[desc]:
Large memory VM starts slowly when using -mem-prealloc, and
there are some areas to optimize in current method;
1、mmap will be used to alloc threads stack during create page
clearing threads, and it will attempt mm->mmap_sem for write
lock, but clearing threads have hold read lock, this competition
will cause threads createion very slow;
2、methods of calcuating pages for per threads is not well;if we use
64 threads to split 160 hugepage,63 threads clear 2page,1 thread
clear 34 page,so the entire speed is very slow;
to solve the first problem,we add a mutex in thread function,and
start all threads when all threads finished createion;
and the second problem, we spread remainder to other threads,in
situation that 160 hugepage and 64 threads, there are 32 threads
clear 3 pages,and 32 threads clear 2 pages.
[test]:
320G 84c VM start time can be reduced to 10s
680G 84c VM start time can be reduced to 18s
Signed-off-by: bauerchen <bauerchen@tencent.com>
Reviewed-by: Pan Rui <ruippan@tencent.com>
Reviewed-by: Ivan Ren <ivanren@tencent.com>
[Simplify computation of the number of pages per thread. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The virtual-device fuzzer must initialize QOM, prior to running
vl:qemu_init, so that it can use the qos_graph to identify the arguments
required to initialize a guest for libqos-assisted fuzzing. This change
prevents errors when vl:qemu_init tries to (re)initialize the previously
initialized QOM module.
Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Darren Kenny <darren.kenny@oracle.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20200220041118.23264-4-alxndr@bu.edu
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
File descriptor monitoring is O(1) with epoll(7), but
aio_dispatch_handlers() still scans all AioHandlers instead of
dispatching just those that are ready. This makes aio_poll() O(n) with
respect to the total number of registered handlers.
Add a local ready_list to aio_poll() so that each nested aio_poll()
builds a list of handlers ready to be dispatched. Since file descriptor
polling is level-triggered, nested aio_poll() calls also see fds that
were ready in the parent but not yet dispatched. This guarantees that
nested aio_poll() invocations will dispatch all fds, even those that
became ready before the nested invocation.
Since only handlers ready to be dispatched are placed onto the
ready_list, the new aio_dispatch_ready_handlers() function provides O(1)
dispatch.
Note that AioContext polling is still O(n) and currently cannot be fully
disabled. This still needs to be fixed before aio_poll() is fully O(1).
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Sergio Lopez <slp@redhat.com>
Message-id: 20200214171712.541358-6-stefanha@redhat.com
[Fix compilation error on macOS where there is no epoll(87). The
aio_epoll() prototype was out of date and aio_add_ready_list() needed to
be moved outside the ifdef.
--Stefan]
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
It is not necessary to scan all AioHandlers for deletion. Keep a list
of deleted handlers instead of scanning the full list of all handlers.
The AioHandler->deleted field can be dropped. Let's check if the
handler has been inserted into the deleted list instead. Add a new
QLIST_IS_INSERTED() API for this check.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Sergio Lopez <slp@redhat.com>
Message-id: 20200214171712.541358-5-stefanha@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Don't pass the nanosecond timeout into epoll_wait(), which expects
milliseconds.
The epoll_wait() timeout value does not matter if qemu_poll_ns()
determined that the poll fd is ready, but passing a value in the wrong
units is still ugly. Pass a 0 timeout to epoll_wait() instead.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Sergio Lopez <slp@redhat.com>
Message-id: 20200214171712.541358-3-stefanha@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
epoll_handler is a stack variable and must not be accessed after it goes
out of scope:
if (aio_epoll_check_poll(ctx, pollfds, npfd, timeout)) {
AioHandler epoll_handler;
...
add_pollfd(&epoll_handler);
ret = aio_epoll(ctx, pollfds, npfd, timeout);
} ...
...
/* if we have any readable fds, dispatch event */
if (ret > 0) {
for (i = 0; i < npfd; i++) {
nodes[i]->pfd.revents = pollfds[i].revents;
}
}
nodes[0] is &epoll_handler, which has already gone out of scope.
There is no need to use pollfds[] for epoll. We don't need an
AioHandler for the epoll fd.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Sergio Lopez <slp@redhat.com>
Message-id: 20200214171712.541358-2-stefanha@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
The ctx->first_bh list contains all created BHs, including those that
are not scheduled. The list is iterated by the event loop and therefore
has O(n) time complexity with respected to the number of created BHs.
