Hi Jason,
I have this and a few other network-related patches not reviewed. Can
you review them?
I have the following patches ready for review:
https://patchew.org/QEMU/20230921094851.36295-1-akihiko.od...@daynix.com/
("[PATCH v2] Revert "tap: setting error appropriately when calling
On 2023/12/11 18:43, Marc-André Lureau wrote:
On Mon, Dec 11, 2023 at 12:51 PM Akihiko Odaki wrote:
ui/vnc may have a bound console so pass it to qemu_console_is_graphic()
and qemu_text_console_put_keysym().
Fixes: 1d0d59fe2919 ("vnc: allow binding servers to qemu consoles")
Signed-off-by:
Hi Marc-André, Paolo,
This patch has Reviewed-by: and Tested-by: but not pulled yet. Can
either of you pull this?
Regards,
Akihiko Odaki
On 2023/12/16 17:03, Akihiko Odaki wrote:
-display lists display backends, but does not tell their options.
Use the help messages from qemu-options.def,
Hi,
This patch series has been forgotten for a while but can still be
applied. Can anyone review it?
Regards,
Akihiko Odaki
On 2023/09/21 16:54, Akihiko Odaki wrote:
qemu_get_runtime_dir() returns a dynamically allocated directory path
that is appropriate for storing runtime files. It
With KVM virtualization, debug exception is passthrough to
to guest kernel rather than host mode. Here hypercall
instruction with special hypercall code is used for sw
breakpoint usage.
Now only software breakpoint is supported, and itt is allowed
to insert/remove software breakpoint. Later
Commit 03fcbd9dc508 ("qdev: Check for the availability of a hotplug
controller before adding a device") says:
> The qdev_unplug() function contains a g_assert(hotplug_ctrl)
> statement, so QEMU crashes when the user tries to device_add +
> device_del a device that does not have a corresponding
On 2023/12/20 16:53, Markus Armbruster wrote:
Akihiko Odaki writes:
On 2023/12/18 23:02, Markus Armbruster wrote:
Akihiko Odaki writes:
On 2023/12/11 15:51, Markus Armbruster wrote:
Akihiko Odaki writes:
The HotplugHandler of the machine will be used when the parent bus does
not
The renamed state will not only represent powering state of PFs, but
also represent SR-IOV VF enablement in the future.
Signed-off-by: Akihiko Odaki
---
include/hw/pci/pci.h| 7 ++-
include/hw/pci/pci_device.h | 2 +-
hw/pci/pci.c| 14 +++---
The spec does not NumVFs is reset after disabling VFs except when
resetting the PF. Clearing it is guest visible and out of spec, even
though Linux doesn't rely on this value being preserved, so we never
noticed.
Fixes: 7c0fa8dff811 ("pcie: Add support for Single Root I/O Virtualization
pcie_sriov_pf_disable_vfs() is called when resetting the PF, but it only
disables VFs and does not reset SR-IOV extended capability, leaking the
state and making the VF Enable register inconsistent with the actual
state.
Replace pcie_sriov_pf_disable_vfs() with pcie_sriov_pf_reset(), which
does
Disable SR-IOV VF devices by reusing code to power down PCI devices
instead of removing them when the guest requests to disable VFs. This
allows to realize devices and report VF realization errors at PF
realization time.
Signed-off-by: Akihiko Odaki
---
docs/pcie_sriov.txt | 8 ++--
I submitted a RFC series[1] to add support for SR-IOV emulation to
virtio-net-pci. During the development of the series, I fixed some
trivial bugs and made improvements that I think are independently
useful. This series extracts those fixes and improvements from the RFC
series.
[1]:
Call pcie_sriov_pf_reset() from pci_do_device_reset() just as we do
for msi_reset() and msix_reset() to prevent duplicating code for each
SR-IOV PF.
