| 8f4b056427a9d2321e3aa4f21ca8ffb18b3e5ae6 |
|
02-Mar-2018 |
David Sehr <sehr@google.com> |
Move most of runtime/base to libartbase/base
Enforce the layering that code in runtime/base should not depend on
runtime by separating it into libartbase. Some of the code in
runtime/base depends on the Runtime class, so it cannot be moved yet.
Also, some of the tests depend on CommonRuntimeTest, which itself needs
to be factored (in a subsequent CL).
Bug: 22322814
Test: make -j 50 checkbuild
make -j 50 test-art-host
Change-Id: I8b096c1e2542f829eb456b4b057c71421b77d7e2
Merged-In: c431b9dc4b23cc950eb313695258df5d89f53b22
(cherry picked from commit c431b9dc4b23cc950eb313695258df5d89f53b22)
|
| 928033d81f63912f669acd27a72897694eb31db6 |
|
07-Feb-2018 |
Orion Hodson <oth@google.com> |
Revert^2 "ART: Interpreter support for VarHandle accessors"
This reverts commit c4d3bf4eaff9720ed0bb7775db6fc24a1bd755dd.
Bug: 65872996
Test: art/test/testrunner.py -b --host
Change-Id: I2d1017c45cee5b2fd2dd05e5eeff0a72146663b1
|
| 001eff919009f60af4f8184ab37c10ef56275c68 |
|
24-Jan-2018 |
Roland Levillain <rpl@google.com> |
Improve to-space invariant error reporting in the CC collector.
In particular, distinguish references in an unused region from
the region space (which should never be encountered) from a
reference in a non-moving space.
Test: art/test/testrunner/testrunner.py
Bug: 72758079
Bug: 73004523
Change-Id: Iac632ae71dce9a36b33c14586eeb709dc2d7e7ce
|
| c4d3bf4eaff9720ed0bb7775db6fc24a1bd755dd |
|
06-Feb-2018 |
Orion Hodson <oth@google.com> |
Revert "ART: Interpreter support for VarHandle accessors"
This reverts commit becc83b59ff703c02198f832dbc3216a0ab8a022.
Reason for revert: breaks on the ART buildbots due to localized toolchain differences.
Change-Id: Ieb1d93b0fc7116ddc542d8bbfee1975bcb30a8f7
|
| becc83b59ff703c02198f832dbc3216a0ab8a022 |
|
08-Dec-2017 |
Orion Hodson <oth@google.com> |
ART: Interpreter support for VarHandle accessors
Add support for invoking VarHandle accessors in the interpreter and
forces code with VarHandles accessor invocations to use the interpreter.
Bug: 65872996
Test: art/test/run-test --host 712
Change-Id: I9ee3ad6aef6a3bc73d90cec0a8e023e5db42b7a2
|
| 4557b3858a66aa20e42bce937e1f0620aad880a2 |
|
03-Jan-2018 |
Orion Hodson <oth@google.com> |
ART: Rename Atomic::CompareExchange methods
Renames Atomic::CompareExchange methods to Atomic::CompareAndSet
equivalents. These methods return a boolean and do not get the witness
value. This makes space for Atomic::CompareAndExchange methods in a
later commit that will return a boolean and get the witness value.
This is pre-work for VarHandle accessors which require both forms.
Bug: 65872996
Test: art/test.py --host -j32
Change-Id: I9c691250e5556cbfde7811381b06d2920247f1a1
|
| 09659c22dc2f2c85a0ade965d1fc5160944b8692 |
|
19-Sep-2017 |
Andreas Gampe <agampe@google.com> |
ART: Remove heap poisoning from globals.h
Remove mostly-unused include and move it to its users.
Test: m
Change-Id: Ibb40f919db64a490290c6e18cf1123aaf44199fc
|
| cc55e1dcdd2ec669c635420468b3cc4c99740f0a |
|
28-Jul-2017 |
Hans Boehm <hboehm@google.com> |
Don't use fences to implement volatiles
Mixing the fence-based implementation with acquire/release instructions
on ARMv8 is not just ugly but incorrect. A volatile store; volatile
load sequence implemented as a release store followed by ld; dmb
does not prevent reordering.
This should remove the last places we were using fences to implement
volatiles.
The HeapReference representation is changed to be an Atomic,
thereby avoiding many casts. We no longer inherit from ObjectReference,
which was documented to be a value type. HeapReference is not, since
it contains an atomic.
Disentangle HeapReference and ObjectReference/CompressedReference
uses sufficiently to get the code to compile again. They were
previously used somewhat interchangably in a few places, in spite
of the different intended semantics (value-type vs. a concurrently-
updateable field). Further disentanglement might be useful.
Flag a strange fence use I haven't yet understood.
Test: Booted AOSP. Ran default tests. Some object code inspection.
Bug: 31023171
Test: Built AOSP
Change-Id: I7b3c3e624f480994541c8e3a79e585071c122a3d
|
| 895f92218f705ff8ad9c47b8be0c093130d9fbbc |
|
05-Jul-2017 |
Andreas Gampe <agampe@google.com> |
ART: Fix up small header includes
Test: m
Change-Id: I6978d6eb4b95a6ee810e5a48ca6f5d6c590d4ce1
|
| 8cf9cb386cd9286d67e879f1ee501ec00d72a4e1 |
|
19-Jul-2017 |
Andreas Gampe <agampe@google.com> |
ART: Include cleanup
Let clang-format reorder the header includes.
Derived with:
* .clang-format:
BasedOnStyle: Google
IncludeIsMainRegex: '(_test|-inl)?$'
* Steps:
find . -name '*.cc' -o -name '*.h' | xargs sed -i.bak -e 's/^#include/ #include/' ; git commit -a -m 'ART: Include cleanup'
git-clang-format -style=file HEAD^
manual inspection
git commit -a --amend
Test: mmma art
Change-Id: Ia963a8ce3ce5f96b5e78acd587e26908c7a70d02
|
| 6d3e76884710e639e79a46d505395bc83ad5ec7d |
|
11-Jul-2017 |
Chang Xing <chxing@google.com> |
Abstract SetField functions with Transaction check
Abstracted several functions which set objects' fields with a check to
test whether transaction is active to determine one template argument,
this make more object modification operations support transaction.
Replaced some function calls with transaction checking version to
support clinit for app image to run under transaction.
Tests shows not breaking anything, testcase cover this situation will be
added shortly.
Test: make test-art-host -j64
Change-Id: Ic15a4428a7c7cce6832651e1894eb760354de009
|
| 2ff3b97500b717f69415ae474d5cfb1613cd15ee |
|
06-Jun-2017 |
Andreas Gampe <agampe@google.com> |
ART: More header cleanups
Clean up some more includes. Move some -inls from object-inl to
object-refvisitor-inl. Fix up transitives.
Test: m test-art-host
Change-Id: I0e3e268d95a1c71b0a80fc963d13d191a97b2283
|
| fdd513d6c522841e82f1dc144d19a8d60269b9f7 |
|
01-Jun-2017 |
Mathieu Chartier <mathieuc@google.com> |
Move to release CAS for updating object fields
Relaxed cas is not sufficient to make sure threads that read the
field will see the copied contents of objects.
Bug: 37187694
Bug: 62240510
Test: test-art-host
(cherry picked from commit a1f20c3f8d0dabb9723acccf3ba760acf3ebe62d)
Change-Id: I8bff8a67c2c52eb131714b52e6d842c8c08dd70a
|
| a1f20c3f8d0dabb9723acccf3ba760acf3ebe62d |
|
01-Jun-2017 |
Mathieu Chartier <mathieuc@google.com> |
Move to release CAS for updating object fields
Relaxed cas is not sufficient to make sure threads that read the
field will see the copied contents of objects.
Bug: 37187694
Bug: 62240510
Test: test-art-host
Change-Id: I1239817e2b63e0e43ac3ae3148b73487408b378b
|
| c15a2f4f45661a7f5f542e406282c146ea1a968d |
|
21-Apr-2017 |
Andreas Gampe <agampe@google.com> |
ART: Add object-readbarrier-inl.h
Move some read-barrier code into a new header. This prunes the
include tree for the concurrent-copying collector. Clean up other
related includes.
