On 10/05/2023 16.39, Juan Quintela wrote:
Any good reason why this still exist?
I can understand u* and __u* to be linux kernel like, but ulong?
Signed-off-by: Juan Quintela
---
pc-bios/s390-ccw/helper.h| 2 +-
pc-bios/s390-ccw/s390-ccw.h | 7 +++
pc-bios/s390-ccw/virtio-
On Mon, 2023-05-15 at 21:57 +0200, Juan Quintela wrote:
> Remove the one in control_save_page().
>
> Signed-off-by: Juan Quintela
> ---
> migration/ram.c | 7 ---
> 1 file changed, 7 deletions(-)
>
> diff --git a/migration/ram.c b/migration/ram.c
> index a706edecc0..67ed49b387 100644
> ---
On Mon, 2023-05-15 at 21:57 +0200, Juan Quintela wrote:
> Signed-off-by: Juan Quintela
> ---
> migration/rdma.c | 23 +++
> 1 file changed, 11 insertions(+), 12 deletions(-)
>
> diff --git a/migration/rdma.c b/migration/rdma.c
> index 074456f9df..416dec00a2 100644
> --- a/mig
On 23/05/2023 10.49, Tommy Wu wrote:
Add some simple tests of the watchdog timer in the always-on domain device
of HiFive 1 rev b.
Signed-off-by: Tommy Wu
Reviewed-by: Frank Chang
---
tests/qtest/meson.build | 3 +
tests/qtest/sifive-e-aon-watchdog-test.c | 450 ++
On Mon, 2023-05-15 at 21:57 +0200, Juan Quintela wrote:
> RDMA protocol is completely asynchronous, so in qemu_rdma_save_page()
> they "invent" that a byte has been transferred. And then they call
> qemu_file_credit_transfer() and ram_transferred_add() with that byte.
> Just remove that calls as n
On Mon, 2023-05-15 at 21:57 +0200, Juan Quintela wrote:
> After this change, nothing abuses QEMUFile to account for data
> transferrefd during migration.
>
> Signed-off-by: Juan Quintela
> ---
> migration/qemu-file.h | 8
> migration/qemu-file.c | 5 -
> 2 files changed, 13 deletion
Queued, thanks.
Paolo
Queued, thanks.
Paolo
On 24/5/23 23:10, Mark Cave-Ayland wrote:
Signed-off-by: Mark Cave-Ayland
---
hw/m68k/q800.c | 13 +
include/hw/m68k/q800.h | 1 +
2 files changed, 6 insertions(+), 8 deletions(-)
Reviewed-by: Philippe Mathieu-Daudé
On 24/5/23 23:10, Mark Cave-Ayland wrote:
Signed-off-by: Mark Cave-Ayland
---
hw/m68k/q800.c | 10 +-
include/hw/m68k/q800.h | 4 +++-
2 files changed, 8 insertions(+), 6 deletions(-)
diff --git a/hw/m68k/q800.c b/hw/m68k/q800.c
index 976da06231..ee6175ceb4 100644
--- a/hw/
The "usb-storage" device might not have been compiled into the binary
(e.g. when compiling with "--without-default-devices"), so we have to
check first before using it.
Signed-off-by: Thomas Huth
---
tests/qtest/usb-hcd-uhci-test.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff -
Here are some more patches that are required for running the qtests
with builds that have been configured with "--without-default-devices".
We need to check whether the required devices are really available
in the binaries before we can use them, otherwise the tests will
fail.
Thomas Huth (4):
t
Though we are already using CONFIG_RTL8139_PCI in the meson.build file
for testing whether the rtl8139 device is available or not, this is not
enough: The CONFIG switch might have been selected by another target
(e.g. the mips fuloong2e machine has the rtl8139 chip soldered on the
board), so CONFIG
On Mon, 2023-05-15 at 21:57 +0200, Juan Quintela wrote:
> When we sent a page through QEMUFile hooks (RDMA) there are three
> posiblities:
> - We are not using RDMA. return RAM_SAVE_CONTROL_DELAYED and
> control_save_page() returns false to let anything else to proceed.
