On 28/11/2024 02:37, Alistair Francis wrote:
> On Thu, Nov 21, 2024 at 6:28 PM Clément Léger wrote:
>>
>>
>>
>> On 20/11/2024 06:02, Alistair Francis wrote:
>>> On Tue, Nov 19, 2024 at 9:27 PM Clément Léger wrote:
On 19/11/2024 05:16, Alistair Francis wrote:
> On Thu, N
This adds `virt` virtual register on debug interface so that users
can access current virtualization mode for debugging purposes.
Signed-off-by: Yanfeng Liu
---
gdb-xml/riscv-32bit-virtual.xml | 1 +
gdb-xml/riscv-64bit-virtual.xml | 1 +
target/riscv/gdbstub.c | 18 --
On Thu, 2024-11-28 at 14:46 +1000, Alistair Francis wrote:
> On Thu, Nov 28, 2024 at 2:27 PM Yanfeng wrote:
> >
> > On Thu, 2024-11-28 at 14:10 +1000, Alistair Francis wrote:
> > > On Thu, Nov 28, 2024 at 2:05 PM Yanfeng wrote:
> > > >
> > > > Alistair.
> > > >
> > > > My initial `git send-ema
On 28/11/2024 02.35, bibo mao wrote:
Hi Thomas and Philippe,
Could you do me a favor to merge this bugfix?
Gao Song is offline and cannot work on computer. I have no merge permission
and I am applying maintainer permission.
Sure, I've put it in my current pull request!
Regards,
Thomas
Hi, thank you for your comment.
On 2024/11/27 20:23, Peter Maydell wrote:
On Wed, 27 Nov 2024 at 04:34, Tomoyuki HIROSE
wrote:
I would be happy to receive your comments.
ping.
Hi; this one is on my to-review list (along, sadly, with 23 other
series); I had a quick look a while back and it see
On 27/11/24 23:35, Richard Henderson wrote:
On 11/27/24 13:19, Philippe Mathieu-Daudé wrote:
There is no vCPU within the QTest accelerator (well, they
are stubs doing nothing, see dummy_cpu_thread_fn).
Directly use the cpu_physical_memory_rw() API -- which
amusingly prefixed 'cpu_' does not use
Hi Philippe,
> Subject: Re: [PATCH v3 0/3] Introduce a new Write Protected pin inverted
> property
>
> On 27/11/24 10:44, Cédric Le Goater wrote:
> > On 11/14/24 10:48, Jamin Lin wrote:
> >> change from v1:
> >> 1. Support RTC for AST2700.
> >> 2. Support SDHCI write protected pin inverted for AS
On Thu, Nov 28, 2024 at 2:27 PM Yanfeng wrote:
>
> On Thu, 2024-11-28 at 14:10 +1000, Alistair Francis wrote:
> > On Thu, Nov 28, 2024 at 2:05 PM Yanfeng wrote:
> > >
> > > Alistair.
> > >
> > > My initial `git send-email" on Ubuntu 22.04 wasn't lukcy:
> > >
> > > ```shell
> > > $ git send-email
On Wed, Nov 27, 2024 at 11:29:06PM -0500, Donald Dutile wrote:
> On 11/27/24 5:21 AM, Shameerali Kolothum Thodi wrote:
> > > > W.r.t naming, maybe something related to "hardware-accelerated"?
> > > >
> > > Given that 'accel' has been used for hw-acceleration elsewhere, that seems
> > > like a reas
On 11/27/24 5:21 AM, Shameerali Kolothum Thodi wrote:
-Original Message-
From: Donald Dutile
Sent: Tuesday, November 26, 2024 6:29 PM
To: Nicolin Chen ; Eric Auger
Cc: Shameerali Kolothum Thodi
; qemu-...@nongnu.org;
qemu-devel@nongnu.org; peter.mayd...@linaro.org; j...@nvidia.com
On Thu, 2024-11-28 at 14:10 +1000, Alistair Francis wrote:
> On Thu, Nov 28, 2024 at 2:05 PM Yanfeng wrote:
> >
> > Alistair.