Test: mmma art
Change-Id: I40ce4e74f2e5d4c692529ffb4df933230b6fd73e
|
| c6ea7d00ad069a2736f603daa3d8eaa9a1f8ea11 |
|
02-Feb-2017 |
Andreas Gampe <agampe@google.com> |
ART: Clean up art_method.h
Clean up the header. Fix up other headers including the -inl file,
in an effort to prune the include graph. Fix broken transitive
includes by making includes explicit. Introduce new -inl files
for method handles and reference visiting.
Test: source build/envsetup.sh && lunch aosp_angler-userdebug && mmma art
Test: source build/envsetup.sh && lunch aosp_mips64-userdebug && mmma art
Change-Id: I8f60f1160c2a702fdf3598149dae38f6fa6bc851
|
| d08f66f519f6cf75c2eb05a489b6e87f739badff |
|
13-Apr-2017 |
Mathieu Chartier <mathieuc@google.com> |
Various debug ART GC performance improvements
Removed unnecessary read barriers in SizeOf and a few other places.
Disabled disallow read barrier check.
Before:
GC time: 15.817s
Real 0m26.113s
user 1m16.780s
sys 0m3.152s
After:
GC time: 9.212s
real 0m19.875s
user 1m9.916s
sys 0m1.916s
Bug: 35644369
Test: test-art-host
Change-Id: I79a65259deff2a478a96e02ae69b14730b6dcbe6
|
| 1595815c2a914a78df7dfb6f0082f47d4e82bb36 |
|
10-Feb-2017 |
Alexey Frunze <Alexey.Frunze@imgtec.com> |
MIPS: Implement read barriers.
This is the core functionality. Further improvements
will be done separately.
This also adds/moves memory barriers where they belong and
removes the UnsafeGetLongVolatile and UnsafePutLongVolatile
MIPS32 intrinsics as they need to load/store a pair of
registers atomically, which is not supported directly by
the CPU.
Test: booted MIPS32R2 in QEMU
Test: test-art-target-run-test
Test: booted MIPS64 (with 2nd arch MIPS32R6) in QEMU
Test: "testrunner.py --target --optimizing -j1"
Test: same MIPS64 boot/test with ART_READ_BARRIER_TYPE=TABLELOOKUP
Test: "testrunner.py --target --optimizing --32 -j2" on CI20
Test: same CI20 test with ART_READ_BARRIER_TYPE=TABLELOOKUP
Change-Id: I0ff91525fefba3ec1cc019f50316478a888acced
|
| 90b936ddda63139ff46a6755c3b83ad6e4ab4ac5 |
|
31-Jan-2017 |
Andreas Gampe <agampe@google.com> |
ART: Refactor verify_object.h
Move the actual VerifyObject check into a new cc file, as we
commonly don't enable the check at all. This allows to cut the
-inl include from almost all current users.
This also exposes missing -inl includes. Also fix up some of our old
mess where .h defined functions require -inl.h defined functions.
Test: m
Change-Id: I3dd821bbe2015564a29bf1ed9be00f7a7276ad61
|
| b3da36c182329f228b8c78d306a57fa11a6004ea |
|
15-Dec-2016 |
Hans Boehm <hboehm@google.com> |
Reduce the number of fences needed for monitors
Add the necessary CasWeakAcquire primitives for LockWords.
Have MonitorEnter initially read the lockword using a
memory_order_relaxed operation. In the unlikely case we need more,
compensate with an explicit fence.
In the uncontended case, install the thin lock with Acquire,
rather than SequentiallyConsistent semantics.
Have MonitorExit use a Release instead of SequentiallyConsistent
CAS in the ReadBarrier case. Add TODO for the other case.
Together, these should usually eliminate 3 fences (or acq/rel)
per critical section.
Have Install() only use Release ordering.
Add TODO for inflation spinning, which looks to me like it could be
improved appreciably.
Drive-by fix:
GetMaxSpinsBeforeThinLockInflation spelling
Test: Build for several targets, boot, m art-test-host art-test-target
Change-Id: I2cab09723252065f6365e4234ee3249c69ece888
|
| 12b58b23de974232e991c650405f929f8b0dcc9f |
|
01-Nov-2016 |
Hiroshi Yamauchi <yamauchi@google.com> |
Clean up the runtime read barrier and fix fake address dependency.
- Rename GetReadBarrierPointer to GetReadBarrierState.
- Change its return type to uint32_t.
- Fix the runtime fake address dependency for arm/arm64 using inline
asm.
- Drop ReadBarrier::black_ptr_ and some brooks code.
Bug: 12687968
Test: test-art with CC, Ritz EAAC, libartd boot on N9.
Change-Id: I595970db825db5be2e98ee1fcbd7696d5501af55
|
| ef41db7a3f322a1feb305fdb457410c4cea94d00 |
|
26-Oct-2016 |
Mathieu Chartier <mathieuc@google.com> |
Move interpreter to ObjPtr
Moved most of interpreter, interpreter_common,
interpreter_switch_impl, and some of mterp to ObjPtr.
Bug: 31113334
Test: test-art-host ART_TEST_INTERPRETER=true
Test: art/tools/run-libcore-tests.sh '--mode=host' '--variant=X32' --debug
Change-Id: I0935d18287e1332205c17c5a018aa167788ab897
|
| 709b070044354d9f47641f273edacaeeb0240ab7 |
|
13-Oct-2016 |
David Sehr <sehr@google.com> |
Remove mirror:: and ArtMethod deps in utils.{h,cc}
The latest chapter in the ongoing saga of attempting to dump a DEX
file without having to start a whole runtime instance. This episode
finds us removing references to ArtMethod/ArtField/mirror.
One aspect of this change that I would like to call out specfically
is that the utils versions of the "Pretty*" functions all were written
to accept nullptr as an argument. I have split these functions up as
follows:
1) an instance method, such as PrettyClass that obviously requires
this != nullptr.
2) a static method, that behaves the same way as the util method, but
calls the instance method if p != nullptr.
This requires using a full class qualifier for the static methods,
which isn't exactly beautiful. I have tried to remove as many cases
as possible where it was clear p != nullptr.
Bug: 22322814
Test: test-art-host
Change-Id: I21adee3614aa697aa580cd1b86b72d9206e1cb24
|
| 31e88225b2ef68e7f32f11186acf922c74ddabab |
|
15-Oct-2016 |
Mathieu Chartier <mathieuc@google.com> |
Move most mirror:: args to ObjPtr
Fixed possible moving GC bugs in ClinitImageUpdate class.
Bug: 31113334
Test: test-art-host
Change-Id: I0bf6578553d58b944aaa17665f1350bdf5ed15ec
|
| a058fdf0cf7435a13a6e8cae5e3a9bfa1513828d |
|
07-Oct-2016 |
Mathieu Chartier <mathieuc@google.com> |
Move mirror::Object setters to ObjPtr
Bug: 31113334
Test: test-art-host
Change-Id: I2c4c84645e194c3c435a4a6fd670176b0e98671f
|
| a59d9b228b1eda3bf71a81b6201ec64e26086c23 |
|
27-Sep-2016 |
Mathieu Chartier <mathieuc@google.com> |
Use ObjPtr for reflection.cc/h/inl
Changed Pretty helpers to use this to reduce usage of Decode. The
eventual goal is not have almost any calls to ObjPtr::Decode.
Moved ObjPtr out of mirror namespace for convenience. Added more
PoisonObjectPointers calls in class linker, thread suspension.
Bug: 31113334
Test: test-art-host
Change-Id: I44d08db5143d95ed1b65e2f00f9749ef5cf379f7
|
| bdf7f1c3ab65ccb70f62db5ab31dba060632d458 |
|
31-Aug-2016 |
Andreas Gampe <agampe@google.com> |
ART: SHARED_REQUIRES to REQUIRES_SHARED
This coincides with the actual attribute name and upstream usage.
