-----BEGIN PGP SIGNATURE-----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=o8eA
-----END PGP SIGNATURE-----
Merge remote-tracking branch 'remotes/armbru/tags/pull-util-2017-02-23' into staging
option cutils: Fix and clean up number conversions
# gpg: Signature made Thu 23 Feb 2017 19:41:17 GMT
# gpg: using RSA key 0x3870B400EB918653
# gpg: Good signature from "Markus Armbruster <armbru@redhat.com>"
# gpg: aka "Markus Armbruster <armbru@pond.sub.org>"
# Primary key fingerprint: 354B C8B3 D7EB 2A6B 6867 4E5F 3870 B400 EB91 8653
* remotes/armbru/tags/pull-util-2017-02-23: (24 commits)
option: Fix checking of sizes for overflow and trailing crap
util/cutils: Change qemu_strtosz*() from int64_t to uint64_t
util/cutils: Return qemu_strtosz*() error and value separately
util/cutils: Let qemu_strtosz*() optionally reject trailing crap
qemu-img: Wrap cvtnum() around qemu_strtosz()
test-cutils: Drop suffix from test_qemu_strtosz_simple()
test-cutils: Use qemu_strtosz() more often
util/cutils: Drop QEMU_STRTOSZ_DEFSUFFIX_* macros
util/cutils: New qemu_strtosz()
util/cutils: Rename qemu_strtosz() to qemu_strtosz_MiB()
util/cutils: New qemu_strtosz_metric()
test-cutils: Cover qemu_strtosz() around range limits
test-cutils: Cover qemu_strtosz() with trailing crap
test-cutils: Cover qemu_strtosz() invalid input
test-cutils: Add missing qemu_strtosz()... endptr checks
option: Fix to reject invalid and overflowing numbers
util/cutils: Clean up control flow around qemu_strtol() a bit
util/cutils: Clean up variable names around qemu_strtol()
util/cutils: Rename qemu_strtoll(), qemu_strtoull()
util/cutils: Rewrite documentation of qemu_strtol() & friends
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Rbd can do readv and writev directly, so wo do not need to transform
iov to buf or vice versa any more.
Signed-off-by: tianqing <tianqing@unitedstack.com>
Reviewed-by: Jeff Cody <jcody@redhat.com>
Signed-off-by: Jeff Cody <jcody@redhat.com>
if the passed qiov contains exactly one iov we can
pass the buffer directly.
Signed-off-by: Peter Lieven <pl@kamp.de>
Reviewed-by: Jeff Cody <jcody@redhat.com>
Message-id: 1487349541-10201-3-git-send-email-pl@kamp.de
Signed-off-by: Jeff Cody <jcody@redhat.com>
Signed-off-by: Peter Lieven <pl@kamp.de>
Reviewed-by: Jeff Cody <jcody@redhat.com>
Message-id: 1487349541-10201-2-git-send-email-pl@kamp.de
Signed-off-by: Jeff Cody <jcody@redhat.com>
blk_insert_bs() and block job related functions will soon require an
opened block node (permission calculations will involve the block
driver), so let our tests be consistent with the real users in this
respect.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
We should not try to assign a not yet opened node as the backing file,
because as soon as the permission system is added it will fail. The
just added bdrv_new_open_driver() function is the right tool to open a
file with an internal driver, use it.
In case anyone wonders whether that magic fake backing file to trigger a
special action on 'commit' actually works today: No, not for me. One
reason is that we've been adding a raw format driver on top for several
years now and raw doesn't support commit. Other reasons include that the
backing file isn't writable and the driver doesn't support reopen, and
it's also size 0 and the driver doesn't support bdrv_truncate. All of
these are easily fixable, but then 'commit' ended up in an infinite loop
deep in the vvfat code for me, so I thought I'd best leave it alone. I'm
not really sure what it was supposed to do anyway.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
This function allows to create more or less normal BlockDriverStates
even for BlockDrivers that aren't globally registered (e.g. helper
filters for block jobs).
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
This is a function that doesn't do any option parsing, but just does
some basic BlockDriverState setup and calls the .bdrv_open() function of
the block driver.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
This fixes the use of a parent-less BdrvChild in bdrv_open_inherit() by
converting it into a BlockBackend. Which is exactly what it should be,
image probing is an external, standalone user of a node. The requests
can't be considered to originate from the format driver node because
that one isn't even opened yet.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
This is the part of bdrv_open_child() that opens a BDS with option
inheritance, but doesn't attach it as a child to the parent yet.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
The way that attaching bs->file worked was a bit unusual in that it was
the only child that would be attached to a node which is not opened yet.
