Peter Maydell [Mon, 11 Nov 2019 13:58:47 +0000 (13:58 +0000)]
Merge remote-tracking branch 'remotes/pmaydell/tags/pull-target-arm-20191111' into staging
target-arm queue:
* Remove old unassigned_access CPU hook API
* Remove old ptimer_init_with_bh() API
* hw/arm/boot: Set NSACR.{CP11, CP10} in dummy SMC setup routine
* remotes/pmaydell/tags/pull-target-arm-20191111:
hw/arm/boot: Set NSACR.{CP11, CP10} in dummy SMC setup routine
Remove unassigned_access CPU hook
ptimer: Remove old ptimer_init_with_bh() API
hw/arm/boot: Set NSACR.{CP11, CP10} in dummy SMC setup routine
The boot.c code usually puts the CPU into NS mode directly when it is
booting a kernel. Since fc1120a7f5f2d4b6 this has included a
requirement to set NSACR to give NS state access to the FPU; we fixed
that for the usual code path in ece628fcf6. However, it is also
possible for a board model to request an alternative mode of booting,
where its 'board_setup' code hook runs in Secure state and is
responsible for doing the S->NS transition after it has done whatever
work it must do in Secure state. In this situation the board_setup
code now also needs to update NSACR.
This affects all boards which set info->secure_board_setup, which is
currently the 'raspi' and 'highbank' families. They both use the
common arm_write_secure_board_setup_dummy_smc().
Set the NSACR CP11 and CP10 bits in the code written by that
function, to allow FPU access in Non-Secure state when using dummy
SMC setup routine. Otherwise an AArch32 kernel booted on the
highbank or raspi boards will UNDEF as soon as it tries to use the
FPU.
Update the comment describing secure_board_setup to note the new
requirements on users of it.
This fixes a kernel panic when booting raspbian on raspi2.
Peter Maydell [Mon, 11 Nov 2019 13:44:16 +0000 (13:44 +0000)]
Remove unassigned_access CPU hook
All targets have now migrated away from the old unassigned_access
hook to the new do_transaction_failed hook. This means we can remove
the core-code infrastructure for that hook and the code that calls it.
Peter Maydell [Mon, 11 Nov 2019 13:44:16 +0000 (13:44 +0000)]
ptimer: Remove old ptimer_init_with_bh() API
Now all the users of ptimers have converted to the transaction-based
API, we can remove ptimer_init_with_bh() and all the code paths
that are used only by bottom-half based ptimers, and tidy up the
documentation comments to consider the transaction-based API the
only possibility.
The code changes result from:
* s->bh no longer exists
* s->callback is now always non-NULL
Laurent Vivier [Wed, 6 Nov 2019 11:23:41 +0000 (12:23 +0100)]
dp8393x: fix dp8393x_receive()
RXpkt.in_use is always 16 bit wide, but when the bus access mode is 32bit
and the endianness is big, we must access the second word and not the
first. This patch adjusts the offset according to the size and endianness.
Peter Maydell [Thu, 7 Nov 2019 14:45:36 +0000 (14:45 +0000)]
Merge remote-tracking branch 'remotes/marcel/tags/rdma-pull-request' into staging
RDMA queue
* better memory registration performance
# gpg: Signature made Wed 06 Nov 2019 14:37:47 GMT
# gpg: using RSA key 36D4C0F0CF2FE46D
# gpg: Good signature from "Marcel Apfelbaum <[email protected]>" [marginal]
# gpg: aka "Marcel Apfelbaum <[email protected]>" [marginal]
# gpg: aka "Marcel Apfelbaum <[email protected]>" [marginal]
# gpg: WARNING: This key is not certified with sufficiently trusted signatures!
# gpg: It is not certain that the signature belongs to the owner.
# Primary key fingerprint: B1C6 3A57 F92E 08F2 640F 31F5 36D4 C0F0 CF2F E46D
* remotes/marcel/tags/rdma-pull-request:
hw/rdma: Utilize ibv_reg_mr_iova for memory registration
configure: Check if we can use ibv_reg_mr_iova
Tuguoyi [Fri, 1 Nov 2019 07:37:35 +0000 (07:37 +0000)]
qcow2-bitmap: Fix uint64_t left-shift overflow
There are two issues in In check_constraints_on_bitmap(),
1) The sanity check on the granularity will cause uint64_t
integer left-shift overflow when cluster_size is 2M and the
granularity is BIGGER than 32K.
