From: Laurent Vivier
This allows to cover the virtio tests with a 32bit big-endian
virtio-mmio machine.
Signed-off-by: Laurent Vivier
Reviewed-by: Philippe Mathieu-Daudé
Reviewed-by: Cornelia Huck
Message-Id: <20210319202335.2397060-6-laur...@vivier.eu>
Signed-off-by: Alex Bennée
---
tests/
Recent QEMU versions now sometimes exit cleanly with an error message
that a bus is not available for a specified device. Don't flag those
as an error in the device-crash-test script.
Signed-off-by: Thomas Huth
---
scripts/device-crash-test | 1 +
1 file changed, 1 insertion(+)
diff --git a/scr
Eric Blake writes:
> On 3/23/21 4:40 AM, Markus Armbruster wrote:
>> Naming rules differ for the various kinds of names. To prepare
>> enforcing them, define functions to check them: check_name_upper(),
>> check_name_lower(), and check_name_camel(). For now, these merely
>> wrap around check_na
You need a kernel with a the following fix for POWER9:
commit 25edcc50d76c834479d11fcc7de46f3da4d95121
Author: Fabiano Rosas
Date: Thu Feb 4 17:05:17 2021 -0300
KVM: PPC: Book3S HV: Save and restore FSCR in the P9 path
The Facility Status and Control Register is a privileged SPR t
In case anyone is interested in fixing those, there are two regressions with
qemu-system-ppc64 in the current master branch:
$ ./qemu-system-ppc64 -M ppce500 -device macio-oldworld
qemu-system-ppc64: ../../devel/qemu/softmmu/memory.c:2443:
memory_region_add_subregion_common: Assertion `!subr
On Tue, 23 Mar 2021 at 01:59, Alistair Francis wrote:
>
> The following changes since commit c95bd5ff1660883d15ad6e0005e4c8571604f51a:
>
> Merge remote-tracking branch 'remotes/philmd/tags/mips-fixes-20210322' into
> staging (2021-03-22 14:26:13 +)
>
> are available in the Git repository at
On 23/03/21 16:51, Philippe Mathieu-Daudé wrote:
install_edk2_blobs = install_edk2_blobs or target in edk2_targets
install_blobs_alpha = install_blobs_alpha or target in ['alpha-softmmu']
+install_blobs_arm = install_blobs_hppa or target in ['arm-softmmu',
'aarch64-softmmu']
On 3/22/21 10:12 PM, David Gibson wrote:
On Fri, Mar 12, 2021 at 05:07:36PM -0300, Daniel Henrique Barboza wrote:
Hi,
This series adds 2 new QAPI events, DEVICE_NOT_DELETED and
DEVICE_UNPLUG_ERROR. They were (and are still being) discussed in [1].
Patches 1 and 3 are independent of the ppc
Some packaged versions of Sphinx (fedora33/alpine so far) have issues
with the annotated C code that kernel-doc spits out. Without knowing
about things like QEMU_PLUGIN_EXPORT it chokes trying to understand
the code. Evidently this is a problem for the kernel as well as the
long stream of regex sub
I confused myself wandering if this had been merged by looking at the
help output. It seems fuse_opt doesn't automagically add to help
output so lets do it now.
Signed-off-by: Alex Bennée
Reviewed-by: Connor Kuehl
Reviewed-by: Stefan Hajnoczi
Updates: f6698f2b03 ("tools/virtiofsd: add support f
Eric Blake writes:
> On 3/23/21 4:40 AM, Markus Armbruster wrote:
>> Rename pragma returns-whitelist to command-returns-exceptions, and
>> name-case-whitelist to member-name-case-exceptions.
>>
>> Signed-off-by: Markus Armbruster
>> ---
> Doesn't mention anything about the motivation for this i
We have kerneldoc tags for the headers so we might as well extract
them into our developer documentation whilst we are at it.