> >
> > My initial `git send-email" on Ubuntu 22.04 wasn't lukcy:
> >
> > ```shell
> > $ git send-email
> > git: 'send-email' is not a git command. See 'git --help'.
> >
Alistair.
My initial `git send-email" on Ubuntu 22.04 wasn't lukcy:
```shell
$ git send-email
git: 'send-email' is not a git command. See 'git --help'.
```
I need time to find a solution.
As for seperate `virt` thing, my concern is that though the V and P status looks
orthogonal, they are not
On Thu, Nov 28, 2024 at 2:05 PM Yanfeng wrote:
>
> Alistair.
>
> My initial `git send-email" on Ubuntu 22.04 wasn't lukcy:
>
> ```shell
> $ git send-email
> git: 'send-email' is not a git command. See 'git --help'.
> ```
>
> I need time to find a solution.
You will need to install git-send-email
Hi, Eric
On Thu, 28 Nov 2024 at 00:06, Eric Auger wrote:
> > Yeah, there is no live migration support yet in the SMMU qmeu driver,
> > AFAIK?
> the non accelerated SMMU QEMU device does support migration.
Could you clarify more about this?
The migration is not supported if using viommu (SMMU QE
Hi, Joao
On Fri, Jun 23, 2023 at 5:51 AM Joao Martins wrote:
>
> Hey,
>
> This series introduces support for vIOMMU with VFIO device migration,
> particurlarly related to how we do the dirty page tracking.
>
> Today vIOMMUs serve two purposes: 1) enable interrupt remaping 2)
> provide dma transla
On 11/27/24 19:01, Pierrick Bouvier wrote:
On 11/27/24 09:46, Richard Henderson wrote:
On 11/26/24 18:15, Pierrick Bouvier wrote:
Except for saving memory in *very* specific case (a structure allocated tens of
millions
times for example), I hardly see a benefit vs using integer types.
Even t
On Thu, Nov 28, 2024 at 11:43 AM Yanfeng wrote:
>
> On Thu, 2024-11-28 at 10:39 +1000, Alistair Francis wrote:
> > On Thu, Nov 28, 2024 at 12:09 AM Yanfeng wrote:
> > >
> > >
> > > When debugging hypervisor extension based programs, it is convenient to
> > > see
> > > the
> > > current virtualiz
Add logic cpu index output parameter for function cpu_by_arch_id,
CPUState::cpu_index is logic cpu slot index for possible_cpus.
However it is logic cpu index with LoongsonIPICommonState::IPICore,
here hide access for CPUState::cpu_index directly, it comes from
function cpu_by_arch_id().
Signed-of
There is cpu mapping from physical cpu id inside, cpu_by_arch_id
can be optimized from cpu mapping table.
Signed-off-by: Bibo Mao
---
hw/intc/loongarch_ipi.c | 47 ++---
1 file changed, 11 insertions(+), 36 deletions(-)
diff --git a/hw/intc/loongarch_ipi.c b/
Add logic cpu allocation and cpu mapping with cpu hotplug interface.
When cpu is added, connect ipi gpio irq to CPU IRQ_IPI irq pin.
Also use hotplug_handler_plug() to nofity ipi object when cold-plug
is created.
Signed-off-by: Bibo Mao
---
hw/intc/loongarch_ipi.c | 46 +
LoongArch ipi can send interrupt to multiple CPUs, interrupt routing
to CPU comes from destination physical cpu id. Here hotplug interface
is added for IPI object, sot that logic cpu mapping from physical cpu
id can be setup.
Here only basic hotplug framework is added, it is stub function.
Signed
LoongArch IPI can send intterrupt to multiple CPUs, physical cpu id is
used to route interrupt for CPUs.
With cpu hotplug feature in future, logic cpu index depends on
CPUState::cpu_index heavily. Here add cpu hotplug notification interface
in IPI driver, cpu mapping logic is created inside. CPUSt
Bitmap table present_cpu_map is added, it is to allocate logical cpu
when CPU object is created. Also present_cpu array is added to get
logical cpu from physical cpu id.
