| 558511a782baf60547ce9a93377ef0eae89e01ad |
|
02-Mar-2018 |
Mathieu Chartier <mathieuc@google.com> |
Remove mandatory address for non moving space / zygote
This used to not be safe when the immune region didn't correctly
handle gaps between the regions.
The immune spaces handles this correctly but may cause the GC to be
slightly slower when it occurs.
Test: test-art-host
Bug: 74062530
(cherry picked from commit fa4ea8243f1ed949bb76cf088a4b34275c622696)
Merged-In: Ia24db1c0a8c7b3667f3f893d0d9e9ead33d6a248
Change-Id: Ie6191f76dd082a6a22496e2855ba623ce594f68e
|
| 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)
|
| 67bf42e89592c3a1c648f927f2ce3ccb189a1161 |
|
27-Feb-2018 |
David Sehr <sehr@google.com> |
Header library to remove dependence on runtime/
Add a new header library to remove libdexfile and others' dependence on
runtime (typically runtime/base) includes in libdexfile. Also a small step
to tease dexlayout and profman away from relying on these as well.
Bug: 22322814
Test: make -j 50 checkbuild
make -j 50 test-art-host-gtest
Change-Id: I38e2fe399a75f4bc6318c77a71954c00ea73ec2b
|
| 99bd16bfd0cebb338dd33473cdf71176d560d6a3 |
|
21-Feb-2018 |
Roland Levillain <rpl@google.com> |
Extend error reporting in art::gc::Heap::GrowForUtilization.
Test: art/test.py
Bug: 73712520
Change-Id: I425368fd0bb69241e986b38c40a541e8f5c05c59
|
| 8f7ea9ab1703ef52c0c5ca3490e2913ac67f2a16 |
|
26-Jan-2018 |
Roland Levillain <rpl@google.com> |
Various ART GC documentation updates.
Also some cosmetic and stylistic changes.
Test: mmma art
Change-Id: I411cc45c6b5cb2a4b0652eeb9c4a6f4a3a274bd6
|
| ef012225dcec8f736fc8fd848f5d8ee01c3c280f |
|
21-Jun-2017 |
Roland Levillain <rpl@google.com> |
Fix typos.
Test: art/test/testrunner/testrunner.py
Change-Id: I5f5e052c940650a3adb3a8717b84260e6b306500
|
| c06d6c60fcd8e0a66d9e8dc51a8f35e8f87454c8 |
|
24-Jan-2018 |
Roland Levillain <rpl@google.com> |
Remove art::gc::Heap::allocation_rate_.
This member is no longer used anywhere.
Test: mmma art
Change-Id: Ieb11ea74b27c3df9d818c420c50cc04d0f4cfb26
|
| f4eedfe973638ab22116acf6b85a1cf8a439db97 |
|
29-Dec-2017 |
Richard Uhler <ruhler@google.com> |
Remove blocking case from RegisterNativeAllocation.
To avoid causing jank for dubious reasons.
Test: art/test/testrunner/testrunner.py -b -t 004-NativeAllocations --host
Test: vogar --mode host libcore/luni/src/test/java/libcore/libcore/util/NativeAllocationRegistryTest.java
Bug: 70831911
Change-Id: I310e3daf5bd3b65097b6011d6a96fb42ac50132b
|
| 94c589db78da2b66bc681c6480819ca4bd4d3326 |
|
27-Dec-2017 |
Andreas Gampe <agampe@google.com> |
ART: Mark Dbg GCs as debugger
Change Heap::CollectGarbage to accept explicit GcCause, but implicitly
default to kGcCauseExplicit.
Change Dbg functions that run an explicit GC to set the cause to
kGcCauseDebugger.
Test: m test-art-host
Change-Id: I53d4073fca01c1de78d14a58dff33004c7971981
|
| 170331f0e44a0e07fcfe0b5932517e0500f5cd1f |
|
08-Dec-2017 |
Andreas Gampe <agampe@google.com> |
ART: Remove base/logging from heap-inl.h
Hide the single uncommon VLOG and remove the include. Fix up
transitive includes.
Test: m
Change-Id: I917df597cb62c57040c1fb0e0079df4d95e5a658
|
| dcc528d2c7d5ac2cc075d4c965fdf702421d0f43 |
|
07-Dec-2017 |
Andreas Gampe <agampe@google.com> |
ART: Move runtime-debug flags to own files
To reduce the need for base/logging.h and separate out concerns.
Test: m
Change-Id: Ib373357325c6e622f608ada341594c3bea2fce2e
|
| 57943810cfc789da890d73621741729da5feaaf8 |
|
07-Dec-2017 |
Andreas Gampe <agampe@google.com> |
ART: Replace base/logging with android-base/logging
Replace wherever possible. ART's base/logging is now mainly VLOG
and initialization code that is unnecessary to pull in and makes
changes to verbose logging more painful than they have to be.
Test: m test-art-host
Change-Id: I3e3a4672ba5b621e57590a526c7d1c8b749e4f6e
|
| 660be6ff67870d2512230b13effefdbc5abe43e4 |
|
22-Nov-2017 |
Richard Uhler <ruhler@google.com> |
Add VMDebug.getInstancesOfClasses API.
The API can be used to iterate over instances of a given type on the
heap.
The GetInstancesOfClassesBenchmark run on bullhead shows the run time
of the API grows linearly in the number of classes, C, to get instances
of and the number of total instances, N, allocated on the heap.
C N=~2^18 N=~2^19
1 13ms 21ms
2 26ms 43ms
4 53ms 87ms
Bug: 69729799
Test: ./test/testrunner/testrunner.py -t 099-vmdebug -b --host
Change-Id: Ied053d19760e656012e2577776f75a1cc0a14ac3
|
| 2ffb703bf431d74326c88266b4ddaf225eb3c6ad |
|
08-Nov-2017 |
Igor Murashkin <iam@google.com> |
cpplint: Cleanup errors
Cleanup errors from upstream cpplint in preparation
for moving art's cpplint fork to upstream tip-of-tree cpplint.
