On 11/2/25 22:43, Kevin Wolf wrote:
This adds a separate block driver for the bochs image format called
'bochs-rs' so that for the moment both the C implementation and the Rust
implementation can be present in the same build. The intention is to
remove the C implementation eventually and rename t
On 11/2/25 22:43, Kevin Wolf wrote:
This is the equivalent of bdrv_open_file_child() to be used in cases
where the caller is QAPI based and has a BlockdevRef rather than a
filename and an options QDict.
Signed-off-by: Kevin Wolf
---
include/block/block-global-state.h | 4
block.c
On 11/2/25 22:43, Kevin Wolf wrote:
Signed-off-by: Kevin Wolf
---
meson.build| 36
storage-daemon/meson.build | 2 +-
2 files changed, 33 insertions(+), 5 deletions(-)
if have_rust and target_type == 'system'
-target_rust = ru
On 12/2/25 02:56, Xianglai Li wrote:
When the KVM_REG_LOONGARCH_VCPU_RESET command word
is sent to the kernel through the kvm_set_one_reg interface,
the parameter source needs to be a legal address,
otherwise the kernel will return an error and the command word
will fail to be sent.
Signed-off-b
On 5/2/25 05:03, Richard Henderson wrote:
Since we no longer support 64-bit guests on 32-bit hosts,
we can use a 32-bit type on a 32-bit host.
Signed-off-by: Richard Henderson
---
include/exec/vaddr.h | 16 +---
1 file changed, 9 insertions(+), 7 deletions(-)
Reviewed-by: Phili
On 5/2/25 05:03, Richard Henderson wrote:
The declarations use vaddr for size.
Which seems dubious, since TARGET_PAGE_SIZE is int IIUC.
Signed-off-by: Richard Henderson
---
accel/tcg/cputlb.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/accel/tcg/cputlb.c b/acc
On 5/2/25 05:03, Richard Henderson wrote:
The declaration uses uint64_t for addr.
Fixes: 595cd9ce2ec ("plugins: add plugin API to read guest memory")
Signed-off-by: Richard Henderson
---
plugins/api.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/plugins/api.c b/plu
On 5/2/25 05:03, Richard Henderson wrote:
There is now always only one guest address register.
Signed-off-by: Richard Henderson
---
tcg/tcg.c| 18 +-
tcg/aarch64/tcg-target.c.inc | 4 ++--
tcg/arm/tcg-target.c.inc | 4 ++--
tcg/i386/tcg
Hi everyone,
The QEMU v7.2.16 stable release is now available.
You can grab the tarball from our download page here:
https://www.qemu.org/download/#source
https://download.qemu.org/qemu-7.2.16.tar.xz
https://download.qemu.org/qemu-7.2.16.tar.xz.sig (signature)
v7.2.16 is now tagged in t
Hi everyone,
The QEMU v9.1.3 stable release is now available.
You can grab the tarball from our download page here:
https://www.qemu.org/download/#source
https://download.qemu.org/qemu-9.1.3.tar.xz
https://download.qemu.org/qemu-9.1.3.tar.xz.sig (signature)
v9.1.3 is now tagged in the o
Hi everyone,
The QEMU v9.2.1 stable release is now available.
You can grab the tarball from our download page here:
https://www.qemu.org/download/#source
https://download.qemu.org/qemu-9.2.1.tar.xz
https://download.qemu.org/qemu-9.2.1.tar.xz.sig (signature)
v9.2.1 is now tagged in the o
Hi everyone,
The QEMU v8.2.9 stable release is now available.
You can grab the tarball from our download page here:
https://www.qemu.org/download/#source
https://download.qemu.org/qemu-8.2.9.tar.xz
https://download.qemu.org/qemu-8.2.9.tar.xz.sig (signature)
v8.2.9 is now tagged in the o
Add migration support for AMD IOMMU model by saving necessary AMDVIState
parameters for MMIO registers, device table, command buffer, and event
buffers.
Signed-off-by: Suravee Suthikulpanit
---
hw/i386/amd_iommu.c | 58 -
1 file changed, 52 insertions(
Currently, amd-iommu device does not support migration. This series addresses
an issue due hidden AMDVI-PCI device enumeration. Then introduces migratable
VMStateDescription, which saves necessary parameters for the device.
