On Wed, Dec 08 2021, Cornelia Huck wrote:
> Add 7.0 machine types for arm/i440fx/q35/s390x/spapr.
>
> Signed-off-by: Cornelia Huck
> ---
> hw/arm/virt.c | 9 -
> hw/core/machine.c | 3 +++
> hw/i386/pc.c | 3 +++
> hw/i386/pc_piix.c | 14 +
On Fri, Dec 17, 2021 at 02:31:14AM +, Jag Raman wrote:
>
>
> > On Dec 16, 2021, at 4:58 AM, Stefan Hajnoczi wrote:
> >
> > On Wed, Dec 15, 2021 at 10:35:28AM -0500, Jagannathan Raman wrote:
> >> +static void vfu_object_set_socket(Object *obj, Visitor *v, const char
> >> *name,
> >> +
On Fri, Dec 17, 2021 at 12:26:53PM +0800, Jason Wang wrote:
Dave: You created the VIRTIO vmstate infrastructure in QEMU. Please see
the bottom of this email about moving to a standard VIRTIO device
save/load format defined by the VIRTIO spec in the future.
> On Thu, Dec 16, 2021 at 5:10 PM Stefan
On Dec 17 06:27, Qiuhao Li wrote:
> Thanks Alex. It seems this patch sets and checks if the destination device is
> busy. But how about the data transfers not triggered directly by PMIO/MMIO
> handlers? For example:
>
> 1. Device A Timer's callback -> Device A MMIO handler
> 2. Device A BH's cal
Yes, it still works. Now it looks orthodox:
cat << EOF | ./qemu-system-x86_64 -display none -machine accel=qtest \
-machine q35 -nodefaults -drive file=null-co://,if=none,format=raw,id=disk0 \
-device nvme,drive=disk0,serial=1 -qtest stdio \
outl 0xcf8 0x8810 /* MLBAR (BAR0) – M
Meson is planning to change the default of the "check" argument to
run_command (from false to true). Be explicit and include it in
all invocations.
Signed-off-by: Paolo Bonzini
---
docs/meson.build | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/docs/meson.build b/docs
From: Paolo Montesel
Signed-off-by: Alessandro Di Federico
Signed-off-by: Paolo Montesel
Signed-off-by: Anton Johansson
---
target/hexagon/idef-parser/idef-parser.h | 259 ++
target/hexagon/idef-parser/idef-parser.lex | 571 +
target/hexagon/meson.build
From: Alessandro Di Federico
Signed-off-by: Alessandro Di Federico
Signed-off-by: Anton Johansson
---
target/hexagon/README | 5 +
target/hexagon/idef-parser/README.rst | 722 ++
2 files changed, 727 insertions(+)
create mode 100644 target/hexagon/ide
From: Paolo Montesel
Make certain helper functions non-static, making them available outside
genptr.c. These functions are required by code generated by the
idef-parser.
This commit also makes some op_helper.c non-static in order to avoid
having them marked as unused when using the idef-parser g
From: Paolo Montesel
Signed-off-by: Alessandro Di Federico
Signed-off-by: Paolo Montesel
Reviewed-by: Richard Henderson
Reviewed-by: Taylor Simpson
---
target/hexagon/translate.c | 3 ++-
target/hexagon/translate.h | 1 +
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/target/h
From: Alessandro Di Federico
Extend gen_tcg_funcs.py in order to emit calls to the functions emitted
by the idef-parser, if available. An option is also added to fully
disable the output of the idef-parser, which is useful for debugging purposes.
Signed-off-by: Alessandro Di Federico
Signed-off
From: Paolo Montesel
Signed-off-by: Alessandro Di Federico
Signed-off-by: Paolo Montesel
Acked-by: Richard Henderson
---
target/hexagon/genptr.c | 24 +---
target/hexagon/macros.h | 2 +-
2 files changed, 14 insertions(+), 12 deletions(-)
diff --git a/target/hexagon/genp
From: Paolo Montesel
Signed-off-by: Alessandro Di Federico
Signed-off-by: Paolo Montesel
Signed-off-by: Anton Johansson
---
.gitlab-ci.d/windows.yml | 2 ++
tests/docker/dockerfiles/alpine.docker | 2 ++
tests/docker/dockerfiles/centos8.docker
From: Niccolò Izzo
Signed-off-by: Alessandro Di Federico
Signed-off-by: Niccolò Izzo
---
tests/tcg/hexagon/Makefile.target | 30 +-
tests/tcg/hexagon/crt.S| 26
tests/tcg/hexagon/test_abs.S | 20 ++
tests/tcg/hexagon/test_bitcnt.S| 42 +++
From: Niccolò Izzo
These helpers will be employed by the idef-parser generated code.