Test: cd art && mm
Bug: 68951293
Change-Id: I15faed4594cbcb8399850f8bdee39d42c0c5b956
|
| ed56b5e1b92ab11a2a1414807d1a8a1c9123f455 |
|
19-Oct-2017 |
Andreas Gampe <agampe@google.com> |
ART: Only stress background transition in slow-debug
To lighten the load when starting new processes, by default keep
the wait time until running background compaction.
Bug: 35644369
Test: m test-art-host
Change-Id: I932684b02f3095bc011e5e8346eea103e458bfc5
|
| 217488a9ddf351033c1688198c492b9c40c36d8a |
|
18-Sep-2017 |
Andreas Gampe <agampe@google.com> |
ART: Move read barrier config out of globals
Reduce the global dependencies by refactoring where the read
barrier constants are defined.
Rename read_barier_c.h to read_barier_config.h and ifdef the
C++ parts to have a common header for both C/asm and C++.
Put heap poisoning configuration into its own minimal header.
Fix up transitive includes.
Test: m
Change-Id: I159669ec61e3d1c4c7ddcd79e63b023a0519717a
|
| 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
|
| 0c18338ebae6b3597c882887f8354b64abb5e90f |
|
14-Jul-2017 |
Andreas Gampe <agampe@google.com> |
ART: Replace or remove some ObjectCallback Walk functions
Replace with visitor functions in RegionSpace and SpaceBitmap. Remove
old ObjectCallback version in HeapBitmap. Fix up users. Move some
thread-safety annotations.
Move ObjectCallback definition to the only remaining user (ModUnionTable).
Test: m
Change-Id: I10307aeacad0c60d21fbade2081ec040d6a6ac4c
|
| 10d39085d5e064ae0afdc898eb32741a660c4b9d |
|
14-Jul-2017 |
Andreas Gampe <agampe@google.com> |
ART: Remove old visit functions
Remove now-unused ObjectCallback-based VisitObjects functions.
Test: m
Change-Id: Iefccd6ff28654d86f5254a411e7c30263a85b9aa
|
| 351c44765279142d15333e2ae02b8a423d195b1b |
|
13-Jul-2017 |
Andreas Gampe <agampe@google.com> |
ART: Add templated VisitObjects
Add templated versions of VisitObjects that accept visitors. This
allows to use more modern lambdas.
Test: m
Change-Id: I71a7f59bcae02090b9493bf8b477bb8b6ba649de
|
| 34afcde6e57f6d8a6077c0f23c1481c1122d3be9 |
|
01-Jul-2017 |
Mathieu Chartier <mathieuc@google.com> |
Trace heap size for allocations
Trace the heap size when we get bulk allocated bytes. This doesn't
happen often for RosAlloc/TLAB and should not hurt
performance.
The motivation is to enable systrace to see allocation behavior of
apps.
Test: test-art-host
Change-Id: I1e373b29507cc64bd5b6ecc9b4e9278399adf4b1
|
| 40112dd31260c624337f353ec00c444df64458d4 |
|
27-Jun-2017 |
Mathieu Chartier <mathieuc@google.com> |
Make waiting for blocking GC more intuitive
Now tells what GC cause blocked on what other GC cause.
Example output:
WaitForGcToComplete blocked HeapTrim on ProfileSaver for 396.361ms
Bug: 62941975
Test: test-art-host
(cherry picked from commit 87a619f43d4cb79b8723b42dc53cb10f7577f1de)
Change-Id: I1e3a35244db16974487dd8b82baae81669ae9aee
|
| 87a619f43d4cb79b8723b42dc53cb10f7577f1de |
|
27-Jun-2017 |
Mathieu Chartier <mathieuc@google.com> |
Make waiting for blocking GC more intuitive
Now tells what GC cause blocked on what other GC cause.
Example output:
WaitForGcToComplete blocked HeapTrim on ProfileSaver for 396.361ms
Bug: 62941975
Test: test-art-host
Change-Id: I66832d7f1daf565129c27ad23fc25852d7782733
|
| 5d08fcc21a14190c4dba3028a3e35d281afdd016 |
|
06-Jun-2017 |
Andreas Gampe <agampe@google.com> |
ART: More header cleanups
Pull out more dependencies through forward declarations.
Test: m test-art-host
Change-Id: I7d86726928937f788b956ec9eac91532d66d57ae
|
| d49012909625c3bf87bf51138fe79315ce1b1bdc |
|
31-May-2017 |
Andreas Gampe <agampe@google.com> |
ART: Clean up heap headers
Use more forward declarations for accounting structures and spaces.
Factor out structs to reduce header surface. Remove heap include where
unnecessary. Fix up transitive users. Move some debug-only code out
of line.
Test: m test-art-host
Change-Id: I16db4aaa803f39e155ce6e1b0778b7e393dcbb17
|
| da1da8a78c9df45890f48b4f0197ed18fa1de3c5 |
|
16-May-2017 |
Richard Uhler <ruhler@google.com> |
Revert "Revert "RegisterNativeAllocation: Avoid case of double blocking gc.""
This reverts commit 602b1c1421028310fdb1662fcec0aa8ac5e77604.
Change-Id: I02b698074845977cc799a0d0234627f7c8ef4eeb
Test: ART_DEFAULT_GC_TYPE=GSS ART_USE_READ_BARRIER=false ./test/testrunner/testrunner.py -b -j32 --target -t 004-NativeAllocations
Bug: 36851903
|
| 602b1c1421028310fdb1662fcec0aa8ac5e77604 |
|
16-May-2017 |
Richard Uhler <ruhler@google.com> |
Revert "RegisterNativeAllocation: Avoid case of double blocking gc."