Signed-off-by: Akihiko Odaki
---
hw/net/igb.c | 2 --
hw/nvme/ctrl.c | 4
hw/pci/pci.c | 1 +
3 files changed, 1 insertion(+), 6
nvme_sriov_pre_write_ctrl() used to directly inspect SR-IOV
configurations to know the number of VFs being disabled due to SR-IOV
configuration writes, but the logic was flawed and resulted in
out-of-bound memory access.
It assumed PCI_SRIOV_NUM_VF always has the number of currently enabled
VFs,
It is no longer used.
Signed-off-by: Akihiko Odaki
Reviewed-by: Philippe Mathieu-Daudé
---
include/hw/qdev-core.h | 4
hw/core/qdev.c | 1 -
system/qdev-monitor.c | 12 +++-
3 files changed, 7 insertions(+), 10 deletions(-)
diff --git a/include/hw/qdev-core.h
The guest may write NumVFs greater than TotalVFs and that can lead
to buffer overflow in VF implementations.
Cc: qemu-sta...@nongnu.org
Fixes: 7c0fa8dff811 ("pcie: Add support for Single Root I/O Virtualization
(SR/IOV)")
Signed-off-by: Akihiko Odaki
---
hw/pci/pcie_sriov.c | 3 +++
1 file
vfio determines if rombar is explicitly enabled by inspecting QDict.
Inspecting QDict is not nice because QDict is untyped and depends on the
details on the external interface. Add an infrastructure to determine if
rombar is explicitly enabled to hw/pci.
Signed-off-by: Akihiko Odaki
---
Use pci_rom_bar_explicitly_enabled() to determine if rombar is explicitly
enabled.
Signed-off-by: Akihiko Odaki
---
hw/vfio/pci.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c
index 4fa387f0430d..647f15b2a060 100644
--- a/hw/vfio/pci.c
+++
Currently there is no way to distinguish the case that rombar is
explicitly specified as 1 and the case that rombar is not specified.
Set rombar -1 by default to distinguish these cases just as it is done
for addr and romsize. It was confirmed that changing the default value
to -1 will not change
Hi Michael,
Understood. Thanks.
Best Regards,
Rick Zhong
-邮件原件-
发件人: Michael S. Tsirkin
发送时间: 2024年2月13日 17:53
收件人: Rick Zhong
抄送: Eugenio Perez Martin ; qemu-devel@nongnu.org; Jason
Wang ; Peter Xu ; Guo Zhi
; Xinying Yu ; Wentao Jia
; Shujing Dong ; Kyle Xu
主题: Re: 回复: FW:
Still wondering if there are any plans to apply my patch or if you would like
to change anything in the patch?
Being able to set parameters in smb.conf would be really useful these days for
people running old versions of Windows like Windows XP in a qemu guest. Today,
the default settings of
On 2/16/24 05:35, Philippe Mathieu-Daudé wrote:
sysbus_address_space() is not more used, remove it.
Signed-off-by: Philippe Mathieu-Daudé
---
include/hw/sysbus.h | 1 -
hw/core/sysbus.c| 5 -
2 files changed, 6 deletions(-)
Reviewed-by: Richard Henderson
r~
On 2/16/24 05:35, Philippe Mathieu-Daudé wrote:
sysbus_address_space(...) is a simple wrapper to
get_system_memory(). Use it in place, since KVM
VAPIC doesn't distinct address spaces.
Rename the 'as' variable as 'mr' since it is a
MemoryRegion type, not an AddressSpace one.
Signed-off-by:
On 2/16/24 05:35, Philippe Mathieu-Daudé wrote:
Add the Exynos4210fimdState::'framebuffer-memory' property. Have
the board set it. We don't need to call sysbus_address_space()
anymore.
Signed-off-by: Philippe Mathieu-Daudé
---
hw/display/exynos4210_fimd.c | 19 ---
1 file
On 2/16/24 05:35, Philippe Mathieu-Daudé wrote:
We want to set another qdev property (a link) for the FIMD
device, we can not use sysbus_create_varargs() which only
passes sysbus base address and IRQs as arguments. Inline
it so we can set the link property in the next commit.