"Helper" can here mean two things, a helper in the QEMU sense added to
`helper.h` and `op_helper.c`, but also helper functions providing a
manual TCG implementation of a certain features.
Signed-off-by: Alessand
From: Alessandro Di Federico
This commit is necessary in order to use container built by the current
run of the CI. If we don't do this, we use official containers which are
not affected by the additional dependencies we're introducing.
This should be considered as a temporary solution in order
From: Alessandro Di Federico
Introduce infrastructure necessary to produce a file suitable for being
parsed by the idef-parser.
Signed-off-by: Alessandro Di Federico
Signed-off-by: Anton Johansson
---
target/hexagon/gen_idef_parser_funcs.py | 123 +
target/hexagon/idef-par
Signed-off-by: Alessandro Di Federico
Signed-off-by: Paolo Montesel
Signed-off-by: Anton Johansson
---
target/hexagon/idef-parser/idef-parser.y| 1064
target/hexagon/idef-parser/parser-helpers.c | 2548 +++
target/hexagon/idef-parser/parser-helpers.h | 379 +++
tar
This patchset introduces the idef-parser for target/hexagon.
It's the seventh iteration of the patchset and includes fixes suggested in
previous iterations.
`idef-parser` is a build-time tool built using flex and bison. Its aim
is to generate a large part of the tiny code generator frontend for
H
On 12/16/21 11:05, Paolo Bonzini wrote:
Just use sizeof, avoiding the need to write down all the ABIs twice.
Cc: Laurent Vivier
Signed-off-by: Paolo Bonzini
This breaks bsd-user, I'll resend when "[PATCH v7 00/15] linux-user:
simplify safe signal handling" is in.
Paolo
---
include/exe
From: Alessandro Di Federico
Signed-off-by: Alessandro Di Federico
Signed-off-by: Anton Johansson
Reviewed-by: Richard Henderson
---
MAINTAINERS | 11 ++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index 7543eb4d59..f6fbc5f664 100644
--- a/
Am 15.12.21 um 16:53 schrieb Matthew Rosato:
On 12/15/21 2:35 AM, Pierre Morel wrote:
On 12/7/21 22:04, Matthew Rosato wrote:
Note: The first 3 patches of this series are included as pre-reqs, but
should be pulled via a separate series. Also, patch 5 is needed to
support 5.16+ linux head
On 12/15/21 11:29, Philippe Mathieu-Daudé wrote:
Hi Laurent,
This patch is reviewed, can it go via your trivial tree?
Queued, thanks.
Paolo
On 11/22/21 11:47, Philippe Mathieu-Daudé wrote:
Commit 739e95f5741 ("scsi: Replace scsi_bus_new() with
scsi_bus_init(), scsi_bus_init_named()") forgo
On 11/19/21 21:11, Philippe Mathieu-Daudé wrote:
Fix issue #521 reported by Alex some months ago:
https://gitlab.com/qemu-project/qemu/-/issues/521
Philippe Mathieu-Daudé (2):
hw/scsi/megasas: Fails command if SGL buffer overflows
tests/qtest/fuzz-megasas-test: Add test for GitLab issue #5
On 12/1/21 23:32, Philippe Mathieu-Daudé wrote:
If the 'i8042' property is not set, mouse events handled by
vmmouse_mouse_event() end calling i8042_isa_mouse_fake_event()
with a NULL argument, resulting in ps2_mouse_fake_event() being
called with invalid PS2MouseState pointer. Fix by requiring
th
Clean up some unnecessary code
Signed-off-by: Li Zhang
---
migration/multifd.c | 9 ++---
1 file changed, 2 insertions(+), 7 deletions(-)
diff --git a/migration/multifd.c b/migration/multifd.c
index 3242f688e5..1405cf95b8 100644
--- a/migration/multifd.c
+++ b/migration/multifd.c
@@ -854,9
sndio is the native API used by OpenBSD, although it has been ported to
other *BSD's and Linux (packages for Ubuntu, Debian, Void, Arch, etc.).