Preparation for deferring to libbase.
Test: m
Test: m test-art-host
Change-Id: Ia8986b5dfd926ba772bf00b0a35eaf83596d8518
|
| c381c36aacf977f7e314e6a91e47b31b04639f62 |
|
23-Aug-2016 |
Mathieu Chartier <mathieuc@google.com> |
Avoid CAS for marking region space bitmap for baker
Only have the GC thread mark it. This occurs when popping from the
mark stack. The race where an object may be pushed to the mark
stack twice is handled by not scanning if it is already marked.
Also avoid checking is_active when marking from the GC.
EAAC: 1263 -> 1253 (average of 30 runs)
GC time: 7.21s -> 6.83s (average of 18 runs)
Timings on 960 mhz N6P.
Bug: 12687968
Change-Id: I47e98c3e258829d2ba0babd803a219c82a36168c
Test: test-art-host, debug N6P booting with baker CC.
|
| 23da026ec7a7fae22833ed2f61a80d9f9bf7e732 |
|
30-Jul-2016 |
Mathieu Chartier <mathieuc@google.com> |
Revert "Revert "Use try lock to fix class resolution race""
Fix possible deadlock in EnsureResolved caused by interaction with
GC. Since we were sleeping while holding the mutator lock, it could
block thread suspension. This would deadlock if the thread that
had locked h_class is already suspended since we would spin forever
and not make progress.
Bug: 27417671
Bug: 30500547
Test: test-art-host ART_TEST_GC_STRESS=true
This reverts commit 69bf969c055c31a75d17ea92aeee756042678114.
(cherry picked from commit 4b0ef1c980a1f3b0201d77e33bdb2f7df12c9114)
Change-Id: If5766c2c3c8a130cbb83735cdb9970038570dafd
|
| adc538a57ad1f771051cd52afc216c3e5f0270cd |
|
29-Jul-2016 |
Mathieu Chartier <mathieuc@google.com> |
Revert "Use try lock to fix class resolution race"
This reverts commit a704eda0078989a73cac111ed309aca50d2e289b.
Bug: 27417671
Bug: 30500547
(cherry picked from commit 69bf969c055c31a75d17ea92aeee756042678114)
Change-Id: I4354d1c9f1c554f054e99efd7aa52d8a2c5d402c
|
| 1386f8619bb9cd338e51a9b6b5121bc8443bda76 |
|
13-Jul-2016 |
Mathieu Chartier <mathieuc@google.com> |
Use try lock to fix class resolution race
There was some possible deadlocks related to EnsureResolved caused by
acquiring an object lock.
Scenario:
Thread 1 acquires lock on obj1
Thread 1 begins to resolve / initialize class1
Thread 1 blocks since it sees that class1 is already being resolved and
gets preempted before it can acquire the object lock on class1
Thread 2 finishes resolving and initializing class1 and locks class1
Thread 2 blocks attempting to lock obj1
Thread 1 blocks attempting to lock class1
Deadlock
Fixed the deadlock by changing EnsureResolved to use a try lock for the
unresolved case.
Added a test.
Test: Device boot, test-art-host, monitor_test
Bug: 27417671
(cherry picked from commit a704eda0078989a73cac111ed309aca50d2e289b)
Change-Id: I1150b19bdc1a5cc87ae95eda4f2b6b4bca215a60
|
| d6a595baf0f389a7eb29fa3d94390c18a29c28da |
|
04-Aug-2016 |
Mathieu Chartier <mathieuc@google.com> |
Fix CC table lookup tests
Only set the mark bits for baker barrier, not sure why this
doesn't work for table lookup barrier yet.
Bug: 30162165
Test: test-art-host with CC table lookup
Change-Id: I4309ac22bfc0f8a5c9bf081365e3e38da43eee30
|
| 36a270ae4f288e49493432b7128f899ad579849e |
|
29-Jul-2016 |
Mathieu Chartier <mathieuc@google.com> |
Change one read barrier bit to mark bit
Optimization to help slow path performance. When the GC marks an
object through the read barrier slow path. The GC sets the mark bit
in the lock word of that reference. This bit is checked from the
assembly entrypoint the common case is that it is set. If the bit is
set, the read barrier knows the object is already marked and there is
no work to do.
To prevent dirty pages in zygote and image, the bit is set by the
image writer and zygote space creation.
EAAC score (lower is better):
N9: 777 -> 700 (average 31 of runs)
N6P (960000 mhz): 1737.48 -> 1442.31 (average of 25 runs)
Bug: 30162165
Bug: 12687968
Test: N9, N6P booting, test-art-host, test-art-target all with CC
Change-Id: Iae0cacfae221e33151d3c0ab65338d1c822ab63d
|
| 542451cc546779f5c67840e105c51205a1b0a8fd |
|
26-Jul-2016 |
Andreas Gampe <agampe@google.com> |
ART: Convert pointer size to enum
Move away from size_t to dedicated enum (class).
Bug: 30373134
Bug: 30419309
Test: m test-art-host
Change-Id: Id453c330f1065012e7d4f9fc24ac477cc9bb9269
|
| 4b0ef1c980a1f3b0201d77e33bdb2f7df12c9114 |
|
30-Jul-2016 |
Mathieu Chartier <mathieuc@google.com> |
Revert "Revert "Use try lock to fix class resolution race""
Fix possible deadlock in EnsureResolved caused by interaction with
GC. Since we were sleeping while holding the mutator lock, it could
block thread suspension. This would deadlock if the thread that
had locked h_class is already suspended since we would spin forever
and not make progress.
Bug: 27417671
Bug: 30500547
Test: test-art-host ART_TEST_GC_STRESS=true
This reverts commit 69bf969c055c31a75d17ea92aeee756042678114.
Change-Id: Id8750df065dc3b9ef7dc874f2eb2cc2c58e5d1eb
|
| 69bf969c055c31a75d17ea92aeee756042678114 |
|
29-Jul-2016 |
Mathieu Chartier <mathieuc@google.com> |
Revert "Use try lock to fix class resolution race"
This reverts commit a704eda0078989a73cac111ed309aca50d2e289b.
Bug: 27417671
Bug: 30500547
Change-Id: Ieea05236b9e61c722660cd9497c9d55d13ccd010
|
| a704eda0078989a73cac111ed309aca50d2e289b |
|
13-Jul-2016 |
Mathieu Chartier <mathieuc@google.com> |
Use try lock to fix class resolution race
There was some possible deadlocks related to EnsureResolved caused by
acquiring an object lock.
Scenario:
Thread 1 acquires lock on obj1
Thread 1 begins to resolve / initialize class1
Thread 1 blocks since it sees that class1 is already being resolved and
gets preempted before it can acquire the object lock on class1
Thread 2 finishes resolving and initializing class1 and locks class1
Thread 2 blocks attempting to lock obj1
Thread 1 blocks attempting to lock class1
Deadlock
Fixed the deadlock by changing EnsureResolved to use a try lock for the
unresolved case.
Added a test.
Test: Device boot, test-art-host, monitor_test
Bug: 27417671
Change-Id: Ic6e1c3ca6f45490cf8a7bf8e137dee71ac83ff64
|
| 5496f69c0a4c2cc357a065f57b7f4ff5d9ad2fa9 |
|
17-Feb-2016 |
Hiroshi Yamauchi <yamauchi@google.com> |
Disable read barriers in GC VisitReferences calls.
It's safe to disable read barriers in the CC collector's
VisitReferences calls. It speeds up the collector by some amount,
eg. Ritzperf EAAC GC time on N5 decreases by ~4.2%.
This is based on Mathieu Chartier's idea.
Bug: 12687968
Change-Id: I884dee7018f92761fad54269f4cd01db4fa82f2a
|
| dfe02f6aafee264478d510b9742ee266ea52e8a8 |
|
02-Feb-2016 |
Mathieu Chartier <mathieuc@google.com> |
Fix remaining read barrier issues in image relocation
Added a way to disallow read barriers, this makes it easy to find
the issues.
