This file begins tracking the files that will be the code base for HVF
support in QEMU. This code base is part of Google's QEMU version of
their Android emulator, and can be found at
https://android.googlesource.com/platform/external/qemu/+/emu-master-dev
This code is based on Veertu Inc's vdhh (Veertu Desktop Hosted
Hypervisor), found at https://github.com/veertuinc/vdhh. Everything is
appropriately licensed under GPL v2-or-later, except for the code inside
x86_task.c and x86_task.h, which, deriving from KVM (the Linux kernel),
is licensed GPL v2-only.
This code base already implements a very great deal of functionality,
although Google's version removed from Vertuu's the support for APIC
page and hyperv-related stuff. According to the Android Emulator Release
Notes, Revision 26.1.3 (August 2017), "Hypervisor.framework is now
enabled by default on macOS for 32-bit x86 images to improve performance
and macOS compatibility", although we better use with caution for, as the
same Revision warns us, "If you experience issues with it specifically,
please file a bug report...". The code hasn't seen much update in the
last 5 months, so I think that we can further develop the code with
occasional visiting Google's repository to see if there has been any
update.
On top of Google's code, the following changes were made:
- add code to the configure script to support the --enable-hvf argument.
If the OS is Darwin, it checks for presence of HVF in the system. The
patch also adds strings related to HVF in the file qemu-options.hx.
QEMU will only support the modern syntax style '-M accel=hvf' no enable
hvf; the legacy '-enable-hvf' will not be supported.
- fix styling issues
- add glue code to cpus.c
- move HVFX86EmulatorState field to CPUX86State, changing the
the emulation functions to have a parameter with signature 'CPUX86State *'
instead of 'CPUState *' so we don't have to get the 'env'.
Signed-off-by: Sergio Andres Gomez Del Real <Sergio.G.DelReal@gmail.com>
Message-Id: <20170913090522.4022-2-Sergio.G.DelReal@gmail.com>
Message-Id: <20170913090522.4022-3-Sergio.G.DelReal@gmail.com>
Message-Id: <20170913090522.4022-5-Sergio.G.DelReal@gmail.com>
Message-Id: <20170913090522.4022-6-Sergio.G.DelReal@gmail.com>
Message-Id: <20170905035457.3753-7-Sergio.G.DelReal@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Normally we create an address space for that CPU and pass that address
space into the function. Let's just do it inside to unify address space
creations. It'll simplify my next patch to rename those address spaces.
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20171123092333.16085-3-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
pause_all_cpus() is sometimes called from a VCPU thread (e.g. s390x
during special reset). It cannot deal with multiple VCPUs per Thread
(single threaded TCG) yet.
Booting an s390x guest with -smp 2 and single threaded TCG from disk
currently fails. The DIAG 308 will issue a pause_all_cpus() and wait
forever for the CPUs to actually stop. But it is waiting for itself.
So let's stop all VCPUs belonging to the current thread. Factor out
stopping of a VCPU.
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20171129191215.11323-1-david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This enables parallel TCG code generation. However, we do not take
advantage of it yet since tb_lock is still held during tb_gen_code.
In user-mode we use a single TCG context; see the documentation
added to tcg_region_init for the rationale.
Note that targets do not need any conversion: targets initialize a
TCGContext (e.g. defining TCG globals), and after this initialization
has finished, the context is cloned by the vCPU threads, each of
them keeping a separate copy.
TCG threads claim one entry in tcg_ctxs[] by atomically increasing
n_tcg_ctxs. Do not be too annoyed by the subsequent atomic_read's
of that variable and tcg_ctxs; they are there just to play nice with
analysis tools such as thread sanitizer.
Note that we do not allocate an array of contexts (we allocate
an array of pointers instead) because when tcg_context_init
is called, we do not know yet how many contexts we'll use since
the bool behind qemu_tcg_mttcg_enabled() isn't set yet.