Because of this, the block layer couldn't know yet which permissions the
driver would eventually need.
This patch moves the point where bs->file is attached to the beginning
of the individual .bdrv_open() implementations, so drivers already know
what they are going to do with the child. This is also more consistent
with how driver-specific children work.
For a moment, bdrv_open() gets its own BdrvChild to perform image
probing, but instead of directly assigning this BdrvChild to the BDS, it
becomes a temporary one and the node name is passed as an option to the
drivers, so that they can simply use bdrv_open_child() to create another
reference for their own use.
This duplicated child for (the not opened yet) bs is not the final
state, a follow-up patch will change the image probing code to use a
BlockBackend, which is completely independent of bs.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
This is more consistent with the commit block job, and it moves the code
to a place where we already have the necessary BlockBackends to resize
the base image when bdrv_truncate() is changed to require a BdrvChild.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
In order to able to convert bdrv_truncate() to take a BdrvChild and
later to correctly check the resize permission here, we need to use a
BlockBackend for resizing the image.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
In order to be able to do permission checking and to keep working with
the BdrvChild based bdrv_truncate() that this involves, we need to
create a temporary BlockBackend to resize the image.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
We can't rely on a non-paused job to be present and running for us.
Assume that if the job is not present that it completed already.
Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Now that we are truncating the file in both PREALLOC_MODE_FULL and
PREALLOC_MODE_OFF, not truncating in PREALLOC_MODE_FALLOC looks odd.
Add a comment explaining why we do not truncate in this case.
Signed-off-by: Nir Soffer <nirsof@gmail.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
In a previous commit (qemu-img: Do not truncate before preallocation) we
moved truncate to the PREALLOC_MODE_OFF branch to avoid slowdown in
posix_fallocate().
However this change is not optimal when using PREALLOC_MODE_FULL, since
knowing the final size from the beginning could allow the file system
driver to do less allocations and possibly avoid fragmentation of the
file.
Now we truncate also before doing full preallocation.
Signed-off-by: Nir Soffer <nirsof@gmail.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Add tests for creating raw image with and without the preallocation
option.
Signed-off-by: Nir Soffer <nirsof@gmail.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
When using file system that does not support fallocate() (e.g. NFS <
4.2), truncating the file only when preallocation=OFF speeds up creating
raw file.
Here is example run, tested on Fedora 24 machine, creating raw file on
NFS version 3 server.
$ time ./qemu-img-master create -f raw -o preallocation=falloc mnt/test 1g
Formatting 'mnt/test', fmt=raw size=1073741824 preallocation=falloc
real 0m21.185s
user 0m0.022s
sys 0m0.574s
$ time ./qemu-img-fix create -f raw -o preallocation=falloc mnt/test 1g
Formatting 'mnt/test', fmt=raw size=1073741824 preallocation=falloc
real 0m11.601s
user 0m0.016s
sys 0m0.525s
$ time dd if=/dev/zero of=mnt/test bs=1M count=1024 oflag=direct
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 15.6627 s, 68.6 MB/s
real 0m16.104s
user 0m0.009s
sys 0m0.220s
Running with strace we can see that without this change we do one
pread() and one pwrite() for each block. With this change, we do only
one pwrite() per block.
$ strace ./qemu-img-master create -f raw -o preallocation=falloc mnt/test 8192
...
pread64(9, "\0", 1, 4095) = 1
pwrite64(9, "\0", 1, 4095) = 1
pread64(9, "\0", 1, 8191) = 1
pwrite64(9, "\0", 1, 8191) = 1
$ strace ./qemu-img-fix create -f raw -o preallocation=falloc mnt/test 8192
...
pwrite64(9, "\0", 1, 4095) = 1
pwrite64(9, "\0", 1, 8191) = 1
This happens because posix_fallocate is checking if each block is
allocated before writing a byte to the block, and when truncating the
file before preallocation, all blocks are unallocated.