Signed-off-by: Bibo Mao
---
hw/intc/loongarch_ipi.c | 24
include/hw/intc/loongarch_ipi.h |
On Thu, 2024-11-28 at 10:39 +1000, Alistair Francis wrote:
> On Thu, Nov 28, 2024 at 12:09 AM Yanfeng wrote:
> >
> >
> > When debugging hypervisor extension based programs, it is convenient to see
> > the
> > current virtualization mode from GDB debugger.
> >
> > This patch shares the virtualiz
On Thu, Nov 21, 2024 at 6:28 PM Clément Léger wrote:
>
>
>
> On 20/11/2024 06:02, Alistair Francis wrote:
> > On Tue, Nov 19, 2024 at 9:27 PM Clément Léger wrote:
> >>
> >>
> >>
> >> On 19/11/2024 05:16, Alistair Francis wrote:
> >>> On Thu, Nov 14, 2024 at 7:14 PM Clément Léger wrote:
>
>
Hi Thomas and Philippe,
Could you do me a favor to merge this bugfix?
Gao Song is offline and cannot work on computer. I have no merge
permission and I am applying maintainer permission.
Regards
Bibo Mao
On 2024/11/27 上午9:34, Xianglai Li wrote:
There is a bug in the process of resolving the
On Mon, Nov 25, 2024 at 10:38 PM Mohammad Javad Sekonji
wrote:
>
> Dear QEMU developer team
>
> I hope this email finds you well. My name is Mohammad Javad Sekonji, and I am
> a master’s student specializing in electronic digital systems. I am currently
> working on a project involving the insta
On Mon, Nov 25, 2024 at 11:47 PM Sai Pavan Boddu
wrote:
>
> Add a basic board with interrupt controller (intc), timer, serial
> (uartlite), small memory called LMB@0 (128kB) and DDR@0x8000
> (configured via command line eg. -m 2g).
> This is basic configuration which matches HW generated out o
On 11/27/24 09:46, Richard Henderson wrote:
On 11/26/24 18:15, Pierrick Bouvier wrote:
Except for saving memory in *very* specific case (a structure allocated tens of
millions
times for example), I hardly see a benefit vs using integer types.
Even then, 'uint32_t flags' can be just as easy to
On 11/26/24 22:38, Thomas Huth wrote:
On 26/11/2024 22.17, Pierrick Bouvier wrote:
Signed-off-by: Pierrick Bouvier
---
docs/devel/style.rst | 10 ++
1 file changed, 10 insertions(+)
diff --git a/docs/devel/style.rst b/docs/devel/style.rst
index 2f68b500798..13cb1ef626b 100644
---
On Mon, Nov 25, 2024 at 11:47 PM Sai Pavan Boddu
wrote:
>
> Add a basic board with interrupt controller (intc), timer, serial
> (uartlite), small memory called LMB@0 (128kB) and DDR@0x8000
> (configured via command line eg. -m 2g).
> This is basic configuration which matches HW generated out o
On 11/27/24 01:20, Stefano Garzarella wrote:
On Tue, Nov 26, 2024 at 01:17:34PM -0800, Pierrick Bouvier wrote:
This attribute is not recognized by clang.
An investigation has been performed to ensure this attribute has no
effect on layout of structures we use in QEMU [1], so it's safe to
remove
On Wed, Nov 27, 2024 at 3:38 PM Alexey Baturo wrote:
>
> Hi Daniel,
>
> Thanks for the suggestion. I've resubmitted it with the v11 tag and put some
> reviewed-bys.
Thanks! It's a lot of time and effort to review code, so it's helpful
to keep existing tags. Note that it's fine (and you should) t
On Thu, Nov 28, 2024 at 2:09 AM Philippe Mathieu-Daudé
wrote:
>
> Hi,
>
> [very old patch merged as commit e481a1f63c93]
>
> On 30/9/16 02:25, Alistair Francis wrote:
> > Add a generic loader to QEMU which can be used to load images or set
> > memory values.
> >
> > Internally inside QEMU this is
On Thu, Nov 28, 2024 at 12:09 AM Yanfeng wrote:
>
>
> When debugging hypervisor extension based programs, it is convenient to see
> the
> current virtualization mode from GDB debugger.