Signed-off-by: Alex Bennée
Reviewed-by: Aaron Lindsay
Message-Id: <20210312172821.31647-4-alex.ben...@linaro.org>
Message-Id: <20210320133706.21475-3-alex.ben...@linaro.o
This aims to provide a bit more guidance for those who take on one of
our "clean up memory allocation" bite-sized tasks.
Signed-off-by: Alex Bennée
Reviewed-by: Richard Henderson
Message-Id: <20210320133706.21475-4-alex.ben...@linaro.org>
---
docs/devel/style.rst | 46 ++
23.03.2021 16:06, Paolo Bonzini wrote:
Right now there is no easy way for "check" to print a reproducer command.
Because such a reproducer command line would be huge, we can instead teach
check to start a command of our choice. This can be for example a Python
unit test with arguments to only ru
This change was made by a bot.
** Changed in: linux (Ubuntu)
Status: New => Confirmed
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1920784
Title:
qemu-system-ppc64le fails with kvm acceler
It may be arm-compat-semihosting but more than one architecture uses
it so lets move the tests into the multiarch area. We gate it on the
feature and split the semicall.h header between the arches.
Also clean-up a bit of the Makefile messing about to one common set of
runners.
Signed-off-by: Alex
23.03.2021 16:06, Paolo Bonzini wrote:
Due to a typo, in this case the SOCK_DIR was not being created.
Signed-off-by: Paolo Bonzini
---
tests/qemu-iotests/testenv.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/qemu-iotests/testenv.py b/tests/qemu-iotests/testenv
From: Thomas Huth
FreeBSD version 12.1 is out of service now, and the task in the
Cirrus-CI is failing. Update to 12.2 to get it working again.
Unfortunately, there is a bug in libtasn1 that triggers with the
new version of Clang that is used there (see this thread for details:
https://lists.gnu.
23.03.2021 20:22, Paolo Bonzini wrote:
On 23/03/21 18:11, Vladimir Sementsov-Ogievskiy wrote:
If you have positional arguments that must begin with - and don’t look like
negative numbers, you can insert the pseudo-argument '--' which tells
parse_args() that everything after that is a positiona
From: Eric Blake
Our tests were not validating the return value in all cases, nor was
it guaranteeing our documented claim that 'res' is unchanged on error.
For that matter, it wasn't as thorough as the existing tests for
qemu_strtoi() and friends for proving that endptr and res are sanely
set.
From: Thomas Huth
Our gitlab-ci got quite slow in the past weeks, due to the immense amount
of jobs that we have, so we should try to reduce the number of jobs.
There is no real good reason for having separate jobs just to test the
trace backends, we can do this just fine in other jobs, too.
Sig
Query the SYS_HEAPINFO semicall and do some basic verification of the
information via libc calls.
Signed-off-by: Alex Bennée
Message-Id: <20210320133706.21475-10-alex.ben...@linaro.org>
---
v2
- expand test as suggested by Richard
---
.../multiarch/arm-compat-semi/semihosting.c | 55 +++
Remove dependency on qiochannel by removing yank_generic_iochannel and
letting migration and chardev use their own yank function for
iochannel.
Signed-off-by: Lukas Straub
---
MAINTAINERS | 1 +
chardev/char-socket.c | 21 ++---
include/qemu/yank.h
23.03.2021 18:29, Paolo Bonzini wrote:
On 23/03/21 15:34, Vladimir Sementsov-Ogievskiy wrote:
+ def __init__(self, *args, **kwargs):
+ super().__init__(stream=ReproducibleTextTestRunner.output, *args,
**kwargs)
over-79 line (PEP8)
Ok, thanks.
+
+def execute_unittest(argv, debug=
Thanks for the feedback, it helped us improve the implementation quite a bit.
> > +| rvalue QMARK rvalue COLON rvalue
> > +{
> > +@1.last_column = @5.last_column;
> > +bool is_64bit = ($3.bit_width == 64) || ($5.bit_width == 64);
> > +int bit_width = (is_64bit) ? 64 : 32;
> > +if (
Yank now only depends on util and can be always linked in. Also remove
the stubs as they are not needed anymore.
