Peter Maydell [Fri, 23 Oct 2015 09:24:08 +0000 (10:24 +0100)]
Merge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging
vhost: build fix
Fix build breakages when using older gcc.
Signed-off-by: Michael S. Tsirkin <[email protected]>
# gpg: Signature made Thu 22 Oct 2015 20:36:07 BST using RSA key ID D28D5469
# gpg: Good signature from "Michael S. Tsirkin <[email protected]>"
# gpg: aka "Michael S. Tsirkin <[email protected]>"
* remotes/mst/tags/for_upstream:
vhost-user: fix up rhel6 build
Move the target_disas() alpha specifics to the CPUClass::disas_set_info()
hook and delete the #ifdef specific code in disas.c.
This also makes monitor_disas() consistent with target_disas(), as
monitor_disas() was missing a set of the BFD (This was an omission from
commit b9bec751c8c8b08d8055da32306eb105db03031b).
Peter Maydell [Thu, 22 Oct 2015 11:41:44 +0000 (12:41 +0100)]
Merge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging
vhost, pc, virtio features, fixes, cleanups
New features:
VT-d support for devices behind a bridge
vhost-user migration support
Signed-off-by: Michael S. Tsirkin <[email protected]>
# gpg: Signature made Thu 22 Oct 2015 12:39:19 BST using RSA key ID D28D5469
# gpg: Good signature from "Michael S. Tsirkin <[email protected]>"
# gpg: aka "Michael S. Tsirkin <[email protected]>"
* remotes/mst/tags/for_upstream: (37 commits)
hw/isa/lpc_ich9: inject the SMI on the VCPU that is writing to APM_CNT
i386: keep cpu_model field in MachineState uptodate
vhost: set the correct queue index in case of migration with multiqueue
piix: fix resource leak reported by Coverity
seccomp: add memfd_create to whitelist
vhost-user-test: check ownership during migration
vhost-user-test: add live-migration test
vhost-user-test: learn to tweak various qemu arguments
vhost-user-test: wrap server in TestServer struct
vhost-user-test: remove useless static check
vhost-user-test: move wait_for_fds() out
vhost: add migration block if memfd failed
vhost-user: use an enum helper for features mask
vhost user: add rarp sending after live migration for legacy guest
vhost user: add support of live migration
net: add trace_vhost_user_event
vhost-user: document migration log
vhost: use a function for each call
vhost-user: add a migration blocker
vhost-user: send log shm fd along with log_base
...
Laszlo Ersek [Tue, 20 Oct 2015 18:14:00 +0000 (20:14 +0200)]
hw/isa/lpc_ich9: inject the SMI on the VCPU that is writing to APM_CNT
Commit 4d00636e97b7 ("ich9: Add the lpc chip", Nov 14 2012) added the
ich9_apm_ctrl_changed() ioport write callback function such that it would
inject the SMI, in response to a write to the APM_CNT register, on the
first CPU, invariably.
Since this register is used by guest code to trigger an SMI synchronously,
the interrupt should be injected on the VCPU that is performing the write.
apm_ioport_writeb() is the .write callback of the "apm_ops"
MemoryRegionOps [hw/isa/apm.c]; it is parametrized to call
ich9_apm_ctrl_changed() by ich9_lpc_init() [hw/isa/lpc_ich9.c], via
apm_init(). Therefore this change affects no other board.
ich9_generate_smi() is an unrelated function that is called by the TCO
watchdog; a watchdog is likely in its right to (asynchronously) inject
interrupts on the first CPU only.
This patch allows the combined edk2/OVMF SMM driver stack to work with
multiple VCPUs on TCG, using both qemu-system-i386 and qemu-system-x86_64.
Thibaut Collet [Mon, 19 Oct 2015 12:59:27 +0000 (14:59 +0200)]
vhost: set the correct queue index in case of migration with multiqueue
When a live migration is started the log address to mark dirty pages is provided
to the vhost backend through the vhost_dev_set_log function.
This function is called for each queue pairs but the queue index is wrongly set:
always set to the first queue pair. Then vhost backend lost descriptor addresses
of the queue pairs greater than 1 and behaviour of the vhost backend is
unpredictable.
