On Mon, Feb 21, 2022 at 01:25:53PM +, David Edmondson wrote:
> On Wednesday, 2022-02-16 at 22:04:32 -08, Yang Zhong wrote:
>
> > From: Jing Liu
> >
> > When dynamic xfeatures (e.g. AMX) are used by the guest, the xsave
> > area would be larger than 4KB. KVM_GET_XSAVE2 and KVM_SET_XSAVE
> >
On Mon, Feb 21, 2022 at 01:00:41PM +, David Edmondson wrote:
> On Wednesday, 2022-02-16 at 22:04:30 -08, Yang Zhong wrote:
>
> > From: Jing Liu
> >
> > Intel introduces XFD faulting mechanism for extended
> > XSAVE features to dynamically enable the features in
> > runtime. If CPUID
On 24/02/22 9:48 pm, Daniel P. Berrangé wrote:
On Thu, Feb 24, 2022 at 11:59:08AM +, Kshitij Suri wrote:
Currently screendump only supports PPM format, which is un-compressed and not
standard. Added a "format" parameter to qemu monitor screendump capabilites
to support PNG image capture
On 24/02/22 9:32 pm, Eric Blake wrote:
On Thu, Feb 24, 2022 at 11:59:08AM +, Kshitij Suri wrote:
Currently screendump only supports PPM format, which is un-compressed and not
standard. Added a "format" parameter to qemu monitor screendump capabilites
to support PNG image capture using
On Thu, Feb 24, 2022 at 08:34:40PM +, Joao Martins wrote:
> On 2/24/22 20:12, Michael S. Tsirkin wrote:
> > On Thu, Feb 24, 2022 at 08:04:48PM +, Joao Martins wrote:
> >> On 2/24/22 19:54, Michael S. Tsirkin wrote:
> >>> On Thu, Feb 24, 2022 at 07:44:26PM +, Joao Martins wrote:
>
Hi all,
I find a potential Use-after-free in QEMU 6.2.0, which is in
test_blockjob_common_drain_node() (./tests/unit/test-bdrv-drain.c).
Specifically, at line 880, the variable 'scr' is released by the bdrv_unref().
However, at line 881, it is subsequently used as the 1st parameter of the
On Thu, Feb 24, 2022 at 03:58:16PM -0300, Fabiano Rosas wrote:
> When saving the guest "timebase" we look to the first_cpu for its
> tb_offset. If that CPU happens to be running a nested guest at this
> time, the tb_offset will have the nested guest value.
>
> This was caught by code inspection.
在 2022/2/23 下午11:57, Stefan Hajnoczi 写道:
When ->poll() succeeds the AioHandler is placed on the ready list with
revents set to the magic value 0. This magic value causes
aio_dispatch_handler() to invoke ->poll_ready() instead of ->io_read()
for G_IO_IN or ->io_write() for G_IO_OUT.
This magic
>
> The fix is correct. We just need the submission formatted properly, with
> your
> Signed-off-by tag. When re-formatting, you can add my
>
> Reviewed-by: Richard Henderson
>
> r~
Hi guys,
Thank you for waiting for me.
Here is a new patch with Signed-off-by tags.
Best,
Wentao
From
> On Feb 24, 2022, at 12:52 PM, Paolo Bonzini wrote:
>
> On 2/22/22 20:05, Jag Raman wrote:
>>> -defaults[prefix + 'COMPILER'] = exe_list
>>> +defaults[f'{prefix}COMPILER'] = [exe_list[0]]
>>> +for i in range(1, len(exe_list)):
>>> +
Hi all,
Here is a new patch with Signed-off-by tags.
The old one is wrong for it did't have Signed-off-by tags.
I am looking forward to your confirmation.
Thanks,
Wentao
From 8ece42bda1099a9a0df584cac2478ec5a6e83924 Mon Sep 17 00:00:00 2001
From: Wentao_Liang
Date: Fri, 25 Feb 2022 11:49:54
On Thu, Feb 24, 2022 at 09:00:24PM +, Mark Cave-Ayland wrote:
> On 24/02/2022 18:58, Fabiano Rosas wrote:
>
> > This series implements the migration for a TCG pseries guest running a
> > nested KVM guest. This is just like migrating a pseries TCG guest, but
> > with some extra state to allow
On Thu, Feb 24, 2022 at 03:58:17PM -0300, Fabiano Rosas wrote:
> This adds migration support for TCG pseries machines running a KVM-HV
> guest.
