewed-by: Julian Ganz
Reviewed-by: Pierrick Bouvier
Signed-off-by: Rowan Hart
---
g
From: novafacing
This patch adds functions to the plugins API to allow reading and
writing memory via virtual addresses. These functions only permit doing
so on the current CPU, because there is no way to ensure consistency if
plugins are allowed to read or write to other CPUs that aren't current
From: novafacing
This patch adds a plugin that exercises the virtual and hardware memory
read-write API functions added in a previous patch. The plugin takes a
target and patch byte sequence, and will overwrite any instruction
matching the target byte sequence with the patch.
Signed-off-by: Rowa
This patch adds functionality to enforce the requested QEMU_PLUGIN_CB_
flags level passed when registering a callback function using the
plugins API. Each time a callback is about to be invoked, a thread-local
variable will be updated with the level that callback requested. Then,
called API functio
I forgot one additional update here that was necessary because the
plugin_cb_flags getter/setters are now on the hot path and need
inlining, and left a typo, both of which I fixed in v11
(https://lore.kernel.org/qemu-devel/20250609193841.348076-1-rowanbh...@gmail.com/T/#t).
On 6/9/25 12:23 P
Before this patch, the blocktime context can be created very early, because
postcopy_ram_supported_by_host() <- migrate_caps_check() can happen during
migration object init.
The trick here is the blocktime context needs system vCPU information,
which seems to be possible to change after that point
Il lun 9 giu 2025, 20:27 Stefan Hajnoczi ha scritto:
> > If you disagree with this change we can certainly live without them---I
> > asked Tanish to start with this as an exercise to get familiar with
> > tracetool, and he's learnt a bunch of things around git anyway so it's
> > all good.
>
> A m
On Mon, Jun 09, 2025 at 11:37:02AM -0300, Fabiano Rosas wrote:
> Peter Xu writes:
>
> > On Fri, Jun 06, 2025 at 05:23:18PM -0300, Fabiano Rosas wrote:
> >> Peter Xu writes:
> >>
> >> > On Mon, Jun 02, 2025 at 10:38:08PM -0300, Fabiano Rosas wrote:
> >> >> Allow the migrate and migrate_incoming
On Mon, Jun 09, 2025 at 04:43:40PM +0100, Daniel P. Berrangé wrote:
> On Mon, Jun 09, 2025 at 11:33:14AM -0400, Peter Xu wrote:
> >
> > Now I think I know part of what I've missed: I used to think the "config"
> > of per-QMP-migrate-command can be totally temporary for a specific
> > migration req
On Mon, Jun 09, 2025 at 11:53:47AM -0400, Peter Xu wrote:
> On Mon, Jun 09, 2025 at 04:43:40PM +0100, Daniel P. Berrangé wrote:
> > On Mon, Jun 09, 2025 at 11:33:14AM -0400, Peter Xu wrote:
> > >
> > > Now I think I know part of what I've missed: I used to think the "config"
> > > of per-QMP-migra
The CCI and Fabric Manager APIs are used to configure CXL switches and
devices. DMTF has defined an MCTP binding specification to carry these
messages. The end goal of this work is to hook this up to emulated CXL
switches and devices to allow control of the configuration.
Since this relies on i2c
1 - 100 of 204 matches
Mail list logo