Bug: 26786304
Change-Id: I7ebb50832686d03e096a979aae9741239371683f
|
| fbc31087932a65e036a153afab3049dc5298656a |
|
24-Jan-2016 |
Mathieu Chartier <mathieuc@google.com> |
Revert "Revert "Load app images""
This reverts commit 1bc977cf2f8199311a97f2ba9431a184540e3e9c.
Bug: 22858531
Change-Id: Ide00bf3a73a02cba3bb364177204ad1b13f70295
|
| 1bc977cf2f8199311a97f2ba9431a184540e3e9c |
|
23-Jan-2016 |
Nicolas Geoffray <ngeoffray@google.com> |
Revert "Load app images"
Fails when a method is duplicated (see test 097-duplicate-method)
Bug: 22858531
This reverts commit f7fd970244f143b1abb956e29794c446e4d57f46.
Change-Id: Ib30ae5be00cc568e799290be6b3c8f29cbbe4c20
|
| f7fd970244f143b1abb956e29794c446e4d57f46 |
|
09-Nov-2015 |
Mathieu Chartier <mathieuc@google.com> |
Load app images
Support in-place patching of the app image based on boot image
location and app oat location. Only loads for art run test so far
since we do not automatically generate app images for app installs.
N5 maps launch time (~200 runs):
Before: 930ms
After: 878.18ms
After + image class table: 864.57ms
TODO:
Oatdump support.
Store class loaders as class roots in image.
Bug: 22858531
Change-Id: I9cbc645645e62ea2ed1ad8e139e91af7d88514c1
|
| ed70b4a439acea537d55266bb9b1f309de8cbec9 |
|
18-Nov-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
Relax CAS in some uses of Object::AtomicSetReadBarrierPointer.
Aside from the call to change the object from gray to black, the CAS in
AtomicSetReadBarrierPointer doesn't need to be a release CAS.
CC collector Ritz EAAC GC time: 28.4 -> 26.5s (-6.7%) on N5.
Bug: 12687968
Change-Id: Ic162717981b98e7f1e762fd6b27bd0c146187ca4
|
| 5abc6f913bc77aa3b9c05ad07905d4c1c28b2661 |
|
30-Jun-2015 |
Przemyslaw Szczepaniak <pszczepaniak@google.com> |
Add Object::Get/SetField methods for 8 and 16 bit fields.
For simplicity new methods lack support for transactions.
Those new netods are used to implement sun.misc.Unsafe.get/set
methods for 8/16 bit wide primitives. Those methods are used
by the java primitive serialization/deserialization methods.
Change-Id: Iaa133b32beac996642f691e1611536f5fd11a260
|
| fed3e2fd4919b58f8e4f8cbc317ee101f3b9af49 |
|
20-Oct-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
Relax some CASes for the CC collector.
That is, removing some unnecessary memory fences.
We can use the relaxed CAS for the mark bitmap and reference field/GC
root updates because only the atomicity of the updated word matters
there.
We can use the release CAS for the read barrier bits in the lock word
because it needs to make sure the reference field updates are visible
when the object changes black from gray (the field update stores won't
be reordered after the CAS.)
The CC collector's Ritz EAAC GC time decreases from 34.7s to
29.1s (-16%) on N5.
Bug: 12687968
Change-Id: If082d5911a25fac695df66263a8f55ce8149b199
|
| 05792b98980741111b4d0a24d68cff2a8e070a3a |
|
03-Aug-2015 |
Vladimir Marko <vmarko@google.com> |
ART: Move DexCache arrays to native.
This CL has a companion CL in libcore/
https://android-review.googlesource.com/162985
Change-Id: Icbc9e20ad1b565e603195b12714762bb446515fa
|
| 66c2d2d64d7ea75602eb63de7ae9bd2eaeb0a3c2 |
|
25-Aug-2015 |
Mathieu Chartier <mathieuc@google.com> |
Address some comments for class flags
Change-Id: I354f48aefc37ce92c4d02cfce1723db0e28907bf
|
| 52a7f5caebdf359ab877f1928aad59f1e9ad29fa |
|
19-Aug-2015 |
Mathieu Chartier <mathieuc@google.com> |
Add class flags to class to help GC scanning
Reduces GC time and pauses by reducing the number of loads required
to scan an object.
Average total GC time before on EvaluateAndApplyChanges (EAAC): 7.452s
After: 7.144s
Average GC pause times before on EAAC: 860.67us
After: 722.75us
Adding the class flags field cause a memory increase of ~24k system
wide on low memory devices.
Change-Id: I3f04212d5787bfbf5e55026584d149f55476105e
|
| 059ef3ddb2088f926ac452889e0953fdcd646a5e |
|
18-Aug-2015 |
Mathieu Chartier <mathieuc@google.com> |
Always visit object class from VisitReferences
We don't want to unload classes which have instances.
Slight increase in CMS GC time from ~6.5s to ~7.3s on
EvaluateAndApplyChanges.
Bug: 22720414
Change-Id: I467ff9c9d55163d2a90b999aef3bdd7b3f648bac
|
| e4275c07e9852a6944f47efa9d0591fceb8e8e36 |
|
07-Aug-2015 |
Mathieu Chartier <mathieuc@google.com> |
Visit class roots from ClassLoader::VisitReferences
This causes the classes of a class loader to get marked when that
class loader gets marked instead of during class root visiting.
Bug: 22720414
Change-Id: If53f042aff1d9f7bf94ecbe6886601edda029b7d
|
| 90443477f9a0061581c420775ce3b7eeae7468bc |
|
17-Jul-2015 |
Mathieu Chartier <mathieuc@google.com> |
Move to newer clang annotations
Also enable -Wthread-safety-negative.
Changes:
Switch to capabilities and negative capabilities.
Future work:
Use capabilities to implement uninterruptible annotations to work
with AssertNoThreadSuspension.
Bug: 20072211
Change-Id: I42fcbe0300d98a831c89d1eff3ecd5a7e99ebf33
|
| 3d21bdf8894e780d349c481e5c9e29fe1556051c |
|
22-Apr-2015 |
Mathieu Chartier <mathieuc@google.com> |
Move mirror::ArtMethod to native
Optimizing + quick tests are passing, devices boot.
TODO: Test and fix bugs in mips64.
Saves 16 bytes per most ArtMethod, 7.5MB reduction in system PSS.
Some of the savings are from removal of virtual methods and direct
methods object arrays.
Bug: 19264997
(cherry picked from commit e401d146407d61eeb99f8d6176b2ac13c4df1e33)
Change-Id: I622469a0cfa0e7082a2119f3d6a9491eb61e3f3d
Fix some ArtMethod related bugs
Added root visiting for runtime methods, not currently required
since the GcRoots in these methods are null.
Added missing GetInterfaceMethodIfProxy in GetMethodLine, fixes
--trace run-tests 005, 044.
Fixed optimizing compiler bug where we used a normal stack location
instead of double on ARM64, this fixes the debuggable tests.
TODO: Fix JDWP tests.
Bug: 19264997
Change-Id: I7c55f69c61d1b45351fd0dc7185ffe5efad82bd3
ART: Fix casts for 64-bit pointers on 32-bit compiler.
Bug: 19264997
Change-Id: Ief45cdd4bae5a43fc8bfdfa7cf744e2c57529457
Fix JDWP tests after ArtMethod change
Fixes Throwable::GetStackDepth for exception event detection after
internal stack trace representation change.
Adds missing ArtMethod::GetInterfaceMethodIfProxy call in case of
proxy method.
Bug: 19264997
Change-Id: I363e293796848c3ec491c963813f62d868da44d2
Fix accidental IMT and root marking regression
Was always using the conflict trampoline. Also included fix for
regression in GC time caused by extra roots. Most of the regression
was IMT.
Fixed bug in DumpGcPerformanceInfo where we would get SIGABRT due to
detached thread.