>
> The state that needs to be migrated is:
>
> - the nested PTCR value;
> - the in_nested flag;
> - the nested_tb_offset.
> - the saved host
On Thu, Feb 24, 2022 at 03:58:14PM -0300, Fabiano Rosas wrote:
> These two were not migrated so the remote end was starting with the
> decrementer expired.
>
> I am seeing less frequent crashes with this patch (tested with -smp 4
> and -smp 32). It certainly doesn't fix all issues but it looks
On Thu, Feb 24, 2022 at 03:58:15PM -0300, Fabiano Rosas wrote:
> I'm seeing some stack traces in the migrated guest going through cede
> and some hangs at the plpar_hcall_norets so let's make sure everything
> related to cede/prod is being migrated just in case.
This is a poor approach in
On Fri, Feb 25, 2022 at 2:30 AM Michael S. Tsirkin wrote:
>
> On Thu, Feb 24, 2022 at 05:54:58PM +, Joao Martins wrote:
> > On 2/24/22 17:23, Michael S. Tsirkin wrote:
> > > On Thu, Feb 24, 2022 at 04:07:22PM +, Joao Martins wrote:
> > >> On 2/23/22 23:35, Joao Martins wrote:
> > >>> On
On 2022/02/24 21:53, Peter Maydell wrote:
On Sun, 13 Feb 2022 at 03:58, Akihiko Odaki wrote:
Support the latest PSCI on TCG and HVF. A 64-bit function called from
AArch32 now returns NOT_SUPPORTED, which is necessary to adhere to SMC
Calling Convention 1.0. It is still not compliant with
>
> yes. Could you please send a patch using g_autofree ?
>
> Thanks,
>
> C.
Here is the new patch.
Thanks,
WentaoFrom 8ed76446f78ab1b4152403fdb9dd6f349d6fd52e Mon Sep 17 00:00:00 2001
From: Wentao_Liang
Date: Fri, 25 Feb 2022 11:17:33 +0800
Subject: [PATCH] Fix a potential memory leak
>-Original Message-
>From: Kevin Wolf
>Sent: Thursday, February 24, 2022 7:31 PM
>To: Duan, Zhenzhong
>Cc: qemu-devel@nongnu.org; pbonz...@redhat.com; ebl...@redhat.com;
>m...@redhat.com; pkre...@redhat.com; ler...@redhat.com
>Subject: Re: [PATCH] vl: transform QemuOpts device to JSON
On 2/24/22 14:23, Richard Henderson wrote:
On 2/14/22 03:26, Luc Michel wrote:
In some cases, cpu->exit_request can be false after handling the
interrupt, leading to another TB being executed instead of returning
to the main loop.
Fix this by returning true unconditionally when in single-step
Excerpts from Fabiano Rosas's message of February 25, 2022 4:58 am:
> This adds migration support for TCG pseries machines running a KVM-HV
> guest.
>
> The state that needs to be migrated is:
>
> - the nested PTCR value;
> - the in_nested flag;
> - the nested_tb_offset.
> - the saved host
On 2/14/22 03:26, Luc Michel wrote:
In some cases, cpu->exit_request can be false after handling the
interrupt, leading to another TB being executed instead of returning
to the main loop.
Fix this by returning true unconditionally when in single-step mode.
Fixes:
On 2/14/22 08:31, Philippe Mathieu-Daudé wrote:
ArchCPU is our interface with target-specific code. Use it as
a forward-declared opaque pointer (abstract type), having its
structure defined by each target.
Signed-off-by: Philippe Mathieu-Daudé
---
include/hw/core/cpu.h | 4 ++--
On 2/14/22 08:31, Philippe Mathieu-Daudé wrote:
Replace the boilerplate code to declare CPU QOM types
and macros, and forward-declare the CPU instance type.
Signed-off-by: Philippe Mathieu-Daudé
---
Reviewed-by: Richard Henderson
r~
On 2/14/22 08:31, Philippe Mathieu-Daudé wrote:
HexagonCPU field parent_class is of type CPUClass, which
is declared in "hw/core/cpu.h".