2) The way to calculate image size that the maximum bitmap
supported can map to is a bit incorrect.
This patch fix it by add a helper function to calculate the
number of bytes needed by a normal bitmap in image and compare
it to the maximum bitmap bytes supported by qemu.
Peter Maydell [Thu, 7 Nov 2019 11:56:18 +0000 (11:56 +0000)]
Merge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging
virtio, pci: fixes
A couple of bugfixes.
Signed-off-by: Michael S. Tsirkin <[email protected]>
# gpg: Signature made Wed 06 Nov 2019 12:00:19 GMT
# gpg: using RSA key 281F0DB8D28D5469
# gpg: Good signature from "Michael S. Tsirkin <[email protected]>" [full]
# gpg: aka "Michael S. Tsirkin <[email protected]>" [full]
# Primary key fingerprint: 0270 606B 6F3C DF3D 0B17 0970 C350 3912 AFBE 8E67
# Subkey fingerprint: 5D09 FD08 71C8 F85B 94CA 8A0D 281F 0DB8 D28D 5469
* remotes/mst/tags/for_upstream:
virtio: notify virtqueue via host notifier when available
hw/i386: AMD-Vi IVRS DMA alias support
pci: Use PCI aliases when determining device IOMMU address space
* remotes/vivier2/tags/linux-user-for-4.2-pull-request:
linux-user/alpha: Set r20 secondary return value
linux-user/sparc: Fix cpu_clone_regs_*
linux-user: Introduce cpu_clone_regs_parent
linux-user: Rename cpu_clone_regs to cpu_clone_regs_child
linux-user/sparc64: Fix target_signal_frame
linux-user/sparc: Fix WREG usage in setup_frame
linux-user/sparc: Use WREG_SP constant in sparc/signal.c
linux-user/sparc: Begin using WREG constants in sparc/signal.c
linux-user/sparc: Use WREG constants in sparc/target_cpu.h
target/sparc: Define an enumeration for accessing env->regwptr
tests/tcg/multiarch/linux-test: Fix error check for shmat
scripts/qemu-binfmt-conf: Update for sparc64
linux-user: Support for NETLINK socket options
Peter Maydell [Wed, 6 Nov 2019 17:52:14 +0000 (17:52 +0000)]
Merge remote-tracking branch 'remotes/philmd-gitlab/tags/mips-next-20191105' into staging
The i440FX northbridge is only used by the PC machine, while the
PIIX southbridge is also used by the Malta MIPS machine.
Split the PIIX3 southbridge from i440FX northbridge.
# gpg: Signature made Tue 05 Nov 2019 22:48:12 GMT
# gpg: using RSA key 89C1E78F601EE86C867495CBA2A3FD6EDEADC0DE
# gpg: Good signature from "Philippe Mathieu-Daudé (Phil) <[email protected]>" [marginal]
# gpg: WARNING: This key is not certified with sufficiently trusted signatures!
# gpg: It is not certain that the signature belongs to the owner.
# Primary key fingerprint: 89C1 E78F 601E E86C 8674 95CB A2A3 FD6E DEAD C0DE
* remotes/philmd-gitlab/tags/mips-next-20191105: (21 commits)
hw/pci-host/i440fx: Remove the last PIIX3 traces
hw/pci-host: Rename incorrectly named 'piix' as 'i440fx'
hw/pci-host/piix: Extract PIIX3 functions to hw/isa/piix3.c
hw/pci-host/piix: Fix code style issues
hw/pci-host/piix: Move i440FX declarations to hw/pci-host/i440fx.h
hw/pci-host/piix: Define and use the PIIX IRQ Route Control Registers
hw/pci-host/piix: Move RCR_IOPORT register definition
hw/pci-host/piix: Extract piix3_create()
hw/i386: Remove obsolete LoadStateHandler::load_state_old handlers
hw/isa/piix4: Move piix4_create() to hw/isa/piix4.c
hw/mips/mips_malta: Extract the PIIX4 creation code as piix4_create()
hw/mips/mips_malta: Create IDE hard drive array dynamically
piix4: Add a MC146818 RTC Controller as specified in datasheet
piix4: Add an i8254 PIT Controller as specified in datasheet
piix4: Add an i8257 DMA Controller as specified in datasheet
piix4: Rename PIIX4 object to piix4-isa
Revert "irq: introduce qemu_irq_proxy()"
piix4: Add an i8259 Interrupt Controller as specified in datasheet
piix4: Add the Reset Control Register
MAINTAINERS: Keep PIIX4 South Bridge separate from PC Chipsets
...