Rewrite BHs so that only scheduled or deleted BHs are enqueued.
Only BHs that actually require action will be iterated.
One semantic change is required: qemu_bh_delete() enqueues the BH and
therefore invokes aio_notify(). The
tests/test-aio.c:test_source_bh_delete_from_cb() test case assumed that
g_main_context_iteration(NULL, false) returns false after
qemu_bh_delete() but it now returns true for one iteration. Fix up the
test case.
This patch makes aio_compute_timeout() and aio_bh_poll() drop from a CPU
profile reported by perf-top(1). Previously they combined to 9% CPU
utilization when AioContext polling is commented out and the guest has 2
virtio-blk,num-queues=1 and 99 virtio-blk,num-queues=32 devices.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-id: 20200221093951.1414693-1-stefanha@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
The first rcu_read_lock/unlock() is expensive. Nested calls are cheap.
This optimization increases IOPS from 73k to 162k with a Linux guest
that has 2 virtio-blk,num-queues=1 and 99 virtio-blk,num-queues=32
devices.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-id: 20200218182708.914552-1-stefanha@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
ppc patch queue 2020-02-21
Here's the next patch of ppc target patches. Highlights are:
* Some fixes for CAS / unplug interactions
* Remove some leaks of device trees
* Some fixes for the PHB3 and PHB4 devices
* Support for NVDIMMs on the pseries machine type
* Assorted other fixes and cleanups
# gpg: Signature made Fri 21 Feb 2020 03:35:40 GMT
# gpg: using RSA key 75F46586AE61A66CC44E87DC6C38CACA20D9B392
# gpg: Good signature from "David Gibson <david@gibson.dropbear.id.au>" [full]
# gpg: aka "David Gibson (Red Hat) <dgibson@redhat.com>" [full]
# gpg: aka "David Gibson (ozlabs.org) <dgibson@ozlabs.org>" [full]
# gpg: aka "David Gibson (kernel.org) <dwg@kernel.org>" [unknown]
# Primary key fingerprint: 75F4 6586 AE61 A66C C44E 87DC 6C38 CACA 20D9 B392
* remotes/dgibson/tags/ppc-for-5.0-20200221:
hw/ppc/virtex_ml507:fix leak of fdevice tree blob
spapr: Fix handling of unplugged devices during CAS and migration
spapr: Don't use spapr_drc_needed() in CAS code
ppc: free 'fdt' after reset the machine
target/ppc/cpu.h: Clean up comments in the struct CPUPPCState definition
target/ppc/cpu.h: Move fpu related members closer in cpu env
target/ppc: Fix typo in comments
spapr: Allow changing offset for -kernel image
pnv/phb3: Add missing break statement
pnv/phb4: Fix error path in pnv_pec_realize()
pnv/phb3: Convert 1u to 1ull
target/ppc/cpu.h: Remove duplicate includes
spapr: Add Hcalls to support PAPR NVDIMM device
spapr: Add NVDIMM device support
nvdimm: add uuid property to nvdimm
mem: move nvdimm_device_list to utilities
ppc: function to setup latest class options
ppc/pnv: Fix PCI_EXPRESS dependency
qtest: Fix rtas dependencies
spapr/rtas: Print message from "ibm,os-term"
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Implement membarrier, SO_RCVTIMEO and SO_SNDTIMEO
Disable by default build of fdt, slirp and tools with linux-user
Improve strace and use qemu_log to send trace to a file
Add partial ALSA ioctl supports
# gpg: Signature made Thu 20 Feb 2020 09:20:20 GMT
# gpg: using RSA key CD2F75DDC8E3A4DC2E4F5173F30C38BD3F2FBE3C
# gpg: issuer "laurent@vivier.eu"
# gpg: Good signature from "Laurent Vivier <lvivier@redhat.com>" [full]
# gpg: aka "Laurent Vivier <laurent@vivier.eu>" [full]
# gpg: aka "Laurent Vivier (Red Hat) <lvivier@redhat.com>" [full]
# Primary key fingerprint: CD2F 75DD C8E3 A4DC 2E4F 5173 F30C 38BD 3F2F BE3C
* remotes/vivier2/tags/linux-user-for-5.0-pull-request:
linux-user: Add support for selected alsa timer instructions using ioctls
linux-user: Add support for getting/setting selected alsa timer parameters using ioctls
linux-user: Add support for selecting alsa timer using ioctl
linux-user: Add support for getting/setting specified alsa timer parameters using ioctls
linux-user: Add support for getting alsa timer version and id
linux-user: remove gemu_log from the linux-user tree
linux-user: Use `qemu_log' for strace
linux-user: Use `qemu_log' for non-strace logging
configure: Avoid compiling system tools on user build by default
linux-user/strace: Improve output of various syscalls
configure: linux-user doesn't need neither fdt nor slirp
linux-user: implement getsockopt SO_RCVTIMEO and SO_SNDTIMEO
linux-user: Implement membarrier syscall
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
This change switches linux-user strace logging to use the newer `qemu_log`
logging subsystem rather than the older `gemu_log` (notice the "g")
logger. `qemu_log` has several advantages, namely that it allows logging
to a file, and provides a more unified interface for configuration
of logging (via the QEMU_LOG environment variable or options).