The queue index is computed by taking account of the vq_index (to retrieve the
queue pair index) and calling the vhost_get_vq_index method of the backend.
Thibaut Collet [Fri, 9 Oct 2015 15:17:32 +0000 (17:17 +0200)]
vhost user: add rarp sending after live migration for legacy guest
A new vhost user message is added to allow QEMU to ask to vhost user backend to
broadcast a fake RARP after live migration for guest without GUEST_ANNOUNCE
capability.
This new message is sent only if the backend supports the new
VHOST_USER_PROTOCOL_F_RARP protocol feature.
The payload of this new message is the MAC address of the guest (not known by
the backend). The MAC address is copied in the first 6 bytes of a u64 to avoid
to create a new payload message type.
This new message has no equivalent ioctl so a new callback is added in the
userOps structure to send the request.
Upon reception of this new message the vhost user backend must generate and
broadcast a fake RARP request to notify the migration is terminated.
Thibaut Collet [Fri, 9 Oct 2015 15:17:31 +0000 (17:17 +0200)]
vhost user: add support of live migration
Some vhost user backends are able to support live migration.
To provide this service the following features must be added:
1. Add the VIRTIO_NET_F_GUEST_ANNOUNCE capability to vhost-net when netdev
backend is vhost-user.
2. Provide a nop receive callback to vhost-user.
This callback is called by:
* qemu_announce_self after a migration to send fake RARP to avoid network
outage for peers talking to the migrated guest.
- For guest with GUEST_ANNOUNCE capabilities, guest already sends GARP
when the bit VIRTIO_NET_S_ANNOUNCE is set.
=> These packets must be discarded.
- For guest without GUEST_ANNOUNCE capabilities, migration termination
is notified when the guest sends packets.
=> These packets can be discarded.
* virtio_net_tx_bh with a dummy boot to send fake bootp/dhcp request.
BIOS guest manages virtio driver to send 4 bootp/dhcp request in case of
dummy boot.
=> These packets must be discarded.
Replace the generic vhost_call() by specific functions for each
function call to help with type safety and changing arguments.
While doing this, I found that "unsigned long long" and "uint64_t" were
used interchangeably and causing compilation warnings, using uint64_t
instead, as the vhost & protocol specifies.
Send the shm for the dirty pages logging if the backend supports
VHOST_USER_PROTOCOL_F_LOG_SHMFD. Wait for a reply to make sure
the old log is no longer used.
Implement memfd_create() fallback if not available in system libc.
memfd_create() is still not included in glibc today, atlhough it's been
available since Linux 3.17 in Oct 2014.
memfd has numerous advantages over traditional shm/mmap for ipc memory
sharing with fd handler, which we are going to make use of for
vhost-user logging memory in following patches.
The next patches are going to introduce helpers to use best practices of
memfd usage and provide some compatibility fallback. memfd.c is thus
temporarily useless and eventually empty if memfd_create() is provided
by the system.
Igor Mammedov [Tue, 6 Oct 2015 08:37:28 +0000 (10:37 +0200)]
pc-dimm: add vhost slots limit check before commiting to hotplug
it allows safely cancel memory hotplug if vhost backend
doesn't support necessary amount of memory slots and prevents
QEMU crashing in vhost due to hitting vhost limit on amount
of supported memory ranges.
Igor Mammedov [Tue, 6 Oct 2015 08:37:27 +0000 (10:37 +0200)]
vhost: add vhost_has_free_slot() interface
it will allow for other parts of QEMU check if it's safe
to map memory region during hotplug/runtime.
That way hotplug path will have a chance to cancel
hotplug operation instead of crashing in vhost_commit().
Peter Maydell [Wed, 21 Oct 2015 20:21:29 +0000 (21:21 +0100)]
Merge remote-tracking branch 'remotes/xtensa/tags/20151021-xtensa' into staging
Xtensa updates:
- fix register window overflow with l32e/s32e instructions;
- make MMU events logging dependent on CPU_LOG_MMU;
- attach FLASH to system I/O region on XTFPGA boards;
- implement depbits and l32nb instructions.