Signed-off-by: Philippe Mathieu-Daudé
---
target/hexagon/cpu.h | 1 +
1 file changed, 1 insertion(+)
Reviewed-by: Richard Henderson
r~
On 2/24/22 03:48, Amir Gonnen wrote:
+static void nios2_cpu_set_eic_irq(Nios2CPU *cpu, int level)
+{
+CPUNios2State *env = >env;
+CPUState *cs = CPU(cpu);
+
+env->irq_pending = level;
+
+if (env->irq_pending && nios2_take_eic_irq(cpu)) {
+env->irq_pending = 0;
+
On 2/24/22 03:48, Amir Gonnen wrote:
+void helper_eret(CPUNios2State *env, uint32_t new_pc)
+{
+uint32_t crs = cpu_get_crs(env);
+if (crs == 0) {
+env->regs[CR_STATUS] = env->regs[CR_ESTATUS];
+} else {
+env->regs[CR_STATUS] = env->regs[R_SSTATUS];
+}
+
On 24/2/22 22:22, Stefan Weil wrote:
Am 07.01.22 um 16:49 schrieb Stefan Weil:
Signed-off-by: Stefan Weil
---
LGTM has some more alerts which need attention:
https://lgtm.com/projects/g/qemu/qemu/
Regards,
Stefan
subprojects/libvhost-user/libvhost-user.c | 2 +-
1 file changed, 1
On 2/24/22 03:48, Amir Gonnen wrote:
@@ -88,7 +93,9 @@ struct Nios2CPUClass {
#define CR_STATUS_IH (1 << 3)
#define CR_STATUS_IL (63 << 4)
#define CR_STATUS_CRS (63 << 10)
+#define CR_STATUS_CRS_OFFSET 10
#define CR_STATUS_PRS (63 << 16)
+#define CR_STATUS_PRS_OFFSET 16
Hi
On Fri, Feb 25, 2022 at 1:41 AM Stefan Weil wrote:
> Am 24.02.22 um 20:12 schrieb Peter Maydell:
>
> > On Thu, 24 Feb 2022 at 18:38, wrote:
> >> From: Marc-André Lureau
> >>
> >> This workaround was added in commit 95df51a4 ("w32: Always use standard
> >> instead of native format
On 2/24/22 10:19, Daniel Henrique Barboza wrote:
+void helper_ebb_perfm_excp(CPUPPCState *env)
Please reserve the prefix "helper_" for something that is called from tcg generated code,
which this is not.
r~
On Thu, 24 Feb 2022 20:34:40 +
Joao Martins wrote:
> On 2/24/22 20:12, Michael S. Tsirkin wrote:
> > On Thu, Feb 24, 2022 at 08:04:48PM +, Joao Martins wrote:
> >> On 2/24/22 19:54, Michael S. Tsirkin wrote:
> >>> On Thu, Feb 24, 2022 at 07:44:26PM +, Joao Martins wrote:
>
Am 24.02.22 um 20:12 schrieb Peter Maydell:
On Thu, 24 Feb 2022 at 18:38, wrote:
From: Marc-André Lureau
This workaround was added in commit 95df51a4 ("w32: Always use standard
instead of native format strings"), as it claimed glib was using
__printf__ attribute. This is surprising, since
On 2/24/22 10:18, Daniel Henrique Barboza wrote:
This is an exclusive TCG helper. Gating it with CONFIG_TCG and changing
meson.build accordingly will prevent problems --disable-tcg and
--disable-linux-user later on.
We're also changing the uses of !kvm_enabled() to tcg_enabled() to avoid
adding
On 2/24/22 10:23, Matheus K. Ferst wrote:
You don't want to use tcg_gen_rotlv_vec directly, but tcg_gen_rotlv_vec.
I guess there is a typo here. Did you mean tcg_gen_gvec_rotlv? Or
tcg_gen_rotlv_mod_vec?
Dangit. Paste-paste error. The first: tcg_gen_gvec_rotlv.
r~
On 2/24/22 09:16, Víctor Colombo wrote:
Could you please elaborate more on how do you think using
float*_compare and its FloatRelation result would work here?
I noticed do_scalar_cmp modifies CR and sets FPCC flag, which
is not what VSX_SCALAR_CMP do. Using that function would require a
rework.