> - There is one error but
The arm "virt" machine needs "virtio-blk-pci" for devices that get attached
via the "-cdrom" option. Since this is an optional device that might not
be available in the binary, we should check for the availability of this
device first before using it.
Signed-off-by: Thomas Huth
---
tests/qtest/b
The devices might not be available in the binary (e.g. when compiling
with "--without-default-devices"), so we have to check before we can
use them.
Signed-off-by: Thomas Huth
---
tests/qtest/usb-hcd-ehci-test.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/tests/qtest/usb-hcd-ehci-te
On 24/5/23 23:10, Mark Cave-Ayland wrote:
The djMEMC controller is used to store information related to the physical
memory
configuration.
Co-developed-by: Laurent Vivier
Signed-off-by: Mark Cave-Ayland
---
MAINTAINERS | 2 +
hw/m68k/Kconfig | 1 +
hw/m68k/q800.c
On 24/5/23 23:10, Mark Cave-Ayland wrote:
MacOS reads this address to identify the hardware.
This is a basic implementation returning the ID of Quadra 800.
Details:
http://mess.redump.net/mess/driver_info/mac_technical_notes
"There are 3 ID schemes [...]
The third and most scalable is a
On 24/5/23 23:10, Mark Cave-Ayland wrote:
MacOS attempts a series of writes and reads over the entire RAM area in order
to determine the amount of RAM within the machine. Allow accesses to the
entire RAM area ignoring writes and always reading zero for areas where there
is no physical RAM install
On Mon, 2023-05-15 at 21:57 +0200, Juan Quintela wrote:
> In the past, we had to put the in the main thread all the operations
> related with sizes due to qemu_file not beeing thread safe. As now
> all counters are atomic, we can update the counters just after the
> do the write. As an aditional
On Wed, May 24, 2023 at 02:39:50PM -0700, Sean Christopherson wrote:
> On Wed, May 24, 2023, Peter Zijlstra wrote:
> > On Wed, May 24, 2023 at 01:16:03PM -0700, Sean Christopherson wrote:
> > > Of course, the only accesses outside of mmu_lock are reads, so on x86 that
> > > "atomic" access is just
On 24/05/2023 18:38, Cédric Le Goater wrote:
External email: Use caution opening links or attachments
Hello Avihai,
On 5/24/23 14:49, Avihai Horon wrote:
On 23/05/2023 17:56, Cédric Le Goater wrote:
External email: Use caution opening links or attachments
Hello Avihai,
On 5/21/23 17:18
On Thu, 25 May 2023 at 07:45, Enze Li wrote:
>
> I noticed that in the latest version, the copyright string is still
> 2022, even though 2023 is halfway through. This patch fixes that and
> fixes the documentation along with it.
>
> Signed-off-by: Enze Li
Whoops, thanks for catching this. These
> On 25-May-2023, at 1:40 PM, Thomas Huth wrote:
>
> The devices might not be available in the binary (e.g. when compiling
> with "--without-default-devices"), so we have to check before we can
> use them.
>
> Signed-off-by: Thomas Huth
Reviewed-by: Ani Sinha
> ---
> tests/qtest/usb-hcd-e
> On 25-May-2023, at 1:40 PM, Thomas Huth wrote:
>
> The "usb-storage" device might not have been compiled into the binary
> (e.g. when compiling with "--without-default-devices"), so we have to
> check first before using it.
>
> Signed-off-by: Thomas Huth
Reviewed-by: Ani Sinha
> ---
> t
Avihai Horon writes:
> Migration downtime estimation is calculated based on bandwidth and
> remaining migration data. This assumes that loading of migration data in
> the destination takes a negligible amount of time and that downtime
> depends only on network speed.
>
> While this may be true fo
On Mon, 15 May 2023 at 21:25, Niek Linnenbank wrote:
>
> Hi Qianfan,
>
> Thanks for sending the v4. I've reviewed all patches and replied to each
> individual patch.
>
> Great work so far!