>
> This patch shares the virtualization mode as part of the existing "priv"
> virtual
> register, or more spec
On 11/27/24 11:00 AM, Jason Gunthorpe wrote:
On Wed, Nov 27, 2024 at 10:21:24AM +, Shameerali Kolothum Thodi wrote:
For SMMUv3, NVIDIA-specific vCMDQ, it needs a parameter to state that
specifically,
since I'm concluding from reading the SMMUv3 version G.a spec, that
ECMDQ was added
to be
On 11/27/24 13:19, Philippe Mathieu-Daudé wrote:
There is no vCPU within the QTest accelerator (well, they
are stubs doing nothing, see dummy_cpu_thread_fn).
Directly use the cpu_physical_memory_rw() API -- which
amusingly prefixed 'cpu_' does not use vCPU -- to access
memory. This reduces access
On 27.11.2024 10:13, Cédric Le Goater wrote:
On 11/17/24 20:20, Maciej S. Szmigiero wrote:
From: "Maciej S. Szmigiero"
Some drivers might want to make use of auxiliary helper threads during VM
state loading, for example to make sure that their blocking (sync) I/O
operations don't block the res
On 27/11/24 19:44, Pierrick Bouvier wrote:
On 11/26/24 23:40, Дмитрий Фролов wrote:
Hello, Daniel
On 26.11.2024 21:32, Daniel P. Berrangé wrote:
On Wed, Nov 13, 2024 at 12:43:40PM +0300, Dmitry Frolov wrote:
Some tests need more time when qemu is built with
"--enable-asan --enable-ubsan"
As
There is no vCPU within the QTest accelerator (well, they
are stubs doing nothing, see dummy_cpu_thread_fn).
Directly use the cpu_physical_memory_rw() API -- which
amusingly prefixed 'cpu_' does not use vCPU -- to access
memory. This reduces accesses to the global 'first_cpu'.
Signed-off-by: Phili
The current build structure for migration tests is confusing. There is
the tests/migration directory, which contains two different guest code
implementations, one for the qtests (a-b-{bootblock|kernel}.S) and
another for the guestperf script (stress.c). One uses a Makefile,
while the other uses mes
On 11/27/24 10:27, Richard Henderson wrote:
On 11/27/24 11:57, Pierrick Bouvier wrote:
I noticed that it was redundant (for user-mode at least), but it seemed too
implicit to
rely on this.
As well, I didn't observe such a flush in system-mode, does it work the same as
user-mode
(regarding the
On 11/26/24 23:40, Дмитрий Фролов wrote:
Hello, Daniel
On 26.11.2024 21:32, Daniel P. Berrangé wrote:
On Wed, Nov 13, 2024 at 12:43:40PM +0300, Dmitry Frolov wrote:
Some tests need more time when qemu is built with
"--enable-asan --enable-ubsan"
As was discussed here:
https://patchew.org/QEMU
We current have a bunch of non-test functions in migration-test.c and
some others in migration-helpers.c. In order to split migration-test.c
into separate test binaries, these helpers need to go somewhere
else.
To avoid making migration-helpers even larger, move all QMP-related
functions into a ne
The migration tests have a set of core infrastructure routines. These
are functions that are called by (almost) all tests and centralize the
common operations of: starting migration on both sides, waiting for
guests to boot, performing guest initialization and teardown, guest
memory validation, etc
Test frameworks usually prefix "test_" to the entry point of the test
code. Having every function prefixed with test_ makes it hard to
understand the code and to grep for the actual tests.
Remove the "test" prefix from everything that is not a test.