Changes from v2:
(https://lore.kernel.org/all/20250206051856.323651-1-sura
Current amd-iommu model internally creates an AMDVI-PCI device. Here is
a snippet from info qtree:
bus: main-system-bus
type System
dev: amd-iommu, id ""
xtsup = false
pci-id = ""
intremap = "on"
device-iotlb = false
pt = true
...
dev: q35-pcihost, i
On Wed, Feb 12, 2025 at 02:39:16AM +0100, Thomas Huth wrote:
> On 12/02/2025 01.41, Thomas Huth wrote:
> > On 11/02/2025 20.03, Philippe Mathieu-Daudé wrote:
> >> On 11/2/25 19:53, Philippe Mathieu-Daudé wrote:
> >>> On 11/2/25 19:48, Philippe Mathieu-Daudé wrote:
> On 11/2/25 19:26, Stefan Ha
In legacy doorbell mode, the command descriptor slot matched
the UTRD slot. To maintain consistency in MCQ mode, command descriptor
slot allocation and deallocation now use a bitmap-based approach.
Signed-off-by: Jeuk Kim
---
tests/qtest/ufs-test.c | 562 +
This patch tests whether MCQ initialization and basic read-write
operations work correctly when the MCQ parameter of hw/ufs is enabled.
Signed-off-by: Jeuk Kim
---
tests/qtest/ufs-test.c | 171 ++---
1 file changed, 142 insertions(+), 29 deletions(-)
diff --g
Removed dead code related to the unimplemented task
management request.
Signed-off-by: Jeuk Kim
---
tests/qtest/ufs-test.c | 11 +--
1 file changed, 1 insertion(+), 10 deletions(-)
diff --git a/tests/qtest/ufs-test.c b/tests/qtest/ufs-test.c
index 1f860b41c0..ce8b398c6b 100644
--- a/tes
This series adds test support for UFS Multi-Circular Queue (MCQ) mode in QEMU.
It includes MCQ initialization tests, a bitmap-based slot allocation method,
and cleanup of unused code.
Jeuk Kim (3):
tests/qtest/ufs-test: Cleanup unused code
tests/qtest/ufs-test: Prepare for MCQ test
tests
Replace target-specific 'Z' with generic 'z'.
Signed-off-by: Richard Henderson
---
tcg/loongarch64/tcg-target-con-set.h | 15 ++---
tcg/loongarch64/tcg-target-con-str.h | 1 -
tcg/loongarch64/tcg-target.c.inc | 32
3 files changed, 21 insertions(+), 27 d
Replace target-specific 'Z' with generic 'z'.
Signed-off-by: Richard Henderson
---
tcg/riscv/tcg-target-con-set.h | 10 +-
tcg/riscv/tcg-target-con-str.h | 1 -
tcg/riscv/tcg-target.c.inc | 12 ++--
3 files changed, 11 insertions(+), 12 deletions(-)
diff --git a/tcg/riscv/t
Replace target-specific 'Z' with generic 'z'.
Signed-off-by: Richard Henderson
---
tcg/sparc64/tcg-target-con-set.h | 12 ++--
tcg/sparc64/tcg-target-con-str.h | 1 -
tcg/sparc64/tcg-target.c.inc | 12 ++--
3 files changed, 12 insertions(+), 13 deletions(-)
diff --git a/tcg
Replace target-specific 'Z' with generic 'z'.
Signed-off-by: Richard Henderson
---
tcg/mips/tcg-target-con-set.h | 26 ++---
tcg/mips/tcg-target-con-str.h | 1 -
tcg/mips/tcg-target.c.inc | 44 ++-
3 files changed, 31 insertions(+), 40 deletio
Based-on: 20250205040341.2056361-1-richard.hender...@linaro.org
("[PATCH 00/11] tcg: Cleanups after disallowing 64-on-32")
Introduce a new general-purpose constraint which maps 0
to TCG_REG_ZERO, if defined. This differs from existing
constant constraints in that const_arg[*] is recorded as
false
For loongarch, mips, riscv and sparc, a zero register is
available all the time. For aarch64, register index 31
depends on context: sometimes it is the stack pointer,
and sometimes it is the zero register.