This reverts commit 8df0f36b4fd5db6da67066da62eccc1e0b81e028.
004-NativeAllocations test fails on art-gss-gc-tlab configuration.
Change-Id: I0fb0969c8e4af0bcd5f8481ce828ac4cf258c089
Bug: 36851903
|
| 8df0f36b4fd5db6da67066da62eccc1e0b81e028 |
|
12-May-2017 |
Richard Uhler <ruhler@google.com> |
RegisterNativeAllocation: Avoid case of double blocking gc.
If multiple threads call RegisterNativeAllocation causing the blocking
watermark to be exceeded while a background GC is in progress, then
two back-to-back blocking GCs would be performed when a single
blocking GC is sufficient. For example:
1. Thread A RegisterNativeAllocation triggers background GC1
2. Thread A RegisterNativeAllocation triggers blocking GC2
3. Thread A's GC2 waits for GC1 to complete.
4. Thread B RegisterNativeAllocation sees GC2 in progress and waits for
it to complete before triggering a second blocking GC3.
5. GC1 completes.
Because thread B's RegisterNativeAllocation was called before GC1
completed and GC2 had not begun for real, thread B can simply wait for
GC2 to start and finish before returning, rather than waiting for GC2 to
start and finish and then doing a second blocking GC.
This change fixes the behavior so that only a single blocking GC is
performed in this case.
Bug: 36851903
Test: 004-NativeAllocations run test.
Change-Id: I1e178b9ee7bb68703bdc9a09b1041a982de8b2ce
|
| 1ca689096b532e007dc9f8ba16db4731e6afd719 |
|
18-Apr-2017 |
Mathieu Chartier <mathieuc@google.com> |
More robust GC verification and corruption dumping
Added a test for GC heap corruption dumping, added more info to the
dump like adjacent bytes and card table.
Added heap corruption detection in
ConcurrentCopying::MarkNonMoving().
Bug: 37187694
Bug: 12687968
Test: mm test-art-host-gtest-verification_test -j20
Change-Id: I8c90e45796d0784265aa091b2f8082f0cfb62719
|
| 35b59a262f660adaf30db11bc0c3d14bb56ebf79 |
|
18-Apr-2017 |
Mathieu Chartier <mathieuc@google.com> |
Differentiate between native alloc and normal background GC
Added a new GC cause kGcCauseForNativeAllocBackground.
Bug: 35872915
Test: test-art-host
Change-Id: I94e17f8bd53af29f2862b9910bd8abd2df97e229
|
| d6b17d4eacd8a1a46a0d4837285735a071c0beb7 |
|
17-Feb-2017 |
Mathieu Chartier <mathieuc@google.com> |
Move thread_running_gc_ logic to VisitObjectsInternalRegionSpace
Cleaner to have the logic in the caller.
Test: test-art-host CC
Change-Id: I93a16f8baf327b6692cce5c6141d1c361ce53f16
|
| 183009a616b9d0d77591651f8e97f48ed4d924db |
|
17-Feb-2017 |
Mathieu Chartier <mathieuc@google.com> |
Change IsMovingGCDisabled to take into account GC critical sections
Fixes CC test 130.
Test: test-art-host
Change-Id: I39a0d6b88a7e651a5fb2320fd34c5a8bdc615345
|
| caaa2b05cf581d5c5fc4253723ddd3299b3c3e25 |
|
01-Feb-2017 |
Richard Uhler <ruhler@google.com> |
Revert "Revert "Redesign implementation of RegisterNativeAllocation.""
This reverts commit 36bdbd2bf2ee36662f700b63474b546a7abecfa3.
Bug: 29156652
Bug: 32576211
Test: 004-NativeAllocations in a loop with high machine load.
Change-Id: I4470222c66aef4e0daa7612c84177b6c35bd28a9
|
| 36bdbd2bf2ee36662f700b63474b546a7abecfa3 |
|
24-Jan-2017 |
Richard Uhler <ruhler@google.com> |
Revert "Redesign implementation of RegisterNativeAllocation."
This reverts commit 6bfc37d875b2556b3f1d95b00a785a09dc94ccb0.
Test failures in 004-NativeAllocations.
Bug: 29156652
Bug: 32576211
Change-Id: Ic54c67caf913024485a5c0621857d68a6fb710fa
|
| 6bfc37d875b2556b3f1d95b00a785a09dc94ccb0 |
|
24-Nov-2016 |
Richard Uhler <ruhler@google.com> |
Redesign implementation of RegisterNativeAllocation.
Improve the performance and remove the issue with GCing apps to death when
they reach the growth_limit_.
This implements the REDESIGN approach described in detail in the
document at go/understanding-register-native-allocation.
Bug: 29156652
Bug: 32576211
Test: m test-art-host
Test: vogar libcore/luni/src/test/java/libcore/util/NativeAllocationRegistryTest.java
Test: Device boots.
Test: adb bugreport and look for "Registered native bytes allocated"
Change-Id: I09176b2b0e4f0c401fe1947830fa5271060f4e61
|
| 5ace201d84adb7753680bf4c7877b3b71558da82 |
|
30-Nov-2016 |
Mathieu Chartier <mathieuc@google.com> |
Revert "Revert CC related changes."
Disable entrypoint switching in ResetQuickAllocEntryPointsForThread
instead of callers. Fixes bug where instrumentation would switch
to non CC entrypoints for non X86_64 architectures causing aborts.
Bug: 31018974
Test: test-art-host
Test: test/run-test 099
This reverts commit 96172e0172c5fca6e9a5ad4b857a24d8c7b064e5.
Change-Id: If206694ae35ff4446c6a8a97bfbcbf2dac35e3f9
|
| 96172e0172c5fca6e9a5ad4b857a24d8c7b064e5 |
|
30-Nov-2016 |
Nicolas Geoffray <ngeoffray@google.com> |
Revert CC related changes.