Am 07.01.22 um 16:49 schrieb Stefan Weil:
Signed-off-by: Stefan Weil
---
LGTM has some more alerts which need attention:
https://lgtm.com/projects/g/qemu/qemu/
Regards,
Stefan
subprojects/libvhost-user/libvhost-user.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
On 24/02/2022 18:58, Fabiano Rosas wrote:
This series implements the migration for a TCG pseries guest running a
nested KVM guest. This is just like migrating a pseries TCG guest, but
with some extra state to allow a nested guest to continue to run on
the destination.
Unfortunately the regular
On 2/24/22 20:12, Michael S. Tsirkin wrote:
> On Thu, Feb 24, 2022 at 08:04:48PM +, Joao Martins wrote:
>> On 2/24/22 19:54, Michael S. Tsirkin wrote:
>>> On Thu, Feb 24, 2022 at 07:44:26PM +, Joao Martins wrote:
On 2/24/22 18:30, Michael S. Tsirkin wrote:
> On Thu, Feb 24, 2022
On 23/02/2022 19:19, Richard Henderson wrote:
On 2/23/22 11:43, Matheus K. Ferst wrote:
Note that rotlv does the masking itself:
/*
* Expand D = A << (B % element bits)
*
* Unlike scalar shifts, where it is easy for the target front end
* to include the modulo as part of the expansion.
This patch adds the EBB exception support that are triggered by
Performance Monitor alerts. This happens when a Performance Monitor
alert occurs and MMCR0_EBE, BESCR_PME and BESCR_GE are set.
fire_PMC_interrupt() will execute a new ebb_perfm_excp() helper that
will check for MMCR0_EBE, BESCR_PME
This is an exclusive TCG helper. Gating it with CONFIG_TCG and changing
meson.build accordingly will prevent problems --disable-tcg and
--disable-linux-user later on.
We're also changing the uses of !kvm_enabled() to tcg_enabled() to avoid
adding "defined(CONFIG_TCG)" ifdefs, since tcg_enabled()
There are still PMU exclusive bits to handle in fire_PMC_interrupt()
before implementing the EBB support. Let's finalize it now to avoid
dealing with PMU and EBB logic at the same time in the next patches.
fire_PMC_interrupt() will fire an Performance Monitor alert depending on
MMCR0_PMAE. If we
PPC_INTERRUPT_EBB is a new interrupt that will be used to deliver EBB
exceptions that had to be postponed because the thread wasn't in problem
state at the time the event-based branch was supposed to occur.
ISA 3.1 also defines two EBB exceptions: Performance Monitor EBB
exception and External
Hi,
This new version contains a change in patch 1 (former 2) that was
proposed by Richard in the v12 review.
Changes from v12:
- former patch 1: dropped, no longer applicable
- patch 1 (former 2):
* use tcg_enabled() instead of !kvm_enabled() to avoid defined(CONFIG_TCG)
ifdefs
- v12 link:
On Thu, 24 Feb 2022 at 19:50, Marc-André Lureau
wrote:
> On Thu, Feb 24, 2022 at 11:23 PM Peter Maydell
> wrote:
>> You're probably building with a newer glib, and possibly also
>> a newer mingw.
>>
>> I've cc'd Stefan Weil who might know whether we can drop this
>> workaround as far as the
On 2/17/22 19:17, Richard Henderson wrote:
On 2/16/22 21:10, Daniel Henrique Barboza wrote:
static void init_tcg_pmu_power8(CPUPPCState *env)
{
-#if defined(TARGET_PPC64) && !defined(CONFIG_USER_ONLY)
+#if defined(CONFIG_TCG)
/* Init PMU overflow timers */
if (!kvm_enabled())
On Thu, Feb 24, 2022 at 08:04:48PM +, Joao Martins wrote:
>
>
> On 2/24/22 19:54, Michael S. Tsirkin wrote:
> > On Thu, Feb 24, 2022 at 07:44:26PM +, Joao Martins wrote:
> >> On 2/24/22 18:30, Michael S. Tsirkin wrote:
> >>> On Thu, Feb 24, 2022 at 05:54:58PM +, Joao Martins wrote:
>
On 2/24/22 08:58, Fabiano Rosas wrote:
These two were not migrated so the remote end was starting with the
decrementer expired.