Introduce a new general-purpose constraint which maps 0
to TCG_REG_ZERO, if defined. This
Note that 'Z' is still used for addsub2.
Signed-off-by: Richard Henderson
---
tcg/aarch64/tcg-target-con-set.h | 12 -
tcg/aarch64/tcg-target.c.inc | 46 ++--
2 files changed, 26 insertions(+), 32 deletions(-)
diff --git a/tcg/aarch64/tcg-target-con-set.h
Field ipistate in LoongArch CPU object is not used any more,
remove it here.
Signed-off-by: Bibo Mao
Reviewed-by: Bibo Mao
---
hw/loongarch/virt.c| 5 -
target/loongarch/cpu.h | 2 --
2 files changed, 7 deletions(-)
diff --git a/hw/loongarch/virt.c b/hw/loongarch/virt.c
index db4821722
File acpi-build.c is relative with virt machine type, rename it with
virt-acpi-build.c
Signed-off-by: Bibo Mao
Reviewed-by: Philippe Mathieu-Daudé
Tested-by: Philippe Mathieu-Daudé
---
hw/loongarch/meson.build | 2 +-
hw/loongarch/{acpi-build.c => virt-acpi-build.c} | 0
There is only one iocsr address space for the whole virt-machine
board. When CPU is created, the one of percpu points to that of
the board.
Here set iocsr address space when CPU is created rather than IPI
creation stage.
Signed-off-by: Bibo Mao
Reviewed-by: Bibo Mao
---
hw/loongarch/virt.c | 2
The following changes since commit ffaf7f0376f8040ce9068d71ae9ae8722505c42e:
Merge tag 'pull-10.0-testing-and-gdstub-updates-100225-1' of
https://gitlab.com/stsquad/qemu into staging (2025-02-10 13:26:17 -0500)
are available in the Git repository at:
https://gitlab.com/bibo-mao/qemu.git tag
Similiar with virt-acpi-build.c, file virt-fdt-build.c is added here.
And move functions relative with fdt table building to the file.
It is only code movement and there is no function change.
Signed-off-by: Bibo Mao
Reviewed-by: Philippe Mathieu-Daudé
Tested-by: Philippe Mathieu-Daudé
---
hw
Replace function prefix name loongarch_xxx with virt_xxx in file
virt-acpi-build.c
Signed-off-by: Bibo Mao
Reviewed-by: Philippe Mathieu-Daudé
Tested-by: Philippe Mathieu-Daudé
---
hw/loongarch/virt-acpi-build.c | 6 +++---
hw/loongarch/virt.c| 2 +-
include/hw/loongarch/virt.h
Interrupt controller extioi and ipi connect to CPU with irq line method.
With command -smp x, -device la464-loongarch-cpu, smp.cpus is not
accurate for all possible CPU objects, possible_cpu_arch_ids() is used.
Signed-off-by: Bibo Mao
Reviewed-by: Bibo Mao
---
hw/loongarch/virt.c | 60 +
Daniel,
On 2/6/2025 4:01 PM, Daniel P. Berrangé wrote:
On Thu, Feb 06, 2025 at 05:18:56AM +, Suravee Suthikulpanit wrote:
Add migration support for AMD IOMMU model by saving necessary AMDVIState
parameters for MMIO registers, device table, command buffer, and event
buffers.
Signed-off-by:
On 2/6/25 01:13, Dmitry Osipenko wrote:
> On 2/5/25 23:08, Dmitry Osipenko wrote:
>>> Thanks for showing me the video. I will take a look and check what would go
>>> wrong here. I kinda understand corruption may happen
>>> in some scenario but I don't know what could cause the upside down image.
When the KVM_REG_LOONGARCH_VCPU_RESET command word
is sent to the kernel through the kvm_set_one_reg interface,
the parameter source needs to be a legal address,
otherwise the kernel will return an error and the command word
will fail to be sent.