Signed-off-by: Lukas Straub
---
MAINTAINERS | 1 -
stubs/meson.build | 1 -
stubs/yank.c | 23 ---
util/meson.build | 2 +-
4 files changed, 1 inser
From: Eric Blake
Mingw recognizes that "0x" has value 0 without setting errno, but
fails to advance endptr to the trailing garbage 'x'. This in turn
showed up in our recent testsuite additions for qemu_strtosz (commit
1657ba44b4 utils: Enhance testsuite for do_strtosz()); adjust our
remaining te
These builds are running very close to the default build limit and as
they are already pared down the only other option is to extend the
timeout a little to give some breathing room.
Signed-off-by: Alex Bennée
Reviewed-by: Thomas Huth
Reviewed-by: Willian Rampazzo
Reviewed-by: Philippe Mathieu-
Hi,
I've added a testing patch to extend the CFI times, Laurent's series
to fix the iotest regressions currently keeping the CI from going
green and an optimisation to the --enable-docs build to try and start
bringing the total CI time down a bit. rc0 is being tagged today so
I'll probably roll a
From: Laurent Vivier
Similarly to 5f629d943cb0 ("s390x: fix s390 virtio aliases"),
define the virtio aliases.
This allows to start machines with virtio devices without
knowledge of the implementation type.
For instance, we can use "-device virtio-scsi" on
m68k, s390x or PC, and the device will
From: Laurent Vivier
Commit f1d5516ab583 introduces a test in some iotests to check if
the machine is a s390-ccw-virtio and to select virtio-*-ccw rather
than virtio-*-pci.
We don't need that because QEMU already provides aliases to use the correct
virtio interface according to the machine type.
As per the spec:
the PARAMETER REGISTER contains the address of a pointer to a
four-field data block.
So we need to follow arg0 and place the results of SYS_HEAPINFO there.
Fixes: 3c37cfe0b1 ("semihosting: Change internal common-semi interfaces to use
CPUState *")
Signed-off-by: Alex Bennée
On 3/12/21 2:07 PM, Daniel Henrique Barboza wrote:
> This new event informs QAPI listeners that a previously issued
> 'device_del' command failed to delete the device from the machine.
>
> Note that no assertion can be made about the failure reason. The goal of
> this event is to inform management
On 23/03/21 17:43, Vladimir Sementsov-Ogievskiy wrote:
Interesting that REMAINDER documentation disappeared from latest (3.9)
python documentation https://docs.python.org/3.9/library/argparse.html ,
but exists here https://docs.python.org/3.8/library/argparse.html (and
no mark of deprecation
From: Thomas Huth
The test for 128-bit atomics is causing trouble with FreeBSD 12.2 and
--enable-werror:
cc -Werror -fPIE -DPIE -std=gnu99 -Wall -m64 -mcx16 -D_GNU_SOURCE
-D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -Wstrict-prototypes
-Wredundant-decls -Wundef -Wwrite-strings -Wmissing-prototy
On 23/03/2021 18.52, Lukas Straub wrote:
Yank now only depends on util and can be always linked in. Also remove
the stubs as they are not needed anymore.
Signed-off-by: Lukas Straub
---
MAINTAINERS | 1 -
stubs/meson.build | 1 -
stubs/yank.c | 23 ---
util
On 3/22/21 5:54 PM, David Gibson wrote:
On Mon, Mar 22, 2021 at 10:55:46AM -0600, Richard Henderson wrote:
On 3/21/21 9:52 PM, David Gibson wrote:
+/*
+ * Bits for env->hflags.
+ *
+ * Most of these bits overlap with corresponding bits in MSR,
+ * but some come from other sources. Be cautious
From: Laurent Vivier
This is used to define virtio-*-pci and virtio-*-ccw aliases
rather than substracting the CCW architecture from all the others.