Revert: "X86_64: Add allocation entrypoint switching for CC is_marking"
Revert: "Fix mips build in InitEntryPoints"
Revert: "Fix mac build in ResetQuickAllocEntryPoints"
Test: test-art-target-run-test
Change-Id: If38d44edf8c5def5c4d8c9419e4af0cd8d3be724
|
| f5de23265360e15fcfceb7d07bdadca0e5bb5f0a |
|
16-Nov-2016 |
Mathieu Chartier <mathieuc@google.com> |
X86_64: Add allocation entrypoint switching for CC is_marking
Only X86_64 done so far. Use normal TLAB allocators if GC is not
marking.
Allocation speed goes up by ~8% based on perf sampling.
Without change:
1.19%: art_quick_alloc_object_region_tlab
With change:
0.63%: art_quick_alloc_object_tlab
0.47%: art_quick_alloc_object_region_tlab
Bug: 31018974
Bug: 12687968
Test: test-art-host-run-test
Change-Id: I4c4d9eb229d4ad2f41b856ba5c2958a5eb3b7ffa
|
| 3049324f4ef71b5d7a3de49bd77c75f07dbf8f3a |
|
03-Nov-2016 |
Hiroshi Yamauchi <yamauchi@google.com> |
Make empty checkpoint work while weak ref access is disabled.
Fix a potential race on PushOntoMarkStack for CC by running an empty
checkpoint (while weak ref access is disabled).
Bug: 32508093
Bug: 12687968
Test: test-art-host with CC/CMS, libartd boot with N9, Ritz EAAC.
Change-Id: I3749bb525e7734804307ee16262355f3fc730312
|
| 9b8c5880de9b0f7dd8b1b863520714a9700a8dc2 |
|
22-Oct-2016 |
Andreas Gampe <agampe@google.com> |
ART: Add GC callbacks
Add start and end of pause callbacks.
Fix setup and missing functions in test 907.
Bug: 31684633
Test: m test-art-host
Test: m test-art-host-run-test-908-gc-start-finish
Change-Id: I1d8872ef9cd6914de7fb033cd873f8d5cb48ea17
|
| 2d8559517759dcfac749a28d6713d36809ec556f |
|
13-Oct-2016 |
Mathieu Chartier <mathieuc@google.com> |
Move Heap::GetInstances to use Handles
Prevent moving GC bugs from VisitObjects. Fix JDWP tests.
Test: art/tools/run-jdwp-tests.sh '--mode=host' '--variant=X32' --debug
Bug: 31113334
Change-Id: Ie7f51f1980ec0c1eddc3b59c3e49564eacb3be85
|
| aea9bfb051bb197fdfb63d56ff0f3d63b36eb499 |
|
13-Oct-2016 |
Mathieu Chartier <mathieuc@google.com> |
Use handles for GetReferringObjects
Required since VisitObjects may cause thread suspension. Fixes
failing jdwp tests.
Test: test-art-host
Test: art/tools/run-jdwp-tests.sh '--mode=host' '--variant=X32' --debug
Bug: 31113334
Change-Id: Ic11780b3778c83e7812bc90c0723a293537f376c
|
| 9d156d500801accee919b6d51e22d6ddcdcd4a05 |
|
07-Oct-2016 |
Mathieu Chartier <mathieuc@google.com> |
Move Heap parameters to ObjPtr
Deleted some unused object dumping code.
Test: test-art-host
Bug: 31113334
Change-Id: I747220caafe6679591fd4b361d7f50383a046164
|
| 27fa96c285c4aa2f64e9bb63a3f38ffdc98c282a |
|
08-Oct-2016 |
Andreas Gampe <agampe@google.com> |
ART: Add allocation callback
Bug: 31684277
Test: m test-art-host
Change-Id: I959f44e23ca5fe55ed678315708895faf0aadb04
|
| a058fdf0cf7435a13a6e8cae5e3a9bfa1513828d |
|
07-Oct-2016 |
Mathieu Chartier <mathieuc@google.com> |
Move mirror::Object setters to ObjPtr
Bug: 31113334
Test: test-art-host
Change-Id: I2c4c84645e194c3c435a4a6fd670176b0e98671f
|
| 0795f23920ee9aabf28e45c63cd592dcccf00216 |
|
28-Sep-2016 |
Mathieu Chartier <mathieuc@google.com> |
Clean up ScopedThreadStateChange to use ObjPtr
Also fixed inclusion of -inl.h files in .h files by adding
scoped_object_access-inl.h and scoped_fast_natvie_object_access-inl.h
Changed AddLocalReference / Decode to use ObjPtr.
Changed libartbenchmark to be debug to avoid linkage errors.
Bug: 31113334
Test: test-art-host
Change-Id: I4d2e160483a29d21e1e0e440585ed328b9811483
|
| 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
|
| 60985b7a56d4fa7170721808734093a3affc41c6 |
|
24-Aug-2016 |
Hiroshi Yamauchi <yamauchi@google.com> |
Background full compaction for CC.
Invoke a full compaction with the CC collector when an app goes to the
background like the HSpace compaction for the CMS collector.
Bug: 31039431
Bug: 12687968
Test: test-art, Ritz EAAC, N9 libartd.so device boot with CC
Change-Id: I119aa26c1d3c167b12983fffcb16164929bf8f68
|
| 8261d02f9523b95013108f271b82bb157ef6f71d |
|
08-Aug-2016 |
Mathieu Chartier <mathieuc@google.com> |
Revert "Revert "ARM64 asm for region space array allocation""
Also added missing large object check. No regression from the check
N6P CC EAAC time at 1313 for 10 samples vs 1314 before reverts.