Previous patches folded some TCG globals into TCGContext. The non-const
globals remaining are only set at init time, i.e. before the TCG
threads are spawned. Here is a list of these set-at-init-time globals
under tcg/:
Only written by tcg_context_init:
- indirect_reg_alloc_order
- tcg_op_defs
Only written by tcg_target_init (called from tcg_context_init):
- tcg_target_available_regs
- tcg_target_call_clobber_regs
- arm: arm_arch, use_idiv_instructions
- i386: have_cmov, have_bmi1, have_bmi2, have_lzcnt,
have_movbe, have_popcnt
- mips: use_movnz_instructions, use_mips32_instructions,
use_mips32r2_instructions, got_sigill (tcg_target_detect_isa)
- ppc: have_isa_2_06, have_isa_3_00, tb_ret_addr
- s390: tb_ret_addr, s390_facilities
- sparc: qemu_ld_trampoline, qemu_st_trampoline (build_trampolines),
use_vis3_instructions
Only written by tcg_prologue_init:
- 'struct jit_code_entry one_entry'
- aarch64: tb_ret_addr
- arm: tb_ret_addr
- i386: tb_ret_addr, guest_base_flags
- ia64: tb_ret_addr
- mips: tb_ret_addr, bswap32_addr, bswap32u_addr, bswap64_addr
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This is groundwork for supporting multiple TCG contexts.
The naive solution here is to split code_gen_buffer statically
among the TCG threads; this however results in poor utilization
if translation needs are different across TCG threads.
What we do here is to add an extra layer of indirection, assigning
regions that act just like pages do in virtual memory allocation.
(BTW if you are wondering about the chosen naming, I did not want
to use blocks or pages because those are already heavily used in QEMU).
We use a global lock to serialize allocations as well as statistics
reporting (we now export the size of the used code_gen_buffer with
tcg_code_size()). Note that for the allocator we could just use
a counter and atomic_inc; however, that would complicate the gathering
of tcg_code_size()-like stats. So given that the region operations are
not a fast path, a lock seems the most reasonable choice.
The effectiveness of this approach is clear after seeing some numbers.
I used the bootup+shutdown of debian-arm with '-tb-size 80' as a benchmark.
Note that I'm evaluating this after enabling per-thread TCG (which
is done by a subsequent commit).
* -smp 1, 1 region (entire buffer):
qemu: flush code_size=83885014 nb_tbs=154739 avg_tb_size=357
qemu: flush code_size=83884902 nb_tbs=153136 avg_tb_size=363
qemu: flush code_size=83885014 nb_tbs=152777 avg_tb_size=364
qemu: flush code_size=83884950 nb_tbs=150057 avg_tb_size=373
qemu: flush code_size=83884998 nb_tbs=150234 avg_tb_size=373
qemu: flush code_size=83885014 nb_tbs=154009 avg_tb_size=360
qemu: flush code_size=83885014 nb_tbs=151007 avg_tb_size=370
qemu: flush code_size=83885014 nb_tbs=151816 avg_tb_size=367
That is, 8 flushes.
* -smp 8, 32 regions (80/32 MB per region) [i.e. this patch]:
qemu: flush code_size=76328008 nb_tbs=141040 avg_tb_size=356
qemu: flush code_size=75366534 nb_tbs=138000 avg_tb_size=361
qemu: flush code_size=76864546 nb_tbs=140653 avg_tb_size=361
qemu: flush code_size=76309084 nb_tbs=135945 avg_tb_size=375
qemu: flush code_size=74581856 nb_tbs=132909 avg_tb_size=375
qemu: flush code_size=73927256 nb_tbs=135616 avg_tb_size=360
qemu: flush code_size=78629426 nb_tbs=142896 avg_tb_size=365
qemu: flush code_size=76667052 nb_tbs=138508 avg_tb_size=368
Again, 8 flushes. Note how buffer utilization is not 100%, but it
is close. Smaller region sizes would yield higher utilization,
but we want region allocation to be rare (it acquires a lock), so
we do not want to go too small.