Signed-off-by:
On 2/16/24 05:35, Philippe Mathieu-Daudé wrote:
Add the PL110::'framebuffer-memory' property. Have the different
ARM boards set it. We don't need to call sysbus_address_space()
anymore.
Signed-off-by: Philippe Mathieu-Daudé
---
hw/arm/realview.c| 2 ++
hw/arm/versatilepb.c | 2 ++
On 2/16/24 05:35, Philippe Mathieu-Daudé wrote:
We want to set another qdev property (a link) for the pl110
and pl111 devices, we can not use sysbus_create_simple() which
only passes sysbus base address and IRQs as arguments. Inline
it so we can set the link property in the next commit.
On 2/16/24 03:05, Ilya Leoshkevich wrote:
The upcoming follow-fork-mode child support requires advertising the
fork-events feature, which is user-specific. Introduce a user-specific
hook for this.
Signed-off-by: Ilya Leoshkevich
---
gdbstub/gdbstub.c | 12 +---
gdbstub/internals.h
On 2/16/24 03:05, Ilya Leoshkevich wrote:
The upcoming follow-fork-mode child support requires post-fork message
exchange between the parent and the child. Prepare gdbserver_fork() for
this purpose. Rename it to gdbserver_fork_end() to better reflect its
purpose.
Signed-off-by: Ilya Leoshkevich
On 2/16/24 03:05, Ilya Leoshkevich wrote:
The upcoming follow-fork-mode child support requires knowing the child
pid. Pass it down.
Signed-off-by: Ilya Leoshkevich
---
bsd-user/main.c| 2 +-
gdbstub/user.c | 2 +-
include/gdbstub/user.h | 2 +-
linux-user/main.c | 2 +-
On 2/16/24 03:05, Ilya Leoshkevich wrote:
diff --git a/linux-user/user-internals.h b/linux-user/user-internals.h
index c63ef45fc78..9014014d920 100644
--- a/linux-user/user-internals.h
+++ b/linux-user/user-internals.h
@@ -71,7 +71,7 @@ const char *target_strerror(int err);
int
On 2/16/24 07:45, Alex Bennée wrote:
Ilya Leoshkevich writes:
Currently ts_tid contains the parent tid after fork(), which is not
correct. So far it has not affected anything, but the upcoming
follow-fork-mode child support relies on the correct value, so fix it.
Signed-off-by: Ilya
On 2/16/24 03:05, Ilya Leoshkevich wrote:
The upcoming follow-fork-mode child support will require disabling
gdbstub in the parent process, which may have multiple threads (which
are represented as CPUs).
Loop over all CPUs in order to remove breakpoints and disable
single-step. Move the
On 2/17/24 00:53, Daniel Henrique Barboza wrote:
This patch is replacing mark_vs_dirty() with finalize(), that does call
mark_vs_dirty() and
set start_eq_zero = true, but it's missing the start_eq_zero update for store
functions
because of these ifs.
We could just remove these ifs and
We want to add a RISC-V 'virt' libqos machine to increase our test
coverage. Some of the tests will try to plug a virtio-iommu-pci
device into the board and do some tests with it.
Enable virtio-iommu-pci in the 'virt' machine.
Signed-off-by: Daniel Henrique Barboza
Acked-by: Alistair Francis
Add a RISC-V 'virt' machine to the graph. This implementation is a
modified copy of the existing arm machine in arm-virt-machine.c
It contains a virtio-mmio and a generic-pcihost controller. The
generic-pcihost controller hardcodes assumptions from the ARM 'virt'
machine, like ecam and pio_base
Hotplugged FDT nodes will attempt to write this node that, at this
moment, is being created only in create_fdt_pcie() during
finalize_fdt().
Create it earlier.