>
> One thing I wanted to mention is that when you receive a 'Reviewed-by:' or
> 'Tested-by:' line from a r
On 24/05/2023 22:32, Peter Xu wrote:
External email: Use caution opening links or attachments
On Sun, May 21, 2023 at 06:18:03PM +0300, Avihai Horon wrote:
Implement switchover ack logic. This prevents the source from stopping
the VM and completing the migration until an ACK is received from
On Wed, 24 May 2023 at 18:38, Richard Henderson
wrote:
>
> We currently print FAIL for the failure of a succ_* test, but don't
> return a failure exit code. Instead, convert the script to emit
> Test Anything Protocol, which gives visibility into each subtest
> as well as not relying on exit code
On Tue, 23 May 2023 at 19:41, Thomas Huth wrote:
>
> On 23/05/2023 19.30, Marcin Juszkiewicz wrote:
> ...
> >> (is there a reason why it can't be disabled with "-vga none" or
> >> "-nodefaults"?)
> >
> > That's something I need to check how it should be done.
>
> Other boards set mc->default_displ
On Wed, 24 May 2023 at 12:33, Marcin Juszkiewicz
wrote:
>
> About year ago support for setting platform version for sbsa-ref was
> added with "0.0" value.
>
> This patchset bumps it to "0.1" as we export basic GIC information via
> DeviceTree to the firmware. TF-A will then take it and provide to
On Wed, 24 May 2023 at 12:14, Marcin Juszkiewicz
wrote:
>
> arm/sbsa-ref uses Bochs-display graphics card and without it being
> present in vga_interfaces "-vga none" argument handling cannot be added.
>
> Signed-off-by: Marcin Juszkiewicz
> ---
> hw/pci/pci.c| 2 ++
> include/sysemu
On Wed, 24 May 2023 at 15:37, Clément Chigot wrote:
>
> When passing --smp with a number lower than XLNX_ZYNQMP_NUM_APU_CPUS,
> the expression (ms->smp.cpus - XLNX_ZYNQMP_NUM_APU_CPUS) will result
> in a positive number as ms->smp.cpus is a unsigned int.
> This will raise the following error after
Hello Stefan, folks,
the perf-ci detected and bisected the 6d740fb - aio-posix: do not nest poll
handlers - as a performance improvement when using multiple concurrent jobs and
4k (22%) as well as 1024k (63%) blocks on aarch64 (on a slow rotational disk).
https://ldoktor.github.io/tmp/RedH
A few fixes for things that have gotten inconsistent or redundant
over many changes to Makefile.
Paolo
Paolo Bonzini (3):
tests/docker: simplify HOST_ARCH definition
tests/vm: fix and simplify HOST_ARCH definition
Makefile: remove $(TESTS_PYTHON)
tests/Makefile.include| 8 +++-
It is now the same as $(PYTHON), since the latter always points at
pyvenv/bin/python3.
Signed-off-by: Paolo Bonzini
---
tests/Makefile.include| 8 +++-
tests/vm/Makefile.include | 2 +-
2 files changed, 4 insertions(+), 6 deletions(-)
diff --git a/tests/Makefile.include b/tests/Makefil
ARCH is always empty, so just define HOST_ARCH as the result of uname.
The incorrect definition was not being used because the "ifeq" statement
is wrong; replace it with the same idiom based on $(realpath) that the
main Makefile uses.
With this change, vm-build-netbsd in a configured tree will not
ARCH is always empty, so just define HOST_ARCH as the result of uname.
Signed-off-by: Paolo Bonzini
---
tests/docker/Makefile.include | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/docker/Makefile.include b/tests/docker/Makefile.include
index 94015253254c..142e8605eee9
On 25/05/2023 12.05, Peter Maydell wrote:
On Tue, 23 May 2023 at 19:41, Thomas Huth wrote:
On 23/05/2023 19.30, Marcin Juszkiewicz wrote:
...
(is there a reason why it can't be disabled with "-vga none" or
"-nodefaults"?)
That's something I need to check how it should be done.