Signed-off-by: Brad Smith
Signed-off-by: Alexandre Ratchov
---
Thank you for the reviews and all the comments. Here's a second diff
with all the sugges
Le 09/11/2021 à 15:45, Philippe Mathieu-Daudé a écrit :
When using the MinGW toolchain, we use the .exe suffix for the
executable name. We also need to use it for the symlinks in the
build directory.
Signed-off-by: Philippe Mathieu-Daudé
---
configure | 2 +-
1 file changed, 1 insertion(+),
Hi Li,
the full function for context:
static void multifd_new_send_channel_async(QIOTask *task, gpointer opaque)
{
MultiFDSendParams *p = opaque;
QIOChannel *sioc = QIO_CHANNEL(qio_task_get_source(task));
Error *local_err = NULL;
trace_multifd_new_send_channel_async(p->id);
i
Le 16/11/2021 à 20:39, Philippe Mathieu-Daudé a écrit :
QEMU coding style mandates to not use Linux kernel internal
types for scalars types. Replace __u32 by uint32_t.
Signed-off-by: Philippe Mathieu-Daudé
---
target/i386/kvm/kvm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff
On Fri, Dec 17, 2021 at 7:28 AM Qiuhao Li wrote:
>
> Thanks Alex. It seems this patch sets and checks if the destination device is
> busy. But how about the data transfers not triggered directly by PMIO/MMIO
> handlers? For example:
>
> 1. Device A Timer's callback -> Device A MMIO handler
> 2.
Le 17/11/2021 à 18:45, Philippe Mathieu-Daudé a écrit :
Signed-off-by: Philippe Mathieu-Daudé
---
qemu-keymap.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/qemu-keymap.c b/qemu-keymap.c
index 536e8f2385d..4095b654a60 100644
--- a/qemu-keymap.c
+++ b/qemu-keymap.c
@@ -187,6 +187,7 @@ i
Le 05/12/2021 à 23:41, Philippe Mathieu-Daudé a écrit :
TYPE_AVR_CPU inherits TYPE_CPU, which itself inherits TYPE_DEVICE.
TYPE_DEVICE instances are realized using qdev_realize(), we don't
need to access QOM internal values.
Signed-off-by: Philippe Mathieu-Daudé
---
hw/avr/atmega.c | 2 +-
1
Le 23/11/2021 à 12:48, lagar...@linux.ibm.com a écrit :
From: Leonardo Garcia
Signed-off-by: Leonardo Garcia
Reviewed-by: Laurent Vivier
Reviewed-by: Philippe Mathieu-Daudé
---
hw/virtio/vhost.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/virtio/vhost.c b/hw/vir
On 12/17/21 10:39 AM, Claudio Fontana wrote:
Hi Li,
the full function for context:
static void multifd_new_send_channel_async(QIOTask *task, gpointer opaque)
{
MultiFDSendParams *p = opaque;
QIOChannel *sioc = QIO_CHANNEL(qio_task_get_source(task));
Error *local_err = NULL;
On 12/16/21 16:16, Matthew Rosato wrote:
On 12/16/21 3:15 AM, Pierre Morel wrote:
On 12/7/21 22:04, Matthew Rosato wrote:
Let's use the reserved pool of simulated PCI groups to allow intercept
devices to have separate groups from interpreted devices as some group
values may be different. I
Le 08/11/2021 à 21:07, Stefan Weil a écrit :
A build with gcc (Debian 10.2.1-6) 10.2.1 20210110 fails:
../../../softmmu/device_tree.c: In function ‘qemu_fdt_add_path’:
../../../softmmu/device_tree.c:560:18: error: ‘retval’ may be used
uninitialized in this function [-Werror=maybe-uninitialized]
Clean up some unnecessary code
Signed-off-by: Li Zhang
---
migration/multifd.c | 12 +++-
1 file changed, 3 insertions(+), 9 deletions(-)
diff --git a/migration/multifd.c b/migration/multifd.c
index 3242f688e5..212be1ed04 100644
--- a/migration/multifd.c
+++ b/migration/multifd.c
@@ -85
Le 18/10/2021 à 10:50, Zhang Chen a écrit :
This demo not correct, the original childs1 can't pass the
the bdrv_is_root_node check in replcation_start().