Signed-off-by: Daniel Henrique Barboza
Reviewed-by: Alistair Francis
---
hw/riscv/virt.c | 9 -
1 file changed, 8
The loop isn't setting the values for the last element. Every other
element is being initialized with addr = 0, flags = VRING_DESC_F_NEXT
and next = i + 1. The last elem is never touched.
This became a problem when enabling a RISC-V 'virt' libqos machine in
the 'indirect' test of
In qvring_init() we're writing vq->used->avail_event at "vq->used + 2 +
array_size". The struct pointed by vq->used is, from virtio_ring.h
Linux header):
* // A ring of used descriptor heads with free-running index.
* __virtio16 used_flags;
* __virtio16 used_idx;
* struct
The 'virt' machine makes assumptions on the Advanced Core-Local
Interruptor, or aclint, based on 'tcg_enabled()' conditionals. This
will impact MSI related tests support when adding a RISC-V 'virt' libqos
machine. The accelerator used in that case, 'qtest', isn't being
accounted for and we'll
Hi,
This second version was rebased with current
alistair/riscv.to.apply-next and has typo fixes in patch 1. No other
changes were made.
All patches reviewed/acked.
Changes from v1:
- patch 1: typos in the commit message fixed
- v1 link:
On 2/16/24 23:49, Alexander Monakov wrote:
On Fri, 16 Feb 2024, Richard Henderson wrote:
Benchmark each acceleration function vs an aligned buffer of zeros.
Signed-off-by: Richard Henderson
---
+
+static void test(const void *opaque)
+{
+size_t len = 64 * KiB;
This exceeds L1 cache
On 2/17/24 01:33, Alexander Monakov wrote:
On Fri, 16 Feb 2024, Richard Henderson wrote:
Because non-embedded aarch64 is expected to have AdvSIMD enabled, merely
double-check with the compiler flags for __ARM_NEON and don't bother with
a runtime check. Otherwise, model the loop after the x86
On 2/17/24 02:13, Alexander Monakov wrote:
On Fri, 16 Feb 2024, Richard Henderson wrote:
Split less-than and greater-than 256 cases.
Use unaligned accesses for head and tail.
Avoid using out-of-bounds pointers in loop boundary conditions.
I guess it did not carry
typedef uint64_t
Hi, I am just starting out on QEMU development, and I would like to know which
tools to use. I would really appreciate it.
Additionally, how is the support for LLVM toolchain on QEMU?
Best regards
*
Isaac
Hi Richard,
After I tried to reduce the overhead of unnecessary segment flow and
memory plugin callbacks, I observed that there several functions consume
most of runtime from perf report.
So these three inline patches and previous patch were created to reduce
the overhead without modifying
macOS Sonoma changes the NSView.clipsToBounds to false by default where it was
true in
earlier version of macOS. This causes the window contents to be obscured by the
window
frame. This fixes the issue by conditionally setting the clipping on Sonoma to
true.
Resolves:
On 2/17/24 01:06, Michael Tokarev wrote:
28.02.2023 20:11, Guenter Roeck wrote:
Host drivers do not necessarily set cdb_len in megasas io commands.
With commits 6d1511cea0 ("scsi: Reject commands if the CDB length
exceeds buf_len") and fe9d8927e2 ("scsi: Add buf_len parameter to
configure --help currently outputs the following line for the tools
option:
-->8---
░░tcg░TCG░support░░
tools build support utilities that come with QEMU
On Fri, 16 Feb 2024, Richard Henderson wrote:
> Split less-than and greater-than 256 cases.
> Use unaligned accesses for head and tail.
> Avoid using out-of-bounds pointers in loop boundary conditions.