EvaluateAndApplyChanges:
From ~2500 -> ~1980
GC time: 8.2s -> 7.2s due to 1s less of MarkConcurrentRoots
Bug: 19264997
Change-Id: I4333e80a8268c2ed1284f87f25b9f113d4f2c7e0
Fix bogus image test assert
Previously we were comparing the size of the non moving space to
size of the image file.
Now we properly compare the size of the image space against the size
of the image file.
Bug: 19264997
Change-Id: I7359f1f73ae3df60c5147245935a24431c04808a
[MIPS64] Fix art_quick_invoke_stub argument offsets.
ArtMethod reference's size got bigger, so we need to move other args
and leave enough space for ArtMethod* and 'this' pointer.
This fixes mips64 boot.
Bug: 19264997
Change-Id: I47198d5f39a4caab30b3b77479d5eedaad5006ab
|
| e401d146407d61eeb99f8d6176b2ac13c4df1e33 |
|
22-Apr-2015 |
Mathieu Chartier <mathieuc@google.com> |
Move mirror::ArtMethod to native
Optimizing + quick tests are passing, devices boot.
TODO: Test and fix bugs in mips64.
Saves 16 bytes per most ArtMethod, 7.5MB reduction in system PSS.
Some of the savings are from removal of virtual methods and direct
methods object arrays.
Bug: 19264997
Change-Id: I622469a0cfa0e7082a2119f3d6a9491eb61e3f3d
|
| 3b45ef277e4a5f7e0314d5df7ef82e480156ba75 |
|
27-May-2015 |
Andreas Gampe <agampe@google.com> |
ART: Fix VerifyObject runtime verification
Update some bit-rotted code to work again. Most tests now work, for
some the verification overhead results in a timeout.
Change-Id: Ieab4f2de474a05e915e24abc93da3c2eeed996eb
|
| 848f70a3d73833fc1bf3032a9ff6812e429661d9 |
|
15-Jan-2014 |
Jeff Hao <jeffhao@google.com> |
Replace String CharArray with internal uint16_t array.
Summary of high level changes:
- Adds compiler inliner support to identify string init methods
- Adds compiler support (quick & optimizing) with new invoke code path
that calls method off the thread pointer
- Adds thread entrypoints for all string init methods
- Adds map to verifier to log when receiver of string init has been
copied to other registers. used by compiler and interpreter
Change-Id: I797b992a8feb566f9ad73060011ab6f51eb7ce01
|
| 60f63f53c01cb38ca18a815603282e802a6cf918 |
|
24-Apr-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
Use the lock word bits for Baker-style read barrier.
This enables the standard object header to be used with the
Baker-style read barrier.
Bug: 19355854
Bug: 12687968
Change-Id: Ie552b6e1dfe30e96cb1d0895bd0dff25f9d7d015
|
| 2cebb24bfc3247d3e9be138a3350106737455918 |
|
22-Apr-2015 |
Mathieu Chartier <mathieuc@google.com> |
Replace NULL with nullptr
Also fixed some lines that were too long, and a few other minor
details.
Change-Id: I6efba5fb6e03eb5d0a300fddb2a75bf8e2f175cb
|
| c785344b87221f5e4e6473e5b762e4e61fe65dcf |
|
27-Mar-2015 |
Mathieu Chartier <mathieuc@google.com> |
Move ArtField to native
Add linear alloc. Moved ArtField to be native object. Changed image
writer to put ArtFields after the mirror section.
Savings:
2MB on low ram devices
4MB on normal devices
Total PSS measurements before (normal N5, 95s after shell start):
Image size: 7729152 bytes
23112 kB: .NonMoving
23212 kB: .NonMoving
22868 kB: .NonMoving
23072 kB: .NonMoving
22836 kB: .NonMoving
19618 kB: .Zygote
19850 kB: .Zygote
19623 kB: .Zygote
19924 kB: .Zygote
19612 kB: .Zygote
Avg: 42745.4 kB
After:
Image size: 7462912 bytes
17440 kB: .NonMoving
16776 kB: .NonMoving
16804 kB: .NonMoving
17812 kB: .NonMoving
16820 kB: .NonMoving
18788 kB: .Zygote
18856 kB: .Zygote
19064 kB: .Zygote
18841 kB: .Zygote
18629 kB: .Zygote
3499 kB: .LinearAlloc
3408 kB: .LinearAlloc
3424 kB: .LinearAlloc
3600 kB: .LinearAlloc
3436 kB: .LinearAlloc
Avg: 39439.4 kB
No reflection performance changes.
Bug: 19264997
Bug: 17643507
Change-Id: I10c73a37913332080aeb978c7c94713bdfe4fe1c
|
| 2cd334ae2d4287216523882f0d298cf3901b7ab1 |
|
09-Jan-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
More of the concurrent copying collector.
Bug: 12687968
Change-Id: I62f70274d47df6d6cab714df95c518b750ce3105
|
| 2d7210188805292e463be4bcf7a133b654d7e0ea |
|
10-Nov-2014 |
Mathieu Chartier <mathieuc@google.com> |
Change 64 bit ArtMethod fields to be pointer sized
Changed the 64 bit entrypoint and gc map fields in ArtMethod to be
pointer sized. This saves a large amount of memory on 32 bit systems.
Reduces ArtMethod size by 16 bytes on 32 bit.
Total number of ArtMethod on low memory mako: 169957
Image size: 49203 methods -> 787248 image size reduction.
Zygote space size: 1070 methods -> 17120 size reduction.
App methods: ~120k -> 2 MB savings.
Savings per app on low memory mako: 125K+ per app
(less active apps -> more image methods per app).
Savings depend on how often the shared methods are on dirty pages vs
shared.
TODO in another CL, delete gc map field from ArtMethod since we
should be able to get it from the Oat method header.
Bug: 17643507
Change-Id: Ie9508f05907a9f693882d4d32a564460bf273ee8
(cherry picked from commit e832e64a7e82d7f72aedbd7d798fb929d458ee8f)
|
| e832e64a7e82d7f72aedbd7d798fb929d458ee8f |
|
10-Nov-2014 |
Mathieu Chartier <mathieuc@google.com> |
Change 64 bit ArtMethod fields to be pointer sized
Changed the 64 bit entrypoint and gc map fields in ArtMethod to be
pointer sized. This saves a large amount of memory on 32 bit systems.
Reduces ArtMethod size by 16 bytes on 32 bit.
Total number of ArtMethod on low memory mako: 169957
Image size: 49203 methods -> 787248 image size reduction.
Zygote space size: 1070 methods -> 17120 size reduction.
App methods: ~120k -> 2 MB savings.
Savings per app on low memory mako: 125K+ per app
(less active apps -> more image methods per app).
Savings depend on how often the shared methods are on dirty pages vs
shared.
TODO in another CL, delete gc map field from ArtMethod since we
should be able to get it from the Oat method header.
Bug: 17643507
Change-Id: Ie9508f05907a9f693882d4d32a564460bf273ee8
|
| 76649e8d775519fe19f2b14d18ac488c13296054 |
|
10-Nov-2014 |
Vladimir Marko <vmarko@google.com> |
Keep original order of fields in Class.
The fields of a class are ordered alphabetically in the dex
file. Keep the same order in the field arrays so that we can
do binary search lookups by name. Those lookups will be
implemented in a subsequent change in libcore/.
Bug: 18211592
(cherry picked from commit bfa3ed0ad988e1da13626ddbaf6dcae0c58ea79e)
Change-Id: I8f979de62ffe37d1c7d5c721717d2f3501e7c9e6
|
| bfa3ed0ad988e1da13626ddbaf6dcae0c58ea79e |
|
10-Nov-2014 |
Vladimir Marko <vmarko@google.com> |
Keep original order of fields in Class.
The fields of a class are ordered alphabetically in the dex
file. Keep the same order in the field arrays so that we can
do binary search lookups by name. Those lookups will be
implemented in a subsequent change in libcore/.
Bug: 18211592
Change-Id: I8f979de62ffe37d1c7d5c721717d2f3501e7c9e6
|
| 6a3c1fcb4ba42ad4d5d142c17a3712a6ddd3866f |
|
31-Oct-2014 |
Ian Rogers <irogers@google.com> |
Remove -Wno-unused-parameter and -Wno-sign-promo from base cflags.