In order to still keep some namespacing, stick
Standardize the hook names:
- change the names to .start|end_hook to match
test_migrate_start|end()
- use the migrate_hook_start_ and migrate_hook_end_ prefixes
Reviewed-by: Peter Xu
Signed-off-by: Fabiano Rosas
---
tests/qtest/migration-test.c | 342 ++-
1 f
On 11/27/24 01:06, Daniel P. Berrangé wrote:
On Tue, Nov 26, 2024 at 04:54:18PM -0600, Richard Henderson wrote:
On 11/26/24 11:52, Thomas Huth wrote:
I think we want to continue to maek failing downloads as test failures,
otherwise we'll never notice when an asset is not available from the
inte
Move kvm_dirty_ring_supported() to migration-util.c. Similarly to
ufd_version_check(), this function is used during test definition so
put it in the utils file.
Reviewed-by: Peter Xu
Signed-off-by: Fabiano Rosas
---
tests/qtest/migration-test.c | 29 --
tests/q
Continuing the split of groups of tests from migration-test.c, split
the compression tests into their own file.
Reviewed-by: Peter Xu
Signed-off-by: Fabiano Rosas
---
tests/qtest/meson.build | 1 +
tests/qtest/migration-test.c | 161 +--
tests/qtest/
Move the remaining tests into a misc-tests.c file. These tests are
mostly about validation of input and should be in the future replaced
by unit testing.
Reviewed-by: Peter Xu
Signed-off-by: Fabiano Rosas
---
tests/qtest/meson.build| 1 +
tests/qtest/migration-test.c | 278 +
Move the mode/reboot test into a separate file to hold all the CPR
tests. Currently there's just one test, but we're adding more CPR
modes and the feature is different enough from live migration that
it's worth it to have a separate file for it.
Reviewed-by: Peter Xu
Signed-off-by: Fabiano Rosas
Move ufd_version_check() to migration-util.c file. This is a helper
function that is used during tests definition so it should be
available outside of migration-test.c
Since the function is moving to another file, change it to stop
accessing the global uffd_feature_thread_id and take it as argumen
v3:
- Fixed some more hook names I had missed;
- Move addition of migration-qmp.c in meson to correct patch;
- Moved QEMU_ENV_* variables to framework.c where they are used;
- patch 7: improve commit message;
- s/test-framework/framework/
- meson.build style changes;
- d
Rename migration-helpers.c to migration-util.c to make its purpose
more explicit and avoid the "helper" terminology.
Move the file to the qtest/migration/ directory along with the rest of
the migration files.
Signed-off-by: Fabiano Rosas
---
tests/qtest/meson.build
Split the file tests from migration-test.c. These are being moved to
their own file due to being special enough compared with the regular
stream migration. There is also the entire mapped-ram feature which
depends on file migration.
Reviewed-by: Peter Xu
Signed-off-by: Fabiano Rosas
---
tests/q
The migration-test.c file has become unwieldy large. It's quite
confusing to navigate with all the test definitions mixed with hook
definitions. The TLS tests make this worse with ifdef'ery.
Since we're planning on having a smaller set of tests to run as smoke
testing on all architectures, I'm tak
Move the code that creates the guest binary out of migration-test and
into the qtest/migration/ directory, along with the rest of the
a-b-kernel code.
That code is part of the basic infrastructure of migration tests, it
shouldn't be among the tests themselves.
Also take the chance and rename migr
Split the next group of tests from migration-test.c, the postcopy
tests. This is another well-defined group of tests and postcopy is a
unique enough feature that it deserves it's own file.
Reviewed-by: Peter Xu
Signed-off-by: Fabiano Rosas
---
tests/qtest/meson.build| 1 +
tes
We currently have some environment validation to perform and flags to
set during the initialization of the tests. To be able to add more
migration test binaries, we'll need these tasks to be in their own
function so they can be called from more than one place.
Move the initialization code to a fun
Split the precopy tests from migration-test.c. This is the largest
group of tests and the more difficult one to break into smaller
groups, so move all of it.
Reviewed-by: Peter Xu
Signed-off-by: Fabiano Rosas
---
tests/qtest/meson.build |1 +
tests/qtest/migration-test.c
On 11/27/24 11:57, Pierrick Bouvier wrote:
I noticed that it was redundant (for user-mode at least), but it seemed too implicit to
rely on this.
As well, I didn't observe such a flush in system-mode, does it work the same as user-mode
(regarding the CF_PARALLEL flag)?