Keep consistent with docs/COLO-FT.txt
Signed-off-by: Zhang Chen
---
docs/block-replication.txt | 2 +-
1 file changed, 1 insertion(+), 1 d
Le 15/12/2021 à 11:24, Philippe Mathieu-Daudé a écrit :
Ping?
On 11/22/21 11:51, Philippe Mathieu-Daudé wrote:
While both MII_ANAR_TX / MII_ANLPAR_TX have the same value,
the ANAR and ANLPAR registers are different, so technically
the ANLPAR register have to use the MII_ANLPAR_TX definition.
S
Le 18/10/2021 à 10:50, Zhang Chen a écrit :
This demo not correct, the original childs1 can't pass the
the bdrv_is_root_node check in replcation_start().
Keep consistent with docs/COLO-FT.txt
Signed-off-by: Zhang Chen
---
docs/block-replication.txt | 2 +-
1 file changed, 1 insertion(+), 1 d
"meson test" starting with version 0.57 is just as capable and easy to
use as QEMU's own TAP driver. All existing options for "make check"
work. The only required code change involves how to mark "slow" tests;
they need to belong additionally to the "slow" suite.
The rules for .tap output are re
"if (tcg_enabled())" allows elision of the code inside it; we only need
the prototype to exist, so that the code compile even for the --disable-tcg
case.
Signed-off-by: Paolo Bonzini
---
cpu.c | 5 -
include/exec/cpu-all.h | 2 --
2 files changed, 7 deletions(-)
diff --git
The test was disabled when CONFIG_EPOLL_CREATE1 was moved out
of config-host.mak. Fix the condition.
Signed-off-by: Paolo Bonzini
---
tests/unit/meson.build | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/unit/meson.build b/tests/unit/meson.build
index acac3622ed..90ac
Le 03/09/2021 à 19:44, Philippe Mathieu-Daudé a écrit :
When experimenting raising GLIB_VERSION_MIN_REQUIRED to 2.68
(Fedora 34 provides GLib 2.68.1) we get:
hw/virtio/virtio-crypto.c:245:24: error: 'g_memdup' is deprecated: Use
'g_memdup2' instead [-Werror,-Wdeprecated-declarations]
...
Le 03/09/2021 à 19:45, Philippe Mathieu-Daudé a écrit :
Per
https://discourse.gnome.org/t/port-your-module-from-g-memdup-to-g-memdup2-now/5538
The old API took the size of the memory to duplicate as a guint,
whereas most memory functions take memory sizes as a gsize. This
made it easy
Le 03/09/2021 à 19:45, Philippe Mathieu-Daudé a écrit :
g_memdup() is insecure and as been deprecated in GLib 2.68.
QEMU provides the safely equivalent g_memdup2() wrapper.
Do not allow more g_memdup() calls in the repository, provide
a hint to use g_memdup2().
Signed-off-by: Philippe Mathieu-D
Le 15/12/2021 à 17:54, Philippe Mathieu-Daudé a écrit :
Hi Laurent,
On 9/3/21 19:44, Philippe Mathieu-Daudé wrote:
This series provides the safely equivalent g_memdup2() wrapper,
and replace all g_memdup() calls by it.
Philippe Mathieu-Daudé (28):
hw/hyperv/vmbus: Remove unused vmbus_loa
On 24.11.21 07:44, Emanuele Giuseppe Esposito wrote:
bdrv_co_invalidate_cache is special: it is an I/O function,
I still don’t believe it is, but well.
(Yes, it is called by a test in an iothread, but I believe we’ve seen
that the tests simply sometimes test things that shouldn’t be allowed.)
Philippe Mathieu-Daudé writes:
> On 12/16/21 15:11, Alex Bennée wrote:
>> Philippe Mathieu-Daudé writes:
>>
>>> When experimenting raising GLIB_VERSION_MIN_REQUIRED to 2.68
>>> (Fedora 34 provides GLib 2.68.1) we get:
>>>
>>> hw/virtio/virtio-crypto.c:245:24: error: 'g_memdup' is deprecated
Alex,
I've added this patch to my trivial patches branch, do you want I drop it?
