Singh, Brijesh [Mon, 1 Oct 2018 19:44:32 +0000 (19:44 +0000)]
x86_iommu/amd: remove V=1 check from amdvi_validate_dte()
Currently, the amdvi_validate_dte() assumes that a valid DTE will
always have V=1. This is not true. The V=1 means that bit[127:1] are
valid. A valid DTE can have IV=1 and V=0 (i.e address translation
disabled and interrupt remapping enabled)
Remove the V=1 check from amdvi_validate_dte(), make the caller
responsible to check for V or IV bits.
This also fixes a bug in existing code that when error is
detected during the translation we'll fail the translation
instead of assuming a passthrough mode.
Singh, Brijesh [Mon, 1 Oct 2018 19:44:29 +0000 (19:44 +0000)]
x86_iommu: move vtd_generate_msi_message in common file
The vtd_generate_msi_message() in intel-iommu is used to construct a MSI
Message from IRQ. A similar function will be needed when we add interrupt
remapping support in amd-iommu. Moving the function in common file to
avoid the code duplication. Rename it to x86_iommu_irq_to_msi_message().
There is no logic changes in the code flow.
Yongji Xie [Wed, 6 Jun 2018 13:24:48 +0000 (21:24 +0800)]
vhost-user-blk: start vhost when guest kicks
Some old guests (before commit 7a11370e5: "virtio_blk: enable VQs early")
kick virtqueue before setting VIRTIO_CONFIG_S_DRIVER_OK. This violates
the virtio spec. But virtio 1.0 transitional devices support this behaviour.
So we should start vhost when guest kicks in this case.
Peter Xu [Tue, 9 Oct 2018 07:45:43 +0000 (15:45 +0800)]
intel_iommu: handle invalid ce for shadow sync
We should handle VTD_FR_CONTEXT_ENTRY_P properly when synchronizing
shadow page tables. Having invalid context entry there is perfectly
valid when we move a device out of an existing domain. When that
happens, instead of posting an error we invalidate the whole region.
Without this patch, QEMU will crash if we do these steps:
(1) start QEMU with VT-d IOMMU and two 10G NICs (ixgbe)
(2) bind the NICs with vfio-pci in the guest
(3) start testpmd with the NICs applied
(4) stop testpmd
(5) rebind the NIC back to ixgbe kernel driver
Peter Xu [Tue, 9 Oct 2018 07:45:42 +0000 (15:45 +0800)]
intel_iommu: move ce fetching out when sync shadow
There are two callers for vtd_sync_shadow_page_table_range(): one
provided a valid context entry and one not. Move that fetching
operation into the caller vtd_sync_shadow_page_table() where we need to
fetch the context entry.
Meanwhile, remove the error_report_once() directly since we're already
tracing all the error cases in the previous call. Instead, return error
number back to caller. This will not change anything functional since
callers are dropping it after all.
We do this move majorly because we want to do something more later in
vtd_sync_shadow_page_table().
Peter Xu [Sat, 29 Sep 2018 03:36:15 +0000 (11:36 +0800)]
intel_iommu: better handling of dmar state switch
QEMU is not handling the global DMAR switch well, especially when from
"on" to "off".
Let's first take the example of system reset.
Assuming that a guest has IOMMU enabled. When it reboots, we will drop
all the existing DMAR mappings to handle the system reset, however we'll
still keep the existing memory layouts which has the IOMMU memory region
enabled. So after the reboot and before the kernel reloads again, there
will be no mapping at all for the host device. That's problematic since
any software (for example, SeaBIOS) that runs earlier than the kernel
after the reboot will assume the IOMMU is disabled, so any DMA from the
software will fail.
For example, a guest that boots on an assigned NVMe device might fail to
find the boot device after a system reboot/reset and we'll be able to
observe SeaBIOS errors if we capture the debugging log:
WARNING - Timeout at nvme_wait:144!
Meanwhile, we should see DMAR errors on the host of that NVMe device.
It's the DMA fault that caused a NVMe driver timeout.
The correct fix should be that we do proper switching of device DMA
address spaces when system resets, which will setup correct memory
regions and notify the backend of the devices. This might not affect
much on non-assigned devices since QEMU VT-d emulation will assume a
default passthrough mapping if DMAR is not enabled in the GCMD
register (please refer to vtd_iommu_translate). However that's required
for an assigned devices, since that'll rebuild the correct GPA to HPA
mapping that is needed for any DMA operation during guest bootstrap.