I am seeing less frequent crashes with this patch (tested with -smp 4
and -smp 32). It certainly doesn't fix all issues but it looks like it
helps.
Signed-off-by:
On 2/24/22 19:54, Michael S. Tsirkin wrote:
> On Thu, Feb 24, 2022 at 07:44:26PM +, Joao Martins wrote:
>> On 2/24/22 18:30, Michael S. Tsirkin wrote:
>>> On Thu, Feb 24, 2022 at 05:54:58PM +, Joao Martins wrote:
On 2/24/22 17:23, Michael S. Tsirkin wrote:
> On Thu, Feb 24,
On 2/24/22 08:37, marcandre.lur...@redhat.com wrote:
From: Marc-André Lureau
Move qemu_ether_ntoa() which is only needed in net/.
Signed-off-by: Marc-André Lureau
---
include/qemu-common.h | 1 -
net/announce.c| 13 +
util/cutils.c | 14 --
3 files
On 2/24/22 08:37, marcandre.lur...@redhat.com wrote:
From: Marc-André Lureau
Signed-off-by: Marc-André Lureau
---
qapi/qmp-dispatch.c | 1 -
1 file changed, 1 deletion(-)
Reviewed-by: Richard Henderson
r~
On 2/24/22 08:36, marcandre.lur...@redhat.com wrote:
From: Marc-André Lureau
Signed-off-by: Marc-André Lureau
---
util/cutils.c | 2 --
1 file changed, 2 deletions(-)
Reviewed-by: Richard Henderson
r~
Hi
On Thu, Feb 24, 2022 at 11:43 PM Richard Henderson <
richard.hender...@linaro.org> wrote:
> On 2/24/22 08:36, marcandre.lur...@redhat.com wrote:
> > From: Marc-André Lureau
> >
> > Move the macro and declaration so it can use glib in the following
> > patch.
> >
> > Signed-off-by: Marc-André
On 2/24/22 08:36, marcandre.lur...@redhat.com wrote:
From: Marc-André Lureau
This will help to make common code independent.
Signed-off-by: Marc-André Lureau
---
include/qemu/compiler.h | 3 +++
include/qemu/osdep.h| 3 ---
2 files changed, 3 insertions(+), 3 deletions(-)
Hi
On Thu, Feb 24, 2022 at 11:41 PM Richard Henderson <
richard.hender...@linaro.org> wrote:
> On 2/24/22 08:36, marcandre.lur...@redhat.com wrote:
> > From: Marc-André Lureau
> >
> > Replace a config-time define with a compile time condition
> > define (compatible with clang and gcc) that must
On 2/24/22 08:36, marcandre.lur...@redhat.com wrote:
-#if UINTPTR_MAX == UINT32_MAX
-# define HOST_LONG_BITS 32
-#elif UINTPTR_MAX == UINT64_MAX
-# define HOST_LONG_BITS 64
-#else
-# error Unknown pointer size
-#endif
+#define HOST_LONG_BITS (__SIZEOF_POINTER__ * 8)
I guess. I'll note that
Hi
On Thu, Feb 24, 2022 at 11:37 PM Richard Henderson <
richard.hender...@linaro.org> wrote:
> On 2/24/22 08:36, marcandre.lur...@redhat.com wrote:
> > From: Marc-André Lureau
> >
> > G_NORETURN was introduced in glib 2.68, fallback to G_GNUC_NORETURN in
> > glib-compat.
> >
> > Note that this
On Thu, Feb 24, 2022 at 07:44:26PM +, Joao Martins wrote:
> On 2/24/22 18:30, Michael S. Tsirkin wrote:
> > On Thu, Feb 24, 2022 at 05:54:58PM +, Joao Martins wrote:
> >> On 2/24/22 17:23, Michael S. Tsirkin wrote:
> >>> On Thu, Feb 24, 2022 at 04:07:22PM +, Joao Martins wrote:
>
Hi Peter
On Thu, Feb 24, 2022 at 11:23 PM Peter Maydell
wrote:
> On Thu, 24 Feb 2022 at 18:38, wrote:
> >
> > From: Marc-André Lureau
> >
> > This workaround was added in commit 95df51a4 ("w32: Always use standard
> > instead of native format strings"), as it claimed glib was using
> >
On 2/24/22 18:30, Michael S. Tsirkin wrote:
> On Thu, Feb 24, 2022 at 05:54:58PM +, Joao Martins wrote:
>> On 2/24/22 17:23, Michael S. Tsirkin wrote:
>>> On Thu, Feb 24, 2022 at 04:07:22PM +, Joao Martins wrote:
On 2/23/22 23:35, Joao Martins wrote:
> On 2/23/22 21:22, Michael S.