# gpg: Signature made Wed 21 Oct 2015 19:34:02 BST using RSA key ID F83FA044
# gpg: Good signature from "Max Filippov <[email protected]>"
# gpg: aka "Max Filippov <[email protected]>"
* remotes/xtensa/tags/20151021-xtensa:
target-xtensa: implement S32NB
target-xtensa: implement depbits instruction
target-xtensa: xtfpga: attach FLASH to system IO
target-xtensa: use CPU_LOG_MMU for MMU event logging
target-xtensa: add window overflow check to L32E/S32E
Max Filippov [Sun, 19 Jul 2015 06:49:00 +0000 (09:49 +0300)]
target-xtensa: implement S32NB
S32NB provides the same functionality as S32I with two exceptions.
First, when its operation leaves the processor, the external transaction
is marked Non-Bufferable. Second, it may not be used to write to
Instruction RAM.
In QEMU S32NB is equivalent to S32I.
Max Filippov [Sat, 11 Jul 2015 23:10:17 +0000 (02:10 +0300)]
target-xtensa: implement depbits instruction
This option provides an instruction for depositing a bit field from the
least significant position of one register to an arbitrary position in
another register.
Max Filippov [Sun, 27 Sep 2015 15:21:19 +0000 (18:21 +0300)]
target-xtensa: xtfpga: attach FLASH to system IO
XTFPGA FLASH is tied to XTFPGA system IO block. It's not very important
for systems with MMU where system IO block is visible at single
location, but it's important for noMMU systems, where system IO block is
accessible through two separate physical address ranges.
Map XTFPGA FLASH to system IO block and fix offsets used for mapping.
Create and initialize FLASH device with series of qdev_prop_set_* as
that's the preferred interface now. Keep initialization in a separate
function.
Max Filippov [Sun, 19 Jul 2015 07:02:37 +0000 (10:02 +0300)]
target-xtensa: add window overflow check to L32E/S32E
Despite L32E and S32E primary use is for window underflow and overflow
exception handlers they are just normal instructions, and thus need to
check for window overflow.
Peter Maydell [Wed, 21 Oct 2015 14:07:42 +0000 (15:07 +0100)]
Merge remote-tracking branch 'remotes/cohuck/tags/s390x-20151021-v2' into staging
More s390x patches. The first ones are fixes: A regression, missed
compat and a missed part of the SIMD support. The others contain
optimizations and cleanup.
# gpg: Signature made Wed 21 Oct 2015 11:24:48 BST using RSA key ID C6F02FAF
# gpg: Good signature from "Cornelia Huck <[email protected]>"
# gpg: aka "Cornelia Huck <[email protected]>"
* remotes/cohuck/tags/s390x-20151021-v2:
s390x/cmma: clean up cmma reset
s390x: reset crypto only on clear reset and QEMU reset
s390x: machine reset function with new ipl cpu handling
s390x/ipl: we always have an ipl device
s390x: unify device reset during subsystem_reset()
s390x: flagify mcic values
s390x/kvm: Fix vector validity bit in device machine checks
s390x/virtio-ccw: fix 2.4 virtio compat
util/qemu-config: fix missing machine command line options
The cmma reset is per VM, so we don't need a cpu object. We can
directly make use of kvm_state, as it is already available when
the reset is called. By moving the cmma reset in our machine reset
function, we can avoid a manual reset handler.
s390x: reset crypto only on clear reset and QEMU reset
Initializing VM crypto in initial cpu reset has multiple problems
1. We call the exact same function #VCPU times, although one time is enough
2. On SIGP initial cpu reset, we exchange the wrapping key while
other VCPUs are running. Bad!
3. It is simply wrong. According to the Pop, a reset happens only during a
clear reset.
So, we have to reset the keys
- on modified clear reset
- on load clear (QEMU reset - via machine reset)
- on qemu start (via machine reset)
s390x: machine reset function with new ipl cpu handling
Current implementation depends on the order of resets getting triggered.
If a cpu reset is triggered after the ipl device reset, the CPU is stopped and
the VM will not run. In fact, that hinders us from converting the ipl device
into a TYPE_DEVICE. Let's change that by manually configuring the ipl cpu
during a system reset, so we have full control and can demangle that code.
Also remove the superflous cpu parameter from s390_update_iplstate on the way.
commit 5be7d9f1b1452613b95c6ba70b8d7ad3d0797991
vhost-net: tell tap backend about the vnet endianness
makes vhost net always try to set LE - even if that matches the
native endian-ness.