Besides the system reset, we have some other places that might change
the global DMAR status and we'd better do the same thing there. For
example, when we change the state of GCMD register, or the DMAR root
pointer. Do the same refresh for all these places. For these two
places we'll also need to explicitly invalidate the context entry cache
and iotlb cache.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1625173 CC: QEMU Stable <[email protected]> Reported-by: Cong Li <[email protected]> Signed-off-by: Peter Xu <[email protected]>
--
v2:
- do the same for GCMD write, or root pointer update [Alex]
- test is carried out by me this time, by observing the
vtd_switch_address_space tracepoint after system reboot
v3:
- rewrite commit message as suggested by Alex Signed-off-by: Peter Xu <[email protected]> Reviewed-by: Eric Auger <[email protected]> Reviewed-by: Jason Wang <[email protected]> Reviewed-by: Michael S. Tsirkin <[email protected]> Signed-off-by: Michael S. Tsirkin <[email protected]>
Peter Maydell [Fri, 2 Nov 2018 11:52:39 +0000 (11:52 +0000)]
configure: Use LINKS loop for all build tree symlinks
A few places in configure were doing ad-hoc calls to
the symlink function to set up symlinks from the build tree
back to the source tree. We have a loop that does this
already for all files and directories listed in the LINKS
environment variable; use that instead.
Peter Maydell [Fri, 2 Nov 2018 11:52:38 +0000 (11:52 +0000)]
configure: Rename FILES variable to LINKS
The FILES variable is used to accumulate a list of things to symlink
from the source tree into the build tree. These don't have to be
individual files; symlinking an entire directory of data files is
also fine. Rename it to something less confusing before we add a few
directories to it.
Improve the comment to clarify what DIRS and LINKS do and why
it's not a good idea to add things to LINKS with wildcarding.
Peter Maydell [Fri, 2 Nov 2018 11:52:37 +0000 (11:52 +0000)]
tests: Move tests/hex-loader-check-data/ to tests/data/hex-loader/
Currently tests/hex-loader-check-data contains data files used
by the hexloader-test, and configure individually symlinks those
data files into the build directory using a wildcard.
Using a wildcard like this is a bad idea, because if a new
data file is added, nothing causes configure to be rerun,
and so no symlink is added for the new file. This can cause
tests to spuriously fail when they can't find their data.
Instead, it's better to symlink an entire directory of
data files. We already have such a directory: tests/data.
Move the data files from tests/hex-loader-check-data/ to
tests/data/hex-loader/, and remove the unnecessary symlinking.
Peter Maydell [Fri, 2 Nov 2018 11:52:36 +0000 (11:52 +0000)]
tests: Move tests/acpi-test-data/ to tests/data/acpi/
Currently tests/acpi-test-data contains data files used by the
bios-tables-test, and configure individually symlinks those
data files into the build directory using a wildcard.
Using a wildcard like this is a bad idea, because if a new
data file is added, nothing causes configure to be rerun,
and so no symlink is added for the new file. This can cause
tests to spuriously fail when they can't find their data.
Instead, it's better to symlink an entire directory of
data files. We already have such a directory: tests/data.
Move the data files from tests/acpi-test-data/ to
tests/data/acpi/, and remove the unnecessary symlinking.
We can remove entirely the note in rebuild-expected-aml.sh
about copying any new data files, because now they will
be in the source directory, not the build directory, and
no copying is required.
(We can't just change the existing tests/acpi-test-data/
to being a symlinked directory, because if we did that and
a developer switched git branches from one after that change
to one before it then configure would end up trashing all
the test files by making them symlinks to themselves.
Changing their path avoids this annoyance.)
Peter Maydell [Fri, 2 Nov 2018 17:17:12 +0000 (17:17 +0000)]
Merge remote-tracking branch 'remotes/pmaydell/tags/pull-target-arm-20181102' into staging
target-arm queue:
* microbit: Add the UART to our nRF51 SoC model
* Add a virtual Xilinx Versal board "xlnx-versal-virt"
* hw/arm/virt: Set VIRT_COMPAT_3_0 compat
* MAINTAINERS: Remove bouncing email in ARM ACPI
* strongarm: mask off high[31:28] bits from dir and state registers
* target/arm: Conditionalize some asserts on aarch32 support
* hw/arm/xilinx_zynq: Use the ARRAY_SIZE macro
* remotes/pmaydell/tags/pull-target-arm-20181102:
hw/arm: versal: Add a virtual Xilinx Versal board
hw/arm: versal: Add a model of Xilinx Versal SoC
target/arm: Conditionalize some asserts on aarch32 support
hw/arm/xilinx_zynq: Use the ARRAY_SIZE macro
strongarm: mask off high[31:28] bits from dir and state registers
MAINTAINERS: Remove bouncing email in ARM ACPI
tests/boot-serial-test: Add microbit board testcase
hw/arm/nrf51_soc: Connect UART to nRF51 SoC
hw/char: Implement nRF51 SoC UART
hw/arm/virt: Set VIRT_COMPAT_3_0 compat
This board is based on the Xilinx Versal SoC. The exact
details of what peripherals are attached to this board
will remain in control of QEMU. QEMU will generate an
FDT on the fly for Linux and other software to auto-discover
peripherals.