On 2/24/22 08:36, marcandre.lur...@redhat.com wrote:
From: Marc-André Lureau
Move the macro and declaration so it can use glib in the following
patch.
Signed-off-by: Marc-André Lureau
---
include/qemu/compiler.h | 16
include/qemu/osdep.h| 16
2 files
On 2/24/22 08:36, marcandre.lur...@redhat.com wrote:
From: Marc-André Lureau
Replace a config-time define with a compile time condition
define (compatible with clang and gcc) that must be declared prior to
its usage. This avoids having a global configure time define, but also
prevents from bad
On 2/24/22 08:36, marcandre.lur...@redhat.com wrote:
From: Marc-André Lureau
One less qemu-specific macro. It also helps to make some headers/units
only depend on glib, and thus moved in standalone projects eventually.
Signed-off-by: Marc-André Lureau
---
include/qemu/compiler.h| 2 --
On 2/24/22 08:36, marcandre.lur...@redhat.com wrote:
From: Marc-André Lureau
G_NORETURN was introduced in glib 2.68, fallback to G_GNUC_NORETURN in
glib-compat.
Note that this attribute must be placed before the function declaration
(bringing a bit of consistency in qemu codebase usage).
On 2/24/22 08:36, marcandre.lur...@redhat.com wrote:
From: Marc-André Lureau
One less qemu-specific macro. It also helps to make some headers/units
only depend on glib, and thus moved in standalone projects eventually.
Signed-off-by: Marc-André Lureau
---
include/qemu-common.h | 2 +-
On Thu, Jan 6, 2022 at 3:09 PM Titus Rwantare wrote:
> This patch series contains updates to PMBus in QEMU along with some PMBus
> device models for Renesas regulators.
> I have also added myself to MAINTAINERS as this code is in use daily,
> where I am responsible for it.
>
> Shengtan Mao (1):
On 22/02/2022 21:20, Richard Henderson wrote:> On 2/22/22 04:36,
matheus.fe...@eldorado.org.br wrote:
From: Víctor Colombo
Refactor VSX_SCALAR_CMP_DP, changing its name to VSX_SCALAR_CMP and
prepare the helper to be used for quadword comparisons.
Signed-off-by: Víctor Colombo
Signed-off-by:
On Thu, 24 Feb 2022 at 18:39, wrote:
>
> From: Marc-André Lureau
>
> Move qemu_ether_ntoa() which is only needed in net/.
>
> Signed-off-by: Marc-André Lureau
> ---
Reviewed-by: Peter Maydell
thanks
-- PMM
On Thu, Feb 24, 2022 at 2:56 AM Peter Maydell
wrote:
> On Wed, 2 Feb 2022 at 17:57, Patrick Venture wrote:
> >
> > Previously this device created N subdevices which each owned an i2c bus.
> > Now this device simply owns the N i2c busses directly.
> >
> > Tested: Verified devices behind mux are
Clean up some unnecessary code
Signed-off-by: Li Zhang
---
migration/multifd.c | 15 ---
1 file changed, 4 insertions(+), 11 deletions(-)
diff --git a/migration/multifd.c b/migration/multifd.c
index 3242f688e5..d44cc6670f 100644
--- a/migration/multifd.c
+++ b/migration/multifd.c
On Thu, 24 Feb 2022 at 18:38, wrote:
>
> From: Marc-André Lureau
>
> This workaround was added in commit 95df51a4 ("w32: Always use standard
> instead of native format strings"), as it claimed glib was using
> __printf__ attribute. This is surprising, since glib has always used
> G_GNUC_PRINTF
On Thu, Feb 24, 2022 at 10:58:52AM -0800, Patrick Venture wrote:
> On Thu, Jan 6, 2022 at 3:09 PM Titus Rwantare wrote:
>
> > This patch series contains updates to PMBus in QEMU along with some PMBus
> > device models for Renesas regulators.