Yes, we set CF_PARALLEL f
On 11/27/24 00:29, Thomas Huth wrote:
On 26/11/2024 23.54, Richard Henderson wrote:
On 11/26/24 11:52, Thomas Huth wrote:
I think we want to continue to maek failing downloads as test failures, otherwise we'll
never notice when an asset is not available from the internet anymore (since SKIPs ju
On 11/27/24 06:16, Philippe Mathieu-Daudé wrote:
bool tcg_exec_realizefn(CPUState *cpu, Error **errp)
{
-static bool tcg_target_initialized;
Error removing this variable before patch 6.
r~
+const TCGCPUOps *tcg_ops = cpu->cc->tcg_ops;
if (!tcg_target_initialized) {
Hi Richard,
On 11/27/24 09:53, Richard Henderson wrote:
On 11/26/24 13:02, Pierrick Bouvier wrote:
@@ -266,6 +266,19 @@ static void qemu_plugin_vcpu_init__async(CPUState *cpu,
run_on_cpu_data unused)
assert(cpu->cpu_index != UNASSIGNED_CPU_INDEX);
qemu_rec_mutex_lock(&plugin
On 11/26/24 13:02, Pierrick Bouvier wrote:
@@ -266,6 +266,19 @@ static void qemu_plugin_vcpu_init__async(CPUState *cpu,
run_on_cpu_data unused)
assert(cpu->cpu_index != UNASSIGNED_CPU_INDEX);
qemu_rec_mutex_lock(&plugin.lock);
+
+/*
+ * We want to flush tb when a second c
On 11/26/24 18:15, Pierrick Bouvier wrote:
Except for saving memory in *very* specific case (a structure allocated tens of millions
times for example), I hardly see a benefit vs using integer types.
Even then, 'uint32_t flags' can be just as easy to use as unsigned foo:1, bar:1, etc:1.
Plus yo
On 27/11/24, Philippe Mathieu-Daudé wrote:
> Rather than initializing the first random target architecture
> and ignore the following ones when a global boolean is set,
> use a bitmask allowing different frontend targets to be
> initialized.
>
> Signed-off-by: Philippe Mathieu-Daudé
> ---
> acce
On 11/25/24 4:46 PM, Fabiano Rosas wrote:
> Currently a VM that has been target of a migration using
> late-block-activate will crash at the end of a new migration (with it
> as source) when releasing ownership of the disks due to the VM having
> never taken ownership of the disks in the first plac
On 11/27/24 11:28, Anton Johansson wrote:
On 27/11/24, Philippe Mathieu-Daudé wrote:
Rather than initializing the first random target architecture
and ignore the following ones when a global boolean is set,
use a bitmask allowing different frontend targets to be
initialized.
Signed-off-by: Phil
On 11/25/24 7:07 PM, Peter Xu wrote:
> On Mon, Nov 25, 2024 at 11:46:11AM -0300, Fabiano Rosas wrote:
>> Currently a VM that has been target of a migration using
>> late-block-activate will crash at the end of a new migration (with it
>> as source) when releasing ownership of the disks due to the V
On 11/27/24 06:16, Philippe Mathieu-Daudé wrote:
When using multiple TCG frontends concurrently, we need to
be able to discriminate by target architecture. While we
can infer that from the CPUClass,
Can we? I thought the only place this got set is arch_type in
system/arch_init.c.
it can be
On 27/11/24, Philippe Mathieu-Daudé wrote:
> When using multiple TCG frontends concurrently, we need to
> be able to discriminate by target architecture. While we
> can infer that from the CPUClass, it can be quicker to get
> it from a direct field in TCGCPUOps.