This change introduces a new log mask: `LOG_STRACE` which is used for
logging of user-mode strace messages.
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Signed-off-by: Josh Kunz <jkz@google.com>
Message-Id: <20200204025416.111409-3-jkz@google.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Some older parts of QEMU's codebase assume that CLOCK_MONOTONIC
might not be defined by the host OS, and have workarounds to
deal with this. However, more recently (notably in commit
50290c002c for qemu-img in mid-2019, but also much
earlier in 2011 in commit 22795174a3 for ui/spice-display.c)
we've written code that assumes CLOCK_MONOTONIC is always
defined. The only host OS anybody's ever noticed this on
is OSX 10.11 and earlier, which we don't support.
So we can assume that all our host OSes have the #define,
and we can remove some now-unnecessary ifdefs.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20200201172252.6605-1-peter.maydell@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
NULL is a valid log filename used to indicate we want to use stderr
but qemu_set_log_filename (which is called by bsd-user/main.c) was not
handling it correctly.
That also made redundant a couple of NULL checks in calling code which
have been removed.
Signed-off-by: Salvador Fandino <salvador@qindel.com>
Message-Id: <20200123193626.19956-1-salvador@qindel.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
uClibc defines _SC_LEVEL1_ICACHE_LINESIZE and _SC_LEVEL1_DCACHE_LINESIZE
but the corresponding sysconf calls returns -1, which is a valid result,
meaning that the limit is indeterminate.
Handle this situation using the fallback values instead of crashing due
to an assertion failure.
Signed-off-by: Carlos Santos <casantos@redhat.com>
Message-Id: <20191017123713.30192-1-casantos@redhat.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
* Compat machines fix (Denis)
* Command line parsing fixes (Michal, Peter, Xiaoyao)
* Cooperlake CPU model fixes (Xiaoyao)
* i386 gdb fix (mkdolata)
* IOEventHandler cleanup (Philippe)
* icount fix (Pavel)
* RR support for random number sources (Pavel)
* Kconfig fixes (Philippe)
# gpg: Signature made Wed 08 Jan 2020 10:41:00 GMT
# gpg: using RSA key BFFBD25F78C7AE83
# gpg: Good signature from "Paolo Bonzini <bonzini@gnu.org>" [full]
# gpg: aka "Paolo Bonzini <pbonzini@redhat.com>" [full]
# Primary key fingerprint: 46F5 9FBD 57D6 12E7 BFD4 E2F7 7E15 100C CD36 69B1
# Subkey fingerprint: F133 3857 4B66 2389 866C 7682 BFFB D25F 78C7 AE83
* remotes/bonzini/tags/for-upstream: (38 commits)
chardev: Use QEMUChrEvent enum in IOEventHandler typedef
chardev: use QEMUChrEvent instead of int
chardev/char: Explicit we ignore some QEMUChrEvent in IOEventHandler
monitor/hmp: Explicit we ignore a QEMUChrEvent in IOEventHandler
monitor/qmp: Explicit we ignore few QEMUChrEvent in IOEventHandler
virtio-console: Explicit we ignore some QEMUChrEvent in IOEventHandler
vhost-user-blk: Explicit we ignore few QEMUChrEvent in IOEventHandler
vhost-user-net: Explicit we ignore few QEMUChrEvent in IOEventHandler
vhost-user-crypto: Explicit we ignore some QEMUChrEvent in IOEventHandler
ccid-card-passthru: Explicit we ignore QEMUChrEvent in IOEventHandler
hw/usb/redirect: Explicit we ignore few QEMUChrEvent in IOEventHandler
hw/usb/dev-serial: Explicit we ignore few QEMUChrEvent in IOEventHandler
hw/char/terminal3270: Explicit ignored QEMUChrEvent in IOEventHandler
hw/ipmi: Explicit we ignore some QEMUChrEvent in IOEventHandler
hw/ipmi: Remove unnecessary declarations
target/i386: Add missed features to Cooperlake CPU model
target/i386: Add new bit definitions of MSR_IA32_ARCH_CAPABILITIES
target/i386: Fix handling of k_gs_base register in 32-bit mode in gdbstub
hw/rtc/mc146818: Add missing dependency on ISA Bus
hw/nvram/Kconfig: Restrict CHRP NVRAM to machines using OpenBIOS or SLOF
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Internally, qemu may create chardev without ID. Those will not be
looked up with qemu_chr_find(), which prevents using qdev_prop_set_chr().
