xemu/util
Claudio Imbrenda c891c24b1a os-posix: asynchronous teardown for shutdown on Linux
This patch adds support for asynchronously tearing down a VM on Linux.

When qemu terminates, either naturally or because of a fatal signal,
the VM is torn down. If the VM is huge, it can take a considerable
amount of time for it to be cleaned up. In case of a protected VM, it
might take even longer than a non-protected VM (this is the case on
s390x, for example).

Some users might want to shut down a VM and restart it immediately,
without having to wait. This is especially true if management
infrastructure like libvirt is used.

This patch implements a simple trick on Linux to allow qemu to return
immediately, with the teardown of the VM being performed
asynchronously.

If the new commandline option -async-teardown is used, a new process is
spawned from qemu at startup, using the clone syscall, in such way that
it will share its address space with qemu.The new process will have the
name "cleanup/<QEMU_PID>". It will wait until qemu terminates
completely, and then it will exit itself.

This allows qemu to terminate quickly, without having to wait for the
whole address space to be torn down. The cleanup process will exit
after qemu, so it will be the last user of the address space, and
therefore it will take care of the actual teardown. The cleanup
process will share the same cgroups as qemu, so both memory usage and
cpu time will be accounted properly.

If possible, close_range will be used in the cleanup process to close
all open file descriptors. If it is not available or if it fails, /proc
will be used to determine which file descriptors to close.

If the cleanup process is forcefully killed with SIGKILL before the
main qemu process has terminated completely, the mechanism is defeated
and the teardown will not be asynchronous.

This feature can already be used with libvirt by adding the following
to the XML domain definition to pass the parameter to qemu directly:

  <commandline xmlns="http://libvirt.org/schemas/domain/qemu/1.0">
  <arg value='-async-teardown'/>
  </commandline>

Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Reviewed-by: Murilo Opsfelder Araujo <muriloo@linux.ibm.com>
Tested-by: Murilo Opsfelder Araujo <muriloo@linux.ibm.com>
Message-Id: <20220812133453.82671-1-imbrenda@linux.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2022-10-31 09:46:34 +01:00
..
aio-posix.c util/event-loop-base: Introduce options to set the thread pool size 2022-05-09 10:43:23 +01:00
aio-posix.h aio-posix: fix spurious ->poll_ready() callbacks in main loop 2022-03-17 11:23:18 +00:00
aio-wait.c aio_wait_kick: add missing memory barrier 2022-06-24 17:07:06 +02:00
aio-win32.c Remove qemu-common.h include from most units 2022-04-06 14:31:55 +02:00
aiocb.c
async-teardown.c os-posix: asynchronous teardown for shutdown on Linux 2022-10-31 09:46:34 +01:00
async.c replay: notify vCPU when BH is scheduled 2022-06-06 09:26:53 +02:00
atomic64.c osdep: Move memalign-related functions to their own header 2022-03-07 13:16:49 +00:00
base64.c
bitmap.c Replace config-time define HOST_WORDS_BIGENDIAN 2022-04-06 10:50:37 +02:00
bitops.c
block-helpers.c
block-helpers.h
buffer.c
bufferiszero.c cpuid: use unsigned for max cpuid 2022-02-04 09:07:43 -05:00
cacheflush.c util/cacheflush: Optimize flushing when ppc host has coherent icache 2022-06-21 09:28:41 -07:00
compatfd.c util: replace pipe()+cloexec with g_unix_open_pipe() 2022-05-03 15:18:14 +04:00
coroutine-sigaltstack.c Remove qemu-common.h include from most units 2022-04-06 14:31:55 +02:00
coroutine-ucontext.c coroutine-ucontext: use QEMU_DEFINE_STATIC_CO_TLS() 2022-05-04 15:55:23 +02:00
coroutine-win32.c coroutine-win32: use QEMU_DEFINE_STATIC_CO_TLS() 2022-05-04 15:55:23 +02:00
crc-ccitt.c
crc32c.c
cutils.c cutils: Add missing dyld(3) include on macOS 2022-08-12 11:33:52 +01:00
dbus.c
drm.c
envlist.c Use g_new() & friends where that makes obvious sense 2022-03-21 15:44:44 +01:00
error-report.c error-report: fix g_date_time_format assertion 2022-04-28 08:51:56 +02:00
error.c
event_notifier-posix.c Replace qemu_pipe() with g_unix_open_pipe() 2022-05-03 15:17:56 +04:00
event_notifier-win32.c Remove qemu-common.h include from most units 2022-04-06 14:31:55 +02:00
fdmon-epoll.c
fdmon-io_uring.c aio-posix: fix build failure io_uring 2.2 2022-03-17 11:23:18 +00:00
fdmon-poll.c
fifo8.c
filemonitor-inotify.c
filemonitor-stub.c
getauxval.c
guest-random.c util/guest-random: Fix size arg to tail memcpy 2021-07-09 18:42:46 +02:00
hbitmap.c block: simplify handling of try to merge different sized bitmaps 2022-06-24 17:07:06 +02:00
hexdump.c include: move C/util-related declarations to cutils.h 2022-04-06 14:31:43 +02:00
host-utils.c host-utils: Implemented signed 256-by-128 division 2022-06-20 08:38:58 -03:00
id.c
int128.c qemu/int128: addition of div/rem 128-bit operations 2022-01-08 15:46:10 +10:00
iov.c util: make do_send_recv work with partial send/recv 2022-10-12 19:22:01 +04:00
iova-tree.c util: accept iova_tree_remove_parameter by value 2022-09-02 10:22:39 +08:00
keyval.c include: add qemu/keyval.h 2022-04-21 17:03:51 +04:00
lockcnt.c
log.c util/log: Support per-thread log files 2022-04-20 10:51:11 -07:00
main-loop.c util/event-loop-base: Introduce options to set the thread pool size 2022-05-09 10:43:23 +01:00
memalign.c osdep: Move memalign-related functions to their own header 2022-03-07 13:16:49 +00:00
memfd.c
meson.build os-posix: asynchronous teardown for shutdown on Linux 2022-10-31 09:46:34 +01:00
mmap-alloc.c util/mmap-alloc: Remove qemu_mempath_getpagesize() 2022-08-26 13:34:21 +02:00
module.c module: Use bundle mechanism 2022-07-13 16:58:57 +02:00
notify.c
nvdimm-utils.c
osdep.c osdep: make readv_writev() work with partial read/write 2022-10-12 19:22:00 +04:00
oslib-posix.c oslib-posix: Introduce qemu_socketpair() 2022-09-29 14:38:05 +04:00
oslib-win32.c util/win32: simplify qemu_get_local_state_dir() 2022-05-28 11:42:56 +02:00
path.c
qdist.c
qemu-co-shared-resource.c co-shared-resource: protect with a mutex 2021-06-25 14:24:24 +03:00
qemu-co-timeout.c util: add qemu-co-timeout 2022-06-29 10:56:12 +03:00
qemu-config.c s390x/s390-virtio-ccw: add zpcii-disable machine property 2022-09-26 17:23:47 +02:00
qemu-coroutine-io.c Remove qemu-common.h include from most units 2022-04-06 14:31:55 +02:00
qemu-coroutine-lock.c coroutine: add flag to re-queue at front of CoQueue 2022-10-26 14:56:42 -04:00
qemu-coroutine-sleep.c coroutine-sleep: introduce qemu_co_sleep 2021-05-21 18:22:33 +01:00
qemu-coroutine.c coroutine: remove incorrect coroutine_fn annotations 2022-10-07 12:11:40 +02:00
qemu-option.c qemu-option: Allow deleting opts during qemu_opts_foreach() 2021-10-15 16:11:22 +02:00
qemu-print.c
qemu-progress.c include: move progress API to qemu-progress.h 2022-04-06 14:31:43 +02:00
qemu-sockets.c util/qemu-sockets: Enable unix socket support on Windows 2022-09-02 15:54:46 +04:00
qemu-thread-common.h
qemu-thread-posix.c thread-posix: optimize qemu_sem_timedwait with zero timeout 2022-04-06 14:31:56 +02:00
qemu-thread-win32.c win32: set threads name 2022-10-12 19:21:31 +04:00
qemu-timer-common.c
qemu-timer.c qemu-timer: Skip empty timer lists before locking in qemu_clock_deadline_ns_all 2022-06-21 09:24:34 -07:00
qht.c osdep: Move memalign-related functions to their own header 2022-03-07 13:16:49 +00:00
qsp.c qemu/atomic: Add aligned_{int64,uint64}_t types 2021-07-21 07:45:38 -10:00
range.c
rcu.c rcu: use coroutine TLS macros 2022-03-04 18:14:40 +01:00
readline.c
selfmap.c util/selfmap: Discard mapping on error 2021-07-26 07:06:49 -10:00
stats64.c
sys_membarrier.c
systemd.c
thread-pool.c thread-pool: remove stopping variable 2022-05-25 21:26:26 +02:00
throttle.c
timed-average.c
trace-events modules: add tracepoints 2021-07-09 18:20:27 +02:00
trace.h
transactions.c transactions: Invoke clean() after everything else 2021-11-16 09:43:44 +01:00
unicode.c
uri.c util/uri: do not check argument of uri_free() 2021-07-09 12:26:05 +02:00
userfaultfd.c
uuid.c
vfio-helpers.c numa: call ->ram_block_removed() in ram_block_notifer_remove() 2022-10-26 14:56:42 -04:00
vhost-user-server.c util: rename qemu_*block() socket functions 2022-05-03 15:53:20 +04:00
yank.c yank: Remove dependency on qiochannel 2021-04-01 15:27:44 +04:00