Signed-off-by: Laurent Vivier
Reviewed-by: Cornelia Huck
Message-Id: <20210319202335.2397060-2-laur...@vivier.eu>
Signed-off-by: Alex Bennée
---
On 3/23/21 11:44 AM, Markus Armbruster wrote:
John Snow writes:
On 3/23/21 5:40 AM, Markus Armbruster wrote:
Member name 'u' and names starting with 'has-' or 'has_' are reserved
for the generator. check_type() enforces this, covered by tests
reserved-member-u and reserved-member-has.
These
On Tue, 23 Mar 2021 10:13:58 -0400
"Michael S. Tsirkin" wrote:
CCing Philippe,
maybe he has an idea how to fix it from mips side.
> On Tue, Mar 23, 2021 at 10:50:05AM +, Peter Maydell wrote:
> > On Mon, 22 Mar 2021 at 22:56, Michael S. Tsirkin wrote:
> > > On Mon, Mar 22, 2021 at 06:46:0
Right now there is no easy way for "check" to print a reproducer command.
Because such a reproducer command line would be huge, we can instead teach
check to start a command of our choice. This can be for example a Python
unit test with arguments to only run a specific subtest.
Move the trailing
On 3/22/21 10:03 PM, David Gibson wrote:
On Fri, Mar 19, 2021 at 03:34:52PM -0300, Daniel Henrique Barboza wrote:
Kernel commit 4bce545903fa ("powerpc/topology: Update
topology_core_cpumask") cause a regression in the pseries machine when
defining certain SMP topologies [1]. The reasoning beh
23.03.2021 20:00, Paolo Bonzini wrote:
On 23/03/21 17:43, Vladimir Sementsov-Ogievskiy wrote:
Interesting that REMAINDER documentation disappeared from latest (3.9) python
documentation https://docs.python.org/3.9/library/argparse.html , but exists
here https://docs.python.org/3.8/library/arg
On 23/03/21 17:56, Vladimir Sementsov-Ogievskiy wrote:
hmm, just use a separate call of unittest.main, or honestly define a
varaible as Union[] or just Any ?
All the ugliness goes away if the implementation is done properly. :)
For me normal try-finally is a lot more clean than calling atexit(
+Mark
On 3/23/21 5:48 PM, Thomas Huth wrote:
>
> In case anyone is interested in fixing those, there are two regressions
> with qemu-system-ppc64 in the current master branch:
>
> $ ./qemu-system-ppc64 -M ppce500 -device macio-oldworld
> qemu-system-ppc64: ../../devel/qemu/softmmu/memory.c:2443:
On 23.03.2021 17:54, Peter Xu wrote:
On Tue, Mar 23, 2021 at 10:51:57AM +0300, Andrey Gruzdev wrote:
On 22.03.2021 23:17, Peter Xu wrote:
On Fri, Mar 19, 2021 at 05:52:47PM +0300, Andrey Gruzdev wrote:
Added missing qemu_fflush() on buffer file holding precopy device state.
Increased initial Q
On 23/03/21 18:11, Vladimir Sementsov-Ogievskiy wrote:
If you have positional arguments that must begin with - and don’t look
like negative numbers, you can insert the pseudo-argument '--' which
tells parse_args() that everything after that is a positional argument:
So, as I understand argpars
Without this patch, the following patch will triger clan runtime
sanitizer warnings as follows. This patch proactively works around it.
I let v582c686.c maintainer address a correct fix as I'm not sure
about fuloong2e device model.
> MALLOC_PERTURB_=${MALLOC_PERTURB_:-$(( ${RANDOM:-0} % 255 + 1))}
23.03.2021 20:04, Paolo Bonzini wrote:
On 23/03/21 17:56, Vladimir Sementsov-Ogievskiy wrote:
hmm, just use a separate call of unittest.main, or honestly define a varaible
as Union[] or just Any ?