strongarm: mask off high[31:28] bits from dir and state registers
The high[31:28] bits of 'direction' and 'state' registers of
SA-1100/SA-1110 device are reserved. Setting them may lead to
OOB 's->handler[]' array access issue. Mask off [31:28] bits to
avoid it.
Shannon Zhao's email at Huawei is bouncing: remove it.
X-Failed-Recipients: [email protected]
** Address not found **
Your message wasn't delivered to [email protected] because the address couldn't be found, or is unable to receive mail.
Note that the section still contains his personal email (see e59f13d76bb).
Peter Maydell [Fri, 2 Nov 2018 13:16:13 +0000 (13:16 +0000)]
Merge remote-tracking branch 'remotes/riscv/tags/riscv-for-master-3.1-sf1' into staging
RISC-V Patches for the 3.1 Soft Freeze, Part 2
This tag contains a few simple patches that I'd like to target for the
QEMU soft freeze. There's only one code change: a fix to our PMP
implementation that avoids an internal truncation while computing a
partial PMP read.
I also have two updates to the MAINTAINERS file: one to add Alistair as
a RISC-V maintainer, and one to add our newly created mailing list.
# gpg: Signature made Tue 30 Oct 2018 18:17:17 GMT
# gpg: using RSA key EF4CA1502CCBAB41
# gpg: Good signature from "Palmer Dabbelt <[email protected]>"
# gpg: aka "Palmer Dabbelt <[email protected]>"
# 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/riscv/tags/riscv-for-master-3.1-sf1:
Add [email protected] as the RISC-V list
Add Alistair as a RISC-V Maintainer
target/riscv/pmp.c: pmpcfg_csr_read returns bogus value on RV64
Peter Maydell [Fri, 2 Nov 2018 10:53:00 +0000 (10:53 +0000)]
Merge remote-tracking branch 'remotes/elmarco/tags/chrdev-pull-request' into staging
- add websocket support
- socket: make 'fd' incompatible with 'reconnect'
- fix a websocket leak
- unrelated editorconfig patch that missed -trivial (included for
convenience)
- v2: fix commit author field
# gpg: Signature made Thu 01 Nov 2018 08:23:39 GMT
# gpg: using RSA key DAE8E10975969CE5
# gpg: Good signature from "Marc-André Lureau <[email protected]>"
# gpg: aka "Marc-André Lureau <[email protected]>"
# Primary key fingerprint: 87A9 BD93 3F87 C606 D276 F62D DAE8 E109 7596 9CE5
* remotes/elmarco/tags/chrdev-pull-request:
editorconfig: set emacs mode
tests/test-char: Check websocket chardev functionality
chardev: Add websocket support
chardev/char-socket: Function headers refactoring
char-socket: make 'fd' incompatible with 'reconnect'
char-socket: correctly set has_reconnect when parsing QemuOpts
websock: fix handshake leak
Peter Maydell [Fri, 2 Nov 2018 09:49:35 +0000 (09:49 +0000)]
Merge remote-tracking branch 'remotes/dgilbert/tags/pull-migration-20181031a' into staging
Minor migration fixes 2018-10-31
# gpg: Signature made Wed 31 Oct 2018 16:55:40 GMT
# gpg: using RSA key 0516331EBC5BFDE7
# gpg: Good signature from "Dr. David Alan Gilbert (RH2) <[email protected]>"
# Primary key fingerprint: 45F5 C71B 4A0C B7FB 977A 9FA9 0516 331E BC5B FDE7
* remotes/dgilbert/tags/pull-migration-20181031a:
migration: avoid segmentfault when take a snapshot of a VM which being migrated
qapi: Fix COLOStatus and query-colo-status since version
COLO: Fix Colo doc secondeary should be secondary
Peter Maydell [Thu, 1 Nov 2018 17:26:16 +0000 (17:26 +0000)]
Merge remote-tracking branch 'remotes/mdroth/tags/qga-pull-2018-10-30-v3-tag' into staging
qemu-ga patch queue for soft-freeze
* support for --retry-path option for recovering from communication
path failures
* support for serial/device name in guest-get-fsinfo for linux/w32
* support for freezing individual mount points in guest-fsfreeze-*
* fixes for unicode paths on w32, not-present vcpus in guest-get-vcpus,
buffer overflow in guest-get-fsinfo for w32, and other minor fixes
v3:
* remove redundant check for --static in configure
* correct authorship on "qga-win: add debugging information"
v2:
* set libudev=off in configure for static builds
* remotes/mdroth/tags/qga-pull-2018-10-30-v3-tag: (24 commits)
qga-win: changing --retry-path option behavior
qga-win: report specific error when failing to open channel
qga-win: install service with --retry-path set by default
qga: add --retry-path option for re-initializing channel on failure
qga: move w32 service handling out of run_agent()
qga: hang GAConfig/socket_activation off of GAState global
qga: group agent init/cleanup init separate routines
qga: fix an off-by-one issue
qga-win: demystify namespace stripping
qga-win: return disk device in guest-get-fsinfo
qga-win: handle multi-disk volumes
qga-win: refactor disk info
qga-win: report disk serial number
qga-win: refactor disk properties (bus)
qga-win: add debugging information
build: rename CONFIG_QGA_NTDDDISK to CONFIG_QGA_NTDDSCSI
qga-win: fsinfo: pci-info: allow partial info
qga-win: prevent crash when executing fsinfo command
qga: linux: return disk device in guest-get-fsinfo
qga: linux: report disk serial number
...