Signed-off-by: Nir Soffer <nirsof@gmail.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Some iotests (e.g. 174) try to filter the output of _make_test_image by
piping the stdout. Pipe the server stdout to /dev/null, so that filter
pipe does not need to wait until process completion.
Signed-off-by: Jeff Cody <jcody@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Add the ability for shell script tests to exclude specific
protocols. This is useful to allow all protocols except ones known to
not support a feature used in the test (e.g. .bdrv_create).
Signed-off-by: Jeff Cody <jcody@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Since test 137 make uses of qcow2.py, only local files are supported.
Signed-off-by: Jeff Cody <jcody@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
CIRRUS_BLTMODE_MEMSYSSRC blits do NOT check blit destination
and blit width, at all. Oops. Fix it.
Security impact: high.
The missing blit destination check allows to write to host memory.
Basically same as CVE-2014-8106 for the other blit variants.
Cc: qemu-stable@nongnu.org
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Introduce support for emulating the MIPS Boston development board. The
Boston board is built around an FPGA & 3 PCIe controllers, one of which
is connected to an Intel EG20T Platform Controller Hub. It is used
during the development & debug of new CPUs and the software intended to
run on them, and is essentially the successor to the older MIPS Malta
board.
This patch does not implement the EG20T, instead connecting an already
supported ICH-9 AHCI controller. Whilst this isn't accurate it's enough
for typical stock Boston software (eg. Linux kernels) to work with hard
disks given that both the ICH-9 & EG20T implement the AHCI
specification.
Boston boards typically boot kernels in the FIT image format, and this
patch will treat kernels provided to QEMU as such. When loading a kernel
directly, the board code will generate minimal firmware much as the
Malta board code does. This firmware will set up the CM, CPC & GIC
register base addresses then set argument registers & jump to the kernel
entry point. Alternatively, bootloader code may be loaded using the bios
argument in which case no firmware will be generated & execution will
proceed from the start of the boot code at the default MIPS boot
exception vector (offset 0x1fc00000 into (c)kseg1).
Currently real Boston boards are always used with FPGA bitfiles that
include a Global Interrupt Controller (GIC), so the interrupt
configuration is only defined for such cases. Therefore the board will
only allow use of CPUs which implement the CPS components, including the
GIC, and will otherwise exit with a message.
Signed-off-by: Paul Burton <paul.burton@imgtec.com>
Reviewed-by: Yongbok Kim <yongbok.kim@imgtec.com>
[yongbok.kim@imgtec.com:
isolated boston machine support for mips64el.
updated for recent Chardev changes.
ignore missing bios/kernel for qtest.
added default -drive to if=ide explicitly.
changed default memory size into 1G due to make check failure
on 32-bit hosts]
Signed-off-by: Yongbok Kim <yongbok.kim@imgtec.com>
This enables the multi-threaded system emulation by default for ARMv7
and ARMv8 guests using the x86_64 TCG backend. This is because on the
guest side:
- The ARM translate.c/translate-64.c have been converted to
- use MTTCG safe atomic primitives
- emit the appropriate barrier ops
- The ARM machine has been updated to
- hold the BQL when modifying shared cross-vCPU state
- defer powerctl changes to async safe work
All the host backends support the barrier and atomic primitives but
need to provide same-or-better support for normal load/store
operations.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Acked-by: Peter Maydell <peter.maydell@linaro.org>
Tested-by: Pranith Kumar <bobby.prani@gmail.com>
Reviewed-by: Pranith Kumar <bobby.prani@gmail.com>
The arm_reset_cpu/set_cpu_on/set_cpu_off() functions do their work
asynchronously in the target vCPUs context. As a result we need to
ensure the SRC_SCR reset bits correctly report the reset status at the
right time. To do this we defer the clearing of the bit with an async
job which will run after the work queued by ARM powerctl functions.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Previously flushes on other vCPUs would only get serviced when they
exited their TranslationBlocks. While this isn't overly problematic it
violates the semantics of TLB flush from the point of view of source
vCPU.
To solve this we call the cputlb *_all_cpus_synced() functions to do
the flushes which ensures all flushes are completed by the time the
vCPU next schedules its own work. As the TLB instructions are modelled
as CP writes the TB ends at this point meaning cpu->exit_request will
be checked before the next instruction is executed.