This makes it fail on older kernels on x86 without TUNSETVNETLE support.
To fix, make qemu_set_vnet_le/qemu_set_vnet_be skip the
ioctl if it matches the host endian-ness.
Peter Maydell [Tue, 20 Oct 2015 15:51:43 +0000 (16:51 +0100)]
Merge remote-tracking branch 'remotes/berrange/tags/io-channel-3-for-upstream' into staging
Merge io-channels-3 partial branch
# gpg: Signature made Tue 20 Oct 2015 16:36:10 BST using RSA key ID 15104FDF
# gpg: Good signature from "Daniel P. Berrange <[email protected]>"
# gpg: aka "Daniel P. Berrange <[email protected]>"
* remotes/berrange/tags/io-channel-3-for-upstream:
util: pull Buffer code out of VNC module
coroutine: move into libqemuutil.a library
osdep: add qemu_fork() wrapper for safely handling signals
ui: convert VNC startup code to use SocketAddress
sockets: allow port to be NULL when listening on IP address
sockets: move qapi_copy_SocketAddress into qemu-sockets.c
sockets: add helpers for creating SocketAddress from a socket
Cornelia Huck [Thu, 8 Oct 2015 13:05:46 +0000 (15:05 +0200)]
s390x: flagify mcic values
Instead of using magic values when building the machine check
interruption code, add some defines as by chapter 11-14 in the PoP.
This should make it easier to catch problems like the missing vector
register validity bit ("s390x/kvm: Fix vector validity bit in device
machine checks"), and less hassle should we want to generate machine
checks beyond the channel reports we currently support.
s390x/kvm: Fix vector validity bit in device machine checks
Device hotplugs trigger a crw machine check. All machine checks
have validity bits for certain register types. With vector support
we also have to claim that vector registers are valid.
This is a band-aid suitable for stable. Long term we should
create the full mcic value dynamically depending on the active
features in the kernel interrupt handler.
Tony Krowiak [Mon, 12 Oct 2015 15:36:21 +0000 (11:36 -0400)]
util/qemu-config: fix missing machine command line options
Commit 0a7cf217 ("util/qemu-config: fix regression of
qmp_query_command_line_options") aimed to restore parsing of global
machine options, but missed two: "aes-key-wrap" and
"dea-key-wrap" (which were present in the initial version of that
patch). Let's add them to the machine_opts again.
The coroutine files are currently referenced by the block-obj-y
variable. The coroutine functionality though is already used by
more than just the block code. eg migration code uses coroutine
yield. In the future the I/O channel code will also use the
coroutine yield functionality. Since the coroutine code is nicely
self-contained it can be easily built as part of the libqemuutil.a
library, making it widely available.
The headers are also moved into include/qemu, instead of the
include/block directory, since they are now part of the util
codebase, and the impl was never in the block/ directory
either.
osdep: add qemu_fork() wrapper for safely handling signals
When using regular fork() the child process of course inherits
all the parents' signal handlers. If the child then proceeds
to close() any open file descriptors, it may break some of those
registered signal handlers. The child generally does not want to
ever run any of the signal handlers that the parent may have
installed in the short time before it exec's. The parent may also
have blocked various signals which the child process will want
enabled.
This introduces a wrapper qemu_fork() that takes care to sanitize
signal handling across fork. Before forking it blocks all signals
in the parent thread. After fork returns, the parent unblocks the
signals and carries on as usual. The child, however, resets all the
signal handlers back to their defaults before it unblocks signals.
The child process can now exec the binary in a "clean" signal
environment.
The VNC code is currently using QemuOpts to configure the
sockets connections / listeners it needs. Convert it to
use SocketAddress to bring it in line with modern QAPI
based code elsewhere in QEMU.
sockets: allow port to be NULL when listening on IP address
If the port in the SocketAddress struct is NULL, it can allow
the kernel to automatically select a free port. This is useful
in particular in unit tests to avoid a race trying to find a
free port to run a test case on.
sockets: move qapi_copy_SocketAddress into qemu-sockets.c
The qapi_copy_SocketAddress method is going to be useful
in more places than just qemu-char.c, so move it into
the qemu-sockets.c file to allow its reuse.
sockets: add helpers for creating SocketAddress from a socket
Add two helper methods that, given a socket file descriptor,
can return a populated SocketAddress struct containing either
the local or remote address information.