Peter Maydell [Wed, 6 Nov 2019 13:36:42 +0000 (13:36 +0000)]
Merge remote-tracking branch 'remotes/stefanha/tags/block-pull-request' into staging
Pull request
Let's get the image fuzzer Python 3 changes merged in QEMU 4.2.
# gpg: Signature made Tue 05 Nov 2019 15:43:16 GMT
# gpg: using RSA key 8695A8BFD3F97CDAAC35775A9CA4ABB381AB73C8
# gpg: Good signature from "Stefan Hajnoczi <[email protected]>" [full]
# gpg: aka "Stefan Hajnoczi <[email protected]>" [full]
# Primary key fingerprint: 8695 A8BF D3F9 7CDA AC35 775A 9CA4 ABB3 81AB 73C8
* remotes/stefanha/tags/block-pull-request:
image-fuzzer: Use OSerror.strerror instead of tuple subscript
image-fuzzer: Use errors parameter of subprocess.Popen()
image-fuzzer: Run using python3
image-fuzzer: Encode file name and file format to bytes
image-fuzzer: Use bytes constant for field values
image-fuzzer: Return bytes objects on string fuzzing functions
image-fuzzer: Use %r for all fiels at Field.__repr__()
image-fuzzer: Use io.StringIO
image-fuzzer: Explicitly use integer division operator
image-fuzzer: Write bytes instead of string to image file
image-fuzzer: Open image files in binary mode
We failed to set the secondary return value in %o1
we failed to advance the PC past the syscall,
we failed to adjust regwptr into the new structure,
we stored the stack pointer into the wrong register.
We will need a target-specific hook for adjusting registers
in the parent during clone. Add an empty inline function for
each target, and invoke it from the proper places.
linux-user: Rename cpu_clone_regs to cpu_clone_regs_child
We will need a target-specific hook for adjusting registers
in the parent during clone. To avoid confusion, rename the
one we have to make it clear it affects the child.
At the same time, pass in the flags from the clone syscall.
We will need them for correct behaviour for Sparc.
linux-user/sparc: Use WREG_SP constant in sparc/signal.c
s/UREG_FP/WREG_SP/g
This is non-obvious because the UREG_FP constant is fact wrong.
However, the previous search-and-replace patch made it clear that
UREG_FP expands to WREG_O6, and we can see from the enumeration in
target/sparc/cpu.h that WREG_O6 is in fact WREG_SP, the stack pointer.
linux-user/sparc: Use WREG constants in sparc/target_cpu.h
This fixes a naming bug wherein we used "UREG_FP" to access the
stack pointer. OTOH, the "UREG_FP" constant was also defined
incorrectly such that it *did* reference the stack pointer.
Note that the kernel legitimately uses the name "FP", because it
utilizes the rolled stack window in processing the system call.
Gerd Hoffmann [Tue, 15 Oct 2019 06:44:26 +0000 (08:44 +0200)]
usb-host: add option to allow all resets.
Commit 65f14ab98da1 ("usb-host: skip reset for untouched devices")
filters out multiple usb device resets in a row. While this improves
the situation for usb some devices it doesn't work for others :-(
So go add a config option to make the behavior configurable.
Peter Maydell [Wed, 6 Nov 2019 11:56:40 +0000 (11:56 +0000)]
Merge remote-tracking branch 'remotes/mdroth/tags/qga-pull-2019-11-04-tag' into staging
qemu-ga patch queue for hard-freeze
* fix handling of Chinese network device names in
guest-network-get-interfaces
* add missing blacklist entries for guest-get-memory-block-info for
w32/non-linux builds
* remotes/mdroth/tags/qga-pull-2019-11-04-tag:
qga: Add "guest-get-memory-block-info" to blacklist
qga-win: network-get-interfaces command name field bug fix
Stefan Hajnoczi [Tue, 5 Nov 2019 14:09:46 +0000 (15:09 +0100)]
virtio: notify virtqueue via host notifier when available
Host notifiers are used in several cases:
1. Traditional ioeventfd where virtqueue notifications are handled in
the main loop thread.