Other board
On 5/24/23 11:30, Denis V. Lunev wrote:
> On 5/23/23 18:24, Andrey Drobyshev wrote:
>> In case when we're rebasing within one backing chain, and when target
>> image
>> is larger than old backing file, bdrv_is_allocated_above() ends up
>> setting
>> *pnum = 0. As a result, target offset isn't gett
On Thu, 25 May 2023 at 11:32, Thomas Huth wrote:
>
> On 25/05/2023 12.05, Peter Maydell wrote:
> > On Tue, 23 May 2023 at 19:41, Thomas Huth wrote:
> >>
> >> On 23/05/2023 19.30, Marcin Juszkiewicz wrote:
> >> ...
> (is there a reason why it can't be disabled with "-vga none" or
> "-nod
On 25/05/2023 12.44, Peter Maydell wrote:
On Thu, 25 May 2023 at 11:32, Thomas Huth wrote:
On 25/05/2023 12.05, Peter Maydell wrote:
On Tue, 23 May 2023 at 19:41, Thomas Huth wrote:
On 23/05/2023 19.30, Marcin Juszkiewicz wrote:
...
(is there a reason why it can't be disabled with "-vga n
On 25/05/2023 11:44, Peter Maydell wrote:
On Thu, 25 May 2023 at 11:32, Thomas Huth wrote:
On 25/05/2023 12.05, Peter Maydell wrote:
On Tue, 23 May 2023 at 19:41, Thomas Huth wrote:
On 23/05/2023 19.30, Marcin Juszkiewicz wrote:
...
(is there a reason why it can't be disabled with "-vga
Thomas Huth writes:
> The arm "virt" machine needs "virtio-blk-pci" for devices that get attached
> via the "-cdrom" option. Since this is an optional device that might not
> be available in the binary, we should check for the availability of this
> device first before using it.
>
We're still pr
Thomas Huth writes:
> Though we are already using CONFIG_RTL8139_PCI in the meson.build file
> for testing whether the rtl8139 device is available or not, this is not
> enough: The CONFIG switch might have been selected by another target
> (e.g. the mips fuloong2e machine has the rtl8139 chip sol
Alex Bennée writes:
> This makes it a little easier for developers to find where things
> where being generated.
>
> Reviewed-by: Richard Henderson
> Signed-off-by: Alex Bennée
> Message-Id: <20230523125000.3674739-5-alex.ben...@linaro.org>
> ---
> scripts/qapi/gen.py | 8 ++--
> 1 file ch
On Tue, 25 Apr 2023 08:35:38 +0200
Klaus Jensen wrote:
> From: Klaus Jensen
>
> Add an abstract MCTP over I2C endpoint model. This implements MCTP
> control message handling as well as handling the actual I2C transport
> (packetization).
>
> Devices are intended to derive from this and impleme
Hi Peter,
See inline.
>-Original Message-
>From: Peter Xu
>Sent: Thursday, May 25, 2023 12:59 AM
>Subject: Re: [PATCH] intel_iommu: Optimize out some unnecessary UNMAP
>calls
>
>Hi, Zhenzhong,
>
>On Tue, May 23, 2023 at 04:07:02PM +0800, Zhenzhong Duan wrote:
>> Commit 63b88968f1 ("intel-
On Tue, 25 Apr 2023 08:35:39 +0200
Klaus Jensen wrote:
> From: Matt Johnston
>
> The Null Destination ID, 0, is used for MCTP control messages when
> addressing by physical ID. That is used for Get Endpoint ID and
> Set Endpoint ID when querying/assigning an EID to an endpoint.
>
> Signed-off-
On 5/24/23 11:54, Philippe Mathieu-Daudé wrote:
When multiple QOM types are registered in the same file, it
is clearer and simpler to use the the DEFINE_TYPES() macro.
Add a rule to checkpatch.pl to suggest using DEFINE_TYPES()
instead of type_init() / type_register_static().
Nice!