> > I have also added myself to MAINTAINERS as this
* Steve Sistare (steven.sist...@oracle.com) wrote:
> If qemu starts and loads a VM in the suspended state, then a later wakeup
> request will set the state to running, which is not sufficient to initialize
> the vm, as vm_start was never called during this invocation of qemu. See
>
This adds migration support for TCG pseries machines running a KVM-HV
guest.
The state that needs to be migrated is:
- the nested PTCR value;
- the in_nested flag;
- the nested_tb_offset.
- the saved host CPUPPCState structure;
Signed-off-by: Fabiano Rosas
---
(this migrates just fine with L2
From: Shengtan Mao
Reviewed-by: Hao Wu
Reviewed-by: Chris Rauer
Signed-off-by: Shengtan Mao
Signed-off-by: Patrick Venture
Signed-off-by: Hao Wu
---
v4:
* use strncmp to compare fixed length strings
v3:
* fixup compilation from missing macro value
v2:
* update copyright year
* check
When saving the guest "timebase" we look to the first_cpu for its
tb_offset. If that CPU happens to be running a nested guest at this
time, the tb_offset will have the nested guest value.
This was caught by code inspection.
Signed-off-by: Fabiano Rosas
---
hw/ppc/ppc.c | 17 -
From: Marc-André Lureau
Signed-off-by: Marc-André Lureau
---
qapi/qmp-dispatch.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/qapi/qmp-dispatch.c b/qapi/qmp-dispatch.c
index d378bccac73b..0990873ec8ec 100644
--- a/qapi/qmp-dispatch.c
+++ b/qapi/qmp-dispatch.c
@@ -21,7 +21,6 @@
#include
This series implements the migration for a TCG pseries guest running a
nested KVM guest. This is just like migrating a pseries TCG guest, but
with some extra state to allow a nested guest to continue to run on
the destination.
Unfortunately the regular TCG migration scenario (not nested) is not
From: Marc-André Lureau
Move qemu_ether_ntoa() which is only needed in net/.
Signed-off-by: Marc-André Lureau
---
include/qemu-common.h | 1 -
net/announce.c| 13 +
util/cutils.c | 14 --
3 files changed, 13 insertions(+), 15 deletions(-)
diff --git
I'm seeing some stack traces in the migrated guest going through cede
and some hangs at the plpar_hcall_norets so let's make sure everything
related to cede/prod is being migrated just in case.
Signed-off-by: Fabiano Rosas
---
hw/ppc/spapr_cpu_core.c | 1 +
On Wed, 22 Dec 2021 at 19:45, Steve Sistare wrote:
>
> Include CONFIG_DEVICES so that populate_vfio_info is instantiated for
> CONFIG_VFIO.
The commit message says "include CONFIG_DEVICES"...
> Signed-off-by: Steve Sistare
> ---
> migration/target.c | 10 +++---
> 1 file changed, 7
The problem is probably because we read it in using "strcmp". strcmp
compares two strings that end with "\0". But one of the string is read in
using read() so it didn't read in the ending '\0' character.
We should use strncmp to compare the two strings. It probably avoids the
issue.
On Tue, Feb
These two were not migrated so the remote end was starting with the
decrementer expired.
I am seeing less frequent crashes with this patch (tested with -smp 4
and -smp 32). It certainly doesn't fix all issues but it looks like it
helps.
Signed-off-by: Fabiano Rosas
---
target/ppc/machine.c |
From: Marc-André Lureau
G_NORETURN was introduced in glib 2.68, fallback to G_GNUC_NORETURN in
glib-compat.
Note that this attribute must be placed before the function declaration
(bringing a bit of consistency in qemu codebase usage).
Signed-off-by: Marc-André Lureau
---
From: Marc-André Lureau
Signed-off-by: Marc-André Lureau
---
util/cutils.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/util/cutils.c b/util/cutils.c
index c9b91e7535a8..53346138c970 100644
--- a/util/cutils.c
+++ b/util/cutils.c
@@ -27,8 +27,6 @@
#include
#include "qemu-common.h"
From: Marc-André Lureau
Simplify the macro, not depending on headers defines, but compiler
predefined __SIZEOF__POINTER__ only.