Peter Maydell [Thu, 1 Nov 2018 16:32:54 +0000 (16:32 +0000)]
Merge remote-tracking branch 'remotes/ehabkost/tags/x86-next-pull-request' into staging
x86 queue, 2018-10-30
* MSR-based feature support for
MSR_IA32_ARCH_CAPABILITIES bits (Robert Hoo)
* Cascadelake-Server CPU model (Tao Xu)
* Add PKU on Skylake-Server CPU model (Tao Xu)
* Correct cpu_x86_cpuid(0xd) (Sebastian Andrzej Siewior)
* Remove dead code (Peter Maydell)
# gpg: Signature made Wed 31 Oct 2018 14:05:25 GMT
# gpg: using RSA key 2807936F984DC5A6
# gpg: Good signature from "Eduardo Habkost <[email protected]>"
# Primary key fingerprint: 5A32 2FD5 ABC4 D3DB ACCF D1AA 2807 936F 984D C5A6
* remotes/ehabkost/tags/x86-next-pull-request:
i386: Add PKU on Skylake-Server CPU model
i386: Add new model of Cascadelake-Server
x86: define a new MSR based feature word -- FEATURE_WORDS_ARCH_CAPABILITIES
x86: Data structure changes to support MSR based features
kvm: Add support to KVM_GET_MSR_FEATURE_INDEX_LIST and KVM_GET_MSRS system ioctl
target/i386: Remove #ifdeffed-out icebp debugging hack
i386: correct cpu_x86_cpuid(0xd)
Peter Maydell [Thu, 1 Nov 2018 14:38:50 +0000 (14:38 +0000)]
Merge remote-tracking branch 'remotes/berrange/tags/misc-next-pull-request' into staging
Merge misc fixes
# gpg: Signature made Wed 31 Oct 2018 11:36:12 GMT
# gpg: using RSA key BE86EBB415104FDF
# gpg: Good signature from "Daniel P. Berrange <[email protected]>"
# gpg: aka "Daniel P. Berrange <[email protected]>"
# Primary key fingerprint: DAF3 A6FD B26B 6291 2D0E 8E3F BE86 EBB4 1510 4FDF
* remotes/berrange/tags/misc-next-pull-request:
scripts: report on author emails that are mangled by the mailing list
block: drop moderated sheepdog mailing list from MAINTAINERS file
# gpg: Signature made Wed 31 Oct 2018 00:28:39 GMT
# gpg: using RSA key 2807936F984DC5A6
# gpg: Good signature from "Eduardo Habkost <[email protected]>"
# Primary key fingerprint: 5A32 2FD5 ABC4 D3DB ACCF D1AA 2807 936F 984D C5A6
* remotes/ehabkost/tags/python-next-pull-request:
scripts/qemu.py: use a more consistent docstring style
scripts/decodetree.py: fix reference to attributes
Travis support for the acceptance tests
Acceptance tests: add make rule for running them
Bootstrap Python venv for tests
iotests: Unify log outputs between Python 2 and 3
iotests: Modify imports for Python 3
iotests: 'new' module replacement in 169
iotests: Explicitly bequeath FDs in Python
iotests: Different iterator behavior in Python 3
iotests: Use // for Python integer division
iotests: Use Python byte strings where appropriate
iotests: Flush in iotests.py's QemuIoInteractive
iotests: Make nbd-fault-injector flush
scripts/device-crash-test: Remove devices that are not user_creatable anymore
Peter Maydell [Thu, 1 Nov 2018 12:08:10 +0000 (12:08 +0000)]
Merge remote-tracking branch 'remotes/stefanberger/tags/pull-tpm-2018-10-29-2' into staging
Merge tpm 2018/10/29 v2
# gpg: Signature made Tue 30 Oct 2018 21:40:24 GMT
# gpg: using RSA key 75AD65802A0B4211
# gpg: Good signature from "Stefan Berger <[email protected]>"
# 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: B818 B9CA DF90 89C2 D5CE C66B 75AD 6580 2A0B 4211
* remotes/stefanberger/tags/pull-tpm-2018-10-29-2:
tpm: Zero-init structure to avoid uninitialized variables in valgrind log
MAINTAINERS: Change my email address to the new domain
docs: tpm: Mention implemented TPM CRB interface emulation and specs
tests/tpm: Display if swtpm is not found or --tpm2 not supported
tests/tpm: fix tpm_util_swtpm_has_tpm2()
Laurent Vivier [Tue, 30 Oct 2018 16:55:54 +0000 (17:55 +0100)]
