Kevin Wolf writes:
> Am 04.09.2023 um 18:25 hat Daniel P. Berrangé geschrieben:
>> By the time of the 8.2.0 release, it will have been 2 years and 6
>> releases since we accidentally broke setting of array properties
>> for user creatable devices:
>>
>> https://gitlab.com/qemu-project/qemu/-/i
On Thu, 7 Sept 2023 at 10:33, Markus Armbruster wrote:
>
> Kevin Wolf writes:
>
> > Am 04.09.2023 um 18:25 hat Daniel P. Berrangé geschrieben:
> >> By the time of the 8.2.0 release, it will have been 2 years and 6
> >> releases since we accidentally broke setting of array properties
> >> for user
Daniel P. Berrangé writes:
> By the time of the 8.2.0 release, it will have been 2 years and 6
> releases since we accidentally broke setting of array properties
> for user creatable devices:
>
> https://gitlab.com/qemu-project/qemu/-/issues/1090
>
> Some context:
>
> * Initial identification /
On 9/7/2023 5:10 PM, Eric Auger wrote:
> Hi,
>
> On 9/7/23 09:16, Philippe Mathieu-Daudé wrote:
>> Widening Cc to ARM/VFIO.
>>
>> On 4/8/23 11:15, Wu, Fei wrote:
>>> On 8/3/2023 11:07 PM, Andrew Jones wrote:
On Mon, Jul 31, 2023 at 09:53:17AM +0800, Fei Wu wrote:
> riscv virt platform's m
On Thu, Sep 07, 2023 at 10:35:22AM +0100, Peter Maydell wrote:
> On Thu, 7 Sept 2023 at 10:33, Markus Armbruster wrote:
> >
> > Kevin Wolf writes:
> >
> > > Am 04.09.2023 um 18:25 hat Daniel P. Berrangé geschrieben:
> > >> By the time of the 8.2.0 release, it will have been 2 years and 6
> > >> r
On 01/09/2023 17.58, Nina Schoetterl-Glausch wrote:
From: Pierre Morel
A dedicated CPU in vertical polarization can only have
a high entitlement.
Let's check this from both host and guest point of view.
Signed-off-by: Pierre Morel
Co-developed-by: Nina Schoetterl-Glausch
Signed-off-by: Nina
-soname is not needed for runtime-loaded modules. For example, Meson says:
if not isinstance(target, build.SharedModule) or
target.force_soname:
# Add -Wl,-soname arguments on Linux, -install_name on OS X
commands += linker.get_soname_args(
Andrew Jones writes:
> Add instructions for how to cross-compile QEMU for RISC-V. The
> file is named generically because there's no reason not to collect
> other architectures steps into the same file, especially because
> several subsections like those for cross-compiling QEMU dependencies
>
This series has been successfully tested by QE. Specify a directory
for the mem-path of a memory-backend-file object. Check the error
message has been improved without referring to a device, which can
lead to confusion.
Tested-by: Mario Casquero
On Wed, Sep 6, 2023 at 2:07 PM David Hildenbr
On macOS, compiling contrib/plugins incurs several -Wformat
warnings; fix them.
Signed-off-by: Paolo Bonzini
---
contrib/plugins/cache.c| 10 +-
contrib/plugins/drcov.c| 2 +-
contrib/plugins/howvec.c | 4 ++--
contrib/plugins/lockstep.c | 6 +++---
4 files changed, 11 inser
On 07.09.23 12:31, Mario Casquero wrote:
Tested-by: Mario Casquero
Thanks Mario!
--
Cheers,
David / dhildenb
On 7/9/23 12:18, Paolo Bonzini wrote:
-soname is not needed for runtime-loaded modules. For example, Meson says:
if not isinstance(target, build.SharedModule) or
target.force_soname:
# Add -Wl,-soname arguments on Linux, -install_name on OS X
comm
Trivial:
- Fix string formats
- Link with GLib
Alex, Thomas: FWIW, plugins aren't build on CI.