Signed-off-by: Xianglai Li
---
Cc: Bibo Mao
Cc: P
On 12/02/2025 01.41, Thomas Huth wrote:
On 11/02/2025 20.03, Philippe Mathieu-Daudé wrote:
On 11/2/25 19:53, Philippe Mathieu-Daudé wrote:
On 11/2/25 19:48, Philippe Mathieu-Daudé wrote:
On 11/2/25 19:26, Stefan Hajnoczi wrote:
On Mon, Feb 10, 2025 at 3:43 PM Philippe Mathieu-Daudé
wrote:
On 11/02/2025 20.03, Philippe Mathieu-Daudé wrote:
On 11/2/25 19:53, Philippe Mathieu-Daudé wrote:
On 11/2/25 19:48, Philippe Mathieu-Daudé wrote:
On 11/2/25 19:26, Stefan Hajnoczi wrote:
On Mon, Feb 10, 2025 at 3:43 PM Philippe Mathieu-Daudé
wrote:
The following changes since commit
54e91
On 11/02/2025 19.53, Philippe Mathieu-Daudé wrote:
On 11/2/25 19:48, Philippe Mathieu-Daudé wrote:
On 11/2/25 19:26, Stefan Hajnoczi wrote:
On Mon, Feb 10, 2025 at 3:43 PM Philippe Mathieu-Daudé
wrote:
The following changes since commit
54e91d1523b412b4cff7cb36c898fa9dc133e886:
Merge t
f9653d4eb2c target/riscv: Add Smrnmi cpu extension
2d8e8259287 target/riscv: Add Smdbltrp ISA extension enable switch
These patches are supposed to be emitting a warning about disabling an extension that
requires a new OpenSBI image. But they are also emitting the warning for user-only.
Whi
Am 11. Februar 2025 10:06:19 UTC schrieb Peter Maydell
:
>On Mon, 10 Feb 2025 at 22:44, Bernhard Beschow wrote:
>>
>>
>>
>> Am 10. Februar 2025 17:30:01 UTC schrieb Peter Maydell
>> :
>> >On Tue, 4 Feb 2025 at 09:21, Bernhard Beschow wrote:
>> >>
>> >> As a first step, implement the bare min
From: David Hildenbrand
We don't need the MemTxAttrs, so let's simply use the simpler function
variant.
Reviewed-by: Peter Xu
Reviewed-by: Philippe Mathieu-Daudé
Signed-off-by: David Hildenbrand
Link: https://lore.kernel.org/r/20250210084648.33798-7-da...@redhat.com
Signed-off-by: Peter Xu
-
From: David Hildenbrand
As documented in commit 4a2e242bbb306 ("memory: Don't use memcpy for
ram_device regions"), we disallow direct access to RAM DEVICE regions.
This change implies that address_space_write_rom() and
cpu_memory_rw_debug() won't be able to write to RAM DEVICE regions. It
will a
From: David Hildenbrand
As documented in commit 4a2e242bbb306 ("memory: Don't use memcpy for
ram_device regions"), we disallow direct access to RAM DEVICE regions.
Let's make this clearer to prepare for further changes. Note that romd
regions will never be RAM DEVICE at the same time.
Reviewed-
From: William Roche
The list of hwpoison pages used to remap the memory on reset
is based on the backend real page size.
To correctly handle hugetlb, we must mmap(MAP_FIXED) a complete
hugetlb page; hugetlb pages cannot be partially mapped.
Signed-off-by: William Roche
Co-developed-by: David Hi
From: Daniil Tatianin
This will be extended in the future commits, let's move it out of line
right away so that it's easier to read.
Reviewed-by: Vladimir Sementsov-Ogievskiy
Reviewed-by: Peter Xu
Signed-off-by: Daniil Tatianin
Link:
https://lore.kernel.org/r/20250123131944.391886-3-d-tatia.
From: David Hildenbrand
Right now, we only allow for writing to memory regions that allow direct
access using memcpy etc; all other writes are simply ignored. This
implies that debugging guests will not work as expected when writing
to MMIO device regions.
Let's extend cpu_memory_rw_debug() to w
From: David Hildenbrand
Let's factor more of the generic "is this directly accessible" check,
independent of the "write" condition out.