Fix associated errors about unused paramenters and implict sign conversions.
For sign conversion this was largely in the area of enums, so add ostream
operators for the effected enums and fix tools/generate-operator-out.py.
Tidy arena allocation code and arena allocated data types, rather than fixing
new and delete operators.
Remove dead code.
Change-Id: I5b433e722d2f75baacfacae4d32aef4a828bfe1b
|
| 2c4257be8191c5eefde744e8965fcefc80a0a97d |
|
24-Oct-2014 |
Ian Rogers <irogers@google.com> |
Tidy logging code not using UNIMPLEMENTED.
Change-Id: I7a79c1671a6ff8b2040887133b3e0925ef9a3cfe
|
| 13735955f39b3b304c37d2b2840663c131262c18 |
|
08-Oct-2014 |
Ian Rogers <irogers@google.com> |
stdint types all the way!
Change-Id: I4e4ef3a2002fc59ebd9097087f150eaf3f2a7e08
|
| cdc1aaffabbdf417d29b203b2cd2763ed2d623f8 |
|
09-Oct-2014 |
Ian Rogers <irogers@google.com> |
Simplify instance reference offsets.
Don't encode Object's class. Use trailing rather than leading zeroes to give
offset position.
Change-Id: I1ae74e7a01f63696429644adf81cdf6ee58832fe
|
| faff0f05fef90577c9744505555675185832aacd |
|
10-Sep-2014 |
Mingyao Yang <mingyao@google.com> |
Remove reference_static_offsets used for iterating through class static fields.
Since static fields are contiguous in class object and there is no need to
traverse super classes, it's not meaningful to use reference_static_offsets.
Also especially with embedded vtable/imt, static field offset can't be encoded
with an unsigned integer anyway.
A corresponding change is made to Class.java to remove the member field.
Bug: 16236588
Change-Id: I1fde3cd9efce884945876f0658c63d992164fd94
|
| 37f05ef45e0393de812d51261dc293240c17294d |
|
17-Jul-2014 |
Fred Shih <ffred@google.com> |
Reduced memory usage of primitive fields smaller than 4-bytes
Reduced memory used by byte and boolean fields from 4 bytes down to a
single byte and shorts and chars down to two bytes. Fields are now
arranged as Reference followed by decreasing component sizes, with
fields shuffled forward as needed.
Bug: 8135266
Change-Id: I65eaf31ed27e5bd5ba0c7d4606454b720b074752
|
| d85614222fa062ec809af9d65f04ab6b7dc1c248 |
|
11-Jul-2014 |
Fred Shih <ffred@google.com> |
Revert "Revert "Revert "Revert "Add intrinsic for Reference.get()""""
Fixed TargetReg issue causing build failure for x86.
This reverts commit 9e82bd3f0ce9e5f5777bea2f752ff3e251d32f9f.
(cherry picked from commit 4ee7a665e7f9cd2c5ace2d6304e33f64067b209f)
Change-Id: I555f4e06955711262e6b37ffbeabee9698ec695c
|
| 4ee7a665e7f9cd2c5ace2d6304e33f64067b209f |
|
11-Jul-2014 |
Fred Shih <ffred@google.com> |
Revert "Revert "Revert "Revert "Add intrinsic for Reference.get()""""
Fixed TargetReg issue causing build failure for x86.
This reverts commit 9e82bd3f0ce9e5f5777bea2f752ff3e251d32f9f.
Change-Id: I7e6a526954467aaf68deeed999880dfe9aa5f06e
|
| d8434439dc64add41cdfa69ddf96b960af9050de |
|
11-Jul-2014 |
Hans Boehm <hboehm@google.com> |
Call strong CAS from unsafe. Add more CAS versions.
Adds a number of additional CAS versions. Calls the correct
one from sun.misc.unsafe, fixing a recently introduced bug.
Avoid unnecessary ordering constraint when installing hash code.
Change-Id: I7c09d0c95ceb2a549ec28ee34084198ab3107946
|
| 98d1cc8033251c93786e2fa8c59a2e555a9493be |
|
16-May-2014 |
Mingyao Yang <mingyao@google.com> |
Improve performance of invokevirtual/invokeinterface with embedded imt/vtable
Add an embedded version of imt/vtable into class object. Both tables start at
fixed offset within class object so method/entry point can be loaded directly
from class object for invokeinterface/invokevirtual.
Bug: 8142917
Change-Id: I4240d58cfbe9250107c95c0708c036854c455968
|
| 228602f562f1d130d06e60a98752d99c2d467d6a |
|
10-Jul-2014 |
Ian Rogers <irogers@google.com> |
Make CAS operations in Object use art::Atomic.
Make naming consistent with art::Atomic.
Change-Id: If3abdb019ef8b53bd809e3fef3fd5248aeb27e9a
|
| 4f1ebc2b86c8467d1ecb3ec655316e6d7ee8b8b5 |
|
25-Jun-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Add read barriers to the constant roots.
This change makes it possible to concurrently scan the constant roots
(the roots visited by Runtime::VisitConstantRoots()) such as the class
of java.lang.Class by adding read barriers.
Bug: 12687968
Change-Id: If1afea471c4e1093688d2db37b7f1fc2742edeef
|
| 52e4b43d62896b56f8c2bd041e528472bb4a0d8d |
|
10-Jun-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add mark compact collector.
The mark compact collector is a 4 phase collection, doing a normal
full mark_sweep, calculating forwarding addresses of objects in the
from space, updating references of objects in the from space, and
moving the objects in the from space.
Support is diabled by default since it needs to have non movable
classes and field arrays. Performance numbers is around 50% as fast.
The main advantage that this has over semispace is that the worst
case memory usage is 50% since we only need one space isntead of two.
TODO: Make field arrays and classes movable. This causes complication
since Object::VisitReferences relies on these, so if we update the
fields of an object but another future object uses this object to
figure out what fields are reference fields it doesn't work.
Bug: 14059466
Change-Id: I661ed3b71ad4dde124ef80312c95696b4a5665a1
|
| 3035961cb41865b80b927546be0c708b6389cec6 |
|
22-May-2014 |
Hans Boehm <hboehm@google.com> |
Clean up and augment Atomic class. Replace QuasiAtomic MemBars.
Add a number of missing C++11 operations to Atomic class.
Invoke the 64 bit routines in QuasiAtomic when necessary.
Replace QuasiAtomic membars with fences that correspond to C++11 fences.
QuasiAtomic was moved to the top of the file. Only fence implementations
actually changed.
This replaces some buggy uses of MembarStoreStore, as reported
in b/14685856 .
Avoid some redundant fences for long volatile operations.
Incompletely converts low-level memory access operations to Atomic.
Change-Id: Iea828431a0cea46540eb74fcaa02071cab6fdcda
|
| 61c5ebc6aee2cac1c363de6fbdac25ada1697fdb |
|
06-Jun-2014 |
Mathieu Chartier <mathieuc@google.com> |
Change FieldHelper to use a handle.
Fixed compaction bugs related to FieldHelper::GetType in:
artSet32InstanceFromCode
SetFieldValueImpl
CheckReceiver
Field_set
interpreter::DoFieldPut
MethodVerifier::VerifyISGet
MethodVerifier::VerifyISPut
MethodVerifier::VerifyIGetQuick
Bug: 13077697
Change-Id: I7de9ded2893b5568d43e4daa86fd135bf5508b72
|
| bd0fb61e24270b1f382ecbef4c1260c703550e84 |
|
20-May-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Simplify Class::IsArtFieldClass().
Fix the slight glitch that when ImageSpace::VerifyImageAllocations()
called in ImageSpace::Create(), the ArtField and ArtMethod class roots
weren't set, which were used by DCHECKs in Object::Size(), which
VerifyImageAllocations() calls, by delaying the point of the
VerifyImageAllocations() call to Runtime::Init() at which point the
class linker has set the class roots.