Thanks,
Laurent
On 17/12/2021 12:10, Alex Bennée wrote:
Philippe Mathieu-Daudé writes:
On 12/16/21 15:11, Alex Bennée wrote:
Philippe Mathieu-Daudé writes:
When experimenting raising GLIB_VERSION_MIN_REQUIR
Laurent Vivier writes:
> Alex,
>
> I've added this patch to my trivial patches branch, do you want I drop
> it?
No - the patch itself is fine. I would like to fix up the consistency
later if we can agree on what it should be.
--
Alex Bennée
On 24.11.21 07:44, Emanuele Giuseppe Esposito wrote:
block_crypto_amend_options_generic_luks uses the block layer
permission API, therefore it should be called with the BQL held.
However, the same function is being called ib two BlockDriver
s/ ib / by /
callbacks: bdrv_amend_options (under B
On 10/9/21 23:08, marcandre.lur...@redhat.com wrote:
> From: Marc-André Lureau
>
> Currently, virgl initialization error is silent. Make it verbose instead.
>
> (this is likely going to bug later on, as the device isn't fully
> initialized)
>
> Signed-off-by: Marc-André Lureau
> ---
> hw/disp
On 10/9/21 23:08, marcandre.lur...@redhat.com wrote:
> From: Marc-André Lureau
>
> Signed-off-by: Marc-André Lureau
> ---
> hw/display/virtio-gpu-virgl.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
Reviewed-by: Philippe Mathieu-Daudé
Le 16/12/2021 à 15:44, Alex Bennée a écrit :
The pgb_find_hole function goes to the trouble of taking account of
both mmap_min_addr and any offset we've applied to decide the starting
address of a potential hole. This is especially important for
emulating 32bit ARM in a 32bit build as we have app
With the current wording, users might think that the -F option is not
required as long as the script is placed in the default path[0]. Be
clear that the option is always required.
[0]: https://forum.proxmox.com/threads/82680/post-437608
Signed-off-by: Fabian Ebner
---
I also tried to improve th
Laurent Vivier writes:
> Le 16/12/2021 à 15:44, Alex Bennée a écrit :
>> The pgb_find_hole function goes to the trouble of taking account of
>> both mmap_min_addr and any offset we've applied to decide the starting
>> address of a potential hole. This is especially important for
>> emulating 32
On Thu, Dec 16, 2021 at 11:41:27PM +0200, Dov Murik wrote:
>
>
> On 16/12/2021 18:09, Daniel P. Berrangé wrote:
> > On Thu, Dec 16, 2021 at 12:38:34PM +0200, Dov Murik wrote:
> >>
> >>
> >> On 14/12/2021 20:39, Daniel P. Berrangé wrote:
> >>> On Tue, Dec 14, 2021 at 01:59:10PM +, Dov Murik wr
On 10/9/21 23:08, marcandre.lur...@redhat.com wrote:
> From: Marc-André Lureau
>
> Further meson configuration tests are to be added based on the glib
> version. Also correct the version reporting in the config log.
>
> Signed-off-by: Marc-André Lureau
> ---
> configure | 1 +
> meson.build
On 10/9/21 23:08, marcandre.lur...@redhat.com wrote:
> From: Marc-André Lureau
>
> Signed-off-by: Marc-André Lureau
> ---
> include/chardev/char-socket.h | 84 +++
> chardev/char-socket.c | 58 +---
> 2 files changed, 85 insertions(+),
On Fri, Dec 17, 2021 at 09:13:55AM +0100, Cornelia Huck wrote:
> On Wed, Dec 08 2021, Cornelia Huck wrote:
>
> > Add 7.0 machine types for arm/i440fx/q35/s390x/spapr.
> >
> > Signed-off-by: Cornelia Huck
> > ---
> > hw/arm/virt.c | 9 -
> > hw/core/machine.c | 3
On 10/9/21 23:08, marcandre.lur...@redhat.com wrote:
> From: Marc-André Lureau
>
> The next patch will make use of this function to dissociate
> DisplayChangeListener from GL context.