All the ugliness goes away if the implementation is done properly. :)
For me normal try-finall
If a device model
(a) doesn't set the value to a correct interrupt number and then
(b) triggers an interrupt for itself,
it's device model bug. Add assert on interrupt pin number to catch
this kind of bug more obviously.
Suggested-by: Peter Maydell
Signed-off-by: Isaku Yamahata
---
hw/pci/pci.c
Reinitialize ACPI PM device on reset to get preper device state.
Oppertunistically add assert on PCI interrupt pin logic to detect device model
issues and workaround for found one.
Changes from v1:
- bug work around hw/isa/vt82c686.c
- add assert in hw/pci/pci.c when setting/raising PCI interrupt
Commit 6be8cf56bc8b made sure that SCI is enabled in PM1.CNT
on reset in acpi_only mode by modifying acpi_pm1_cnt_reset() and
that worked for q35 as expected.
The function was introduced by commit
eaba51c573a (acpi, acpi_piix, vt82c686: factor out PM1_CNT logic)
that forgot to actually call it a
On 23/03/21 18:27, Vladimir Sementsov-Ogievskiy wrote:
Sounds good :) We'll see dots appearing dynamically during test run, yes?
Yes, exactly! :)
Paolo
** Also affects: linux (Ubuntu)
Importance: Undecided
Status: New
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1920784
Title:
qemu-system-ppc64le fails with kvm acceleration
Status in Q
On Tue, Mar 23, 2021 at 05:54:47PM +0100, Igor Mammedov wrote:
> Let me hijack this thread for beyond this case scope.
>
> I agree that for this particular bug we've done all we could, but
> there is broader issue to discuss here.
>
> We have machine versions to deal with hw compatibility issues
Add tests for yank with the chardev-change case.
Signed-off-by: Lukas Straub
---
MAINTAINERS| 1 +
tests/unit/meson.build | 3 +-
tests/unit/test-yank.c | 201 +
3 files changed, 204 insertions(+), 1 deletion(-)
create mode 100644 tests/un
Hello Everyone,
These patches remove yank's dependency on qiochannel and always link it in.
Please Review.
Regards,
Lukas Straub
Lukas Straub (2):
yank: Remove dependency on qiochannel
yank: Always link full yank code
MAINTAINERS | 2 +-
chardev/char-socket.c | 21
Extract post_load_update_msr to share between cpu_load_old
and cpu_post_load in updating the msr.
Suggested-by: Cédric Le Goater
Signed-off-by: Richard Henderson
---
target/ppc/machine.c | 30 ++
1 file changed, 14 insertions(+), 16 deletions(-)
diff --git a/target/
On Tue, 23 Mar 2021 at 17:26, Isaku Yamahata wrote:
>
> If a device model
> (a) doesn't set the value to a correct interrupt number and then
> (b) triggers an interrupt for itself,
> it's device model bug. Add assert on interrupt pin number to catch
> this kind of bug more obviously.
>
> Suggested
It will be stored in tb->flags, which is also uint32_t,
so let's use the correct size.
Reviewed-by: Cédric Le Goater
Reviewed-by: David Gibson
Signed-off-by: Richard Henderson
---
target/ppc/cpu.h | 4 ++--
target/ppc/misc_helper.c | 2 +-
target/ppc/translate.c | 2 +-
3 files chang
Move object_property_try_add_child out of chardev_new into it's
callers. This is a preparation for the next patches to fix yank
with the chardev-change case.
Signed-off-by: Lukas Straub
Reviewed-by: Marc-André Lureau
Tested-by: Li Zhang
---
chardev/char.c | 42 -
On 23/03/2021 18.52, Lukas Straub wrote:
Remove dependency on qiochannel by removing yank_generic_iochannel and
letting migration and chardev use their own yank function for
iochannel.
Signed-off-by: Lukas Straub
---
MAINTAINERS | 1 +
chardev/char-socket.c | 21 ++
Because this bit was not in hflags, the privilege check
for tlb instructions was essentially random.