Bug: 30162165
Bug: 12687968
Test: test-art-target with CC + heap poisoning
This reverts commit 6ae7f3a4541e70f04243a6fe469aa3bd51e16d79.
Change-Id: Ie28f652f619898d7d37eeebf3f31a88af8fac949
|
| 56fe25895e91d34a0a017429468829a20bdd5ae4 |
|
14-Jul-2016 |
Mathieu Chartier <mathieuc@google.com> |
Add a way to measure read barrier slow paths
If enabled, this option counts number of slow paths, measures the
total slow path time per GC and records the info into a histogram.
Also added support for systrace to see which threads are performing
slow paths.
Added runtime option -Xgc:measure to enable. The info is dumped
for SIGQUIT.
Test: Volantis boot with CC, test-art-host with CC, run EEAC with CC
and -Xgc:measure
Bug: 30162165
Change-Id: I3c2bdb4156065249c45695f13c77c0579bc8e57a
|
| 9e03b2fbc9843b834bb9d18bc8c3d2ef44a838d8 |
|
27-May-2016 |
Mathieu Chartier <mathieuc@google.com> |
Reduce TLAB size
Prevent pathological cases like threads fighting for TLABs and doing
back to back GC for alloc each allocation.
No performance change on angler, before:
EEAC: 1161, 1171, 1211
MemAllocTest: 1005, 971, 1148
After:
EEAC: 1173, 1175, 1163
MemAllocTest: 1003, 908, 803
There is also a mysterious speed up on volantis:
EEAC: 1563, 1553, 1543
MemAllocTest: 1474, 1486, 1485
After:
EEAC: 709, 714, 709
MemAllocTest: 664, 684, 655
Bug: 28976163
Change-Id: I564908badf395b6f948aa2dafcb66fe353fc1eba
|
| 5d2a3f7bdd23003af78e3f09e96b9f77ff33c4b6 |
|
11-May-2016 |
Mathieu Chartier <mathieuc@google.com> |
Add histogram for native allocations
Shows up in traces.txt, sample output:
Histogram of native allocation 0:4315,131072:33,1179648:3 bucket size 131072
Histogram of native free 0:995,131072:19,1179648:2 bucket size 131072
Bug: 28680116
(cherry picked from commit 0dce75dc6945c221a054eb9c479fb60efd193719)
Change-Id: Iaa07c6b8da0d6bab64a8fd4af8a02fc0b8c70e9e
|
| 0dce75dc6945c221a054eb9c479fb60efd193719 |
|
11-May-2016 |
Mathieu Chartier <mathieuc@google.com> |
Add histogram for native allocations
Shows up in traces.txt, sample output:
Histogram of native allocation 0:4315,131072:33,1179648:3 bucket size 131072
Histogram of native free 0:995,131072:19,1179648:2 bucket size 131072
Bug: 28680116
Change-Id: I584eb85709d3ae28f5142a01af11075a2a6436d7
|
| 458b105a9cbab6156d72acf3b9372291dba9097a |
|
29-Mar-2016 |
Mathieu Chartier <mathieuc@google.com> |
Obtain stack trace outside of critical section
Fixes deadlock if the stack walk does allocations, changed stack
trace format to prevent slowdown.
Added missing GetInterfaceMethodIfProxy to fix a crash in maps.
Bug: 27857910
(cherry picked from commit 23428587d32361736d4c5e0ba7270c7602695a43)
Change-Id: I64373bcd87a68fdd1b58fb855db2b16c9f6ed36b
|
| 23428587d32361736d4c5e0ba7270c7602695a43 |
|
29-Mar-2016 |
Mathieu Chartier <mathieuc@google.com> |
Obtain stack trace outside of critical section
Fixes deadlock if the stack walk does allocations, changed stack
trace format to prevent slowdown.
Added missing GetInterfaceMethodIfProxy to fix a crash in maps.
Bug: 27857910
Change-Id: Iba86b7390a87349c38785297ac76751417b0fc87
|
| f8cb1781a4e4be5df1f845206f7b37ed89092b64 |
|
19-Mar-2016 |
Mathieu Chartier <mathieuc@google.com> |
Move process state into runtime
Clean up.
Bug: 27420435
(cherry picked from commit f8484c8b55f4b423048f94dfabbe44110a039a9b)
Change-Id: Ia20781ee36e6a31c88ca41d3866b26813cff434d
|
| f8484c8b55f4b423048f94dfabbe44110a039a9b |
|
19-Mar-2016 |
Mathieu Chartier <mathieuc@google.com> |
Move process state into runtime
Clean up.
Bug: 27420435
Change-Id: I8fff84ed1b29a12310094b10fb6382268e69d54b
|
| 14b0a5ddc0ae305e3ac152c34e4d6fdd0abb854c |
|
12-Mar-2016 |
Mathieu Chartier <mathieuc@google.com> |
Fix cases where we miss instrumentation changes
Moved allocation stack push to after we record the allocation since
it can cause thread suspension.
Added handling in entrypoint utils for thread suspension cases.
Keep the AllocRecordObjectMap around since we do not want to delete
it if there are any threads waiting on new_record_condition_. The
condition guards adding stack traces while the GC is running. If we
delete the map and there are still waiters that did not resume, it
caused a CHECK failure. This could happen in cases where one thread
disables allocation tracking while other threads are about to
resume from the condition.
Bug: 27506909
(cherry picked from commit 0f394021d6abfdf9ebea6c7f405ec936a812ea62)
Change-Id: I1dc51e9f000684b4032b57beab59d317ece26f06
|
| 0f394021d6abfdf9ebea6c7f405ec936a812ea62 |
|
12-Mar-2016 |
Mathieu Chartier <mathieuc@google.com> |
Fix cases where we miss instrumentation changes
Moved allocation stack push to after we record the allocation since
it can cause thread suspension.
Added handling in entrypoint utils for thread suspension cases.