Peter Maydell [Tue, 20 Oct 2015 10:45:23 +0000 (11:45 +0100)]
Merge remote-tracking branch 'remotes/kraxel/tags/pull-fw_cfg-20151020-1' into staging
fw_cfg: add dma interface, add strings via cmdline.
# gpg: Signature made Tue 20 Oct 2015 07:07:34 BST using RSA key ID D3E87138
# gpg: Good signature from "Gerd Hoffmann (work) <[email protected]>"
# gpg: aka "Gerd Hoffmann <[email protected]>"
# gpg: aka "Gerd Hoffmann (private) <[email protected]>"
* remotes/kraxel/tags/pull-fw_cfg-20151020-1:
fw_cfg: Define a static signature to be returned on DMA port reads
Enable fw_cfg DMA interface for x86
Enable fw_cfg DMA interface for ARM
Implement fw_cfg DMA interface
fw_cfg DMA interface documentation
fw_cfg: document fw_cfg_modify_iXX() update functions
fw_cfg: insert string blobs via qemu cmdline
Peter Maydell [Tue, 20 Oct 2015 08:04:20 +0000 (09:04 +0100)]
Merge remote-tracking branch 'remotes/mdroth/tags/qga-pull-2015-10-14-v4-tag' into staging
qemu-ga patch queue
* add unit tests for qemu-ga
* add guest-exec support for posix/w32 guests
* added 'qemu-ga' target for w32. this allows us to do full MSI build,
without overloading 'qemu-ga.exe' target with uneeded dependencies.
* number of s/g_new/g_malloc/ conversions for qga
v2:
* commit message and qapi documentation spelling fixes
* rename 'inp-data' guest-exec param to 'input-data'
v3:
* fix OSX build errors for test-qga by using PRId64
format in place of glib's, and dropping use of G_SPAWN_DEFAULT
macro for glib 2.22 compat
* fix win32 build warnings for 32-bit builds by avoid int casts
of process HANDLEs
v4:
* assert connect_qga() doesn't fail
* only enable test-qga for linux hosts
* allow get-memory-block-info* to fail if memory blocks aren't exposed in
sysfs
* remotes/mdroth/tags/qga-pull-2015-10-14-v4-tag:
qga: fix uninitialized value warning for win32
qga: guest-exec simple stdin/stdout/stderr redirection
qga: handle G_IO_STATUS_AGAIN in ga_channel_write_all()
qga: handle possible SIGPIPE in guest-file-write
qga: guest exec functionality
qga: drop guest_file_init helper and replace it with static initializers
tests: add a local test for guest agent
qga: guest-get-memory-blocks shouldn't fail for unexposed memory blocks
glib-compat: add 2.38/2.40/2.46 asserts
qtest: add a few fd-level qmp helpers
qga: do not override configuration verbosity
qga: add QGA_CONF environment variable
qga: Use g_new() & friends where that makes obvious sense
build: qemu-ga: add 'qemu-ga' build target for w32
Denis V. Lunev [Tue, 13 Oct 2015 15:41:21 +0000 (18:41 +0300)]
qga: handle possible SIGPIPE in guest-file-write
qemu-ga should not exit on guest-file-write to pipe without read end
but proper error code should be returned. The behavior of the
spawned process should be default thus SIGPIPE processing should be
reset to default after fork() but before exec().
Guest-exec rewritten in platform-independent style with glib spawn.
Child process is spawn asynchronously and exit status can later
be picked up by guest-exec-status command.
stdin/stdout/stderr of the child now is redirected to /dev/null
Later we will add ability to specify stdin in guest-exec command
and to get collected stdout/stderr with guest-exec-status.
Signed-off-by: Yuri Pudgorodskiy <[email protected]> Signed-off-by: Denis V. Lunev <[email protected]> CC: Michael Roth <[email protected]>
* use g_new0 in place of g_malloc for GuestExec struct
* commit msg spelling fixes
* s/inp-data/input-data
* document capture-input mode as false by default
* use GetProcessId() for pids on w32 instead of casting HANDLE Signed-off-by: Michael Roth <[email protected]>
Add some local guest agent tests, as it is better than nothing, only
when CONFIG_POSIX (using unix sockets).