* -smp 8, static partitioning of 8 regions (10 MB per region):
qemu: flush code_size=21936504 nb_tbs=40570 avg_tb_size=354
qemu: flush code_size=11472174 nb_tbs=20633 avg_tb_size=370
qemu: flush code_size=11603976 nb_tbs=21059 avg_tb_size=365
qemu: flush code_size=23254872 nb_tbs=41243 avg_tb_size=377
qemu: flush code_size=28289496 nb_tbs=52057 avg_tb_size=358
qemu: flush code_size=43605160 nb_tbs=78896 avg_tb_size=367
qemu: flush code_size=45166552 nb_tbs=82158 avg_tb_size=364
qemu: flush code_size=63289640 nb_tbs=116494 avg_tb_size=358
qemu: flush code_size=51389960 nb_tbs=93937 avg_tb_size=362
qemu: flush code_size=59665928 nb_tbs=107063 avg_tb_size=372
qemu: flush code_size=38380824 nb_tbs=68597 avg_tb_size=374
qemu: flush code_size=44884568 nb_tbs=79901 avg_tb_size=376
qemu: flush code_size=50782632 nb_tbs=90681 avg_tb_size=374
qemu: flush code_size=39848888 nb_tbs=71433 avg_tb_size=372
qemu: flush code_size=64708840 nb_tbs=119052 avg_tb_size=359
qemu: flush code_size=49830008 nb_tbs=90992 avg_tb_size=362
qemu: flush code_size=68372408 nb_tbs=123442 avg_tb_size=368
qemu: flush code_size=33555560 nb_tbs=59514 avg_tb_size=378
qemu: flush code_size=44748344 nb_tbs=80974 avg_tb_size=367
qemu: flush code_size=37104248 nb_tbs=67609 avg_tb_size=364
That is, 20 flushes. Note how a static partitioning approach uses
the code buffer poorly, leading to many unnecessary flushes.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Since FlatViews are shared now and ASes not, this gets rid of
address_space_init_shareable().
This should cause no behavioural change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-17-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Convert all uses of error_report("warning:"... to use warn_report()
instead. This helps standardise on a single method of printing warnings
to the user.
All of the warnings were changed using these two commands:
find ./* -type f -exec sed -i \
's|error_report(".*warning[,:] |warn_report("|Ig' {} +
Indentation fixed up manually afterwards.
The test-qdev-global-props test case was manually updated to ensure that
this patch passes make check (as the test cases are case sensitive).
Signed-off-by: Alistair Francis <alistair.francis@xilinx.com>
Suggested-by: Thomas Huth <thuth@redhat.com>
Cc: Jeff Cody <jcody@redhat.com>
Cc: Kevin Wolf <kwolf@redhat.com>
Cc: Max Reitz <mreitz@redhat.com>
Cc: Ronnie Sahlberg <ronniesahlberg@gmail.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Peter Lieven <pl@kamp.de>
Cc: Josh Durgin <jdurgin@redhat.com>
Cc: "Richard W.M. Jones" <rjones@redhat.com>
Cc: Markus Armbruster <armbru@redhat.com>
Cc: Peter Crosthwaite <crosthwaite.peter@gmail.com>
Cc: Richard Henderson <rth@twiddle.net>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: Greg Kurz <groug@kaod.org>
Cc: Rob Herring <robh@kernel.org>
Cc: Peter Maydell <peter.maydell@linaro.org>
Cc: Peter Chubb <peter.chubb@nicta.com.au>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Cc: David Gibson <david@gibson.dropbear.id.au>
Cc: Alexander Graf <agraf@suse.de>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Cornelia Huck <cohuck@redhat.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Acked-by: Greg Kurz <groug@kaod.org>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed by: Peter Chubb <peter.chubb@data61.csiro.au>
Acked-by: Max Reitz <mreitz@redhat.com>
Acked-by: Marcel Apfelbaum <marcel@redhat.com>
Message-Id: <e1cfa2cd47087c248dd24caca9c33d9af0c499b0.1499866456.git.alistair.francis@xilinx.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Currently, the throttle_thread_scheduled flag is reset back to 0 before
sleeping (as part of the throttling logic). Given that throttle_timer
(well, any timer) may tick with a slight delay, it so happens that under
heavy throttling (ie. close or on CPU_THROTTLE_PCT_MAX) the tick may
schedule a further cpu_throttle_thread() work item after the flag reset,
but before the previous sleep completed. This results on the vCPU thread
sleeping continuously for potentially several seconds in a row.