I guess it did not carry
typedef uint64_t uint64_a __attribute__((may_alias));
along the
On 2/16/24 20:30, Alex Bennée wrote:
With the new plugin register API we can now track changes to register
values. Currently the implementation is fairly dumb which will slow
down if a large number of register values are being tracked. This
could be improved by only instrumenting instructions
On Fri, 16 Feb 2024, Richard Henderson wrote:
> Because non-embedded aarch64 is expected to have AdvSIMD enabled, merely
> double-check with the compiler flags for __ARM_NEON and don't bother with
> a runtime check. Otherwise, model the loop after the x86 SSE2 function,
> and use VADDV to
On 2024/02/15 0:55, Michael S. Tsirkin wrote:
On Wed, Feb 14, 2024 at 06:53:43PM +0300, Michael Tokarev wrote:
Nope, I don't remember how to request a CVE ;)
https://www.qemu.org/contribute/security-process/
Thanks, I requested CVEs with the form.
QEMU doesn't have any list of features
Evaluate [normalWindow styleMask] & NSWindowStyleMaskResizable instead.
Signed-off-by: Akihiko Odaki
---
ui/cocoa.m | 11 ---
1 file changed, 4 insertions(+), 7 deletions(-)
diff --git a/ui/cocoa.m b/ui/cocoa.m
index 81de8d92669b..401ed0c3f1f5 100644
--- a/ui/cocoa.m
+++ b/ui/cocoa.m
ui/cocoa used to release all mouse buttons when it sees
NSEventTypeLeftMouseUp, NSEventTypeRightMouseUp, or
NSEventTypeOtherMouseUp, but it can instead release specific one
according to the delivered event.
Signed-off-by: Akihiko Odaki
---
ui/cocoa.m | 132
Core Graphics is not accelerated and slow.
Signed-off-by: Akihiko Odaki
---
ui/cocoa.m | 24 ++--
1 file changed, 10 insertions(+), 14 deletions(-)
diff --git a/ui/cocoa.m b/ui/cocoa.m
index fe0eb74b0743..b26669160300 100644
--- a/ui/cocoa.m
+++ b/ui/cocoa.m
@@ -496,10
console_select() can be called anytime so explicitly take the BQL.
Signed-off-by: Akihiko Odaki
---
ui/cocoa.m | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/ui/cocoa.m b/ui/cocoa.m
index 17f290987563..81de8d92669b 100644
--- a/ui/cocoa.m
+++ b/ui/cocoa.m
@@ -1391,7
The window will be resizable when zoom-to-fit is on.
Signed-off-by: Akihiko Odaki
---
ui/cocoa.m | 3 +++
1 file changed, 3 insertions(+)
diff --git a/ui/cocoa.m b/ui/cocoa.m
index 0ebe0f326373..17f290987563 100644
--- a/ui/cocoa.m
+++ b/ui/cocoa.m
@@ -1379,8 +1379,10 @@ - (void)zoomToFit:(id)
It allows making the window full screen by clicking full screen button
provided by the platform (the left-top green button) and save some code.
Signed-off-by: Akihiko Odaki
---
ui/cocoa.m | 423 ++---
1 file changed, 207 insertions(+), 216
V5 -> V6:
Rebased.
Signed-off-by: Akihiko Odaki
---
Changes in v11:
- Ensured to reflect screen size changes to bounds
- Link to v10:
https://lore.kernel.org/r/20240214-cocoa-v10-0-c7ffe5d9f...@daynix.com
Changes in v10:
- Removed relative mouse input scaling.
- Link to v9:
Richard,
On 12/14/23 08:18, Ivan Klokov wrote:
The vstart_qe_zero flag is set at the beginning of the translation
phase from the env->vstart variable. During the execution phase, some
instructions may change env->vstart, but the flag remains the same as
at the start of the block. With some
The interrupt handlers need to be populated before the device is realized since
internal devices such as the RTC are wired during realize(). If the interrupt
handlers aren't populated, devices such as the RTC will be wired with a NULL
interrupt handler, i.e. MC146818RtcState::irq is NULL.
Fixes:
Hi Daniel,
Thank you for the information and suggestion.
Yes, we can do it better if we load/store more bytes at a time.
I'll try to improve the RFC on this way.