Based-on: <20230907101811.469236-1-pbonz...@redhat.com>
Philippe Mathieu-Daudé (5):
contrib/plugins/cache: Fix string format
contrib/plugins/drcov: Fix string format
contrib/plugins/howvec: Fix s
This fixes on Darwin:
plugins/cache.c:550:28: warning: format specifies type 'unsigned long' but
the argument has type 'uint64_t' (aka 'unsigned long long') [-Wformat]
l1_daccess,
^~
plugins/cache.c:551:28: warning: format spec
This fixes on Darwin:
plugins/drcov.c:52:13: warning: format specifies type 'unsigned long' but the
argument has type 'uint64_t' (aka 'unsigned long long') [-Wformat]
start_code, end_code, entry, path);
^~
plugins/drcov.c:52:25: warning: format specifies ty
This fixes on Darwin:
ld: symbol(s) not found for architecture arm64
"_g_string_append_printf", referenced from:
_log_stats in cache.o
_log_top_insns in cache.o
_append_stats_line in cache.o
ld: symbol(s) not found for architecture arm64
"_g_string_free", refere
This fixes on Darwin:
plugins/howvec.c:186:40: warning: format specifies type 'long' but the
argument has type 'uint64_t' (aka 'unsigned long long') [-Wformat]
class->count);
^~~~
plugins/howvec.c:213:36
This fixes on Darwin:
plugins/lockstep.c:138:25: warning: format specifies type 'unsigned long' but
the argument has type 'uint64_t' (aka 'unsigned long long') [-Wformat]
us->pc, them->pc, g_slist_length(divergence_log),
^~
plugins/locks
On 7/9/23 12:34, Paolo Bonzini wrote:
On macOS, compiling contrib/plugins incurs several -Wformat
warnings; fix them.
Signed-off-by: Paolo Bonzini
---
contrib/plugins/cache.c| 10 +-
contrib/plugins/drcov.c| 2 +-
contrib/plugins/howvec.c | 4 ++--
contrib/plugins/locks
On 9/7/23 12:49, Philippe Mathieu-Daudé wrote:
- Fix string formats
- Link with GLib
Alex, Thomas: FWIW, plugins aren't build on CI.
They become tested with my patch "configure: unify recursion into
sub-Makefiles", which is how I got into this business. So I'll queue
these patches and inclu
On Wed, 6 Sep 2023 19:22:19 +0800
Yuquan Wang wrote:
> Hi, Jonathan
> On 2023-09-05 22:34, jonathan.cameron wrote:
> >
> > As I understand it the distinction is more about the format / contents of
> > that memory
> > than how you access them.
>
> Yes, RCH DP RCRB includes registers from PC
On 22/06/2023 22:48, 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 translation services for gues
On 9/6/2023 10:53 PM, John Allen wrote:
From: William Roche
AMD guests can't currently deal with BUS_MCEERR_AO MCE injection
as it panics the VM kernel. We filter this event and provide a
warning message.
Signed-off-by: William Roche
---
v3:
- New patch
---
target/i386/kvm/kvm.c | 13 +++
On 06.09.23 15:18, Stefan Hajnoczi wrote:
On Fri, 1 Sept 2023 at 04:18, Hanna Czenczek wrote:
The following changes since commit f5fe7c17ac4e309e47e78f0f9761aebc8d2f2c81:
Merge tag 'pull-tcg-20230823-2' of https://gitlab.com/rth7680/qemu into
staging (2023-08-28 16:07:04 -0400)
are availa
On 7/9/23 12:57, Paolo Bonzini wrote:
On 9/7/23 12:49, Philippe Mathieu-Daudé wrote:
- Fix string formats
- Link with GLib
Alex, Thomas: FWIW, plugins aren't build on CI.
They become tested with my patch "configure: unify recursion into
sub-Makefiles", which is how I got into this business.
This is with QEMU v8.1.0-391-gc152379422a2.
I use the command line from (scroll to the bottom):
https://github.com/tianocore/edk2/commit/49f06b664018
(with "-full-screen" removed).