The chances of that happening can be drastically minimised by resetting
the flag after the sleep.
Signed-off-by: Felipe Franciosi <felipe@nutanix.com>
Signed-off-by: Malcolm Crossley <malcolm@nutanix.com>
Message-Id: <1495229390-18909-1-git-send-email-felipe@nutanix.com>
Acked-by: Jason J. Herne <jjherne@linux.vnet.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
As a rule, CPU internal state should never be updated when
!cpu->kvm_vcpu_dirty (or the HAX equivalent). If that is done, then
subsequent calls to cpu_synchronize_state() - usually safe and idempotent -
will clobber state.
However, we routinely do this during a loadvm or incoming migration.
Usually this is called shortly after a reset, which will clear all the cpu
dirty flags with cpu_synchronize_all_post_reset(). Nothing is expected
to set the dirty flags again before the cpu state is loaded from the
incoming stream.
This means that it isn't safe to call cpu_synchronize_state() from a
post_load handler, which is non-obvious and potentially inconvenient.
We could cpu_synchronize_all_state() before the loadvm, but that would be
overkill since a) we expect the state to already be synchronized from the
reset and b) we expect to completely rewrite the state with a call to
cpu_synchronize_all_post_init() at the end of qemu_loadvm_state().
To clear this up, this patch introduces cpu_synchronize_pre_loadvm() and
associated helpers, which simply marks the cpu state as dirty without
actually changing anything. i.e. it says we want to discard any existing
KVM (or HAX) state and replace it with what we're going to load.
Cc: Juan Quintela <quintela@redhat.com>
Cc: Dave Gilbert <dgilbert@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Highlights:
* New "-numa cpu" option
* NUMA distance configuration
* migration/i386 vmstatification
-----BEGIN PGP SIGNATURE-----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=48wH
-----END PGP SIGNATURE-----
Merge remote-tracking branch 'ehabkost/tags/x86-and-machine-pull-request' into staging
x86 and machine queue, 2017-05-11
Highlights:
* New "-numa cpu" option
* NUMA distance configuration
* migration/i386 vmstatification
# gpg: Signature made Thu 11 May 2017 08:16:07 PM BST
# gpg: using RSA key 0x2807936F984DC5A6
# gpg: Good signature from "Eduardo Habkost <ehabkost@redhat.com>"
# gpg: Note: This key has expired!
# Primary key fingerprint: 5A32 2FD5 ABC4 D3DB ACCF D1AA 2807 936F 984D C5A6
* ehabkost/tags/x86-and-machine-pull-request: (29 commits)
migration/i386: Remove support for pre-0.12 formats
vmstatification: i386 FPReg
migration/i386: Remove old non-softfloat 64bit FP support
tests: check -numa node,cpu=props_list usecase
numa: add '-numa cpu,...' option for property based node mapping
numa: remove node_cpu bitmaps as they are no longer used
numa: use possible_cpus for not mapped CPUs check
machine: call machine init from wrapper
numa: remove no longer need numa_post_machine_init()
tests: numa: add case for QMP command query-cpus
QMP: include CpuInstanceProperties into query_cpus output output
virt-arm: get numa node mapping from possible_cpus instead of numa_get_node_for_cpu()
spapr: get numa node mapping from possible_cpus instead of numa_get_node_for_cpu()
pc: get numa node mapping from possible_cpus instead of numa_get_node_for_cpu()
numa: do default mapping based on possible_cpus instead of node_cpu bitmaps
numa: mirror cpu to node mapping in MachineState::possible_cpus
numa: add check that board supports cpu_index to node mapping
virt-arm: add node-id property to CPU
pc: add node-id property to CPU
spapr: add node-id property to sPAPR core
...