Available since gcc 4.3 and clang 2.8.
Signed-off-by: Marc-André Lureau
---
include/qemu/osdep.h | 8 +---
1 file changed, 1 insertion(+), 7 deletions(-)
diff
From: Marc-André Lureau
Move the macro and declaration so it can use glib in the following
patch.
Signed-off-by: Marc-André Lureau
---
include/qemu/compiler.h | 16
include/qemu/osdep.h| 16
2 files changed, 16 insertions(+), 16 deletions(-)
diff --git
From: Marc-André Lureau
Signed-off-by: Marc-André Lureau
---
scripts/modinfo-collect.py | 5 -
1 file changed, 5 deletions(-)
diff --git a/scripts/modinfo-collect.py b/scripts/modinfo-collect.py
index 61b90688c6dc..4e7584df6676 100755
--- a/scripts/modinfo-collect.py
+++
From: Marc-André Lureau
This will help to make common code independent.
Signed-off-by: Marc-André Lureau
---
include/qemu/compiler.h | 3 +++
include/qemu/osdep.h| 3 ---
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/include/qemu/compiler.h b/include/qemu/compiler.h
index
From: Marc-André Lureau
Replace a config-time define with a compile time condition
define (compatible with clang and gcc) that must be declared prior to
its usage. This avoids having a global configure time define, but also
prevents from bad usage, if the config header wasn't included before.
From: Marc-André Lureau
This workaround was added in commit 95df51a4 ("w32: Always use standard
instead of native format strings"), as it claimed glib was using
__printf__ attribute. This is surprising, since glib has always used
G_GNUC_PRINTF which, as the name implies, uses __gnu_printf__ when
From: Marc-André Lureau
One less qemu-specific macro. It also helps to make some headers/units
only depend on glib, and thus moved in standalone projects eventually.
Signed-off-by: Marc-André Lureau
---
include/qemu/compiler.h| 2 --
include/qom/object.h | 6 +++---
From: Marc-André Lureau
Hi,
A small collection of patches gleaned while working on different things (more
specifically, I am looking at making qapi code usable outside of qemu, many
things are twisted together, it takes effort but it is hopefully worth it)
Based-on:
From: Marc-André Lureau
One less qemu-specific macro. It also helps to make some headers/units
only depend on glib, and thus moved in standalone projects eventually.
Signed-off-by: Marc-André Lureau
---
include/qemu-common.h | 2 +-
include/qemu/compiler.h| 2 --
On Thu, Feb 24, 2022 at 05:54:58PM +, Joao Martins wrote:
> On 2/24/22 17:23, Michael S. Tsirkin wrote:
> > On Thu, Feb 24, 2022 at 04:07:22PM +, Joao Martins wrote:
> >> On 2/23/22 23:35, Joao Martins wrote:
> >>> On 2/23/22 21:22, Michael S. Tsirkin wrote:
> > +static void
* Steve Sistare (steven.sist...@oracle.com) wrote:
> Add a function that returns an error if any ram_list block represents
> volatile memory.
>
> Signed-off-by: Steve Sistare
> ---
> include/exec/memory.h | 8
> softmmu/memory.c | 26 ++
> 2 files changed,
* Steve Sistare (steven.sist...@oracle.com) wrote:
> Use qemu_file_open to simplify a few functions in savevm.c.
> No functional change.
>
> Signed-off-by: Steve Sistare
So I think this is mostly OK, but a couple of minor tidyups below;
so with the tidies and the renames from the previous
* Steve Sistare (steven.sist...@oracle.com) wrote:
> Add qemu_file_open and qemu_fd_open to create QEMUFile objects for unix
> files and file descriptors.
>
> Signed-off-by: Steve Sistare
> ---
> migration/qemu-file-channel.c | 36
>
On 2/24/22 03:46, Peter Maydell wrote:
When we're using KVM, the PSCI implementation is provided by the
kernel, but QEMU has to tell the guest about it via the device tree.
Currently we look at the KVM_CAP_ARM_PSCI_0_2 capability to determine
if the kernel is providing at least PSCI 0.2, but if
1 - 100 of 256 matches
Mail list logo