target/m68k: use EXCP_ILLEGAL instead of EXCP_UNSUPPORTED
Coldfire defines an "Unsupported instruction" exception if execution
of a valid instruction is attempted but the required hardware is not
present in the processor.
We use it with instructions that are in fact undefined or illegal,
and the exception expected in this case by the kernel is the
illegal exception, so this patch fixes that.
Some time ago, I proposed to use an (eval) in .dir-locals.el to set
the mode for all json files and Makefile. Unfortunately, this isn't
safe, and emacs will prompt the user, which isn't very friendly.
Fortunately, editorconfig provides a special config key which does
allow to set the emacs mode. Add a few missing entries and set the
emacs mode.
Update top comment to provide a short summary about the file and the
IDE plugins while at it.
Test order:
Creating server websocket chardev
Creating usual tcp chardev client
Sending handshake message from client
Receiving handshake reply
Sending ping frame with "hello" payload
Receiving pong reply
Sending binary data "world"
Checking the received data on server side
Checking of closing handshake
Julia Suvorova [Thu, 18 Oct 2018 22:34:59 +0000 (01:34 +0300)]
chardev/char-socket: Function headers refactoring
Upcoming websocket support requires additional parameters in function
headers that are already overloaded. This patch replaces the bunch of
parameters with a single structure pointer.
char-socket: make 'fd' incompatible with 'reconnect'
A chardev socket created with the 'fd=' argument is not going to
handle reconnection properly by recycling the same fd (or not in a
supported way). Let's forbid this case.
char-socket: correctly set has_reconnect when parsing QemuOpts
qemu_chr_parse_socket() fills all ChardevSocket fields, but that
doesn't reflect correctly the arguments given with the options / on
the command line. "reconnect" takes a number as argument, and the
default value is 0, which doesn't help to identify the missing
option. The other arguments have default values that are less
problematic, leave them set by default for now.
==955== 217 bytes in 1 blocks are definitely lost in loss record 275 of 321
==955== at 0x483A965: realloc (vg_replace_malloc.c:785)
==955== by 0x50B6839: __vasprintf_chk (in /usr/lib64/libc-2.28.so)
==955== by 0x49AA05C: g_vasprintf (in /usr/lib64/libglib-2.0.so.0.5800.1)
==955== by 0x4983440: g_strdup_vprintf (in /usr/lib64/libglib-2.0.so.0.5800.1)
==955== by 0x126048: qio_channel_websock_handshake_send_res (channel-websock.c:162)
==955== by 0x1266E6: qio_channel_websock_handshake_send_res_ok (channel-websock.c:362)
==955== by 0x126D3E: qio_channel_websock_handshake_process (channel-websock.c:468)
==955== by 0x126EF2: qio_channel_websock_handshake_read (channel-websock.c:511)
==955== by 0x12715B: qio_channel_websock_handshake_io (channel-websock.c:571)
==955== by 0x125027: qio_channel_fd_source_dispatch (channel-watch.c:84)
==955== by 0x496326C: g_main_context_dispatch (in /usr/lib64/libglib-2.0.so.0.5800.1)
==955== by 0x169EC3: glib_pollfds_poll (main-loop.c:215)
While it would be possible to concatenate input files with make,
passing the original input files to decodetree.py allows us to
generate error messages which allows compilation environments
(read: emacs) to next-error to the correct input file.
decodetree: Remove "insn" argument from trans_* expanders
This allows trans_* expanders to be shared between decoders
for 32 and 16-bit insns, by not tying the expander to the
size of the insn that produced it.