Keep the AllocRecordObjectMap around since we do not want to delete
it if there are any threads waiting on new_record_condition_. The
condition guards adding stack traces while the GC is running. If we
delete the map and there are still waiters that did not resume, it
caused a CHECK failure. This could happen in cases where one thread
disables allocation tracking while other threads are about to
resume from the condition.
Bug: 27506909
Change-Id: I097689ca35ad408121c6b8dabd28e75cce1a43a0
|
| 35fd969d663f8f89ecdbdb14e52e4f03e37d3f86 |
|
02-Mar-2016 |
Mathieu Chartier <mathieuc@google.com> |
Delete alloc tracking map outside of critical section
There can be lock order violations otherwise due to runtime shutdown
lock that may get acquired in the condition variable destructor.
Bug: 27506909
(cherry picked from commit 0b8b4a609120b90081d898dbf3c26f68fe80de96)
Change-Id: I6972c450db8856e30c13e27aea94b01943618f64
|
| bcd9dd7d1223eaa60f188d1a109d26e5e0143e10 |
|
07-Mar-2016 |
Mathieu Chartier <mathieuc@google.com> |
Don't use ScopedObjectAccess in Heap::DumpSpaces
ScopedObjectAccess does not work well if the mutator lock is
excusively held while in a suspend thread state. This caused
deadlocks and DCHECK failures.
Bug: 27493316
(cherry picked from commit 03d21bc5bed887243ff6ce3531179185ffd3532c)
Change-Id: I5d67f74fc7082761e45dc1d7778b0ea7fceaaf8f
|
| 03d21bc5bed887243ff6ce3531179185ffd3532c |
|
07-Mar-2016 |
Mathieu Chartier <mathieuc@google.com> |
Don't use ScopedObjectAccess in Heap::DumpSpaces
ScopedObjectAccess does not work well if the mutator lock is
excusively held while in a suspend thread state. This caused
deadlocks and DCHECK failures.
Bug: 27493316
Change-Id: I75ab223b2d15ab4e53753f1046c580228fc1866e
|
| 0b8b4a609120b90081d898dbf3c26f68fe80de96 |
|
02-Mar-2016 |
Mathieu Chartier <mathieuc@google.com> |
Delete alloc tracking map outside of critical section
There can be lock order violations otherwise due to runtime shutdown
lock that may get acquired in the condition variable destructor.
Change-Id: I23cb2dfe241f5cc6c42bf6766e89042cf06069b6
|
| ab4058c14ad800d20138a6c41b999029d342de2c |
|
01-Mar-2016 |
Mathieu Chartier <mathieuc@google.com> |
Fix allocation tracking race
Check if changed from uninstrumented to instrumented during GC for
alloc. If we changed, retry the allocation with kInstrumented = true.
Added stress test.
Bug: 27337759
(cherry picked from commit eebc3af4453f5c1fb5fd80c710cfd49566080d28)
Change-Id: I8fa50975b558199fcf142c8555476053001ace50
|
| eebc3af4453f5c1fb5fd80c710cfd49566080d28 |
|
01-Mar-2016 |
Mathieu Chartier <mathieuc@google.com> |
Fix allocation tracking race
Check if changed from uninstrumented to instrumented during GC for
alloc. If we changed, retry the allocation with kInstrumented = true.
Added stress test.
Bug: 27337759
Change-Id: Iaad7977693c9ed927f779a66b29cd58341a837da
|
| 6ea1a0e2168c8d9b6d97c075c73a72d84080f45b |
|
29-Jan-2016 |
Mingyao Yang <mingyao@google.com> |
AOT compile framework code as non-debuggable
When a debugger attaches, we patch method entry points in framework
code to interpreter bridge. The code will later be jitted as debuggable.
Change-Id: Id148069ccad95e2339ba214742ae3ef4f084f495
|
| 20a0be0299183b54be9643820294e011d3358823 |
|
20-Feb-2016 |
Hiroshi Yamauchi <yamauchi@google.com> |
Fix CC collector thread flip and JNI critical section deadlocks.
There's a bug in the synchronization between the CC collector's thread
flip and JNI critical sections where it incorrectly attempts to make
the synchronization scheme to be writer (GC) preference to avoid
starvation of GC in the presence of frequent JNI critical section
enter/exit. This could cause a deadlock between them if a thread
enters a nested JNI critical section after a thread flip occurs. This
is reproduced in the added test.
The fix is to use a thread local disable counter in addition to the
global counter to detect a nested enter by the same thread and avoid
waiting if nested.
Bug: 19235243
Bug: 12687968
Change-Id: Idf7720a6906c9ea508219935af3727f76680d2d8
|
| a9d82fe8bc6960b565245b920e99107a824ca515 |
|
26-Jan-2016 |
Mathieu Chartier <mathieuc@google.com> |
Use mutator lock to guard adding and removing heap spaces
Too hard to add a new lock since dlmalloc ArtMoreCore requires
looping through the spaces while holding the allocator lock.
Bug: 22858531
Change-Id: Ieac2136da02c766b6795cd604a58798bee37ef2a
|
| 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
|
| 28b1cf779b8c438b01b28a4adfeb22a4a8ebdb12 |
|
16-Jan-2016 |
Mathieu Chartier <mathieuc@google.com> |
Change space base test to use parameters
Still need to convert remaining space tests to use gtest value
parameters.
Bug: 10245322
Change-Id: Iad3183d46f2fff1ec2aee6923c989433885adf81
|
| dcdc85bbd569f0ee66c331b4219c19304a616214 |
|
04-Dec-2015 |
Jeff Hao <jeffhao@google.com> |
Dex2oat support for multiple oat file and image file outputs.
Multiple changes to dex2oat and the runtime to support a --multi-image
option. This generates a separate oat file and image file output for
each dex file input.