Use id_generate(), to generate an internal name (prefixed with #), so
no conflict exist with user-named chardev.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: xiaoqiang zhao <zxq_yx_007@163.com>
Record/replay feature of icount allows deterministic running of execution
scenarios. Some CPUs and peripheral devices read random numbers from
external sources making deterministic execution impossible.
This patch adds recording and replaying of random read operations
into guest-random module, which is used by the virtual hardware.
Signed-off-by: Pavel Dovgalyuk <Pavel.Dovgaluk@ispras.ru>
Message-Id: <157675984852.14505.15709141760677102489.stgit@pasha-Precision-3630-Tower>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add dbus-vmstate
Hi,
With external processes or helpers participating to the VM support, it
becomes necessary to handle their migration. Various options exist to
transfer their state:
1) as the VM memory, RAM or devices (we could say that's how
vhost-user devices can be handled today, they are expected to
restore from ring state)
2) other "vmstate" (as with TPM emulator state blobs)
3) left to be handled by management layer
1) is not practical, since an external processes may legitimatelly
need arbitrary state date to back a device or a service, or may not
even have an associated device.
2) needs ad-hoc code for each helper, but is simple and working
3) is complicated for management layer, QEMU has the migration timing
The proposed "dbus-vmstate" object will connect to a given D-Bus
address, and save/load from org.qemu.VMState1 owners on migration.
Thus helpers can easily have their state migrated with QEMU, without
implementing ad-hoc support (such as done for TPM emulation)
D-Bus is ubiquitous on Linux (it is systemd IPC), and can be made to
work on various other OSes. There are several implementations and good
bindings for various languages. (the tests/dbus-vmstate-test.c is a
good example of how simple the implementation of services can be, even
in C)
dbus-vmstate is put into use by the libvirt series "[PATCH 00/23] Use
a slirp helper process".
v2:
- fix build with broken mingw-glib
# gpg: Signature made Mon 06 Jan 2020 14:43:35 GMT
# gpg: using RSA key 87A9BD933F87C606D276F62DDAE8E10975969CE5
# gpg: issuer "marcandre.lureau@redhat.com"
# gpg: Good signature from "Marc-André Lureau <marcandre.lureau@redhat.com>" [full]
# gpg: aka "Marc-André Lureau <marcandre.lureau@gmail.com>" [full]
# Primary key fingerprint: 87A9 BD93 3F87 C606 D276 F62D DAE8 E109 7596 9CE5
* remotes/elmarco/tags/dbus-vmstate7-pull-request:
tests: add dbus-vmstate-test
tests: add migration-helpers unit
dockerfiles: add dbus-daemon to some of latest distributions
configure: add GDBUS_CODEGEN
Add dbus-vmstate object
util: add dbus helper unit
docs: start a document to describe D-Bus usage
vmstate: replace DeviceState with VMStateIf
vmstate: add qom interface to get id
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
bauerchenandPaolo Bonzini
Alexander BulekovandStefan Hajnoczi
Peter Maydell
Shivaprasad G BhatandDavid Gibson
Josh KunzandLaurent Vivier
Michal PrivoznikandLaurent Vivier
Salvador FandinoandStefan Hajnoczi
Aarushi MehtaandStefan Hajnoczi
Carlos SantosandRichard Henderson
Pan NengyuanandLaurent Vivier
Marc-André Lureau
Pavel DovgalyukandPaolo Bonzini