2. IOThreads (aio_handle_output) where virtqueue notifications are
handled in an IOThread AioContext.
3. vhost where virtqueue notifications are handled by kernel vhost or
a vhost-user device backend.
Most virtqueue notifications from the guest use the ioeventfd mechanism,
but there are corner cases where QEMU code calls virtio_queue_notify().
This currently honors the host notifier for the IOThreads
aio_handle_output case, but not for the vhost case. The result is that
vhost does not receive virtqueue notifications from QEMU when
virtio_queue_notify() is called.
This patch extends virtio_queue_notify() to set the host notifier
whenever it is enabled instead of calling the vq->(aio_)handle_output()
function directly. We track the host notifier state for each virtqueue
separately since some devices may use it only for certain virtqueues.
This fixes the vhost case although it does add a trip through the
eventfd for the traditional ioeventfd case. I don't think it's worth
adding a fast path for the traditional ioeventfd case because calling
virtio_queue_notify() is rare when ioeventfd is enabled.
Yuval Shaia [Sun, 18 Aug 2019 13:21:07 +0000 (16:21 +0300)]
hw/rdma: Utilize ibv_reg_mr_iova for memory registration
The virtual address that is provided by the guest in post_send and
post_recv operations is related to the guest address space. This address
space is unknown to the HCA resides on host so extra step in these
operations is needed to adjust the address to host virtual address.
This step, which is done in data-path affects performances.
An enhanced verion of MR registration introduced here
https://patchwork.kernel.org/patch/11044467/ can be used so that the
guest virtual address space for this MR is known to the HCA in host.
The PIIX3 is not tied to the i440FX and can even be used without it.
Move its creation to the machine code (pc_piix.c).
We have now removed the last trace of southbridge code in the i440FX
northbridge.
hw/pci-host: Rename incorrectly named 'piix' as 'i440fx'
We moved all the PIIX3 southbridge code out of hw/pci-host/piix.c,
it now only contains i440FX northbridge code.
Rename it to match the chipset modelled.
hw/pci-host/piix: Define and use the PIIX IRQ Route Control Registers
The IRQ Route Control registers definitions belong to the PIIX
chipset. We were only defining the 'A' register. Define the other
B, C and D registers, and use them.
These devices implemented their load_state_old() handler 10 years
ago, previous to QEMU v0.12.
Since commit cc425b5ddf removed the pc-0.10 and pc-0.11 machines,
we can drop this code.
Note: the mips_r4k machine started to use the i8254 device just
after QEMU v0.5.0, but the MIPS machine types are not versioned,
so there is no migration compatibility issue removing this handler.
hw/mips/mips_malta: Create IDE hard drive array dynamically
In the next commit we'll refactor the PIIX4 code out of
mips_malta_init(). As a preliminary step, add the 'ide_drives'
variable and create the drive array dynamically.
piix4: Add an i8257 DMA Controller as specified in datasheet
The i8257 is not a chipset on the Malta board, but is part of
the PIIX4 chipset.
Create the i8257 in the PIIX4 code, remove the one instantiated
in malta board, to not have it twice.
Makefile: Fix config-devices.mak not regenerated when Kconfig updated
When hw/$DIR/Kconfig is changed, the corresponding generated
hw/$DIR/config-devices.mak is not being updated.
Fix this by including all the hw/*/Kconfig files to the prerequisite
names of the rule generating the config-devices.mak files.