>
> Signed-off-by: Marc-André Lureau
> ---
> include/ui/console.h | 3 +++
> ui/console.c | 22 ++
On 10/9/21 23:08, marcandre.lur...@redhat.com wrote:
> From: Marc-André Lureau
>
> Export the default consoles over the D-Bus chardev.
>
> Signed-off-by: Marc-André Lureau
> ---
> ui/dbus.c | 53 +
> 1 file changed, 53 insertions(+)
> +stati
On Fri, Dec 17, 2021 at 11:10:31AM +, Alex Bennée wrote:
>
> Philippe Mathieu-Daudé writes:
>
> > On 12/16/21 15:11, Alex Bennée wrote:
> >> Philippe Mathieu-Daudé writes:
> >>
> >>> When experimenting raising GLIB_VERSION_MIN_REQUIRED to 2.68
> >>> (Fedora 34 provides GLib 2.68.1) we get:
Based-on: 20211208170241.110551-1-coh...@redhat.com
Support VIRTIO_MEM_F_UNPLUGGED_INACCESSIBLE in QEMU, which indicates to
a guest that we don't support reading unplugged memory. We indicate
the feature based on a new "unplugged-inaccessible" property available
for x86 targets only (the only ones
Set the new default to "auto", keeping it set to "off" for compat
machines. This property is only available for x86 targets.
Reviewed-by: Michal Privoznik
Reviewed-by: Pankaj Gupta
Signed-off-by: David Hildenbrand
---
hw/i386/pc.c | 4 +++-
hw/virtio/virtio-mem.c | 2 +-
2 files chan
With VIRTIO_MEM_F_UNPLUGGED_INACCESSIBLE, we signal the VM that reading
unplugged memory is not supported. We have to fail feature negotiation
in case the guest does not support VIRTIO_MEM_F_UNPLUGGED_INACCESSIBLE.
First, VIRTIO_MEM_F_UNPLUGGED_INACCESSIBLE is required to properly handle
memory ba
Let's prepare touch_all_pages() for returning differing errors. Return
an error from the thread and report the last processed error.
Translate SIGBUS to -EFAULT, as a SIGBUS can mean all different kind of
things (memory error, read error, out of memory). When allocating memory
fails via the curren
Let's synchronize the new feature flag, available in Linux since
v5.16-rc1.
Reviewed-by: Michal Privoznik
Signed-off-by: David Hildenbrand
---
include/standard-headers/linux/virtio_mem.h | 9 ++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/include/standard-headers/linux/
Let's simplify the case when we only want a single thread and don't have
to mess with signal handlers.
Reviewed-by: Pankaj Gupta
Reviewed-by: Daniel P. Berrangé
Reviewed-by: Michal Privoznik
Signed-off-by: David Hildenbrand
---
util/oslib-posix.c | 8
1 file changed, 8 insertions(+)
Let's sense support and use it for preallocation. MADV_POPULATE_WRITE
does not require a SIGBUS handler, doesn't actually touch page content,
and avoids context switches; it is, therefore, faster and easier to handle
than our current approach.
While MADV_POPULATE_WRITE is, in general, faster than
On Thu, Dec 16, 2021 at 06:35:23PM -0500, John Snow wrote:
> On Thu, Dec 16, 2021 at 5:48 AM Daniel P. Berrangé
> wrote:
>
> > On Wed, Dec 15, 2021 at 04:06:27PM -0500, John Snow wrote:
> > > Signed-off-by: John Snow
> > > ---
> > > Makefile | 32
> > > 1 file c
Temporarily modifying the SIGBUS handler is really nasty, as we might be
unlucky and receive an MCE SIGBUS while having our handler registered.
Unfortunately, there is no way around messing with SIGBUS when
MADV_POPULATE_WRITE is not applicable or not around.
Let's forward SIGBUS that don't belong
Hi
On Fri, Dec 17, 2021 at 5:31 PM Philippe Mathieu-Daudé
wrote:
> On 10/9/21 23:08, marcandre.lur...@redhat.com wrote:
> > From: Marc-André Lureau
> >
> > Further meson configuration tests are to be added based on the glib
> > version. Also correct the version reporting in the config log.