The crash is as follows:
Unexpected error in object_property_find_err() at
../../src/upstream/qemu/qom/objec
gdb_exit function aims to close gdb sessions and sends the exit code of
the current execution. It's being called by qemu_cleanup once the main
loop is over.
Until now, the exit code sent was always 0. Now that hardware can
shutdown this main loop with custom exit codes, these codes must be
transfer
As of now, the exit code was either EXIT_FAILURE when a panic shutdown
was requested or EXIT_SUCCESS otherwise.
However, some hardware could want to pass more complex exit codes. Thus,
introduce a new shutdown request function allowing that.
Signed-off-by: Clément Chigot
---
include/sysemu/runst
This replaces the exit calls by shutdown requests, ensuring a proper
cleanup of Qemu. Otherwise, some connections like gdb could be broken
before its final packet ("Wxx") is being sent. This part, being done
inside qemu_cleanup function, can be reached only when the main loop
exits after a shutdown
This replaces the exit calls by shutdown requests, ensuring a proper
cleanup of Qemu. Otherwise, some connections like gdb could be broken
before its final packet ("Wxx") is being sent. This part, being done
inside qemu_cleanup function, can be reached only when the main loop
exits after a shutdown
On Tue, 5 Sep 2023 18:55:23 +0200
Philippe Mathieu-Daudé wrote:
> On 5/9/23 17:06, Jonathan Cameron wrote:
> > On Tue, 5 Sep 2023 15:56:39 +0100
> > Jonathan Cameron via wrote:
> >
> >> On Mon, 4 Sep 2023 20:26:59 +0200
> >> Philippe Mathieu-Daudé wrote:
> >>
> >>> On 4/9/23 18:47, Jonatha
This series replaces some of the call to exit in hardware used by
Risc-V boards or made when gdb is requested to exit by shutdown
requests. Otherwise, the gdb connection can be abruptly disconnected
resulting in the last gdb packet "Wxx" being not sent.
For the gdbstub modification, gdb_exit calls
This replaces the exit calls by shutdown requests, ensuring a proper
cleanup of Qemu. Features like net/vhost-vdpa.c are expecting
qemu_cleanup to be called to remove their last residuals.
Signed-off-by: Clément Chigot
---
gdbstub/gdbstub.c | 5 +++--
gdbstub/softmmu.c | 6
On 9/6/23 21:01, Stefan Hajnoczi wrote:
It is not safe to call drain_call_rcu() from qmp_device_add() because
some call stacks are not prepared for drain_call_rcu() to drop the Big
QEMU Lock (BQL).
For example, device emulation code is protected by the BQL but when it
calls aio_poll() -> ... ->
Hi Peter,
On Mon, Sep 4, 2023 at 2:46 PM Clément Chigot wrote:
>
> On Mon, Sep 4, 2023 at 11:42 AM Peter Maydell
> wrote:
> >
> > On Mon, 4 Sept 2023 at 10:36, Clément Chigot wrote:
> > >
> > > On Mon, Sep 4, 2023 at 11:23 AM Peter Maydell
> > > wrote:
> > > >
> > > > On Wed, 23 Aug 2023 at
TARGET_BIG_ENDIAN is *always* defined, either as 0 for little endian
targets or as 1 for big endian targets. So we can use this as a value
directly in places that need such a 0 or 1 for some reason, instead
of taking a detour through an additional local variable or something
similar.
Suggested-by:
v2:
- New patch to push previously static inline functions out of cxl_component.h
- Add the CXL r3.0 expanded set of decoder counts - making it more obviously
why
a switch function makes sense.
- Use computed difference in HDM1 and HDM0 decoder registers sets instead of
0x20 / 4.
- Style
There is no strong justification for keeping these in the header
so push them down into the associated cxl-component-utils.c file.