>
> Signed-off-by: Philippe Mathieu
When FEAT_SEL2 was implemented the SEL2 timers where missed. This
shows up when building the latest Hafnium with SPMC_AT_EL=2. The
actual implementation utilises the same logic as the rest of the
timers so all we need to do is:
- define the timers and their access functions
- conditionally add
On 11/27/24 06:25, Philippe Mathieu-Daudé wrote:
Alternatively we can restrict TCGCPUOps::arch_id to
system emulation, using in the next patch:
-- >8 --
diff --git a/include/hw/core/tcg-cpu-ops.h b/include/hw/core/tcg-cpu-ops.h
index ec3d2b50a9e..6fe0e1a7e97 100644
--- a/include/hw/core/tcg-cpu-
On 11/27/24 06:16, Philippe Mathieu-Daudé wrote:
While QEMU architecture bitmask values are only used by
system emulation code, they can be used in generic code
like TCG accelerator.
Move the declarations to "qemu/arch_id.h" and add the
QemuArch type definition.
Signed-off-by: Philippe Mathieu-
On 11/27/24 06:16, Philippe Mathieu-Daudé wrote:
While this handler can be called multiple times, it will
only be run once. Clarify by renaming the handler name.
Signed-off-by: Philippe Mathieu-Daudé
---
We could "optimize" TGC frontends memory use by passing
a CPUClass argument, and each initia
On 11/27/24 06:16, Philippe Mathieu-Daudé wrote:
Ease reading code by declaring a local 'tcg_ops' variable.
Signed-off-by: Philippe Mathieu-Daudé
---
accel/tcg/cpu-exec.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
Reviewed-by: Richard Henderson
r~
diff --git a/accel
On 11/27/24 06:16, Philippe Mathieu-Daudé wrote:
Signed-off-by: Philippe Mathieu-Daudé
Acked-by: Nicholas Piggin
---
target/ppc/cpu_init.c | 29 ++---
1 file changed, 14 insertions(+), 15 deletions(-)
Reviewed-by: Richard Henderson
r~
On 27/11/24, Philippe Mathieu-Daudé wrote:
> Ease reading code by declaring a local 'tcg_ops' variable.
>
> Signed-off-by: Philippe Mathieu-Daudé
> ---
> accel/tcg/cpu-exec.c | 8
> 1 file changed, 4 insertions(+), 4 deletions(-)
Reviewed-by: Anton Johansson
On 27/11/24, Philippe Mathieu-Daudé wrote:
> Signed-off-by: Philippe Mathieu-Daudé
> Acked-by: Nicholas Piggin
> ---
> target/ppc/cpu_init.c | 29 ++---
> 1 file changed, 14 insertions(+), 15 deletions(-)
Reviewed-by: Anton Johansson
Hi,
[very old patch merged as commit e481a1f63c93]
On 30/9/16 02:25, Alistair Francis wrote:
Add a generic loader to QEMU which can be used to load images or set
memory values.
Internally inside QEMU this is a device. It is a strange device that
provides no hardware interface but allows QEMU t
On 11/26/24 23:35, baturo.ale...@gmail.com wrote:
From: Alexey Baturo
Signed-off-by: Alexey Baturo
---
target/riscv/cpu.h | 2 ++
target/riscv/cpu_helper.c | 19 +++
target/riscv/insn_trans/trans_rvh.c.inc | 11 +++
target/riscv/
On 11/27/24 17:00, Jason Gunthorpe wrote:
> On Wed, Nov 27, 2024 at 10:21:24AM +, Shameerali Kolothum Thodi wrote:
>>> For SMMUv3, NVIDIA-specific vCMDQ, it needs a parameter to state that
>>> specifically,
>>> since I'm concluding from reading the SMMUv3 version G.a spec, that
>>> ECMDQ was
On 11/18/24 11:14, Klaus Jensen wrote:
A set of misc fixes related to SR-IOV compliance and MSI-X.
Signed-off-by: Klaus Jensen
---
Klaus Jensen (4):
hw/nvme: fix msix_uninit with exclusive bar
hw/nvme: fix use/unuse of msix vectors
hw/nvme: SR-IOV VFs must hardwire pci inter
On Wed, Nov 27, 2024 at 10:21:24AM +, Shameerali Kolothum Thodi wrote:
> > For SMMUv3, NVIDIA-specific vCMDQ, it needs a parameter to state that
> > specifically,
> > since I'm concluding from reading the SMMUv3 version G.a spec, that
> > ECMDQ was added
> > to be able to assign an ECMDQ to a V
From: Alexander Graf
Apple has its own virtio-blk PCI device ID where it deviates from the
official virtio-pci spec slightly: It puts a new "apple type"
field at a static offset in config space and introduces a new barrier
command.