Note that the "!mr->rom_device" check in the write case essentially
disallows the memory_region_is_romd() condition again. Further note that
RAM DEVICE regions
From: David Hildenbrand
We want to pass another flag that will be stored in MemTxAttrs. So pass
MemTxAttrs directly.
Reviewed-by: Peter Xu
Reviewed-by: Philippe Mathieu-Daudé
Signed-off-by: David Hildenbrand
Link: https://lore.kernel.org/r/20250210084648.33798-6-da...@redhat.com
Signed-off-by
From: Daniil Tatianin
Locking the memory without MCL_ONFAULT instantly prefaults any mmaped
anonymous memory with a write-fault, which introduces a lot of extra
overhead in terms of memory usage when all you want to do is to prevent
kcompactd from migrating and compacting QEMU pages. Add an optio
From: Daniil Tatianin
Replace the boolean value enable_mlock with an enum and add a helper to
decide whether we should be calling os_mlock.
This is a stepping stone towards introducing a new mlock mode, which
will be the third possible state of this enum.
Reviewed-by: Vladimir Sementsov-Ogievsk
From: William Roche
Repair poisoned memory location(s), calling ram_block_discard_range():
punching a hole in the backend file when necessary and regenerating
a usable memory.
If the kernel doesn't support the madvise calls used by this function
and we are dealing with anonymous memory, fall back
From: William Roche
Punching a hole in a file with fallocate needs to take into account the
fd_offset value for a correct file location.
But guest_memfd internal use doesn't currently consider fd_offset.
Fixes: 4b870dc4d0c0 ("hostmem-file: add offset option")
Signed-off-by: William Roche
Revie
From: Daniil Tatianin
This will be used in the following commits to make it possible to only
lock memory on fault instead of right away.
Reviewed-by: Vladimir Sementsov-Ogievskiy
Reviewed-by: Peter Xu
Signed-off-by: Daniil Tatianin
Link:
https://lore.kernel.org/r/20250123131944.391886-2-d-ta
From: David Hildenbrand
Let's factor the complete "directly accessible" check independent of
the "write" condition out so we can reuse it next.
We can now split up the checks RAM and ROMD check, so we really only check
for RAM DEVICE in case of RAM -- ROM DEVICE is neither RAM not RAM DEVICE.
R
The following changes since commit ffaf7f0376f8040ce9068d71ae9ae8722505c42e:
Merge tag 'pull-10.0-testing-and-gdstub-updates-100225-1' of
https://gitlab.com/stsquad/qemu into staging (2025-02-10 13:26:17 -0500)
are available in the Git repository at:
https://gitlab.com/peterx/qemu.git tags/
On Mon, Feb 10, 2025 at 09:46:41AM +0100, David Hildenbrand wrote:
> This is a follow-up to [1], implementing it by avoiding the use of
> address_space_write_rom() in cpu_memory_rw_debug() completely, and
> teaching address_space_write() about debug access instead, the can also
> write to ROM.
>
>
On Tue, Feb 11, 2025 at 09:27:04PM +, “William Roche wrote:
> From: William Roche
>
> Here is a very simplified version of my fix only dealing with the
> recovery of huge pages on VM reset.
> ---
> This set of patches fixes an existing bug with hardware memory errors
> impacting hugetlbfs me
On Tue, Feb 11, 2025 at 10:22:38PM +0100, William Roche wrote:
> On 2/10/25 17:48, Peter Xu wrote:
> > On Fri, Feb 07, 2025 at 07:02:22PM +0100, William Roche wrote:
> > > [...]
> > > So the main reason is a KVM "weakness" with kvm_send_hwpoison_signal(),
> > > and
> > > the second reason is to ha
Signed-off-by: Kevin Wolf
---
meson.build| 36
storage-daemon/meson.build | 2 +-
2 files changed, 33 insertions(+), 5 deletions(-)
diff --git a/meson.build b/meson.build
index 1f26801b69..30aae6b3c3 100644
--- a/meson.build
+++ b/meson.build
This adds a separate block driver for the bochs image format called
'bochs-rs' so that for the moment both the C implementation and the Rust
implementation can be present in the same build. The intention is to
remove the C implementation eventually and rename this one into 'bochs'.