Change-Id: Ie1d6f0b8afa8aed5790065b8c2eb177990c60129
|
| 073b16c8429d302d5413e8ffc488b03b8f770780 |
|
10-Nov-2015 |
Mathieu Chartier <mathieuc@google.com> |
Image space cleanup for app images
Removed Heap::GetImageSpace, added Heap::GetBootImageSpace.
Generalized some logic in the class linker for image spaces.
Bug: 22858531
Change-Id: Ib3e12bb061a247e232d3dc93b0d6b35eb3a34d25
|
| aa5168291c46f9b418d989bccf2d8e09338a83e6 |
|
03-Oct-2015 |
Mathieu Chartier <mathieuc@google.com> |
Add exclusion between instrumentation and GC
Instrumentation calls VisitClasses while exclusive holding the
mutator lock. This can cause deadlocks since VisitClasses needs to
decode JNI weak globals. If the system weak slow path flag is set,
then we wait holding locks while exclusive holding the mutator lock.
This causes a deadlock since the GC cannot acquire the mutator lock
to sweep system weaks.
This fixes a deadlock seen in one of the tracing tests.
Change-Id: I580152118e068a70f309dcc19df4144afec835dd
|
| fdbd13c7af91a042eda753e436eeebf0e1937250 |
|
03-Sep-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
Some fixes for the CC collector.
- Remove a DCHECK in DisableMarkingCheckpoint, which caused
occasional (false) failures.
- Check the thread-local GetWeakRefAccessEnabled in boxed lambdas weak
access.
- Add missing BroadcastForNewAllocationRecords and
BroadcastForNewWeakBoxedLambdas. The lack of the former caused
occasional deadlocks in the ddmc test.
- Remove the 'ensure system weaks disallowed' calls, which weren't
useful and dead.
Bug: 12687968
Change-Id: I33850c8d12e6e1a3aed1c2bb18eba263cbab76e8
|
| db00eaf64e94ab1b4289952699c3b81f73ca406a |
|
01-Sep-2015 |
Mathieu Chartier <mathieuc@google.com> |
Add GC coverage test for moving GC
Adds testing coverage for collector transitions and homogeneous space
compaction.
Bug: 10808403
Change-Id: Ia79fecb47c33fc95d940243d6cb1068e9ee9dc9a
|
| 76f55b030d2517d434f227bee2363c3fb760e5c6 |
|
22-Aug-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
Synchronize JNI critical calls with the CC collector thread flip.
JNI critical calls (like GetArrayElementsCritical) would need to block
for the whole GC run to finish if the CC collector GC is ongoing. This
CL changes it so that they don't need to block for the GC run, but
only for the duration of the thread flip operation, which is much
shorter. This is valid due to the to-space invariant.
Bug: 12687968
Bug: 19235243
Change-Id: I1b6b4ae4fa539ddc0ec50b10ae8c8709f2a12fe8
|
| 511683764f2580d877725e20514ca4976c91bbfb |
|
13-Aug-2015 |
Mathieu Chartier <mathieuc@google.com> |
Wait for GC to finish in ThreadList::~ThreadList
Added a DisableGC to the heap instead of using the runtime shutdown
boolean. The runtime shutting down boolean is set to true before the
non daemon threads have exited. These threads may still be doing
allocations and we don't want to throw OOME if unnecessary.
Bug: 18577101
Change-Id: Iceb7048e6bd799aa2716099459c54f8dc0fb8feb
|
| e8a69447d845cc5c0b5ccfa9c4d4b1b588271f52 |
|
12-Aug-2015 |
Mathieu Chartier <mathieuc@google.com> |
Delete AllocationTimer
Not very useful, running benchmarks is better.
Bug: 19534862
Change-Id: I725830274e306621bffdabffdee602a03daf016f
|
| 3887c468d731420e929e6ad3acf190d5431e94fc |
|
12-Aug-2015 |
Roland Levillain <rpl@google.com> |
Remove unnecessary `explicit` qualifiers on constructors.
Change-Id: Id12e392ad50f66a6e2251a68662b7959315dc567
|
| a4f6af9b1e6380b31674d7ac645b1732c846ac06 |
|
12-Aug-2015 |
Mathieu Chartier <mathieuc@google.com> |
Some heap cleanup
Bug: 19534862
Change-Id: Ia63f489d26ec8813a263ce877bdbbc8c4e8fe5f4
|
| 4e2cb098017bf073335ebb02b1bc0a36828cd720 |
|
23-Jul-2015 |
Mathieu Chartier <mathieuc@google.com> |
Add uninterruptible role
Example error:
cannot call function 'CopyOf' while mutex 'uninterruptible_' is held
TODO: Add annotation to more locations.
Bug: 20072211
Change-Id: I1bbf5a77e3deeafa5898df529cb7cb53a6d010d2
|
| 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
|
| 97509954404d031594b2ecbda607314d169d512e |
|
13-Jul-2015 |
Mathieu Chartier <mathieuc@google.com> |
Clean up GC callbacks to be virtual methods
Change-Id: Ia08034a4e5931c4fcb329c3bd3c4b1f301135735
|
| 1e13374baf7dfaf442ffbf9809c37c131d681eaf |
|
20-May-2015 |
Evgenii Stepanov <eugenis@google.com> |
Generalize Valgrind annotations in ART to support ASan.
Also add redzones around non-fixed mem_map(s).
Also extend -Wframe-larger-than limit to enable arm64 ASan build.
Change-Id: Ie572481a25fead59fc8978d2c317a33ac418516c
|
| 3cf225386e8129dcbe32b289279ecb87ec255318 |
|
10-Jul-2015 |
Mathieu Chartier <mathieuc@google.com> |
Clean up some includes
Change-Id: Ia03a3b54b235df38d5cfe096fef1aebe2b80eb29
|
| 41656de4b6440bcd1bea5fb785f0710153b2bc68 |
|
07-Jul-2015 |
Man Cao <manc@google.com> |
Fix potential bugs in allocation tracker
Add a missing visit allocation records call in concurrent copying collecter.