> >
Add a mutex to protect the SIGBUS case, as we cannot mess concurrently
with the sigbus handler and we have to manage the global variable
sigbus_memset_context. The MADV_POPULATE_WRITE path can run
concurrently.
Note that page_mutex and page_cond are shared between concurrent
invocations, which sho
Based-on: <20211217134039.29670-1-da...@redhat.com>
Support preallocation of memory to make virtio-mem safe to use with
scarce memory resources such as hugetlb. Before acknowledging a plug
request from the guest, we'll try preallcoating memory. If that fails,
we'll fail the request gracefully and
On Thu, Dec 16, 2021 at 02:11:37PM +, Alex Bennée wrote:
>
> Philippe Mathieu-Daudé writes:
>
> > When experimenting raising GLIB_VERSION_MIN_REQUIRED to 2.68
> > (Fedora 34 provides GLib 2.68.1) we get:
> >
> > hw/virtio/virtio-crypto.c:245:24: error: 'g_memdup' is deprecated: Use
> > 'g
On 12/17/21 04:08, Alexander Bulekov wrote:
> Here's my shot at fixing dma-reentracy issues. This patch adds a flag to
> the DeviceState, which is set/checked when we call an accessor
> associated with the device's IO MRs.
Your approach is exactly what Gerd suggested:
https://www.mail-archive.com/
Let's limit the number of threads to something sane, especially that
- We don't have more threads than the number of pages we have
- We don't have threads that initialize small (< 64 MiB) memory
Reviewed-by: Pankaj Gupta
Reviewed-by: Dr. David Alan Gilbert
Reviewed-by: Daniel P. Berrangé
Review
On 12/17/21 09:53, Paolo Bonzini wrote:
> Meson is planning to change the default of the "check" argument to
> run_command (from false to true). Be explicit and include it in
> all invocations.
>
> Signed-off-by: Paolo Bonzini
> ---
> docs/meson.build | 6 +++---
> 1 file changed, 3 insertions(
Let's minimize the number of global variables to prepare for
os_mem_prealloc() getting called concurrently and make the code a bit
easier to read.
The only consumer that really needs a global variable is the sigbus
handler, which will require protection via a mutex in the future either way
as we c
For scarce memory resources, such as hugetlb, we want to be able to
prealloc such memory resources in order to not crash later on access. On
simple user errors we could otherwise easily run out of memory resources
an crash the VM -- pretty much undesired.
For ordinary memory devices, such as DIMMs
On Thu, 16 Dec 2021 14:16:57 +0100
Nico Boehr wrote:
> Previously, we required bits 5, 6 and 7 to be zero (0x07 == 0b111). But,
> as per the principles of operation, bit 5 is ignored in MSCH and bits 0,
> 1, 6 and 7 need to be zero.
On a second thought, don't we have to make sure then that bit 5
On 211217 0627, Qiuhao Li wrote:
> Thanks Alex. It seems this patch sets and checks if the destination device is
> busy. But how about the data transfers not triggered directly by PMIO/MMIO
> handlers? For example:
>
> 1. Device A Timer's callback -> Device A MMIO handler
> 2. Device A BH's call
On Fri, Dec 17, 2021 at 5:35 PM Philippe Mathieu-Daudé
wrote:
>
> On 10/9/21 23:08, marcandre.lur...@redhat.com wrote:
> > From: Marc-André Lureau
> >
> > Export the default consoles over the D-Bus chardev.
> >
> > Signed-off-by: Marc-André Lureau
> > ---
> > ui/dbus.c | 53
On Fri, Dec 17 2021, Daniel P. Berrangé wrote:
> On Fri, Dec 17, 2021 at 09:13:55AM +0100, Cornelia Huck wrote:
>> On Wed, Dec 08 2021, Cornelia Huck wrote:
>>
>> > Add 7.0 machine types for arm/i440fx/q35/s390x/spapr.
>> >
>> > Signed-off-by: Cornelia Huck
>> > ---
>> > hw/arm/virt.c
On 211217 1458, Philippe Mathieu-Daudé wrote:
> On 12/17/21 04:08, Alexander Bulekov wrote:
> > Here's my shot at fixing dma-reentracy issues. This patch adds a flag to
> > the DeviceState, which is set/checked when we call an accessor
> > associated with the device's IO MRs.