Suggested-by: Philippe Mathieu-Daudé
Signed-off-by: Jonathan Cameron
---
include/hw/cxl/cxl_component.h | 18 ++
hw/cxl/cxl-component-utils.c | 18
As an encoded version of these key configuration parameters is available
in a register, provide functions to extract it again so as to avoid
the need for duplicating the storage.
Whilst here update the _enc() function to include additional values
as defined in the CXL 3.0 specification. Whilst the
Support these decoders in CXL host bridges (pxb-cxl), CXL Switch USP
and CXL Type 3 end points.
Signed-off-by: Jonathan Cameron
---
include/hw/cxl/cxl_component.h | 10
hw/cxl/cxl-component-utils.c | 27 +
hw/cxl/cxl-host.c | 65 ++---
hw/mem/cxl_ty
On 01/09/2023 12.15, Michael Tokarev wrote:
25.08.2023 20:51, Thomas Huth wrote:
The NIC init code of the jazz machines is rather cumbersome, with
a for-loop around it that is always left after the first iteration.
This patch series reworks this a little bit to make the code more
readable and sh
---
.gitlab-ci.d/cirrus/build.yml | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/.gitlab-ci.d/cirrus/build.yml b/.gitlab-ci.d/cirrus/build.yml
index a9444902ec8..8c71c70574a 100644
--- a/.gitlab-ci.d/cirrus/build.yml
+++ b/.gitlab-ci.d/cirrus/build.yml
@@ -29,7 +29,7 @@ build_
Under Darwin, using -shared makes it impossible to have undefined symbols
and -bundle has to be used instead; so detect the OS and use
different options.
Based-on: <20230907101811.469236-1-pbonz...@redhat.com>
Signed-off-by: Paolo Bonzini
---
This replaces Philippe's patch to link with GL
On Thu, Sep 7, 2023 at 10:19 AM Philippe Mathieu-Daudé
wrote:
> On 7/9/23 09:43, Albert Esteve wrote:
> > This API manages objects (in this iteration,
> > dmabuf fds) that can be shared along different
> > virtio devices, associated to a UUID.
> >
> > The API allows the different devices to add,
Peter Xu writes:
> For both save/load we actually share the logic on deciding whether a field
> should exist. Merge the checks into a helper and use it for both save and
> load. When doing so, add documentations and reformat the code to make it
> much easier to read.
>
> The real benefit here (
On Thu, 7 Sept 2023 at 12:35, Thomas Huth wrote:
>
> TARGET_BIG_ENDIAN is *always* defined, either as 0 for little endian
> targets or as 1 for big endian targets. So we can use this as a value
> directly in places that need such a 0 or 1 for some reason, instead
> of taking a detour through an ad
在 2023/9/7 下午4:31, Song Gao 写道:
+static bool gen_xvaddsub_q(DisasContext *ctx, arg_vvv *a,
+ void (*func)(TCGv_i64, TCGv_i64, TCGv_i64,
+TCGv_i64, TCGv_i64, TCGv_i64))
+{
+if (!check_vec(ctx, 32)) {
+return true;
+}
在 2023/8/31 上午11:09, Richard Henderson 写道:
Signed-off-by: Richard Henderson
---
target/arm/tcg/translate.c | 56 ++
1 file changed, 9 insertions(+), 47 deletions(-)
diff --git a/target/arm/tcg/translate.c b/target/arm/tcg/translate.c
index 38ad8dd4bd..89a7
在 2023/9/7 下午3:39, gaosong 写道:
Hi, Richard
在 2023/8/31 上午11:09, Richard Henderson 写道:
Signed-off-by: Richard Henderson
---
accel/tcg/tcg-runtime.h | 25 ++
include/tcg/tcg-op-gvec-common.h | 6 ++
accel/tcg/tcg-runtime-gvec.c | 26 ++
tcg/tcg-op-gvec.c
On 7/9/23 13:25, Laszlo Ersek wrote:
This is with QEMU v8.1.0-391-gc152379422a2.
I use the command line from (scroll to the bottom):
https://github.com/tianocore/edk2/commit/49f06b664018
(with "-full-screen" removed).