Deferring the work until the architectural sync point is a possible
future optimisation.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
The WFE and YIELD instructions are really only hints and in TCG's case
they were useful to move the scheduling on from one vCPU to the next. In
the parallel context (MTTCG) this just causes an unnecessary cpu_exit
and contention of the BQL.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
When switching a new vCPU on we want to complete a bunch of the setup
work before we start scheduling the vCPU thread. To do this cleanly we
defer vCPU setup to async work which will run the vCPUs execution
context as the thread is woken up. The scheduling of the work will kick
the vCPU awake.
This avoids potential races in MTTCG system emulation.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
This introduces support to the cputlb API for flushing all CPUs TLBs
with one call. This avoids the need for target helpers to iterate
through the vCPUs themselves.
An additional variant of the API (_synced) will cause the source vCPUs
work to be scheduled as "safe work". The result will be all the flush
operations will be complete by the time the originating vCPU executes
its safe work. The calling implementation can either end the TB
straight away (which will then pick up the cpu->exit_request on
entering the next block) or defer the exit until the architectural
sync point (usually a barrier instruction).
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
The main use case for tlb_reset_dirty is to set the TLB_NOTDIRTY flags
in TLB entries to force the slow-path on writes. This is used to mark
page ranges containing code which has been translated so it can be
invalidated if written to. To do this safely we need to ensure the TLB
entries in question for all vCPUs are updated before we attempt to run
the code otherwise a race could be introduced.
To achieve this we atomically set the flag in tlb_reset_dirty_range and
take care when setting it when the TLB entry is filled.
On 32 bit systems attempting to emulate 64 bit guests we don't even
bother as we might not have the atomic primitives available. MTTCG is
disabled in this case and can't be forced on. The copy_tlb_helper
function helps keep the atomic semantics in one place to avoid
confusion.
The dirty helper function is made static as it isn't used outside of
cputlb.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
This converts the remaining TLB flush routines to use async work when
detecting a cross-vCPU flush. The only minor complication is having to
serialise the var_list of MMU indexes into a form that can be punted
to an asynchronous job.
The pending_tlb_flush field on QOM's CPU structure also becomes a
bitfield rather than a boolean.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
While the vargs approach was flexible the original MTTCG ended up
having munge the bits to a bitmap so the data could be used in
deferred work helpers. Instead of hiding that in cputlb we push the
change to the API to make it take a bitmap of MMU indexes instead.
For ARM some the resulting flushes end up being quite long so to aid
readability I've tended to move the index shifting to a new line so
all the bits being or-ed together line up nicely, for example:
tlb_flush_page_by_mmuidx(other_cs, pageaddr,
(1 << ARMMMUIdx_S1SE1) |
(1 << ARMMMUIdx_S1SE0));
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
[AT: SPARC parts only]
Reviewed-by: Artyom Tarasenko <atar4qemu@gmail.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
[PM: ARM parts only]
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Some architectures allow to flush the tlb of other VCPUs. This is not a problem
when we have only one thread for all VCPUs but it definitely needs to be an
asynchronous work when we are in true multithreaded work.
We take the tb_lock() when doing this to avoid racing with other threads
which may be invalidating TB's at the same time. The alternative would
be to use proper atomic primitives to clear the tlb entries en-mass.
This patch doesn't do anything to protect other cputlb function being
called in MTTCG mode making cross vCPU changes.
Signed-off-by: KONRAD Frederic <fred.konrad@greensocs.com>
[AJB: remove need for g_malloc on defer, make check fixes, tb_lock]
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
This moves the helper function closer to where it is called and updates
the error message to report via error_report instead of the deprecated
fprintf.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
For SoftMMU the TLB flushes are an example of a task that can be
triggered on one vCPU by another. To deal with this properly we need to
use safe work to ensure these changes are done safely. The new assert
can be enabled while debugging to catch these cases.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
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>
tb_lock() has long been used for linux-user mode to protect code
generation. By enabling it now we prepare for MTTCG and ensure all code
generation is serialised by this lock. The other major structure that
needs protecting is the l1_map and its PageDesc structures. For the
SoftMMU case we also use tb_lock() to protect these structures instead
of linux-user mmap_lock() which as the name suggests serialises updates
to the structure as a result of guest mmap operations.
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>
We'll be using the memory ordering definitions to define values for
both the host and guest. To avoid fighting with circular header
dependencies just move these types into their own minimal header.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