>
> Your approach is
On 17.12.21 13:29, Hanna Reitz wrote:
On 24.11.21 07:44, Emanuele Giuseppe Esposito wrote:
block_crypto_amend_options_generic_luks uses the block layer
permission API, therefore it should be called with the BQL held.
However, the same function is being called ib two BlockDriver
s/ ib / by /
On 12/17/21 14:40, Marc-André Lureau wrote:
> > diff --git a/meson.build b/meson.build
> > index e1cddf5139..9494590aa2 100644
> > --- a/meson.build
> > +++ b/meson.build
> > @@ -311,14 +311,16 @@ endif
> > add_project_arguments(config_host['GLIB_CFLAGS'].split(),
> >
From: Marc-André Lureau
The following changes since commit 29eb5c2c86f935b0e9700fad2ecfe8a32b011d57:
Merge tag 'for-upstream' of https://gitlab.com/bonzini/qemu into staging
(2021-12-16 08:39:20 -0800)
are available in the Git repository at:
https://gitlab.com/marcandre.lureau/qemu.git ta
From: Marc-André Lureau
Signed-off-by: Marc-André Lureau
Acked-by: Gerd Hoffmann
---
meson.build | 5 +
ui/vdagent.c | 8
2 files changed, 13 insertions(+)
diff --git a/meson.build b/meson.build
index ae67ca28ab38..58718691d7fc 100644
--- a/meson.build
+++ b/meson.build
@@ -1487
Here's my shot at fixing dma-reentracy issues. This patch adds a flag to
the DeviceState, which is set/checked when we call an accessor
associated with the device's IO MRs.
The problem, in short, as I understand it: For the vast majority of
cases, we want to prevent a device from accessing it's ow
From: Marc-André Lureau
Use a QemuClipboardNotify union type for extendable clipboard events.
Signed-off-by: Marc-André Lureau
Acked-by: Gerd Hoffmann
---
include/ui/clipboard.h | 32 ++--
ui/clipboard.c | 12
ui/gtk-clipboard.c | 20 ++
Add a flag to the DeviceState, which is set/checked when we call an
accessor associated with the device's IO MRs. This should prevent a
device's MMIO handler from using DMA apis to access it's own MMIO
regions.
Signed-off-by: Alexander Bulekov
---
include/hw/qdev-core.h | 1 +
softmmu/memory.c
DMA reentrancy problems can occur in BHs:
dev_mmio->schedule_bh
dev_bh->dma_write->dev_mmio
This patch attempts to address this scenario by marking the device as
engaged_in_io, when it calls into PCI and SGList DMA APIs.
Signed-off-by: Alexander Bulekov
---
include/hw/pci/pci.h | 6 +-
sof
From: Marc-André Lureau
The Spice agent implements a simple serial mechanism to avoid clipboard
races between client & guest. See:
https://gitlab.freedesktop.org/spice/spice-protocol/-/commit/045a6978d6dbbf7046affc5c321fa8177c8cce56
Signed-off-by: Marc-André Lureau
Acked-by: Gerd Hoffmann
---
From: Marc-André Lureau
Signed-off-by: Marc-André Lureau
Acked-by: Gerd Hoffmann
---
include/ui/clipboard.h | 9 +
ui/clipboard.c | 7 +++
ui/gtk-clipboard.c | 3 +++
ui/vdagent.c | 12
ui/vnc-clipboard.c | 3 +++
ui/cocoa.m |
From: Marc-André Lureau
Currently, virgl initialization error is silent. Make it verbose instead.
(this is likely going to bug later on, as the device isn't fully
initialized)
Signed-off-by: Marc-André Lureau
Acked-by: Gerd Hoffmann
Reviewed-by: Philippe Mathieu-Daudé
---
hw/display/virtio-
From: Marc-André Lureau
Track multiple callers blocking requests.
Signed-off-by: Marc-André Lureau
Acked-by: Gerd Hoffmann
---
ui/console.c | 17 +
1 file changed, 13 insertions(+), 4 deletions(-)
diff --git a/ui/console.c b/ui/console.c
index 87f897e46dc4..39f7b66baf54 10064
1 - 100 of 225 matches
Mail list logo