The crash is as follows:
Unexpected error in object_property_find_er
On 7/9/23 13:44, Paolo Bonzini wrote:
Under Darwin, using -shared makes it impossible to have undefined symbols
and -bundle has to be used instead; so detect the OS and use
different options.
Based-on: <20230907101811.469236-1-pbonz...@redhat.com>
Signed-off-by: Paolo Bonzini
---
This
On Thu, Sep 07, 2023 at 11:20:55AM +0100, Alex Bennée wrote:
>
> Andrew Jones writes:
>
> > Add instructions for how to cross-compile QEMU for RISC-V. The
> > file is named generically because there's no reason not to collect
> > other architectures steps into the same file, especially because
>
On 7/9/23 13:35, Thomas Huth wrote:
TARGET_BIG_ENDIAN is *always* defined, either as 0 for little endian
targets or as 1 for big endian targets. So we can use this as a value
directly in places that need such a 0 or 1 for some reason, instead
of taking a detour through an additional local variabl
On 7/9/23 13:35, Jonathan Cameron wrote:
There is no strong justification for keeping these in the header
so push them down into the associated cxl-component-utils.c file.
Suggested-by: Philippe Mathieu-Daudé
Signed-off-by: Jonathan Cameron
---
include/hw/cxl/cxl_component.h | 18 ++-
On 7/9/23 13:35, Jonathan Cameron wrote:
As an encoded version of these key configuration parameters is available
in a register, provide functions to extract it again so as to avoid
the need for duplicating the storage.
Whilst here update the _enc() function to include additional values
as defin
On Thu, Sep 07, 2023 at 02:31:20PM +0200, Andrew Jones wrote:
> On Thu, Sep 07, 2023 at 11:20:55AM +0100, Alex Bennée wrote:
> >
> > Andrew Jones writes:
> >
> > > Add instructions for how to cross-compile QEMU for RISC-V. The
> > > file is named generically because there's no reason not to coll
On Tue, Sep 5, 2023 at 3:45 PM Peter Xu wrote:
>
> On Tue, Sep 05, 2023 at 09:38:39AM +0200, Mattias Nissler wrote:
> > It would be nice to use a property on the device that originates the
> > DMA operation to configure this. However, I don't see how to do this
> > in a reasonable way without bigg
Hello Joao,
Cedric, you mentioned that you take a look at this after you come back, not sure
if that's still the plan. But it's been a while since the last version, so would
you have me repost/rebase on the latest (post your PR)?
Yes please. That's next on the TODO list (after some downstream
On 7/9/23 13:35, Jonathan Cameron wrote:
Support these decoders in CXL host bridges (pxb-cxl), CXL Switch USP
and CXL Type 3 end points.
Signed-off-by: Jonathan Cameron
---
include/hw/cxl/cxl_component.h | 10
hw/cxl/cxl-component-utils.c | 27 +
hw/cxl/cxl-host.c
From: Michael Tokarev
This will stop linking softmmu-specific os_parse_cmd_args() into every
qemu executable which happens to use other functions from os-posix.c,
such as os_set_line_buffering() or os_setup_signal_handling().
Also, since there's no win32-specific options, *all* option parsing is
From: Michael Tokarev
Signed-off-by: Michael Tokarev
Reviewed-by: Eric Blake
Reviewed-by: Richard Henderson
Message-ID: <20230901101302.3618955-3-...@tls.msk.ru>
Signed-off-by: Paolo Bonzini
---
include/sysemu/os-posix.h | 1 +
os-posix.c| 23 ---
2 files
Otherwise tcg_handle_interrupt() triggers an assertion failure:
#5 0x55c97369 in tcg_handle_interrupt (cpu=0x57434cb0, mask=2)
at ../accel/tcg/tcg-accel-ops.c:83
#6 tcg_handle_interrupt (cpu=0x57434cb0, mask=2) at
../accel/tcg/tcg-accel-ops.c:81
#7 0x55b4d58b in
From: Michael Tokarev
Signed-off-by: Michael Tokarev
Reviewed-by: Eric Blake
Message-ID: <20230901101302.3618955-7-...@tls.msk.ru>
Signed-off-by: Paolo Bonzini
---
os-posix.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/os-posix.c b/os-posix.c
index a8e38c02988..f90dfda9b0d 100644
---
Under Darwin, using -shared makes it impossible to have undefined symbols
and -bundle has to be used instead; so detect the OS and use
different options.