Peter Maydell [Tue, 5 Nov 2019 20:59:47 +0000 (20:59 +0000)]
Merge remote-tracking branch 'remotes/maxreitz/tags/pull-block-2019-11-04' into staging
Block patches for 4.2-rc0:
- Work around XFS write-zeroes bug in file-posix block driver
- Fix backup job with compression
- Fix to the NVMe block driver header
* remotes/maxreitz/tags/pull-block-2019-11-04:
block/file-posix: Let post-EOF fallocate serialize
block: Add bdrv_co_get_self_request()
block: Make wait/mark serialising requests public
block/block-copy: fix s->copy_size for compressed cluster
nvme: fix NSSRS offset in CAP register
Peter Maydell [Tue, 5 Nov 2019 20:17:11 +0000 (20:17 +0000)]
Merge remote-tracking branch 'remotes/philmd-gitlab/tags/fw_cfg-next-pull-request' into staging
Fix the fw_cfg reboot-timeout=-1 special value, add a test for it.
# gpg: Signature made Sun 03 Nov 2019 22:21:02 GMT
# gpg: using RSA key 89C1E78F601EE86C867495CBA2A3FD6EDEADC0DE
# gpg: Good signature from "Philippe Mathieu-Daudé (Phil) <[email protected]>" [marginal]
# gpg: WARNING: This key is not certified with sufficiently trusted signatures!
# gpg: It is not certain that the signature belongs to the owner.
# Primary key fingerprint: 89C1 E78F 601E E86C 8674 95CB A2A3 FD6E DEAD C0DE
* remotes/philmd-gitlab/tags/fw_cfg-next-pull-request:
tests/fw_cfg: Test 'reboot-timeout=-1' special value
fw_cfg: Allow reboot-timeout=-1 again
Laurent Vivier [Mon, 4 Nov 2019 10:15:13 +0000 (11:15 +0100)]
q800: fix I/O memory map
Linux kernel 5.4 will introduce a new memory map for SWIM device.
(aee6bff1c325 ("m68k: mac: Revisit floppy disc controller base addresses"))
Until this release all MMIO are mapped between 0x50f00000 and 0x50f40000,
but it appears that for real hardware 0x50f00000 is not the base address:
the MMIO region spans 0x50000000 through 0x60000000, and 0x50040000 through
0x54000000 is repeated images of 0x50000000 to 0x50040000.
Eduardo Habkost [Wed, 16 Oct 2019 19:24:30 +0000 (16:24 -0300)]
image-fuzzer: Use errors parameter of subprocess.Popen()
Instead of manually encoding stderr and stdout output, use
`errors` parameter of subprocess.Popen(). This will make
process.communicate() return unicode strings instead of bytes
objects.
Eduardo Habkost [Wed, 16 Oct 2019 19:24:29 +0000 (16:24 -0300)]
image-fuzzer: Run using python3
image-fuzzer is now supposed to be ready to run using Python 3.
Remove the __future__ imports and change the interpreter line to
"#!/usr/bin/env python3".
Eduardo Habkost [Wed, 16 Oct 2019 19:24:28 +0000 (16:24 -0300)]
image-fuzzer: Encode file name and file format to bytes
Callers of create_image() will pass strings as arguments, but the
Image class will expect bytes objects to be provided. Encode
them inside create_image().
Eduardo Habkost [Wed, 16 Oct 2019 19:24:24 +0000 (16:24 -0300)]
image-fuzzer: Use io.StringIO
StringIO.StringIO is not available on Python 3, but io.StringIO
is available on both Python 2 and 3. io.StringIO is slightly
different from the Python 2 StringIO module, though, so we need
bytes coming from subprocess.Popen() to be explicitly decoded.
Eduardo Habkost [Wed, 16 Oct 2019 19:24:21 +0000 (16:24 -0300)]
image-fuzzer: Open image files in binary mode
This probably never caused problems because on Linux there's no
actual newline conversion happening, but on Python 3 the
binary/text distinction is stronger and we must explicitly open
the image file in binary mode.
Thomas Huth [Wed, 23 Oct 2019 12:01:28 +0000 (14:01 +0200)]
qemu-options: Rework the help text of the '-display' option
Improve the help text of the "-display" option:
- Only print the options that we have enabled in the binary
(similar to what we do for other options like -netdev already)
- The "frame=on|off" from "-display sdl" has been removed in commit 09bd7ba9f5f7 ("Remove deprecated -no-frame option"), so we should
not show this in the help text anymore
- The "-display egl-headless" line was missing a "\n" at the end
Josh Kunz [Tue, 29 Oct 2019 22:43:10 +0000 (15:43 -0700)]
linux-user: Support for NETLINK socket options
This change includes support for all AF_NETLINK socket options up to about
kernel version 5.4 (5.4 is not formally released at the time of writing).