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
if board supports CpuInstanceProperties, report them for
each CPU thread listed. Main motivation for this is to
provide these properties introspection via QMP interface
for using in test cases to verify numa node to cpu mapping,
which includes not only boards that support cpu hotplug
and have this info in query-hotpluggable-cpus (pc/spapr)
but also for boards that don't not support hotpluggable-cpus
but support numa mapping (virt-arm).
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <1494415802-227633-12-git-send-email-imammedo@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Ensure that the unplugged CPU thread is destroyed and the waiting
thread is notified about it. This is needed for CPU unplug to work
correctly in MTTCG mode.
Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This ensures each time the vCPU thread reads the icount we update the
master timer_state.qemu_icount field. This way as long as updates are
in BQL protected sections (which they should be) the main-loop can
never come to update the log and find time has gone backwards.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
There is no particular reason we shouldn't update the global system
icount time as we exit each TranslationBlock run. This ensures the
main-loop doesn't have to wait until we exit to the outer loop for
executed instructions to be credited to timer_state.
The prepare_icount_for_run function is slightly tweaked to match the
logic we run in cpu_loop_exec_tb.
Based on Paolo's original suggestion.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
By holding off updates to timer_state.qemu_icount we can run into
trouble when the non-vCPU thread needs to know the time. This helper
ensures we atomically update timers_state.qemu_icount based on what
has been currently executed.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Outside of the vCPU thread icount time will only be tracked against
timers_state.qemu_icount. We no longer credit cycles until they have
completed the run. Inside the vCPU thread we adjust for passage of
time by looking at how many have run so far. This is only valid inside
the vCPU thread while it is running.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
As icount is only supported for single-threaded execution due to the
requirement for determinism let's remove it from the common
tcg_exec_cpu path.
Also remove the additional fiddling which shouldn't be required as the
icount counters should all be rectified as you enter the loop.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
The lifetime of current_cpu is now the lifetime of the vCPU thread.
However get_icount_raw() can apply a fudge factor if called while code
is running to take into account the current executed instruction
count.
To ensure this is always the case we also check cpu->running.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
We should never be running in multi-threaded mode with icount enabled.
There is no point calling handle_icount_deadline here so remove it and
assert !use_icount.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
While the configure script generates TARGET_SUPPORTS_MTTCG define, one
of the define is cpus.c is checking wrong name: TARGET_SUPPORT_MTTCG
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
use qemu_mutex_lock_iothread consistently in qemu_hax_cpu_thread_fn() as
done in other _thread_fn functions, instead of grabbing directly the
BQL. This way we ensure that iothread_locked is properly set.
On v2.9.0-rc0, QEMU was dying in an assertion in the mutex code when
running with '--enable-hax' either on OSX or Windows. This bug was triggered
since the code modification for multithreading added new usages of
qemu_mutex_iothread_locked.
This fixes the breakage on both platforms, I can now run again a full
Chromium OS image with HAX kernel acceleration.
Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
Message-Id: <20170320101549.150076-1-vpalatin@chromium.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
icount has become much slower after tcg_cpu_exec has stopped
using the BQL. There is also a latent bug that is masked by
the slowness.
The slowness happens because every occurrence of a QEMU_CLOCK_VIRTUAL
timer now has to wake up the I/O thread and wait for it. The rendez-vous
is mediated by the BQL QemuMutex:
- handle_icount_deadline wakes up the I/O thread with BQL taken
- the I/O thread wakes up and waits on the BQL
- the VCPU thread releases the BQL a little later
- the I/O thread raises an interrupt, which calls qemu_cpu_kick
- the VCPU thread notices the interrupt, takes the BQL to
process it and waits on it
All this back and forth is extremely expensive, causing a 6 to 8-fold
slowdown when icount is turned on.