This change requires adjusting the two existing users to match.
Allow argument sets to be shared between two decoders by avoiding
a re-declaration error. Make sure that anonymous argument sets
and anonymous formats have unique names.
Currently whenever the qemu-ga's service doesn't find the virtio-serial
the run_agent() loops in a QGA_RETRY_INTERVAL (default 5 seconds)
intervals and try to restart the qemu-ga which causes a synchronous loop.
Changed to wait and listen for the serial events by registering for
notifications a proper serial event handler that deals with events:
DBT_DEVICEARRIVAL indicates that the device has been inserted and
is available
DBT_DEVICEREMOVECOMPLETE indicates that the devive has been removed
Which allow us to determine when the channel path is available for the
qemu-ga to restart.
Michael Roth [Sun, 7 Oct 2018 11:02:21 +0000 (14:02 +0300)]
qga-win: install service with --retry-path set by default
It's nicer from a management perspective that the agent can survive
hotplug/unplug of the channel device, or be started prior to the
installation of the channel device's driver without and still be able
to resume normal function afterward. On linux there are alternatives
like systemd to support this, but on w32 --retry-path is the only
option so it makes sense to set it by default when installed as a
w32 service.
Michael Roth [Sun, 7 Oct 2018 11:02:20 +0000 (14:02 +0300)]
qga: add --retry-path option for re-initializing channel on failure
This adds an option to instruct the agent to periodically attempt
re-opening the communication channel after a channel error has
occurred. The main use-case for this is providing an OS-independent
way of allowing the agent to survive situations like hotplug/unplug of
the communication channel, or initial guest set up where the agent may
be installed/started prior to the installation of the channel device's
driver.
There are nicer ways of implementing this functionality via things
like systemd services, but this option is useful for platforms like
*BSD/w32.
Currently a channel error will result in the GSource for that channel
being removed from the GMainLoop, but the main loop continuing to run.
That behavior results in a dead loop when --retry-path isn't set, and
prevents us from knowing when to attempt re-opening the channel when
it is set, so we also force the loop to exit as part of this patch.
Michael Roth [Sun, 7 Oct 2018 11:02:19 +0000 (14:02 +0300)]
qga: move w32 service handling out of run_agent()
Eventually we want a w32 service to be able to restart the qga main
loop from within service_main(). To allow for this we move service
handling out of run_agent() such that service_main() calls
run_agent() instead of the reverse.
Michael Roth [Sun, 7 Oct 2018 11:02:18 +0000 (14:02 +0300)]
qga: hang GAConfig/socket_activation off of GAState global
For w32 services we rely on the global GAState to access resources
associated with the agent within service_main(). Currently this is
sufficient for starting the agent since we open the channel once prior
to calling service_main(), and simply start the GMainLoop to start the
agent from within service_main().
Eventually we want to be able to also [re-]open the communication
channel from within service_main(), which requires access to
config/socket_activation variables, so we hang them off GAState in
preparation for that.
Signed-off-by: Michael Roth <[email protected]> Signed-off-by: Sameeh Jubran <[email protected]>
*dont move GAConfig struct, just the typedef
*fix build bisect for w32 Signed-off-by: Michael Roth <[email protected]>
Michael Roth [Sun, 7 Oct 2018 11:02:17 +0000 (14:02 +0300)]
qga: group agent init/cleanup init separate routines
This patch better separates the init/cleanup routines out into
separate functions to make the start-up procedure a bit easier to
follow. This will be useful when we eventually break out the actual
start/stop of the agent's main loop into separates routines that
can be called multiple times after the init phase.
Probe the volume for disk extents and return list of all disks.
Originally only first disk of composite volume was returned.
Note that the patch changes get_pci_info() from one state of brokenness
into a different state of brokenness. In other words it still does not do
what it's supposed to do (see comment in code). If anyone knows how to
fix it, please step in.
Refactor building of disk info into a function that builds the list and
a function that returns infor for single disk. This will be used in
future commit that will handle multi-disk volumes.
Signed-off-by: Tomáš Golembiovský <[email protected]>
*coding style fix-ups (declarations at beginning of block)
*improve readability for user-visible errors
*cover additional edge-cases with debug statements Signed-off-by: Michael Roth <[email protected]>
Refactor code that queries bus type to be more generic. The function
get_disk_bus_type() has been renamed to build_guest_disk_info().