Recompute hflags when storing to LPCR.
Reviewed-by: David Gibson
Signed-off-by: Richard Henderson
---
target/ppc/cpu.h | 1 +
target/ppc/helper_regs.c | 3 +++
target/ppc/mmu-hash64.c |
Hello Everyone,
These patches increase test coverage for yank, add tests and fix bugs and
crashes in yank in combination with chardev-change.
Please Review.
Regards,
Lukas Straub
Changes:
-v3:
-Base on
([PATCH 0/2] yank: Always link full yank code)
-Drop patch 1 (tests: Use the normal yank c
Always pass the id to chardev_new, since it is needed to register
the yank instance for the chardev. Also, after checking that
nothing calls chardev_new with id=NULL, assert() that id!=NULL.
This fixes a crash when using chardev-change to change a chardev
to chardev-socket, which attempts to regis
When changing from chardev-socket (which supports yank) to
chardev-socket again, it fails, because the new chardev attempts
to register a new yank instance. This in turn fails, as there
still is the yank instance from the current chardev. Also,
the old chardev shouldn't unregister the yank instance
On 3/23/21 3:00 PM, Eric Blake wrote:
On 3/12/21 2:07 PM, Daniel Henrique Barboza wrote:
This new event informs QAPI listeners that a previously issued
'device_del' command failed to delete the device from the machine.
Note that no assertion can be made about the failure reason. The goal of
Claudio Fontana writes:
> Here a new version of the series that enables kvm-only builds.
>
> The goal here is to enable the KVM-only build, but there is
> some additional cleanup too.
First the good news. With --disable-tcg on my AArch64 system I can
successfully run:
make vm-boot-ssh-ubunt
This series adds a few usability improvements to qemu-iotests, in
particular:
- arguments can be passed to Python unittests scripts, for example
to run only a subset of the test cases (patches 1-2)
- it is possible to do "./check -- ../../../tests/qemu-iotests/055 args..."
and specify arbitra
Python test scripts that use unittest consist of multiple tests.
unittest.main allows selecting which tests to run, but currently this
is not possible because the iotests wrapper ignores sys.argv.
unittest.main command line options also allow the user to pick the
desired options for verbosity, fai
In the next patch, "check" will learn how to execute a test script without
going through TestRunner. To enable this, keep only the text output
and subprocess handling in the TestRunner; move into TestEnv the logic
to prepare for running a subprocess.
Reviewed-by: Vladimir Sementsov-Ogievskiy
Sig
23.03.2021 21:19, Paolo Bonzini wrote:
Python test scripts that use unittest consist of multiple tests.
unittest.main allows selecting which tests to run, but currently this
is not possible because the iotests wrapper ignores sys.argv.
unittest.main command line options also allow the user to pi
Due to a typo, in this case the SOCK_DIR was not being created.
Reviewed-by: Vladimir Sementsov-Ogievskiy
Signed-off-by: Paolo Bonzini
---
tests/qemu-iotests/testenv.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/qemu-iotests/testenv.py b/tests/qemu-iotests/testenv
Patchew URL:
https://patchew.org/QEMU/20210323165308.15244-1-alex.ben...@linaro.org/
Hi,
This series seems to have some coding style problems. See output below for
more information:
Type: series
Message-id: 20210323165308.15244-1-alex.ben...@linaro.org
Subject: [PATCH for 6.0 v2 00/22] fixes
On 3/23/21 5:15 PM, Richard Henderson wrote:
> exec-vary.c is about variable page size handling,
> rename it page-vary.c. Currently this file is target
> specific (built once for each target), comment this.
>
> Signed-off-by: Philippe Mathieu-Daudé
> Message-Id: <20210322112427.4045204-2-f4...@am
Instead of buffering the test output into a StringIO, patch it on
the fly by wrapping sys.stdout's write method. This can be
done unconditionally, even if using -d, which makes execute_unittest
a bit simpler.