One may think that the issue is that the VCPU thread is too dependent
on the BQL, but then the latent bug comes in. I first tried removing
the BQL completely from the x86 cpu_exec, only to see everything break.
The only way to fix it (and make everything slow again) was to add a dummy
BQL lock/unlock pair.
This is because in -icount mode you really have to process the events
before the CPU restarts executing the next instruction. Therefore, this
series moves the processing of QEMU_CLOCK_VIRTUAL timers straight in
the vCPU thread when running in icount mode.
The required changes include:
- make the timer notification callback wake up TCG's single vCPU thread
when run from another thread. By using async_run_on_cpu, the callback
can override all_cpu_threads_idle() when the CPU is halted.
- move handle_icount_deadline after qemu_tcg_wait_io_event, so that
the timer notification callback is invoked after the dummy work item
wakes up the vCPU thread
- make handle_icount_deadline run the timers instead of just waking the
I/O thread.
- stop processing the timers in the main loop
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
There is no change for now, because the callback just invokes
qemu_notify_event.
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
While we may fail the memory ordering check later that can be
confusing. So in cases where TARGET_SUPPORT_MTTCG has yet to be
defined we should say so specifically.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
The sense of the test was inverted. Make it simple, if icount is
enabled then we disabled MTTCG by default. If the user tries to force
MTTCG upon us then we tell them "no".
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Call kvm_on_sigbus_vcpu asynchronously from the VCPU thread.
Information for the SIGBUS can be stored in thread-local variables
and processed later in kvm_cpu_exec.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Move the KVM "eat signals" code under CONFIG_LINUX, in preparation
for moving it to kvm-all.c; reraise non-MCE SIGBUS immediately,
without passing it to KVM.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The cast is there because sigbus_handler is invoked via sigfd_handler.
But it feels just wrong to use struct qemu_signalfd_siginfo in the
prototype of a function that is passed to sigaction.
Instead, do a simple-minded conversion of qemu_signalfd_siginfo to
siginfo_t.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The patch enables handling atomic code in the guest. This should be
preferably done in cpu_handle_exception(), but the current assumptions
regarding when we can execute atomic sections cause a deadlock.
The current mechanism discards the flags which were set in atomic
execution. We ensure they are properly saved by calling the
cc->cpu_exec_enter/leave() functions around the loop.
As we are running cpu_exec_step_atomic() from the outermost loop we
need to avoid an abort() when single stepping over atomic code since
debug exception longjmp will point to the the setlongjmp in
cpu_exec(). We do this by setting a new jmp_env so that it jumps back
here on an exception.
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
[AJB: tweak title, merge with new patches, add mmap_lock]
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
CC: Paolo Bonzini <pbonzini@redhat.com>
There are a couple of changes that occur at the same time here:
- introduce a single vCPU qemu_tcg_cpu_thread_fn
One of these is spawned per vCPU with its own Thread and Condition
variables. qemu_tcg_rr_cpu_thread_fn is the new name for the old
single threaded function.
- the TLS current_cpu variable is now live for the lifetime of MTTCG
vCPU threads. This is for future work where async jobs need to know
the vCPU context they are operating in.
The user to switch on multi-thread behaviour and spawn a thread
per-vCPU. For a simple test kvm-unit-test like:
./arm/run ./arm/locking-test.flat -smp 4 -accel tcg,thread=multi
Will now use 4 vCPU threads and have an expected FAIL (instead of the
unexpected PASS) as the default mode of the test has no protection when
incrementing a shared variable.
We enable the parallel_cpus flag to ensure we generate correct barrier
and atomic code if supported by the front and backends. This doesn't
automatically enable MTTCG until default_mttcg_enabled() is updated to
check the configuration is supported.