Following commit(s) will extend this function.
build: rename CONFIG_QGA_NTDDDISK to CONFIG_QGA_NTDDSCSI
There was inconsistency between commits:
50cbebb9a3 configure: add configure check for ntdddisk.h a3ef3b2272 qga: added bus type and disk location path
The first commit added #define CONFIG_QGA_NTDDDISK but the second commit
expected the name to be CONFIG_QGA_NTDDSCSI. As a result the code in
second patch was never used.
Renaming the option to CONFIG_QGA_NTDDSCSI to match the name of header
file that is being checked for.
Sameeh Jubran [Tue, 23 Oct 2018 11:23:14 +0000 (13:23 +0200)]
qga-win: fsinfo: pci-info: allow partial info
The call to SetupDiGetDeviceRegistryProperty might fail because the
value doesn't exist in the registry, in this case we shouldn't exit from
the loop but instead continue to look for other available values in the
registry and set this value as unavailable (-1).
Signed-off-by: Sameeh Jubran <[email protected]> Signed-off-by: Michael Roth <[email protected]> Signed-off-by: Tomáš Golembiovský <[email protected]>
*squash in fix for when get_pci_info() returns NULL pci_controller field
*fix handling for error_set() cases in get_pci_info(), not just NULL return
*force all -1 PCI addr fields if any single one of them isn't found Signed-off-by: Michael Roth <[email protected]>
Sameeh Jubran [Tue, 23 Oct 2018 11:23:13 +0000 (13:23 +0200)]
qga-win: prevent crash when executing fsinfo command
The fsinfo command is currently implemented for Windows only and it's disk
parameter can be enabled by adding the define "CONFIG_QGA_NTDDSCSI" to the qga
code. When enabled and executed the qemu-ga crashed with the following message:
------------------------------------------------
File qapi/qapi-visit-core.c, Line 49
After some digging, turns out that the GuestPCIAddress is null and the
qapi visitor doesn't like that, so we can always allocate it instead and
initiate all it's members to -1.
Especially for guests with large numbers of tlbs, like ARM or PPC,
we may well not use all of them in between flush operations.
Remember which tlbs have been used since the last flush, and
avoid any useless flushing.
Our only statistic so far was "full" tlb flushes, where all mmu_idx
are flushed at the same time.
Now count "partial" tlb flushes where sets of mmu_idx are flushed,
but the set is not maximal. Account one per mmu_idx flushed, as
that is the unit of work performed.
We don't actually count elided flushes yet, but go ahead and change
the interface presented to the monitor all at once.
cputlb: Merge tlb_flush_nocheck into tlb_flush_by_mmuidx_async_work
The difference between the two sets of APIs is now miniscule.
This allows tlb_flush, tlb_flush_all_cpus, and tlb_flush_all_cpus_synced
to be merged with their corresponding by_mmuidx functions as well. For
accounting, consider mmu_idx_bitmask = ALL_MMUIDX_BITS to be a full flush.
The set of large pages in the kernel is probably not the same
as the set of large pages in the application. Forcing one
range to cover both will flush more often than necessary.
This allows tlb_flush_page_async_work to flush just the one
mmu_idx implicated, which in turn allows us to remove
tlb_check_page_and_flush_by_mmuidx_async_work.
cputlb: Move cpu->pending_tlb_flush to env->tlb_c.pending_flush
Protect it with the tlb_lock instead of using atomics.
The move puts it in or near the same cacheline as the lock;
using the lock means we don't need a second atomic operation
in order to perform the update. Which makes it cheap to also
update pending_flush in tlb_flush_by_mmuidx_async_work.
cputlb: Remove tcg_enabled hack from tlb_flush_nocheck
The bugs this was working around were fixed with commits 022d6378c7fd target/unicore32: remove tlb_flush from uc32_init_fn 6e11beecfde0 target/alpha: remove tlb_flush from alpha_cpu_initfn
block: drop moderated sheepdog mailing list from MAINTAINERS file
The sheepdog mailing list is setup to stop and queue messages from
non-subscribers, pending moderator approval. Unfortunately it seems
that the moderation queue is not actively deadlt with. Even when messages
are approved, the sender is never added to the whitelist, so every
future mail the same sender continues to get stopped for moderation.
MAINTAINERS entries should be responsive and not uneccessarily block
mails from QEMU contributors, so drop the sheepdog mailing list.
Jia Lina [Fri, 26 Oct 2018 08:36:20 +0000 (16:36 +0800)]
migration: avoid segmentfault when take a snapshot of a VM which being migrated
During an active background migration, snapshot will trigger a
segmentfault. As snapshot clears the "current_migration" struct
and updates "to_dst_file" before it finds out that there is a
migration task, Migration accesses the null pointer in
"current_migration" struct and qemu crashes eventually.