Sugge
On May 25 12:27, Jonathan Cameron wrote:
> On Tue, 25 Apr 2023 08:35:38 +0200
> Klaus Jensen wrote:
>
> > From: Klaus Jensen
> >
> > Add an abstract MCTP over I2C endpoint model. This implements MCTP
> > control message handling as well as handling the actual I2C transport
> > (packetization).
On Tue, 25 Apr 2023 08:35:40 +0200
Klaus Jensen wrote:
> From: Klaus Jensen
>
> Add the 'nmi-i2c' device that emulates an NVMe Management Interface
> controller.
>
> Initial support is very basic (Read NMI DS, Configuration Get).
>
> This is based on previously posted code by Padmakar Kalghat
Alex Bennée writes:
> I don't think I can remove the parameters directly but certainly mark
> them as deprecated.
>
> Reviewed-by: Stefan Hajnoczi
> Reviewed-by: Richard Henderson
> Reviewed-by: Philippe Mathieu-Daudé
> Signed-off-by: Alex Bennée
> Message-Id: <20230523125000.3674739-6-alex.b
On Thu, 25 May 2023 at 12:06, Mark Cave-Ayland
wrote:
>
> On 25/05/2023 11:44, Peter Maydell wrote:
>
> > On Thu, 25 May 2023 at 11:32, Thomas Huth wrote:
> >
> >> So there needs to be at least some logic dealing with vga_interface_type if
> >> we want to be able to select a different graphics ca
On Fri, 19 May 2023 16:58:01 +0800
Li Zhijian wrote:
> The 'Notes:' in this document mentioned CFMW{0-2}, but the figure missed
> CFMW2.
>
> Signed-off-by: Li Zhijian
> ---
> I'm totally new to CXL, so i have little confidence to this change :)
I believe this one is already fixed upstream by B
In preparation for including the number of dirty pages in the
vfio_get_dirty_bitmap() tracepoint, return the number of dirty pages in
cpu_physical_memory_set_dirty_lebitmap() similar to
cpu_physical_memory_sync_dirty_bitmap().
To avoid counting twice when GLOBAL_DIRTY_RATE is enabled, stash the
nu
Include the number of dirty pages on the vfio_get_dirty_bitmap tracepoint.
These are fetched from the newly added return value in
cpu_physical_memory_set_lebitmap().
Signed-off-by: Joao Martins
---
hw/vfio/common.c | 7 ---
hw/vfio/trace-events | 2 +-
2 files changed, 5 insertions(+), 4
Hey,
This tiny series changes the tracepoint to include the number of
dirty pages via the vfio_get_dirty_bitmap. I find it useful for
observability in general to understand the number of dirty pages in an
IOVA range. With dirty tracking supported by device or IOMMU it's specially
relevant data to
On 25/5/23 12:28, Paolo Bonzini wrote:
ARCH is always empty, so just define HOST_ARCH as the result of uname.
Signed-off-by: Paolo Bonzini
---
tests/docker/Makefile.include | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
Reviewed-by: Philippe Mathieu-Daudé
On 25/05/2023 13.39, Peter Maydell wrote:
On Thu, 25 May 2023 at 12:06, Mark Cave-Ayland
wrote:
On 25/05/2023 11:44, Peter Maydell wrote:
On Thu, 25 May 2023 at 11:32, Thomas Huth wrote:
So there needs to be at least some logic dealing with vga_interface_type if
we want to be able to sele
On Fri, 19 May 2023 16:58:02 +0800
Li Zhijian wrote:
> Using the same style except the 'Topo' abbreviation.
>
> Signed-off-by: Li Zhijian
> ---
> I'm not a native speaker, feel free to correct me.
I've edited slightly and applied to my local staging tree for cxl patches.
Includes fixing docs/c
Hi Joao,
On 25/5/23 13:43, Joao Martins wrote:
Include the number of dirty pages on the vfio_get_dirty_bitmap tracepoint.
These are fetched from the newly added return value in
cpu_physical_memory_set_lebitmap().