With the QGA_TEST_SIDE_EFFECTING environment variable, it will include
tests with side effects, such as freezing/thawing the FS or changing the
time.
(a better test would involve a managed VM (or container), but it might
be better to leave that off to autotest/avocado)
Signed-off-by: Marc-André Lureau <[email protected]>
* use mkdtemp() in placeof g_mkdtemp() for glib 2.22 compat
* drop redundant/conflicting compat defines for
g_assert_{true,false}, since glib-compat has them now.
* build fixes for OSX: use PRId64 instead of glib formats, drop
g_spawn_default usage for glib compat
* assert connect_qga() doesn't fail
* only enable test-qga for linux hosts
* allow get-memory-block-info* to fail Signed-off-by: Michael Roth <[email protected]>
Michael Roth [Mon, 19 Oct 2015 21:38:25 +0000 (16:38 -0500)]
qga: guest-get-memory-blocks shouldn't fail for unexposed memory blocks
Some guests don't expose memory blocks via sysfs at all. This
shouldn't be a failure, instead just return an empty list. For
other access failures we still report an error.
Having a environment variable allows to override default configuration
path, useful for testing. Note that this can't easily be an argument,
since loading config is done before parsing the arguments.
qga: Use g_new() & friends where that makes obvious sense
g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
for two reasons. One, it catches multiplication overflowing size_t.
Two, it returns T * rather than void *, which lets the compiler catch
more type errors.
This commit only touches allocations with size arguments of the form
sizeof(T). Same Coccinelle semantic patch as in commit b45c03f.
Michael Roth [Mon, 7 Sep 2015 23:47:05 +0000 (18:47 -0500)]
build: qemu-ga: add 'qemu-ga' build target for w32
Currently POSIX builds rely on 'qemu-ga' target to do qga-only
distributable build. On w32, as with most standalone binary targets,
we rely on 'qemu-ga.exe' target.
Unlike with POSIX, qemu-ga for w32 has a number of related targets
such as VSS DLL and MSI package. We can do the full distributable
qga-only build on w32 with:
make qemu-ga.exe
or:
make msi
To make that work, we tie VSS dependencies onto qemu-ga.exe.
However, in reality the DLL isn't part of the binary, so we use a
filter to pull them out of the LINK recipe, which attempts to link
against prereqs for binary targets. Additionally, it could be argued
that VSS is a separate distributable, and shouldn't be implied by
qemu-ga.exe binary target.
To avoid this, we can tie the VSS dependencies only to the 'msi'
target, but that would make it impossible to do a qga-only build of
the w32 distributable without building the 'msi' package, which was
supported in the past.
An alternative approach is to add a new target to build the whole
distributable. w32 allows us to use the same build target we use
on POSIX, 'qemu-ga', since the current binary-only target on w32
is 'qemu-ga.exe'.
To further simplify the build, we also make 'qemu-ga' build the MSI
package if the appropriate ./configure options are set, making the
full qga-only build the same on both POSIX and w32: `make qemu-ga`
James Hogan [Fri, 2 Oct 2015 12:24:17 +0000 (13:24 +0100)]
tcg/mips: Support r6 SEL{NE, EQ}Z instead of MOVN/MOVZ
Extend MIPS movcond implementation to support the SELNEZ/SELEQZ
instructions introduced in MIPS r6 (where MOVN/MOVZ have been removed).
Whereas the "MOVN/MOVZ rd, rs, rt" instructions have the following
semantics:
rd = [!]rt ? rs : rd
The "SELNEZ/SELEQZ rd, rs, rt" instructions are slightly different:
rd = [!]rt ? rs : 0
First we ensure that if one of the movcond input values is zero that it
comes last (we can swap the input arguments if we invert the condition).
This is so that it can exactly match one of the SELNEZ/SELEQZ
instructions and avoid the need to emit the other one.
Otherwise we emit the opposite instruction first into a temporary
register, and OR that into the result:
SELNEZ/SELEQZ TMP1, v2, c1
SELEQZ/SELNEZ ret, v1, c1
OR ret, ret, TMP1
projects / qemu.git / log