Signed-off-by: Paolo Bonzini
---
tests/qemu-iotests/iotests.py | 68 ++
On Tue, Mar 23, 2021 at 06:52:19PM +0100, Lukas Straub wrote:
> Hello Everyone,
> These patches remove yank's dependency on qiochannel and always link it in.
> Please Review.
It would be useful if the cover letter or commit messages explained
to potential reviewers why this is being changed... The
Verify that hflags was updated correctly whenever we change
cpu state that is used by hflags.
Signed-off-by: Richard Henderson
---
target/ppc/cpu.h | 5 +
target/ppc/helper_regs.c | 29 +++--
2 files changed, 32 insertions(+), 2 deletions(-)
diff --git a/tar
On 3/23/21 5:16 PM, Richard Henderson wrote:
> In the next commit we will extract the generic code out of
> page-vary.c, only keeping the target specific code. Both
> files will use the same TargetPageBits structure, so make
> its declaration in a shared header.
>
> As the common header can not us
When building with --enable-sanitizers we get:
Direct leak of 32 byte(s) in 2 object(s) allocated from:
#0 0x5618479ec7cf in malloc (qemu-system-aarch64+0x233b7cf)
#1 0x7f675745f958 in g_malloc (/lib64/libglib-2.0.so.0+0x58958)
#2 0x561847f02ca2 in usb_packet_init hw/usb/core.c
Hi Isaku,
On 3/23/21 6:24 PM, Isaku Yamahata wrote:
> Without this patch, the following patch will triger clan runtime
> sanitizer warnings as follows. This patch proactively works around it.
> I let v582c686.c maintainer address a correct fix as I'm not sure
> about fuloong2e device model.
>
>>
When building with --enable-sanitizers we get:
Direct leak of 16 byte(s) in 1 object(s) allocated from:
#0 0x5618479ec7cf in malloc (qemu-system-aarch64+0x233b7cf)
#1 0x7f675745f958 in g_malloc (/lib64/libglib-2.0.so.0+0x58958)
#2 0x561847c2dcc9 in xlnx_dp_init hw/display/xlnx_
Copying flags directly from msr has drawbacks: (1) msr bits
mean different things per cpu, (2) msr has 64 bits on 64 cpus
while tb->flags has only 32 bits.
Create a enum to define these bits. Document the origin of each bit
and validate those bits that must match MSR. This fixes the
truncation o
23.03.2021 21:19, Paolo Bonzini wrote:
Right now there is no easy way for "check" to print a reproducer command.
Because such a reproducer command line would be huge, we can instead teach
check to start a command of our choice. This can be for example a Python
unit test with arguments to only ru
Hi Alex,
thanks a lot for looking at this,
On 3/23/21 7:35 PM, Alex Bennée wrote:
>
> Claudio Fontana writes:
>
>> Here a new version of the series that enables kvm-only builds.
>>
>> The goal here is to enable the KVM-only build, but there is
>> some additional cleanup too.
>
> First the goo
23.03.2021 21:19, Paolo Bonzini wrote:
Python test scripts that use unittest consist of multiple tests.
unittest.main allows selecting which tests to run, but currently this
is not possible because the iotests wrapper ignores sys.argv.
unittest.main command line options also allow the user to pi
Perform the test against FSCR_SCV at runtime, in the helper.
This means we can remove the incorrect set against SCV in
ppc_tr_init_disas_context and do not need to add an HFLAGS bit.
Signed-off-by: Richard Henderson
---
target/ppc/helper.h | 1 +
target/ppc/excp_helper.c | 9 +
t
On Tue, Mar 23, 2021 at 08:21:43PM +0300, Andrey Gruzdev wrote:
> > For the long term I think we'd better have a helper:
> >
> > qemu_put_qio_channel_buffer(QEMUFile *file, QIOChannelBuffer *bioc)
> >
> > So as to hide this flush operation, which is tricky. We'll have two users so
> > fa
We weren't recording MSR_GS in hflags, which means that BookE
memory accesses were essentially random vs Guest State.