Tao Xu [Mon, 29 Oct 2018 08:39:53 +0000 (16:39 +0800)]
i386: Add PKU on Skylake-Server CPU model
As the release document ref below link (page 13):
https://software.intel.com/sites/default/files/managed/c5/15/\
architecture-instruction-set-extensions-programming-reference.pdf
PKU is supported in Skylake Server (Only Server) and later, and
on Intel(R) Xeon(R) Processor Scalable Family. So PKU is supposed
to be in Skylake-Server CPU model. And PKU's CPUID has been
exposed to QEMU. But PKU can't be find in Skylake-Server CPU
model in the code. So this patch will fix this issue in
Skylake-Server CPU model.
Tao Xu [Wed, 19 Sep 2018 03:11:22 +0000 (11:11 +0800)]
i386: Add new model of Cascadelake-Server
New CPU models mostly inherit features from ancestor Skylake-Server,
while addin new features: AVX512_VNNI, Intel PT.
SSBD support for speculative execution
side channel mitigations.
Note:
On Cascadelake, some capabilities (RDCL_NO, IBRS_ALL, RSBA,
SKIP_L1DFL_VMENTRY and SSB_NO) are enumerated by MSR.
These features rely on MSR based feature support patch.
Will be added later after that patch's in.
http://lists.nongnu.org/archive/html/qemu-devel/2018-09/msg00074.html
Remove a debugging hack which could be used to cause the
undocumented 'icebp' instruction to enable QEMU internal
debug logging. This code has always been #ifdeffed out
since it was introduced in commit aba9d61e34b9 in 2005;
judging by the rest of that commit (which is entirely
unrelated) it may have even been committed by accident.
(Note that WANT_ICEBP is not defined by default anyway.)
Intel SDM says for CPUID function 0DH, sub-function 0:
| • ECX enumerates the size (in bytes) required by the XSAVE instruction for an
| XSAVE area containing all the user state components supported by this
| processor.
| • EBX enumerates the size (in bytes) required by the XSAVE instruction for an
| XSAVE area containing all the user state components corresponding to bits
| currently set in XCR0.
Cleber Rosa [Thu, 18 Oct 2018 15:31:33 +0000 (11:31 -0400)]
Acceptance tests: add make rule for running them
The acceptance (aka functional, aka Avocado-based) tests are
Python files located in "tests/acceptance" that need to be run
with the Avocado libs and test runner.
Let's provide a convenient way for QEMU developers to run them,
by making use of the tests-venv with the required setup.
Also, while the Avocado test runner will take care of creating a
location to save test results to, it was understood that it's better
if the results are kept within the build tree.
Cleber Rosa [Thu, 18 Oct 2018 15:31:32 +0000 (11:31 -0400)]
Bootstrap Python venv for tests
A number of QEMU tests are written in Python, and may benefit
from an untainted Python venv.
By using make rules, tests that depend on specific Python libs
can set that rule as a requirement, along with rules that require
the presence or installation of specific libraries.
The tests/requirements.txt is supposed to contain the Python
requirements that should be added to the venv created by check-venv.
Max Reitz [Mon, 22 Oct 2018 13:53:07 +0000 (14:53 +0100)]
iotests: Unify log outputs between Python 2 and 3
When dumping an object into the log, there are differences between
Python 2 and 3. First, unicode strings are prefixed by 'u' in Python 2
(they are no longer in 3, because unicode strings are the default
there). Second, the order of keys in dicts may differ. Third,
especially long numbers are longs in Python 2 and thus get an 'L'
suffix, which does not happen in Python 3.
We can get around all of these differences by dumping objects (lists and
dicts) in a language-independent format, namely JSON. The JSON
generator even allows emitting dicts with their keys sorted
alphabetically.
This changes the output of all tests that use these logging functions
(dict keys are ordered now, strings in dicts are now enclosed in double
quotes instead of single quotes, the 'L' suffix of large integers is
dropped, and "true" and "false" are now in lower case).
The quote change necessitates a small change to a filter used in test
207.
Max Reitz [Mon, 22 Oct 2018 13:53:06 +0000 (14:53 +0100)]
iotests: Modify imports for Python 3
There are two imports that need to be modified when running the iotests
under Python 3: One is StringIO, which no longer exists; instead, the
StringIO class comes from the io module, so import it from there (and
use the BytesIO class for Python 2). The other is the ConfigParser,
which has just been renamed to configparser.
Max Reitz [Mon, 22 Oct 2018 13:53:05 +0000 (14:53 +0100)]
iotests: 'new' module replacement in 169
iotest 169 uses the 'new' module to add methods to a class. This module
no longer exists in Python 3. Instead, we can use a lambda. Best of
all, this works in 2.7 just as well.