Signed-off-by: Joao Martins
---
hw/vfio/common.c | 7 ---
hw/vfio/trac
The vinsgr2vr/vpickve2gr instructions need use get_src/get_dst to get
gpr registers value, not cpu_gpr[]. The $zero register does not
have cpu_gpr[0] allocated.
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1662
Signed-off-by: Song Gao
---
target/loongarch/insn_trans/trans_lsx.c.inc |
Hi Tommy,
On 23/5/23 10:49, Tommy Wu wrote:
The watchdog timer is in the always-on domain device of HiFive 1 rev b,
so this patch added the AON device to the sifive_e machine. This patch
only implemented the functionality of the watchdog timer.
Signed-off-by: Tommy Wu
Reviewed-by: Frank Chang
On 23/5/23 10:49, Tommy Wu wrote:
Create the AON device when we realize the sifive_e machine.
This patch only implemented the functionality of the watchdog timer,
not all the functionality of the AON device.
Signed-off-by: Tommy Wu
Reviewed-by: Frank Chang
---
hw/riscv/Kconfig|
On Wed, 24 May 2023 at 09:06, Thomas Huth wrote:
>
> pflash-cfi02-test.c always uses the "musicpal" machine for testing,
> test-arm-mptimer.c always uses the "vexpress-a9" machine, and
> microbit-test.c requires the "microbit" machine, so we should only
> run these tests if the machines have been
On 5/22/23 10:11, Tommy Wu wrote:
Signed-off-by: Frank Chang
Signed-off-by: Tommy Wu
---
hw/riscv/riscv_hart.c | 21 +
include/hw/riscv/riscv_hart.h | 4
target/riscv/cpu.c| 14 ++
target/riscv/cpu.h| 7 +++
tar
On 5/22/23 10:11, Tommy Wu wrote:
Signed-off-by: Frank Chang
Signed-off-by: Tommy Wu
---
Reviewed-by: Daniel Henrique Barboza
target/riscv/cpu.c | 5 +++
target/riscv/cpu.h | 4 ++
target/riscv/cpu_bits.h | 11 ++
target/riscv/csr.c | 82 +++
On Mon, 22 May 2023 at 16:32, Philippe Mathieu-Daudé wrote:
>
> Do not transmit characters when UART or transmitter are disabled.
>
> Signed-off-by: Philippe Mathieu-Daudé
Last time somebody tried to add checks on the tx/rx enable bits
for the PL011 it broke 'make check' because the hand-rolled
在 2023/5/24 上午2:51, David Hildenbrand 写道:
Let's factor the common setup out, to prepare for further changes.
On arm64, we'll add the subregion to system RAM now earlier -- which
shouldn't matter, because the system RAM memory region should already be
alive at that point.
Signed-off-by: David
On 5/22/23 10:11, Tommy Wu wrote:
Signed-off-by: Frank Chang
Signed-off-by: Tommy Wu
---
target/riscv/cpu_helper.c | 60 +++
1 file changed, 55 insertions(+), 5 deletions(-)
diff --git a/target/riscv/cpu_helper.c b/target/riscv/cpu_helper.c
index cc789
On Wed, 24 May 2023 at 20:01, Richard Henderson
wrote:
>
> On 5/24/23 07:58, Philippe Mathieu-Daudé wrote:
> > The 'break' statement terminates the execution of the nearest
> > enclosing 'for' statement in which it appears.
> >
> > Signed-off-by: Philippe Mathieu-Daudé
> > ---
> > hw/arm/realvi
Markus Armbruster writes:
> Alex Bennée writes:
>
>> I don't think I can remove the parameters directly but certainly mark
>> them as deprecated.
>>
>> Reviewed-by: Stefan Hajnoczi
>> Reviewed-by: Richard Henderson
>> Reviewed-by: Philippe Mathieu-Daudé
>> Signed-off-by: Alex Bennée
>> Mes
On Thu, May 25, 2023 at 12:51:46PM +0300, Avihai Horon wrote:
> However, there is one issue -- we can't send the ACK up here [1], as at that
> point the return path has not been created yet.