Thanks,
Max
On 2024/2/16 5:11 AM, Daniel Henrique Barboza wrote:
On 2/15/24 16:28, Max Chou wrote:
In the vector unit-stride
The cdrom test skips to execute on LoongArch system with command
"make check", this patch enables cdrom test for LoongArch virt
machine platform.
With this patch, cdrom test passes to run on LoongArch virt
machine type.
Signed-off-by: Bibo Mao
---
Add virtio-blk-pci checking since LoongArch
Hi Richard,
Thank you for the suggestion and the reference.
I'm trying to follow the reference to implement it and I'll send another
version for this.
Thanks a lot,
Max
On 2024/2/16 4:24 AM, Richard Henderson wrote:
On 2/15/24 09:28, Max Chou wrote:
Hi all,
When glibc with RVV support
On Fri, Feb 16, 2024 at 12:14 PM Philippe Mathieu-Daudé
wrote:
>
> On 13/2/24 16:49, Paolo Bonzini wrote:
> > Allow boards to use the device creation functions even if USB itself
> > is not available; of course the functions will fail inexorably, but
> > this can be okay if the calls are
On Fri, 16 Feb 2024, Richard Henderson wrote:
> Benchmark each acceleration function vs an aligned buffer of zeros.
>
> Signed-off-by: Richard Henderson
> ---
> +
> +static void test(const void *opaque)
> +{
> +size_t len = 64 * KiB;
This exceeds L1 cache capacity, so the performance
Hi Daniel,
I think the the overhead from the plugin callbacks depending on the type
of plugin callbacks.
Because there is only qemu_plugin_vcpu_mem_cb in the perf report of the
glibc memcpy benchtest, so I just looked it up for the the memory
callbacks in this RFC.
I'll try to get more
15.01.2024 05:08, guoguangyao wrote:
When closing PCREL, qemu-system-x86_64 run into error.
Eip modification here leads to the result. Using s->pc
in func gen_update_eip_next() solves the problem.
Fixes: b5e0d5d22fbf("target/i386: Fix 32-bit wrapping of pc/eip computation")
It looks like the
26.07.2023 05:07, Jason Wang wrote:
On Wed, Jul 26, 2023 at 2:21 AM Eugenio Pérez wrote:
The device already has a virtio status set by vhost_vdpa_init by the
time vhost_vdpa_probe_cvq_isolation is called. vhost_vdpa_init set
S_ACKNOWLEDGE and S_DRIVER, so it is invalid to just reset it.
It
17.05.2023 22:48, Stefano Stabellini :
On Wed, 17 May 2023, Michael Tokarev wrote:
17.05.2023 12:47, Chuck Zmudzinski wrote:
On 5/17/2023 2:39 AM, Michael Tokarev wrote:
08.02.2023 05:03, Chuck Zmudzinski wrote:...
Fixes: 998250e97661 ("xen, gfx passthrough: register host bridge
specific to
Hi Richard,
Thank you for the suggestion. I'll do a v2 with this.
Thanks,
Max
On 2024/2/16 4:03 AM, Richard Henderson wrote:
On 2/15/24 09:28, Max Chou wrote:
If there are not any QEMU plugin memory callback functions, checking
before calling the qemu_plugin_vcpu_mem_cb function can reduce
28.02.2023 20:11, Guenter Roeck wrote:
Host drivers do not necessarily set cdb_len in megasas io commands.
With commits 6d1511cea0 ("scsi: Reject commands if the CDB length
exceeds buf_len") and fe9d8927e2 ("scsi: Add buf_len parameter to
scsi_req_new()"), this results in failures to boot Linux
On 2024/02/17 1:30, Alex Bennée wrote:
We can only request a list of registers once the vCPU has been
initialised so the user needs to use either call the get function on
vCPU initialisation or during the translation phase.
We don't expose the reg number to the plugin instead hiding it behind
74 matches
Mail list logo