This can only ha
On Mon, Feb 10, 2025, Tom Lendacky wrote:
> On 2/7/25 17:34, Kim Phillips wrote:
> > @@ -289,6 +291,7 @@ static_assert((X2AVIC_MAX_PHYSICAL_ID &
> > AVIC_PHYSICAL_MAX_INDEX_MASK) == X2AVIC_
> > #define SVM_SEV_FEAT_RESTRICTED_INJECTION BIT(3)
> > #define SVM_SEV_FEAT_ALTERNATE_INJECTION
This adds a map() function to the BlockDriver trait and makes use of it
to implement reading from an image.
Signed-off-by: Kevin Wolf
---
rust/block/src/driver.rs | 100 +++
1 file changed, 100 insertions(+)
diff --git a/rust/block/src/driver.rs b/rust/block/
This is the equivalent of bdrv_open_file_child() to be used in cases
where the caller is QAPI based and has a BlockdevRef rather than a
filename and an options QDict.
Signed-off-by: Kevin Wolf
---
include/block/block-global-state.h | 4
block.c| 30 +
Signed-off-by: Kevin Wolf
---
rust/Cargo.lock| 7 +++
rust/Cargo.toml| 1 +
rust/block/Cargo.toml | 15 +++
rust/block/README.md | 3 +++
rust/block/meson.build | 19 +++
rust/block/src/lib.rs | 1 +
rust/meson.build | 1 +
7 files ch
This adds format probing both to the BlockDriver trait and the bochs-rs
block driver. With this, bochs-rs achieves feature parity with its C
counterpart. Its probe function returns a higher priority so that it is
preferred when both drivers are available.
Signed-off-by: Kevin Wolf
---
rust/block
This adds helper functions that allow running Rust futures to completion
using QEMU's event loops.
Signed-off-by: Kevin Wolf
---
include/qemu/coroutine-rust.h | 24 +++
rust/wrapper.h| 1 +
util/qemu-co-rust-async.c | 55
rust/qemu-api/meson.
This adds a barebones module for a block driver interface. Because there
is no native QAPI support for Rust yet, opening images takes a few
unsafe functions to call into C visitor functions. This should be
cleaned up later.
Signed-off-by: Kevin Wolf
---
rust/block/src/driver.rs | 190 +++
Signed-off-by: Kevin Wolf
---
rust/wrapper.h| 4
meson.build | 1 +
rust/qemu-api/src/zeroable.rs | 5 +++--
3 files changed, 8 insertions(+), 2 deletions(-)
diff --git a/rust/wrapper.h b/rust/wrapper.h
index 41be87adcf..c3e1e6f9cf 100644
--- a/rust/wrapper
Types that implement IoBuffer can be used with safe I/O functions.
Signed-off-by: Kevin Wolf
---
rust/block/src/iobuffer.rs | 94 ++
rust/block/src/lib.rs | 2 +
2 files changed, 96 insertions(+)
create mode 100644 rust/block/src/iobuffer.rs
diff --git
The existing qemu_api library can't be linked into tools because it
contains a few bindings for things that only exist in the system
emulator.
This adds a new "system" feature to the qemu_api crate that enables the
system emulator parts in it, and build the crate twice: qemu_api_tools
is the core
This series adds minimal bindings for writing block drivers in Rust and
converts the bochs block driver as an example. Some parts (such as the
open functions) are still essentially C written in Rust syntax, while
other parts already try to provide somewhat idiomatic abstractions.
The main purpose
Peter Xu writes:
> On Tue, Feb 11, 2025 at 12:01:35PM -0300, Fabiano Rosas wrote:
>> The qmp_migrate_cancel() command is poorly tested and code inspection
>> reveals that there might be concurrency issues with its usage. Add a
>> test that runs a migration and calls qmp_migrate_cancel() at specif
On Tue, Feb 11, 2025 at 06:23:20PM -0300, Fabiano Rosas wrote:
> It gets super boilerplatey:
Let's go with what we have.