Based-on: <20230907101811.469236-1-pbonz...@redhat.com>
Signed-off-by: Paolo Bonzini
---
contrib/plugins/Makefile | 4
1 file changed, 4
-soname is not needed for runtime-loaded modules. For example, Meson says:
if not isinstance(target, build.SharedModule) or
target.force_soname:
# Add -Wl,-soname arguments on Linux, -install_name on OS X
commands += linker.get_soname_args(
From: Philippe Mathieu-Daudé
User emulation doesn't need any KVM declarations.
Signed-off-by: Philippe Mathieu-Daudé
Message-ID: <20230904124325.79040-5-phi...@linaro.org>
Signed-off-by: Paolo Bonzini
---
target/i386/helper.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a
Am 07.09.2023 um 11:33 hat Markus Armbruster geschrieben:
> Kevin Wolf writes:
> > Am 04.09.2023 um 18:25 hat Daniel P. Berrangé geschrieben:
> >> I still think for user creatable devices we'd be better off just
> >> mandating the use of JSON syntax for -device and thus leveraging
> >> the native
Safe signal handling around system calls is mandatory for user-mode
emulation, and requires a small piece of handwritten assembly code.
So refuse to compile unless the common-user/host subdirectory exists
for the host architecture that was detected or selected with --cpu.
Reviewed-by: Richard Hend
Reviewed-by: Richard Henderson
Reviewed-by: Peter Maydell
Signed-off-by: Paolo Bonzini
---
configure | 11 +--
meson.build | 3 ++-
meson_options.txt | 2 ++
scripts/meson-buildoptions.sh | 3 +++
4 files changed, 8 insertions(+), 11
From: Philippe Mathieu-Daudé
This fixes on Darwin:
plugins/cache.c:550:28: warning: format specifies type 'unsigned long' but
the argument has type 'uint64_t' (aka 'unsigned long long') [-Wformat]
l1_daccess,
^~
plugins/cache
Just use $targetos always.
Signed-off-by: Paolo Bonzini
---
configure | 55 +--
1 file changed, 13 insertions(+), 42 deletions(-)
diff --git a/configure b/configure
index 378a0de9fb6..f96f7359a83 100755
--- a/configure
+++ b/configure
@@ -374,
From: Michael Tokarev
this moves code blocks so that functions and variables which
belongs to the same concept are now close to each other.
There's no actual code changes in there.
Signed-off-by: Michael Tokarev
Reviewed-by: Eric Blake
Message-ID: <20230901101302.3618955-6-...@tls.msk.ru>
Sign
From: Philippe Mathieu-Daudé
kvm_arch_get_supported_cpuid() / kvm_arch_get_supported_msr_feature()
are only defined for x86 targets (in target/i386/kvm/kvm.c). Their
declarations are pointless on other targets.
Signed-off-by: Philippe Mathieu-Daudé
Message-ID: <20230904124325.79040-11-phi...@li
Debian 10 has Python 3.7, so it is not possible to use it anymore
now that Python 3.8 is required.
Signed-off-by: Paolo Bonzini
---
python/scripts/mkvenv.py | 74 +---
1 file changed, 16 insertions(+), 58 deletions(-)
diff --git a/python/scripts/mkvenv.py b/p
From: Philippe Mathieu-Daudé
kvm_pc_setup_irq_routing() is only defined for x86 targets (in
hw/i386/kvm/apic.c). Its declaration is pointless on all
other targets.