To completely disable read barriers from Object::SizeOf(), the
ReadBarrierOption template parameter should have been added to
Class::GetInstanceField(), which calls GetFieldObject(), when it's
called from Class::IsArtFieldClass(). This change fixes this by
removing the need for the call, instead of adding the
ReadBarrierOption parameter.
Bug: 12687968
Change-Id: Ibbecc08f4e3b898851805d690dff8ccac55e94f2
|
| 25023c744c4388a6459b21cc3babf8c602b024a2 |
|
09-May-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Make it possible to disable read barriers in Class::GetObjectSize()
This is a leftover from cl/91831 (commit
9103c86a98524e9ddfd14f8cee56e919f68eee9b) that attempted to make it
possible to disable read barriers in Object::SizeOf().
Bug: 12687968
Change-Id: I2b05076832936881ec61bc21b6eb6b7c04e0a1f0
|
| 6a154a419022e2caed4edfd1f311edf733278f35 |
|
02-May-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Update the GetFieldObject calls behind the read barrier macros.
This is a leftover from changes 92535 and 93191.
Bug: 12687968
Change-Id: I4f09d258d037546c716b78d6f1ae6abe5af1137d
|
| 6e83c172f385cb45dd13bbcf41d2df8e410828c6 |
|
02-May-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Replace the bool kDoReadBarrier template parameter with an enum.
Fix one kDoReadBarrier/kIsVolatile mixup in an Object::GetFieldObject
call.
Bug: 12687968
Change-Id: I896b1137b21a20c0504abd2bf3fe6f83805f3300
|
| b0fa5dc7769c1e054032f39de0a3f6d6dd06f8cf |
|
29-Apr-2014 |
Ian Rogers <irogers@google.com> |
Force inlining on trivial accessors.
Make volatility for GetFieldObject a template parameter.
Move some trivial mirror::String routines to a -inl.h.
Bug: 14285442
Change-Id: Ie23b11d4f18cb15a62c3bbb42837a8aaf6b68f92
|
| 9103c86a98524e9ddfd14f8cee56e919f68eee9b |
|
22-Apr-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
More code for the read barrier support.
Make it possible to disable the RB in Object::SizeOf() (and the
functions it calls transitively) which the collector will need to call
to get the size of an object when copying.
Add Object::AtomicSetReadBarrierPointer() for atomic write of a RB
pointer.
Bug: 12687968
Change-Id: Ibedd252860ac7ccd17e4e7d71b377a8892b48ff0
|
| bbd695c71e0bf518f582e84524e1cdeb3de3896c |
|
16-Apr-2014 |
Mathieu Chartier <mathieuc@google.com> |
Replace ObjectSet with LargeObjectBitmap.
Speeds up large object marking since large objects no longer required
a lock. Changed the GCs to use the heap bitmap for marking objects
which aren't in the fast path. This eliminates the need for a
MarkLargeObject function.
Maps before (10 GC iterations):
Mean partial time: 180ms
Mean sticky time: 151ms
Maps after:
Mean partial time: 161ms
Mean sticky time: 101ms
Note: the GC durations are long due to recent ergonomic changes and
because the fast bulk free hasn't yet been enabled. Over 50% of the
GC time is spent in RosAllocSpace::FreeList.
Bug: 13571028
Change-Id: Id8f94718aeaa13052672ccbae1e8edf77d653f62
|
| 4d7f61d44a732cfbc8573e5d93364983fd746888 |
|
17-Apr-2014 |
Mathieu Chartier <mathieuc@google.com> |
Use non volatile lock words in semispace collector.
GSS FormulaEvaluationActions.EvaluateAndApplyChanges:
Before GC time: 9.1s
After GC time: 7.98s
Fixed timing logger errors.
Change-Id: I4193c6ccbbbe7a7220dfaabbf3472a5dcebae616
|
| 800ac2defde5d12b2f1f313c6b6162560cfa6fc7 |
|
03-Apr-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Insert an empty read barrier call.
Bug: 12687968
Change-Id: Ie1d28658e16e09f6a983cb5c1f0d5b375b7ae069
|
| 624468cd401cc1ac0dd70c746301e0788a597759 |
|
01-Apr-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Make the support code for read barriers a bit more general.
Add an option for Baker in addition to Brooks.
Bug: 12687968
Change-Id: I8a31db817ff6686c72951b6534f588228e270b11
|
| 580a8dfda74e7999f6b3e28e239ae8461a39dfcb |
|
26-Mar-2014 |
Mathieu Chartier <mathieuc@google.com> |
Less redundant verification.
~3 less objects verified per object scanned in the GC. Helps the
irogers dogfood experience.
Change-Id: I6efeab7842a6c702adecef73fb573c19291fecf2
|
| 407f702da4f867c074fc3c8c688b8f8c32279eff |
|
18-Feb-2014 |
Mathieu Chartier <mathieuc@google.com> |
Refactor object reference visiting logic.
Refactored the reference visiting logic to be in mirror::Object
instead of MarkSweep.
Change-Id: I773249478dc463d83b465e85c2402320488577c0
|
| 8fa2dad7fe7909c8335101d6c8904ae997cdf29f |
|
13-Mar-2014 |
Mathieu Chartier <mathieuc@google.com> |
Refactor reference code into mirror namespace.
Added two new files: mirror/reference.h and mirror/reference-inl.h.
Change-Id: Ibe3ff6379aef7096ff130594535b7f7c0b7dabce
|
| b48b9eb6d181a1f52e2e605cf26a21505f1d46ed |
|
01-Mar-2014 |
Ian Rogers <irogers@google.com> |
Fix clang to compile and run host tests.
Don't use the computed goto interpreter with clang 3.4 as it causes compilation
to hang.
Avoid inclusion of LLVM_(HOST|DEVICE)_BUILD_MK except for with portable as it
sets clang incompatible cflags.
Most fixes are self-evident, for the quick dex file method inliner the enums
were being used with ostreams, so fix the enums and operator out python script
to allow this.
Note this change effects portable but this is untestable as portable was broken
by ELF file and mc linker changes.
Change-Id: Ia54348f6b1bd3f76d3b71c6e8c5f97626386b903
|
| 9d04a20bde1b1855cefc64aebc1a44e253b1a13b |
|
31-Jan-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
(Experimental) Add Brooks pointers.
This feature is disabled by default.
Verified that the Brooks pointers are installed correctly by using the
CMS/SS collectors.
Change-Id: Ia9be9814ab6e29169ac85edc4792ce8c81d552a9
|
| 4e30541a92381fb280cd0be9a1763b713ee4d64c |
|
19-Feb-2014 |
Mathieu Chartier <mathieuc@google.com> |
Fix and optimize verify object.
VerifyObject no longer resides in heap. You can now enable
VerifyObject for non-debug builds. VerifyStack is still slow, so it
is now guarded by its own flag.
Fixed the image writer to not use verification at places where
verification fails due to invalid reads.
Fixed RosAlloc to use SizeOf which doesn't call verify object.
Added a flag paremeter to some of the mirror getters / setters to
be able to selectively disable VerifyObject on certain calls.
Optimized the GC to not verify each object multiple times during
object scanning if verify object is enabled.
Added 3 verification options: verify reads, verify this, and verify
writes so that you can select how much verification you want for
mirror getters and setters.
Removed some useless DCHECKs which would slow debug builds without
providing any benefits.
TODO: RosAlloc verification doesn't currently work with verify
objects.
Bug: 12934910
Bug: 12879358
Change-Id: Ic61033104dfc334543f89b0fc0ad8cd4f4015d69
|
| d2fe10a3a34af171bf1631219cd2d6ff6b7778b5 |
|
15-Jan-2014 |
Sebastien Hertz <shertz@google.com> |
Remove blacklist
Removes the class initialization blacklist and use transaction to detect and
revert class initialization attempting to invoke native method. This only
concerns class initialization happening at compilation time when generating an
image (like boot.art for the system).
In transactional mode, we log every object's field assignment and array update.