> A possible solution is to check for mis->switchover_ack_pending_num == 0
> when we create the return path
This series started with the last patch, an iotest that tests the fixes
for .bdrv_co_create() that were made in commit b2ab5f545f ('block:
bdrv/blk_co_unref() for calls in coroutine context').
Unfortunately, it turned out that much more is wrong about creating
images in an iothread. In particular,
qcow2_open() doesn't work correctly when opening the 'file' child moves
bs to an iothread, for several reasons:
- It uses BDRV_POLL_WHILE() to wait for the qcow2_open_entry()
coroutine, which involves dropping the AioContext lock for bs when it
is not in the main context - but we don't hold it
bdrv_open_backing_file() calls bdrv_open_inherit(), so all callers must
hold the main AioContext lock.
Signed-off-by: Kevin Wolf
---
block.c| 2 ++
block/mirror.c | 6 ++
2 files changed, 8 insertions(+)
diff --git a/block.c b/block.c
index 79bc9c01de..be9ae364fb 100644
--- a/block.
These functions specify that the caller must hold the "@filename
AioContext lock". This doesn't make sense, file names don't have an
AioContext. New BlockDriverStates always start in the main AioContext,
so this is what we really need here.
Signed-off-by: Kevin Wolf
---
block.c | 10 ++
The function documentation already says that all callers must hold the
main AioContext lock, but not all of them do. This can cause assertion
failures when functions called by bdrv_open() try to drop the lock. Fix
a few more callers to take the lock before calling bdrv_open().
Signed-off-by: Kevin
If blockdev-create references an existing node in an iothread (e.g. as
it's 'file' child), then suddenly all of the image creation code must
run in that AioContext, too. Test that this actually works.
Signed-off-by: Kevin Wolf
---
tests/qemu-iotests/tests/iothreads-create | 67 ++
When opening the 'file' child moves bs to an iothread, we need to hold
the AioContext lock of it before we can call raw_apply_options() (and
more specifically, bdrv_getlength() inside of it).
Signed-off-by: Kevin Wolf
---
block/raw-format.c | 5 +
tests/unit/test-block-iothread
This fixes blk_new_open() to not assume that bs is in the main context.
In particular, the BlockBackend must be created with the right
AioContext because it will refuse to move to a different context
afterwards. (blk->allow_aio_context_change is false.)
Use this opportunity to use blk_insert_bs()
While calling bdrv_new_open_driver_opts(), the main AioContext lock must
be held, not the lock of the AioContext of the block subtree it will be
added to afterwards.
Signed-off-by: Kevin Wolf
---
block.c | 11 +++
1 file changed, 11 insertions(+)
diff --git a/block.c b/block.c
index e36
It has no internal callers, so its only use is being called from
individual test cases. If the name starts with an underscore, it is
considered private and linters warn against calling it. 256 only gets
away with it currently because it's on the exception list for linters.
Signed-off-by: Kevin Wol
All of the functions that currently take a BlockDriverState, BdrvChild
or BlockBackend as their first parameter expect the associated
AioContext to be locked when they are called. In the case of
no_co_wrappers, they are called from bottom halves directly in the main
loop, so no other caller can be
bdrv_refresh_total_sectors() and bdrv_refresh_limits() expect to be
called under the AioContext lock of the node. Take the lock.
Signed-off-by: Kevin Wolf
---
block.c | 7 +++
1 file changed, 7 insertions(+)
diff --git a/block.c b/block.c
index be9ae364fb..e368a43761 100644
--- a/block.c
++
The AioContext lock must not be held for bdrv_open_child(), but it is
necessary for the followig operations, in particular those using nested
event loops in coroutine wrappers.
Temporarily dropping the main AioContext lock is not necessary because
we know we run in the main thread.
Signed-off-by:
On 5/22/23 10:11, Tommy Wu wrote:
Signed-off-by: Frank Chang
Signed-off-by: Tommy Wu
---
target/riscv/helper.h | 1 +
target/riscv/insn32.decode| 3 ++
.../riscv/insn_trans/trans_privileged.c.inc | 12 +
target/riscv/op_helper.c
Peter Maydell writes:
> On Mon, 22 May 2023 at 16:32, Philippe Mathieu-Daudé
> wrote:
>>
>> Do not transmit characters when UART or transmitter are disabled.