Signed-off-by: Philippe Mathieu-Daudé
Message-ID: <20230904124325.79040-14-phi...@linaro.org>
Signed-off-by: Paolo Bonzini
---
hw
Signed-off-by: Paolo Bonzini
---
configure | 8
1 file changed, 8 deletions(-)
diff --git a/configure b/configure
index ef085376922..48ae0fc05ab 100755
--- a/configure
+++ b/configure
@@ -1126,14 +1126,6 @@ else
done
fi
-# see if system emulation was really requested
-case " $ta
Instead of using a single global bounce buffer, give each AddressSpace
its own bounce buffer. The MapClient callback mechanism moves to
AddressSpace accordingly.
This is in preparation for generalizing bounce buffer handling further
to allow multiple bounce buffers, with a total allocation limit
c
From: Tao Su
Latest Intel platform GraniteRapids-D introduces AMX-COMPLEX, which adds
two instructions to perform matrix multiplication of two tiles containing
complex elements and accumulate the results into a packed single precision
tile.
AMX-COMPLEX is enumerated via CPUID.(EAX=7,ECX=1):EDX[b
The initial reason to write this patch was to remove the last use of
CONFIG_DEBUG_TCG from the makefiles; the flags to use to build TCG
plugins are unrelated to --enable-debug-tcg, and instead they should
be the same as those used to build emulators (the plugins are not build
via meson for demonstr
importlib.metadata is included in Python 3.8, so there is no
need to fallback to either importlib-metadata or pkgresources
when generating console script shims.
Signed-off-by: Paolo Bonzini
---
python/scripts/mkvenv.py | 197 ++-
python/setup.cfg | 9
From: Michael Tokarev
qemu-options.h just includes qemu-options.def with some #defines.
We already do this in vl.c in other place. Since no other file
includes qemu-options.h anymore, just inline it in vl.c.
This effectively reverts second half of commit 59a5264b99434.
Signed-off-by: Michael To
CVTPS2PD only loads a half-register for memory, unlike the other
operations under 0x0F 0x5A. "Unpack" the group into separate
emission functions instead of using gen_unary_fp_sse.
Signed-off-by: Paolo Bonzini
---
target/i386/tcg/decode-new.c.inc | 14 --
target/i386/tcg/emit.c.inc
Debian 10 is not anymore a supported distro, since Debian 12 was
released on June 10, 2023. Our supported build platforms as of today
all support at least 3.8 (and all of them except for Ubuntu 20.04
support 3.9):
openSUSE Leap 15.5: 3.6.15 (3.11.2)
CentOS Stream 8:3.6.8 (3.8.13, 3.9.16, 3.1
From: Philippe Mathieu-Daudé
Keep the function accessed by target/i386/ and hw/i386/
exposed, restrict the ones accessed by target/i386/kvm/.
Signed-off-by: Philippe Mathieu-Daudé
Message-ID: <20230904124325.79040-10-phi...@linaro.org>
Signed-off-by: Paolo Bonzini
---
target/i386/kvm/kvm_i386
PCI config space is little-endian, so on a big-endian host we need to
perform byte swaps for values as they are passed to and received from
the generic PCI config space access machinery.
Signed-off-by: Mattias Nissler
---
hw/remote/vfio-user-obj.c | 4 ++--
1 file changed, 2 insertions(+), 2 del
This series adds basic support for message-based DMA in qemu's vfio-user
server. This is useful for cases where the client does not provide file
descriptors for accessing system memory via memory mappings. My motivating use
case is to hook up device models as PCIe endpoints to a hardware design. Th
From: Michael Tokarev
Signed-off-by: Michael Tokarev
Reviewed-by: Eric Blake
Message-ID: <20230901101302.3618955-2-...@tls.msk.ru>
Signed-off-by: Paolo Bonzini
---
include/sysemu/os-posix.h | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/include/sysemu/os-posix.h b/in
From: Philippe Mathieu-Daudé
This fixes on Darwin:
plugins/drcov.c:52:13: warning: format specifies type 'unsigned long' but the
argument has type 'uint64_t' (aka 'unsigned long long') [-Wformat]
start_code, end_code, entry, path);
^~
plugins/drcov.c:52:2
From: Marc-André Lureau
"blob" resources don't have an associated pixman image:
#0 pixman_image_get_stride (image=0x0) at ../pixman/pixman-image.c:921
#1 0x562327c25236 in virtio_gpu_save (f=0x56232bb13b00,
opaque=0x56232b555a60, size=0, field=0x5623289ab6c8 <__compound_literal.3+104>,
v
Unsupported CPU and OSes are not really going away, but the
project simply does not guarantee that they work. Rephrase
the messages accordingly. While at it, move the warning for
TCI performance at the end where it is more visible.