Socket options that were introduced in kernel versions before the oldest
currently stable kernel version are guarded by kernel version macros.
This change has been built under gcc 8.3, and clang 9.0, and it passes
`make check`. The netlink options have been tested by emulating some
non-trival software that uses NETLINK socket options, but they have
not been exaustively verified.
Alex Williamson [Wed, 23 Oct 2019 22:47:28 +0000 (16:47 -0600)]
hw/i386: AMD-Vi IVRS DMA alias support
When we account for DMA aliases in the PCI address space, we can no
longer use a single IVHD entry in the IVRS covering all devices. We
instead need to walk the PCI bus and create alias ranges when we find
a conventional bus. These alias ranges cannot overlap with a "Select
All" range (as currently implemented), so we also need to enumerate
each device with IVHD entries.
Importantly, the IVHD entries used here include a Device ID, which is
simply the PCI BDF (Bus/Device/Function). The guest firmware is
responsible for programming bus numbers, so the final revision of this
table depends on the update mechanism (acpi_build_update) to be called
after guest PCI enumeration.
Alex Williamson [Wed, 23 Oct 2019 22:47:15 +0000 (16:47 -0600)]
pci: Use PCI aliases when determining device IOMMU address space
PCIe requester IDs are used by modern IOMMUs to differentiate devices
in order to provide a unique IOVA address space per device. These
requester IDs are composed of the bus/device/function (BDF) of the
requesting device. Conventional PCI pre-dates this concept and is
simply a shared parallel bus where transactions are claimed by
decoding target ranges rather than the packetized, point-to-point
mechanisms of PCI-express. In order to interface conventional PCI
to PCIe, the PCIe-to-PCI bridge creates and accepts packetized
transactions on behalf of all downstream devices, using one of two
potential forms of a requester ID relating to the bridge itself or its
subordinate bus. All downstream devices are therefore aliased by the
bridge's requester ID and it's not possible for the IOMMU to create
unique IOVA spaces for devices downstream of such buses.
At least that's how it works on bare metal. Until now point we've
ignored this nuance of vIOMMU support in QEMU, creating a unique
AddressSpace per device regardless of the virtual bus topology.
Aside from simply being true to bare metal behavior, there are aspects
of a shared address space that we can use to our advantage when
designing a VM. For instance, a PCI device assignment scenario where
we have the following IOMMU group on the host system:
$ ls /sys/kernel/iommu_groups/1/devices/
0000:00:01.0 0000:01:00.0 0000:01:00.1
An IOMMU group is considered the smallest set of devices which are
fully DMA isolated from other devices by the IOMMU. In this case the
root port at 00:01.0 does not guarantee that it prevents peer to peer
traffic between the endpoints on bus 01: and the devices are therefore
grouped together. VFIO considers an IOMMU group to be the smallest
unit of device ownership and allows only a single shared IOVA space
per group due to the limitations of the isolation.
Therefore, if we attempt to create the following VM, we get an error:
qemu-system-x86_64: -device vfio-pci,host=1:00.1,bus=pcie.1,addr=0.1: vfio \
0000:01:00.1: group 1 used in multiple address spaces
VFIO only allows a single IOVA space (AddressSpace) for both devices,
but we've placed them into a topology where the vIOMMU expects a
separate AddressSpace for each device. On bare metal we know that
a conventional PCI bus would provide the sort of aliasing we need
here, forcing the IOMMU to consider these devices to be part of a
single shared IOVA space. The support provided here does the same
for QEMU, such that we can create a conventional PCI topology to
expose equivalent AddressSpace sharing requirements to the VM:
There are pros and cons to this configuration; it's not necessarily
recommended, it's simply a tool we can use to create configurations
which may provide additional functionality in spite of host hardware
limitations or as a benefit to the guest configuration or resource
usage. An incomplete list of pros and cons:
Cons:
a) Extended PCI configuration space is unavailable to devices
downstream of a conventional PCI bus. The degree to which this
is a drawback depends on the device and guest drivers.
b) Applying this topology to devices which are already isolated by
the host IOMMU (singleton IOMMU groups) will result in devices
which appear to be non-isolated to the VM (non-singleton groups).