Thanks,
--
Peter Xu
From: William Roche
Here is a very simplified version of my fix only dealing with the
recovery of huge pages on VM reset.
---
This set of patches fixes an existing bug with hardware memory errors
impacting hugetlbfs memory backed VMs and its recovery on VM reset.
When using hugetlbfs large pages
From: William Roche
The list of hwpoison pages used to remap the memory on reset
is based on the backend real page size.
To correctly handle hugetlb, we must mmap(MAP_FIXED) a complete
hugetlb page; hugetlb pages cannot be partially mapped.
Signed-off-by: William Roche
Co-developed-by: David Hi
From: William Roche
Generate an x86 similar error injection message on ras enabled ARM
platforms.
ARM qemu only deals with action required memory errors signaled with
SIGBUS/BUS_MCEERR_AR, and will report a message on every memory error
relayed to the VM. A message like:
Guest Memory Error at QEM
From: William Roche
Repair poisoned memory location(s), calling ram_block_discard_range():
punching a hole in the backend file when necessary and regenerating
a usable memory.
If the kernel doesn't support the madvise calls used by this function
and we are dealing with anonymous memory, fall back
On 2/10/25 17:48, Peter Xu wrote:
On Fri, Feb 07, 2025 at 07:02:22PM +0100, William Roche wrote:
[...]
So the main reason is a KVM "weakness" with kvm_send_hwpoison_signal(), and
the second reason is to have richer error messages.
This seems true, and I also remember something when I looked at
Although defaulting the handshake limit to 10 seconds was a nice QoI
change to weed out intentionally slow clients, it can interfere with
integration testing done with manual NBD_OPT commands over 'nbdsh
--opt-mode'. Expose a QMP knob 'handshake-max-secs' to allow the user
to alter the timeout awa
Although defaulting the handshake limit to 10 seconds was a nice QoI
change to weed out intentionally slow clients, it can interfere with
integration testing done with manual NBD_OPT commands over 'nbdsh
--opt-mode'. Expose a command line option to allow the user to alter
the timeout away from the
The following changes since commit ffaf7f0376f8040ce9068d71ae9ae8722505c42e:
Merge tag 'pull-10.0-testing-and-gdstub-updates-100225-1' of
https://gitlab.com/stsquad/qemu into staging (2025-02-10 13:26:17 -0500)
are available in the Git repository at:
https://repo.or.cz/qemu/ericb.git tags/p
Peter Xu writes:
> On Tue, Feb 11, 2025 at 03:04:37PM -0300, Fabiano Rosas wrote:
>> Peter Xu writes:
>>
>> > On Tue, Feb 11, 2025 at 12:01:33PM -0300, Fabiano Rosas wrote:
>> >> It's possible that the migration is cancelled during
>> >> migration_switchover_start(). In that case, don't set the
On Tue, Feb 11, 2025 at 12:01:36PM -0300, Fabiano Rosas wrote:
> Update the migrate_cancel command documentation with a few words about
> postcopy and the expected state of the machine after migration.
>
> Signed-off-by: Fabiano Rosas
Reviewed-by: Peter Xu
--
Peter Xu
On Tue, Feb 11, 2025 at 12:01:35PM -0300, Fabiano Rosas wrote:
> The qmp_migrate_cancel() command is poorly tested and code inspection
> reveals that there might be concurrency issues with its usage. Add a
> test that runs a migration and calls qmp_migrate_cancel() at specific
> moments.
>
> In or
On Tue, Feb 11, 2025 at 12:01:34PM -0300, Fabiano Rosas wrote:
> Introduce a new migration_test_add_suffix to allow programmatic
> creation of tests based on a suffix. Pass the test name into the test
> so it can know which variant to run.
>
> Signed-off-by: Fabiano Rosas
Reviewed-by: Peter Xu
On Tue, Feb 11, 2025 at 03:04:37PM -0300, Fabiano Rosas wrote:
> Peter Xu writes:
>
> > On Tue, Feb 11, 2025 at 12:01:33PM -0300, Fabiano Rosas wrote:
> >> It's possible that the migration is cancelled during
> >> migration_switchover_start(). In that case, don't set the migration
> >> state FAIL
The value of the stride of the strided vector loads and stores is known only
at execution time and this value determines the location of each vector element
to load/store.