Instead of adding this bit directly, record the completed mmu
indexes instead. This makes it obvious that we are recording
exactly the information that we need.
This also means t
Claudio Fontana writes:
> move init, realizefn and reset code into it.
w.r.t my testing this is fingered by bisect for causing:
Running test qtest-aarch64/migration-test
ERROR qtest-aarch64/migration-test - Bail out!
ERROR:../../target/arm/tcg/tcg-cpu.c:233:tcg_arm_init_accel_cpu: assert
Changes for v5:
* Rebase on david's ppc-for-6.0 branch (fcc83aa7359d)
* Extract post_load_update_msr (clg)
* Validate MSR_FOO == HFLAGS_FOO when necessary (david)
Changes for v4:
* Use hregs_recompute_hflags for hw/ppc/ reset.
-- Incorporate Cedric's feedback.
Changes for v3:
* Fixes for
In save_user_regs, there are two bugs where we OR in a bit number
instead of the bit, clobbering the low bits of MSR. However:
The MSR_VR and MSR_SPE bits control the availability of the insns.
If the bits were not already set in MSR, then any attempt to access
those registers would result in SIG
On 3/22/21 8:15 PM, Claudio Fontana wrote:
> Hi, I am starting a cleanup series for s390x,
>
> with the goal of doing similar splits of KVM vs TCG,
One annoying thing is that it appears that the cpu accel model does not offer
much in terms of refactoring
and code simplification opportunities fo
Because these bits were not in hflags, the code generated
for single-stepping on BookE was essentially random.
Recompute hflags when storing to dbcr0.
Reviewed-by: David Gibson
Signed-off-by: Richard Henderson
---
target/ppc/helper_regs.c | 24 +---
target/ppc/misc_helper.c
On Tue, Mar 23, 2021 at 04:04:11PM +0100, Thomas Lamprecht wrote:
> On 23.03.21 15:55, Vitaly Cheptsov wrote:
> >> 23 марта 2021 г., в 17:48, Michael S. Tsirkin написал(а):
> >>
> >> The issue is with people who installed a VM using 5.1 qemu,
> >> migrated to 5.2, booted there and set a config on
23.03.2021 21:19, Paolo Bonzini wrote:
Instead of buffering the test output into a StringIO, patch it on
the fly by wrapping sys.stdout's write method. This can be
done unconditionally, even if using -d, which makes execute_unittest
a bit simpler.
Signed-off-by: Paolo Bonzini
Reviewed-by: Vl
Nothing within the translator -- or anywhere else for that
matter -- checks MSR_SA or MSR_AP on the 602. This may be
a mistake. However, for the moment, we need not record these
bits in hflags.
This allows us to simplify HFLAGS_VSX computation by moving
it to overlap with MSR_VSX.
Reviewed-by:
* huang...@chinatelecom.cn (huang...@chinatelecom.cn) wrote:
> From: Hyman Huang(黄勇)
>
> when execute the following test command:
> $ ./guestperf-batch.py --auto-converge \
> --auto-converge-step {percent} ...
> test aborts and error message be throwed as the following:
> "Parameter 'x-cpu-th
On Tue, Mar 23, 2021 at 05:40:36PM +, Daniel P. Berrangé wrote:
> On Tue, Mar 23, 2021 at 05:54:47PM +0100, Igor Mammedov wrote:
> > Let me hijack this thread for beyond this case scope.
> >
> > I agree that for this particular bug we've done all we could, but
> > there is broader issue to dis
On Tue, Mar 23, 2021 at 10:24:31AM -0700, Isaku Yamahata wrote:
> Commit 6be8cf56bc8b made sure that SCI is enabled in PM1.CNT
> on reset in acpi_only mode by modifying acpi_pm1_cnt_reset() and
> that worked for q35 as expected.
>
> The function was introduced by commit
> eaba51c573a (acpi, acpi
301 - 400 of 562 matches
Mail list logo