Signed-off-by: KONRAD Frederic <fred.konrad@greensocs.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
[AJB: Some fixes, conditionally, commit rewording]
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
There are now only two uses of the global exit_request left.
The first ensures we exit the run_loop when we first start to process
pending work and in the kick handler. This is just as easily done by
setting the first_cpu->exit_request flag.
The second use is in the round robin kick routine. The global
exit_request ensured every vCPU would set its local exit_request and
cause a full exit of the loop. Now the iothread isn't being held while
running we can just rely on the kick handler to push us out as intended.
We lightly re-factor the main vCPU thread to ensure cpu->exit_requests
cause us to exit the main loop and process any IO requests that might
come along. As an cpu->exit_request may legitimately get squashed
while processing the EXCP_INTERRUPT exception we also check
cpu->queued_work_first to ensure queued work is expedited as soon as
possible.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
This finally allows TCG to benefit from the iothread introduction: Drop
the global mutex while running pure TCG CPU code. Reacquire the lock
when entering MMIO or PIO emulation, or when leaving the TCG loop.
We have to revert a few optimization for the current TCG threading
model, namely kicking the TCG thread in qemu_mutex_lock_iothread and not
kicking it in qemu_cpu_kick. We also need to disable RAM block
reordering until we have a more efficient locking mechanism at hand.
Still, a Linux x86 UP guest and my Musicpal ARM model boot fine here.
These numbers demonstrate where we gain something:
20338 jan 20 0 331m 75m 6904 R 99 0.9 0:50.95 qemu-system-arm
20337 jan 20 0 331m 75m 6904 S 20 0.9 0:26.50 qemu-system-arm
The guest CPU was fully loaded, but the iothread could still run mostly
independent on a second core. Without the patch we don't get beyond
32206 jan 20 0 330m 73m 7036 R 82 0.9 1:06.00 qemu-system-arm
32204 jan 20 0 330m 73m 7036 S 21 0.9 0:17.03 qemu-system-arm
We don't benefit significantly, though, when the guest is not fully
loading a host CPU.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Message-Id: <1439220437-23957-10-git-send-email-fred.konrad@greensocs.com>
[FK: Rebase, fix qemu_devices_reset deadlock, rm address_space_* mutex]
Signed-off-by: KONRAD Frederic <fred.konrad@greensocs.com>
[EGC: fixed iothread lock for cpu-exec IRQ handling]
Signed-off-by: Emilio G. Cota <cota@braap.org>
[AJB: -smp single-threaded fix, clean commit msg, BQL fixes]
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Reviewed-by: Pranith Kumar <bobby.prani@gmail.com>
[PM: target-arm changes]
Acked-by: Peter Maydell <peter.maydell@linaro.org>
..and make the definition local to cpus. In preparation for MTTCG the
concept of a global tcg_current_cpu will no longer make sense. However
we still need to keep track of it in the single-threaded case to be able
to exit quickly when required.
qemu_cpu_kick_no_halt() moves and becomes qemu_cpu_kick_rr_cpu() to
emphasise its use-case. qemu_cpu_kick now kicks the relevant cpu as
well as qemu_kick_rr_cpu() which will become a no-op in MTTCG.
For the time being the setting of the global exit_request remains.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Reviewed-by: Pranith Kumar <bobby.prani@gmail.com>
Currently we rely on the side effect of the main loop grabbing the
iothread_mutex to give any long running basic block chains a kick to
ensure the next vCPU is scheduled. As this code is being re-factored and
rationalised we now do it explicitly here.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Reviewed-by: Pranith Kumar <bobby.prani@gmail.com>
We know there will be cases where MTTCG won't work until additional work
is done in the front/back ends to support. It will however be useful to
be able to turn it on.
As a result MTTCG will default to off unless the combination is
supported. However the user can turn it on for the sake of testing.