This patch first creates a mechanism for virtio-blk downstream c
From: Alexander Graf
The VMApple machine exposes AUX and ROOT block devices (as well as USB OTG
emulation) via virtio-pci as well as a special, simple backdoor platform
device.
This patch implements this backdoor platform device to the best of my
understanding. I left out any USB OTG parts; they
macOS's Cocoa event handling must be done on the initial (main) thread
of the process. Furthermore, if library or application code uses
libdispatch, the main dispatch queue must be handling events on the main
thread as well.
So far, this has affected Qemu in both the Cocoa and SDL UIs, although
in
I'm happy to take responsibility for the macOS PV graphics code. As
HVF patches don't seem to get much attention at the moment, I'm also
adding myself as designated reviewer for HVF and x86 HVF to try and
improve that.
I anticipate that the resulting workload should be covered by the
funding I'm r
From: Alexander Graf
Apple defines a new "vmapple" machine type as part of its proprietary
macOS Virtualization.Framework vmm. This machine type is similar to the
virt one, but with subtle differences in base devices, a few special
vmapple device additions and a vastly different boot chain.
This
The virtio_blk_free_request() function has been a 1-liner forwarding
to g_free() for a while now. We may as well call g_free on the request
pointer directly.
Signed-off-by: Phil Dennis-Jordan
Reviewed-by: Akihiko Odaki
---
hw/block/virtio-blk.c | 43 +++---
From: Alexander Graf
Instead of device tree or other more standardized means, VMApple passes
platform configuration to the first stage boot loader in a binary encoded
format that resides at a dedicated RAM region in physical address space.
This patch models this configuration space as a qdev dev
This change wires up the PCI variant of the paravirtualised
graphics device, mainly useful for x86-64 macOS guests, implemented
by macOS's ParavirtualizedGraphics.framework. It builds on code
shared with the vmapple/mmio variant of the PVG device.
Signed-off-by: Phil Dennis-Jordan
---
v4:
* Th
From: Alexander Graf
In addition to the ISA and PCI variants of pvpanic, let's add an MMIO
platform device that we can use in embedded arm environments.
Signed-off-by: Alexander Graf
Reviewed-by: Philippe Mathieu-Daudé
Tested-by: Philippe Mathieu-Daudé
Signed-off-by: Phil Dennis-Jordan
Revie
From: Alexander Graf
VMApple contains an "aes" engine device that it uses to encrypt and
decrypt its nvram. It has trivial hard coded keys it uses for that
purpose.
Add device emulation for this device model.
Signed-off-by: Alexander Graf
Signed-off-by: Phil Dennis-Jordan
Reviewed-by: Akihiko
From: Alexander Graf
MacOS unconditionally disables interrupts of the physical timer on boot
and then continues to use the virtual one. We don't really want to support
a full physical timer emulation, so let's just ignore those writes.
Signed-off-by: Alexander Graf
Signed-off-by: Phil Dennis-Jo
From: Alexander Graf
Some boards such as vmapple don't do real legacy PCI IRQ swizzling.
Instead, they just keep allocating more board IRQ lines for each new
legacy IRQ. Let's support that mode by giving instantiators a new
"nr_irqs" property they can use to support more than 4 legacy IRQ lines.
From: Alexander Graf
We will introduce a number of devices that are specific to the vmapple
target machine. To keep them all tidily together, let's put them into
a single target directory.
Signed-off-by: Alexander Graf
Signed-off-by: Phil Dennis-Jordan
Reviewed-by: Akihiko Odaki
---
MAINTAIN
This change adds a property 'display_modes' on the graphics device
which permits specifying a list of display modes. (screen resolution
and refresh rate)
The property is an array of a custom type to make the syntax slightly
less awkward to use, for example:
-device '{"driver":"apple-gfx-pci", "di
1 - 100 of 189 matches
Mail list logo