This can limit configurations within the guest, such as userspace
drivers or nested device assignment.
Pros:
a) QEMU better emulates bare metal.
b) Configurations as above are now possible.
c) Host IOMMU resources and VM locked memory requirements are reduced
in vIOMMU configurations due to shared IOMMU domains on the host
and avoidance of duplicate locked memory accounting.
qga-win: network-get-interfaces command name field bug fix
Network interface name is fetched as an encoded WCHAR array, (wide
character), then it is decoded using the guest's CP_ACP Windows code
page, which is the default code page as configure in the guest's
Windows, then it is returned as a byte array, (char array).
As stated in the BZ#1733165, when renaming a network interface to a
Chinese name and invoking this command, the returned name field has
the (\ufffd) value for each Chinese character the name had, this
value is an indication that the code page does not have the decoding
information for the given character.
This bug is a result of using the CP_ACP code page for decoding which
is an interchangeable code page, instead CP_UTF8 code page should be
used for decoding the network interface's name.
Max Reitz [Fri, 1 Nov 2019 15:25:10 +0000 (16:25 +0100)]
block/file-posix: Let post-EOF fallocate serialize
The XFS kernel driver has a bug that may cause data corruption for qcow2
images as of qemu commit c8bb23cbdbe32f. We can work around it by
treating post-EOF fallocates as serializing up until infinity (INT64_MAX
in practice).
Peter Maydell [Sat, 2 Nov 2019 17:59:03 +0000 (17:59 +0000)]
Merge remote-tracking branch 'remotes/palmer/tags/palmer-for-master-4.2-sf1' into staging
Update my MAINTAINERS file entry
This contains a single patch to change my email address.
# gpg: Signature made Fri 01 Nov 2019 16:14:45 GMT
# gpg: using RSA key 00CE76D1834960DFCE886DF8EF4CA1502CCBAB41
# gpg: issuer "[email protected]"
# gpg: Good signature from "Palmer Dabbelt <[email protected]>" [unknown]
# gpg: aka "Palmer Dabbelt <[email protected]>" [unknown]
# gpg: WARNING: This key is not certified with a trusted signature!
# gpg: There is no indication that the signature belongs to the owner.
# Primary key fingerprint: 00CE 76D1 8349 60DF CE88 6DF8 EF4C A150 2CCB AB41
* remotes/palmer/tags/palmer-for-master-4.2-sf1:
MAINTAINERS: Change to my personal email address
* remotes/pmaydell/tags/pull-target-arm-20191101-2:
target/arm: Allow reading flags from FPSCR for M-profile
hw/arm/boot: Rebuild hflags when modifying CPUState at boot
target/arm/kvm: host cpu: Add support for sve<N> properties
target/arm/cpu64: max cpu: Support sve properties with KVM
target/arm/kvm: scratch vcpu: Preserve input kvm_vcpu_init features
target/arm/kvm64: max cpu: Enable SVE when available
target/arm/kvm64: Add kvm_arch_get/put_sve
target/arm/cpu64: max cpu: Introduce sve<N> properties
target/arm: Allow SVE to be disabled via a CPU property
tests: arm: Introduce cpu feature tests
target/arm/monitor: Introduce qmp_query_cpu_model_expansion
Christophe Lyon [Fri, 25 Oct 2019 09:57:11 +0000 (11:57 +0200)]
target/arm: Allow reading flags from FPSCR for M-profile
rt==15 is a special case when reading the flags: it means the
destination is APSR. This patch avoids rejecting
vmrs apsr_nzcv, fpscr
as illegal instruction.
Andrew Jones [Thu, 31 Oct 2019 14:27:34 +0000 (15:27 +0100)]
target/arm/kvm: host cpu: Add support for sve<N> properties
Allow cpu 'host' to enable SVE when it's available, unless the
user chooses to disable it with the added 'sve=off' cpu property.
Also give the user the ability to select vector lengths with the
sve<N> properties. We don't adopt 'max' cpu's other sve property,
sve-max-vq, because that property is difficult to use with KVM.