Signed-off-by: KONRAD Frederic <fred.konrad@greensocs.com>
[AJB: move to -accel tcg,thread=multi|single, defaults]
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
This patch:
* moves vm_start to cpus.c.
* exports qemu_vmstop_requested, since it's needed by vm_start.
* extracts vm_prepare_start from vm_start; it does what vm_start did,
except restarting the cpus.
* vm_start now calls vm_prepare_start and then restarts the cpus.
Signed-off-by: Claudio Imbrenda <imbrenda@linux.vnet.ibm.com>
Message-Id: <1487092068-16562-2-git-send-email-imbrenda@linux.vnet.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Use the Intel HAX is kernel-based hardware acceleration module for
Windows (similar to KVM on Linux).
Based on the "target/i386: Add Intel HAX to android emulator" patch
from David Chou <david.j.chou@intel.com>
Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
Message-Id: <7b9cae28a0c379ab459c7a8545c9a39762bd394f.1484045952.git.vpalatin@chromium.org>
[Drop hax_populate_ram stub. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Move the generic cpu_synchronize_ functions to the common hw_accel.h header,
in order to prepare for the addition of a second hardware accelerator.
Signed-off-by: Stefan Weil <sw@weilnetz.de>
Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
Message-Id: <f5c3cffe8d520011df1c2e5437bb814989b48332.1484045952.git.vpalatin@chromium.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This changes the *_run_on_cpu APIs (and helpers) to pass data in a
run_on_cpu_data type instead of a plain void *. This is because we
sometimes want to pass a target address (target_ulong) and this fails on
32 bit hosts emulating 64 bit guests.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20161027151030.20863-24-alex.bennee@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In preparation for adding a MTTCG thread we re-factor out a bit of what
will be common code to handle the QEMU_CLOCK_VIRTUAL expiration.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Message-Id: <20161027151030.20863-18-alex.bennee@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In preparation for multi-threaded TCG we remove tcg_exec_all and move
all the CPU cycling into the main thread function. When MTTCG is enabled
we shall use a separate thread function which only handles one vCPU.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Sergey Fedorov <sergey.fedorov@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Message-Id: <20161027151030.20863-13-alex.bennee@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This is a pure mechanical change in preparation for up-coming
re-factoring. Instead of a forward declaration for tcg_exec_all it and
the associated helper functions are moved in front of the call from
qemu_tcg_cpu_thread_fn.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Message-Id: <20161027151030.20863-12-alex.bennee@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When we cannot emulate an atomic operation within a parallel
context, this exception allows us to stop the world and try
again in a serial context.
Reviewed-by: Emilio G. Cota <cota@braap.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Richard Henderson <rth@twiddle.net>
Reimplement bdrv_flush_all for vm_stop. In contrast to blk_flush_all,
bdrv_flush_all does not have device model restrictions. This allows
us to flush and halt unconditionally without error.
This allows us to do things like migrate when we have a device with
an open tray, but has a node that may need to be flushed, or nodes
that aren't currently attached to any device and need to be flushed.
Specifically, this allows us to migrate when we have a CDROM with
an open tray.
Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Acked-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
This patch fixes bug with stopping and restarting replay
through monitor.
Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
Message-Id: <20160926080815.6992.71818.stgit@PASHA-ISP>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This will serve as the base for async_safe_run_on_cpu. Because
start_exclusive uses CPU_FOREACH, merge exclusive_lock with
qemu_cpu_list_lock: together with a call to exclusive_idle (via
cpu_exec_start/end) in cpu_list_add, this protects exclusive work
against concurrent CPU addition and removal.
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Make CPU work core functions common between system and user-mode
emulation. User-mode does not use run_on_cpu, so do not implement it.
Signed-off-by: Sergey Fedorov <serge.fdrv@gmail.com>
Signed-off-by: Sergey Fedorov <sergey.fedorov@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <1470158864-17651-10-git-send-email-alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