It is then not possible to improve the performance of the emulation of such
instructions by attempting to load/store multiple e
On 11/2/25 19:53, Philippe Mathieu-Daudé wrote:
On 11/2/25 19:48, Philippe Mathieu-Daudé wrote:
On 11/2/25 19:26, Stefan Hajnoczi wrote:
On Mon, Feb 10, 2025 at 3:43 PM Philippe Mathieu-Daudé
wrote:
The following changes since commit
54e91d1523b412b4cff7cb36c898fa9dc133e886:
Merge tag
On Mon, Feb 10, 2025 at 3:43 PM Philippe Mathieu-Daudé
wrote:
>
> The following changes since commit 54e91d1523b412b4cff7cb36c898fa9dc133e886:
>
> Merge tag 'pull-qapi-2025-02-10-v2' of https://repo.or.cz/qemu/armbru into
> staging (2025-02-10 10:47:31 -0500)
>
> are available in the Git reposi
On 2/10/25 3:24 AM, Daniel P. Berrangé wrote:
On Fri, Feb 07, 2025 at 05:33:27PM -0600, Kim Phillips wrote:
The Allowed SEV Features feature allows the host kernel to control
which SEV features it does not want the guest to enable [1].
This has to be explicitly opted-in by the user because it h
This commit improves the performance of QEMU when emulating strided vector
loads and stores by substituting the call for the helper function with the
generation of equivalend TCG operations.
Signed-off-by: Paolo Savini
---
target/riscv/insn_trans/trans_rvv.c.inc | 294
1
On 11/2/25 19:48, Philippe Mathieu-Daudé wrote:
On 11/2/25 19:26, Stefan Hajnoczi wrote:
On Mon, Feb 10, 2025 at 3:43 PM Philippe Mathieu-Daudé
wrote:
The following changes since commit
54e91d1523b412b4cff7cb36c898fa9dc133e886:
Merge tag 'pull-qapi-2025-02-10-v2' of https://repo.or.cz/q
On 11/2/25 19:26, Stefan Hajnoczi wrote:
On Mon, Feb 10, 2025 at 3:43 PM Philippe Mathieu-Daudé
wrote:
The following changes since commit 54e91d1523b412b4cff7cb36c898fa9dc133e886:
Merge tag 'pull-qapi-2025-02-10-v2' of https://repo.or.cz/qemu/armbru into
staging (2025-02-10 10:47:31 -0500
Peter Xu writes:
> On Tue, Feb 11, 2025 at 12:01:33PM -0300, Fabiano Rosas wrote:
>> It's possible that the migration is cancelled during
>> migration_switchover_start(). In that case, don't set the migration
>> state FAILED in migration_completion().
>>
>> Fixes: 3dde8fdbad ("migration: Merge p
On Tue, Feb 11, 2025 at 12:01:33PM -0300, Fabiano Rosas wrote:
> It's possible that the migration is cancelled during
> migration_switchover_start(). In that case, don't set the migration
> state FAILED in migration_completion().
>
> Fixes: 3dde8fdbad ("migration: Merge precopy/postcopy on switcho
On Tue, Feb 11, 2025 at 12:01:29PM -0300, Fabiano Rosas wrote:
> There's no need for two separate functions and this _fd_ is a historic
> artifact that makes little sense nowadays.
>
> Signed-off-by: Fabiano Rosas
Reviewed-by: Peter Xu
--
Peter Xu
On Tue, Feb 11, 2025 at 12:01:32PM -0300, Fabiano Rosas wrote:
> After postcopy has started, it's not possible to recover the source
> machine in case a migration error occurs because the destination has
> already been changing the state of the machine. For that same reason,
> it doesn't make sense
Add .set_vnet_le() function that always returns success, assuming that
vDPA h/w always implements LE data format. Otherwise, QEMU disables vDPA and
outputs the message:
"backend does not support LE vnet headers; falling back on userspace virtio"
Signed-off-by: Konstantin Shkolnyy
---
net/vhost-v
1 - 100 of 278 matches
Mail list logo