>>
>> Signed-off-by: Philippe Mathieu-Daudé
>
> Last time somebody tried to add checks on the tx/rx enable bits
> for the PL011 it bro
On Wed, May 24, 2023 at 11:25 AM Jason Wang wrote:
>
> On Sat, May 20, 2023 at 1:50 AM Viktor Prutyanov wrote:
> >
> > On Thu, May 18, 2023 at 9:14 AM Jason Wang wrote:
> > >
> > > On Fri, May 12, 2023 at 9:51 PM Viktor Prutyanov
> > > wrote:
> > > >
> > > > The guest can disable or never enab
On Thu, 25 May 2023 at 13:52, Alex Bennée wrote:
>
>
> Peter Maydell writes:
>
> > On Mon, 22 May 2023 at 16:32, Philippe Mathieu-Daudé
> > wrote:
> >>
> >> Do not transmit characters when UART or transmitter are disabled.
> >>
> >> Signed-off-by: Philippe Mathieu-Daudé
> >
> > Last time someb
在 2023/5/23 下午7:04, Thomas Huth 写道:
Mark the default NIC via the new MachineClass->default_nic setting
so that the machine-defaults code in vl.c can decide whether the
default NIC is usable or not (for example when compiling with the
"--without-default-devices" configure switch).
Signed-off-b
When IOMMU and vhost are enabled together, QEMU tracks IOTLB or
Device-TLB unmap events depending on whether Device-TLB is enabled. But
even if Device-TLB and PCI ATS is enabled, the guest can reject to use
it. For example, this situation appears when Windows Server 2022 is
running with intel-iommu
The guest can disable or never enable Device-TLB. In these cases,
it can't be used even if enabled in QEMU. So, check Device-TLB state
before registering IOMMU notifier and select unmap flag depending on
that. Also, implement a way to change IOMMU notifier flag if Device-TLB
state is changed.
Bugl
If vhost is enabled for virtio-net, Device-TLB enable/disable events
must be passed to vhost for proper IOMMU unmap flag selection.
Signed-off-by: Viktor Prutyanov
---
hw/net/virtio-net.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.c
index 6df6b732
Hi, Prasad,
On Mon, May 22, 2023 at 11:40:21PM +0530, P J P wrote:
> From: Prasad Pandit
>
> vhost_dev_start function does not release memory objects in case
> of an error. This may crash the guest with:
>
> stack trace of thread 125653:
> Program terminated with signal SIGSEGV, Segmentatio
Andrei Gudkov writes:
> Rewrote calc-dirty-rate documentation. Briefly described
> different modes of dirty page rate measurement. Added some
> examples. Fixed obvious grammar errors.
>
> Signed-off-by: Andrei Gudkov
> ---
> qapi/migration.json | 107 +++-
On 5/25/23 03:00, Peter Maydell wrote:
On Wed, 24 May 2023 at 18:38, Richard Henderson
wrote:
We currently print FAIL for the failure of a succ_* test, but don't
return a failure exit code. Instead, convert the script to emit
Test Anything Protocol, which gives visibility into each subtest
as
I noticed that in the latest version, the copyright string is still
2022, even though 2023 is halfway through. This patch fixes that and
fixes the documentation along with it.
Signed-off-by: Enze Li
---
docs/conf.py | 2 +-
include/qemu/help-texts.h | 2 +-
2 files changed, 2 inser
On Thu, May 25, 2023 at 12:43:20PM +0100, Joao Martins wrote:
> In preparation for including the number of dirty pages in the
> vfio_get_dirty_bitmap() tracepoint, return the number of dirty pages in
> cpu_physical_memory_set_dirty_lebitmap() similar to
> cpu_physical_memory_sync_dirty_bitmap().
T
1 - 100 of 362 matches
Mail list logo