Therefore we're able to abort a transaction to restore values of fields and
array as they were before the transaction starts. We also log changes to the
intern string table so we can restore its state prior to transaction start.
Since transactional mode only happens at compilation time, we don't need to log
all these changes at runtime. In order to reduce the overhead of testing if
transactional mode is on/off, we templatize interfaces of mirror::Object and
mirror::Array, respectively responsible for setting a field and setting an
array element.
For various reasons, we skip some specific fields from transaction:
- Object's class and array's length must remain unchanged so garbage collector
can compute object's size.
- Immutable fields only set during class loading: list of fields, method,
dex caches, vtables, ... as all classes have been loaded and verified before a
transaction occurs.
- Object's monitor for performance reason.
Before generating the image, we browse the heap to collect objects that need to
be written into it. Since the heap may still holds references to unreachable
objects due to aborted transactions, we trigger one collection at the end of
the class preinitialization phase.
Since the transaction is held by the runtime and all compilation threads share
the same runtime, we need to ensure only one compilation thread has exclusive
access to the runtime. To workaround this issue, we force class initialization
phase to run with only one thread. Note this is only done when generating image
so application compilation is not impacted. This issue will be addressed in a
separate CL.
Bug: 9676614
Change-Id: I221910a9183a5ba6c2b99a277f5a5a68bc69b5f9
|
| e7e8a5fea2d852cccc840fa046151a16627f26cd |
|
15-Feb-2014 |
Mathieu Chartier <mathieuc@google.com> |
Remove using mirror::* in jni internal.
Cleans up the code since it helps make it easier to spot moving GC
related bugs.
Also fixed moving GC bugs in MonitorEnter, and FindFieldID.
Change-Id: I794e4fbcc2010875fc6b299c6d4c9fb0f071b71a
|
| 29501cf5663c7951717bae3521815d121a778396 |
|
08-Feb-2014 |
Ian Rogers <irogers@google.com> |
Make primitive getter/setters consistently signed.
Use QuasiAtomic for Cas64, which fixes the MIPS build of ART.
Change-Id: Id1dba264b148a7ce9bcc637159ec733cfbd88b70
|
| ef7d42fca18c16fbaf103822ad16f23246e2905d |
|
06-Jan-2014 |
Ian Rogers <irogers@google.com> |
Object model changes to support 64bit.
Modify mirror objects so that references between them use an ObjectReference
value type rather than an Object* so that functionality to compress larger
references can be captured in the ObjectRefererence implementation.
ObjectReferences are 32bit and all other aspects of object layout remain as
they are currently.
Expand fields in objects holding pointers so they can hold 64bit pointers. Its
expected the size of these will come down by improving where we hold compiler
meta-data.
Stub out x86_64 architecture specific runtime implementation.
Modify OutputStream so that reads and writes are of unsigned quantities.
Make the use of portable or quick code more explicit.
Templatize AtomicInteger to support more than just int32_t as a type.
Add missing, and fix issues relating to, missing annotalysis information on the
mutator lock.
Refactor and share implementations for array copy between System and uses
elsewhere in the runtime.
Fix numerous 64bit build issues.
Change-Id: I1a5694c251a42c9eff71084dfdd4b51fff716822
|
| b122a4bbed34ab22b4c1541ee25e5cf22f12a926 |
|
20-Nov-2013 |
Ian Rogers <irogers@google.com> |
Tidy up memory barriers.
Change-Id: I937ea93e6df1835ecfe2d4bb7d84c24fe7fc097b
|
| ad2541a59c00c2c69e8973088891a2b5257c9780 |
|
25-Oct-2013 |
Mathieu Chartier <mathieuc@google.com> |
Fix object identity hash.
The object identity hash is now stored in the monitor word after
being computed. Hashes are computed by a pseudo random number
generator.
When we write the image, we eagerly compute object hashes to
prevent pages getting dirtied.
Bug: 8981901
Change-Id: Ic8edacbacb0afc7055fd740a52444929f88ed564
|
| 79b4f38dd35b83206e8166aaafb94bd75c3318b3 |
|
24-Oct-2013 |
Mathieu Chartier <mathieuc@google.com> |
Fix incorrect initial dex cache size.
We were previously setting it to be sizeof(DexCacheClass) instead
of sizeof(DexCache). This was causing some problems with
compaction when I relied on the object sizes being accurate to
visit objects in the bump pointer space.
Bug: 8981901
Change-Id: Iede04763aced041986b1b239368fc867143ad70d
|
| d9c4fc94fa618617f94e1de9af5f034549100753 |
|
02-Oct-2013 |
Ian Rogers <irogers@google.com> |
Inflate contended lock word by suspending owner.
Bug 6961405.
Don't inflate monitors for Notify and NotifyAll.
Tidy lock word, handle recursive lock case alongside unlocked case and move
assembly out of line (except for ARM quick). Also handle null in out-of-line
assembly as the test is quick and the enter/exit code is already a safepoint.
To gain ownership of a monitor on behalf of another thread, monitor contenders
must not hold the monitor_lock_, so they wait on a condition variable.
Reduce size of per mutex contention log.
Be consistent in calling thin lock thread ids just thread ids.
Fix potential thread death races caused by the use of FindThreadByThreadId,
make it invariant that returned threads are either self or suspended now.
Code size reduction on ARM boot.oat 0.2%.
Old nexus 7 speedup 0.25%, new nexus 7 speedup 1.4%, nexus 10 speedup 2.24%,
nexus 4 speedup 2.09% on DeltaBlue.
Change-Id: Id52558b914f160d9c8578fdd7fc8199a9598576a
|
| deb43702e611d6c75b459ea352a00f5d98fc0aa9 |
|
04-Sep-2013 |
Jeff Hao <jeffhao@google.com> |
Fix handling of unresolved references in verifier.
The verifier should not treat use of unresolved references as a reason to reject
the entire class. Instead, the verifier treats the instruction as a throw. If
that class is run, the interpreter with extra checks will throw an exception.
Bug: 10457426
(cherry picked from commit a3faaf4bece7f42529c013fe87bd41de59798656)
Change-Id: I161bfdbfa116890ffa9e7a593c756229bd939eb4
|
| a3faaf4bece7f42529c013fe87bd41de59798656 |
|
04-Sep-2013 |
Jeff Hao <jeffhao@google.com> |
Fix handling of unresolved references in verifier.
The verifier should not treat use of unresolved references as a reason to reject
the entire class. Instead, the verifier treats the instruction as a throw. If
that class is run, the interpreter with extra checks will throw an exception.
Bug: 10457426
Change-Id: I3799da843a7ffb3519bbf6dc13a6276519d9cb95
|
| ea46f950e7a51585db293cd7f047de190a482414 |
|
30-Jul-2013 |
Brian Carlstrom <bdc@google.com> |
Refactor java.lang.reflect implementation
Cherry-picked from commit ed41d5c44299ec5d44b8514f6e17f802f48094d1.
Move to ArtMethod/Field instead of AbstractMethod/Field and have
java.lang.reflect APIs delegate to ArtMethod/ArtField.
Bug: 10014286.
Change-Id: Iafc1d8c5b62562c9af8fb9fd8c5e1d61270536e7
|
| fc0e3219edc9a5bf81b166e82fd5db2796eb6a0d |
|
17-Jul-2013 |
Brian Carlstrom <bdc@google.com> |
Fix multiple inclusion guards to match new pathnames
Change-Id: Id7735be1d75bc315733b1773fba45c1deb8ace43
|
| 7940e44f4517de5e2634a7e07d58d0fb26160513 |
|
12-Jul-2013 |
Brian Carlstrom <bdc@google.com> |
Create separate Android.mk for main build targets
The runtime, compiler, dex2oat, and oatdump now are in seperate trees
to prevent dependency creep. They can now be individually built
without rebuilding the rest of the art projects. dalvikvm and jdwpspy
were already this way. Builds in the art directory should behave as
before, building everything including tests.
Change-Id: Ic6b1151e5ed0f823c3dd301afd2b13eb2d8feb81
|