Reviewed-by: Thomas Huth
Signed-off-by: Paolo Bonzini
---
mes
From: Philippe Mathieu-Daudé
kvm_has_pit_state2() is only defined for x86 targets (in
target/i386/kvm/kvm.c). Its declaration is pointless on
all other targets. Have it return a boolean.
Signed-off-by: Philippe Mathieu-Daudé
Message-ID: <20230904124325.79040-13-phi...@linaro.org>
Signed-off-by:
From: Philippe Mathieu-Daudé
This fixes on Darwin:
plugins/lockstep.c:138:25: warning: format specifies type 'unsigned long' but
the argument has type 'uint64_t' (aka 'unsigned long long') [-Wformat]
us->pc, them->pc, g_slist_length(divergence_log),
From: Philippe Mathieu-Daudé
This fixes on Darwin:
plugins/howvec.c:186:40: warning: format specifies type 'long' but the
argument has type 'uint64_t' (aka 'unsigned long long') [-Wformat]
class->count);
^~~
There are no config-host.mak symbols anymore that are needed in
config-host.h; the only symbols that are included in config_host_data via
the foreach loop are:
- CONFIG_DEFAULT_TARGETS, which is not used by C code.
- CONFIG_TCG and CONFIG_TCG_INTERPRETER, which are not part of config-host.mak
So
The argument of --host-cc is not obeyed when cross compiling. To avoid
this issue, place it in a configuration file and pass it to meson
with --native-file.
While at it, clarify that --host-cc is not obeyed anyway when _not_
cross compiling, because cc="$host_cc" is placed before --host-cc is
pro
When DMA memory can't be directly accessed, as is the case when
running the device model in a separate process without shareable DMA
file descriptors, bounce buffering is used.
It is not uncommon for device models to request mapping of several DMA
regions at the same time. Examples include:
* net
CONFIG_SOLARIS is only used to pick tap implementations. But the
target OS is invariant and does not depend on the configuration, so move
away from config_host and just use unconditional rules in softmmu_ss.
Reviewed-by: Philippe Mathieu-Daudé
Signed-off-by: Paolo Bonzini
---
configure |
From: Michael Tokarev
Signed-off-by: Michael Tokarev
Reviewed-by: Eric Blake
Message-ID: <20230901101302.3618955-4-...@tls.msk.ru>
Signed-off-by: Paolo Bonzini
---
include/sysemu/os-posix.h | 1 +
os-posix.c| 9 +++--
2 files changed, 8 insertions(+), 2 deletions(-)
diff
From: Michael Tokarev
Signed-off-by: Michael Tokarev
Reviewed-by: Eric Blake
Message-ID: <20230901101302.3618955-9-...@tls.msk.ru>
Signed-off-by: Paolo Bonzini
---
{util => softmmu}/async-teardown.c | 0
softmmu/meson.build| 1 +
util/meson.build | 1 -
3 fil
This allows building libblkio at the same time as QEMU, if QEMU is
configured with --enable-blkio --enable-download.
Signed-off-by: Paolo Bonzini
---
subprojects/libblkio.wrap | 6 ++
1 file changed, 6 insertions(+)
create mode 100644 subprojects/libblkio.wrap
diff --git a/subprojects/libb
101 - 200 of 450 matches
Mail list logo