That property assumes all vector lengths in the range from 1 up
to and including the specified maximum length are supported, but
there may be optional lengths not supported by the host in that
range. With KVM one must be more specific when enabling vector
lengths.
Andrew Jones [Thu, 31 Oct 2019 14:27:33 +0000 (15:27 +0100)]
target/arm/cpu64: max cpu: Support sve properties with KVM
Extend the SVE vq map initialization and validation with KVM's
supported vector lengths when KVM is enabled. In order to determine
and select supported lengths we add two new KVM functions for getting
and setting the KVM_REG_ARM64_SVE_VLS pseudo-register.
This patch has been co-authored with Richard Henderson, who reworked
the target/arm/cpu64.c changes in order to push all the validation and
auto-enabling/disabling steps into the finalizer, resulting in a nice
LOC reduction.
Andrew Jones [Thu, 31 Oct 2019 14:27:32 +0000 (15:27 +0100)]
target/arm/kvm: scratch vcpu: Preserve input kvm_vcpu_init features
kvm_arm_create_scratch_host_vcpu() takes a struct kvm_vcpu_init
parameter. Rather than just using it as an output parameter to
pass back the preferred target, use it also as an input parameter,
allowing a caller to pass a selected target if they wish and to
also pass cpu features. If the caller doesn't want to select a
target they can pass -1 for the target which indicates they want
to use the preferred target and have it passed back like before.
Andrew Jones [Thu, 31 Oct 2019 14:27:31 +0000 (15:27 +0100)]
target/arm/kvm64: max cpu: Enable SVE when available
Enable SVE in the KVM guest when the 'max' cpu type is configured
and KVM supports it. KVM SVE requires use of the new finalize
vcpu ioctl, so we add that now too. For starters SVE can only be
turned on or off, getting all vector lengths the host CPU supports
when on. We'll add the other SVE CPU properties in later patches.
Andrew Jones [Thu, 31 Oct 2019 14:27:30 +0000 (15:27 +0100)]
target/arm/kvm64: Add kvm_arch_get/put_sve
These are the SVE equivalents to kvm_arch_get/put_fpsimd. Note, the
swabbing is different than it is for fpsmid because the vector format
is a little-endian stream of words.
Andrew Jones [Thu, 31 Oct 2019 14:27:29 +0000 (15:27 +0100)]
target/arm/cpu64: max cpu: Introduce sve<N> properties
Introduce cpu properties to give fine control over SVE vector lengths.
We introduce a property for each valid length up to the current
maximum supported, which is 2048-bits. The properties are named, e.g.
sve128, sve256, sve384, sve512, ..., where the number is the number of
bits. See the updates to docs/arm-cpu-features.rst for a description
of the semantics and for example uses.
Note, as sve-max-vq is still present and we'd like to be able to
support qmp_query_cpu_model_expansion with guests launched with e.g.
-cpu max,sve-max-vq=8 on their command lines, then we do allow
sve-max-vq and sve<N> properties to be provided at the same time, but
this is not recommended, and is why sve-max-vq is not mentioned in the
document. If sve-max-vq is provided then it enables all lengths smaller
than and including the max and disables all lengths larger. It also has
the side-effect that no larger lengths may be enabled and that the max
itself cannot be disabled. Smaller non-power-of-two lengths may,
however, be disabled, e.g. -cpu max,sve-max-vq=4,sve384=off provides a
guest the vector lengths 128, 256, and 512 bits.
This patch has been co-authored with Richard Henderson, who reworked
the target/arm/cpu64.c changes in order to push all the validation and
auto-enabling/disabling steps into the finalizer, resulting in a nice
LOC reduction.
Andrew Jones [Thu, 31 Oct 2019 14:27:28 +0000 (15:27 +0100)]
target/arm: Allow SVE to be disabled via a CPU property
Since 97a28b0eeac14 ("target/arm: Allow VFP and Neon to be disabled via
a CPU property") we can disable the 'max' cpu model's VFP and neon
features, but there's no way to disable SVE. Add the 'sve=on|off'
property to give it that flexibility. We also rename
cpu_max_get/set_sve_vq to cpu_max_get/set_sve_max_vq in order for them
to follow the typical *_get/set_<property-name> pattern.
projects / qemu.git / log