Handle null class objects if we support class unloading, and issues
with disallow and allow new allocation records.
Bug: 20037135
Change-Id: I59b7321c281e0d79a620501b2f43e36d2a576203
|
| 42c3c33df8b6eefc4ba532f1981282510f109928 |
|
24-Jun-2015 |
Man Cao <manc@google.com> |
Make allocation tracker use less memory
The allocation tracker no longer keeps recently allocated objects live.
Instead it just keeps their class objects live as strong roots. This fixed
the gc-stress test failure for 098-ddmc.
Also fixed the issue in DisableNewSystemWeak() for allocation tracker,
by making new allocation to wait until GC's sweeping to complete. I didn't
feel any significant slowdown with this wait.
Bug: 20037135
Change-Id: I6a98188832cf7ee478007e3788e742dc6e18f7b8
|
| 1ed11b9ad5512cf464cb1686640df53201fa5297 |
|
12-Jun-2015 |
Man Cao <manc@google.com> |
Restore DDMS recent allocation tracking's behavior
Make the allocation tracker mark recently allocated objects as roots,
so the allocation records are not swept. Because DDMS needs recent
allocation tracking records even for dead objects. This should fix
the flaky failures for 098-ddmc test, but it cannot pass GC stress
test (OOM). Re-enabled 098-ddmc for other tests.
There should be an option to not mark them as roots, when user only
needs HPROF dump with traces but not DDMS's recent allocation tracking.
Probably need to add a new JNI API function for this option.
There could be another way to keep a second list of recent allocation
records and maintain a type cache for them, so not to make the objects
roots. But it's more complex, and not sure which is better.
Also reduce memory usage for AllocRecordStackTrace objects, and change
default stack depth to 16. Rename the property that controls the stack
depth to "debug.allocTracker.maxStackDepth" so developer can change it.
Bug:20037135
Change-Id: Ic6b9ae87bdcda558be6f14ded8057e763439881c
|
| bad9c7b7f88689133cae59d9ccae231822a2020c |
|
14-Jun-2015 |
Mathieu Chartier <mathieuc@google.com> |
New experimental GC stress mode
Tries to do a GC for every unique call stack (up to 16 frames).
The goal is to catch moving GC bugs and lock violations without being
rediculously slow. Some tests fail on 64 bits, 32 bit host doesn't
work. N5 is booting.
Added runtime -Xgc options: gcstress and nogcstress.
Bug: 21664466
(cherry picked from commit 310008008c90fea246efd00cb99ee7ded97c5209)
Change-Id: Icb8e420f2048e8ee83bcca7937563166a2638f5c
|
| 310008008c90fea246efd00cb99ee7ded97c5209 |
|
14-Jun-2015 |
Mathieu Chartier <mathieuc@google.com> |
New experimental GC stress mode
Tries to do a GC for every unique call stack (up to 16 frames).
The goal is to catch moving GC bugs and lock violations without being
rediculously slow. Some tests fail on 64 bits, 32 bit host doesn't
work. N5 is booting.
Added runtime -Xgc options: gcstress and nogcstress.
Bug: 21664466
Change-Id: Icb8e420f2048e8ee83bcca7937563166a2638f5c
|
| 37670178497df18ae6b2a97a2ea9bf365621f16e |
|
11-Jun-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
Reset GC performance stats at zygote fork.
So GCs before a zygote fork won't be attributed to an app.
(cherry pick commit Ib37bc587e0f039ef8faeabe63dec19de49501863)
Bug: 21491908
Change-Id: Ib37bc587e0f039ef8faeabe63dec19de49501863
|
| 9a7fffb36bad0bf3e7bac21dd223d975bc9dde55 |
|
11-Jun-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
Reset GC performance stats at zygote fork.
So GCs before a zygote fork won't be attributed to an app.
Bug: 21491908
Change-Id: Ib37bc587e0f039ef8faeabe63dec19de49501863
|
| 8c2ff641294715864013737fdec57cdfd410270c |
|
28-May-2015 |
Man Cao <manc@google.com> |
Add allocation stack traces for HPROF dump.
This feature is currently only enabled when DDMS's allocation tracking
is enabled. In the future there should be a way to enable this feature
before an application starts.
Also updates DDMS's recent allocation tracking to use a new backend
data structure that is shared with this feature.
The following system properties controls customizable parameters:
dalvik.vm.allocTrackerMax: max number of objects that have allocation
records, default 512K;
dalvik.vm.recentAllocMax: max number of records that are sent to DDMS
when clicking "Get allocation" button,
default 64K-1 (limit of the protocol);
dalvik.vm.allocStackDepth: max number of stack frames in an allocation
record, default 4.
Bug: 20037135
Change-Id: I26ed378a5613678bd3c43e846025f90470a8e059
|
| b5de3bb65556950d5cce1dea3ea6d80adcc5051d |
|
05-Jun-2015 |
Mathieu Chartier <mathieuc@google.com> |
Use runFinalizationWithTimeout for native allocations
Prevents deadlocks by not waiting longer than 250ms for finalizers
to complete.
(cherry picked from commit 3b532d744034b43ed329a3198f15846d80fec3f0)
Bug: 21544853
Change-Id: I57b2f7ae8b74185922eb3c15ba0ab71a4d2348aa
|
| 3b532d744034b43ed329a3198f15846d80fec3f0 |
|
05-Jun-2015 |
Mathieu Chartier <mathieuc@google.com> |
Use runFinalizationWithTimeout for native allocations
Prevents deadlocks by not waiting longer than 250ms for finalizers
to complete.
Bug: 21544853
Change-Id: I57b2f7ae8b74185922eb3c15ba0ab71a4d2348aa
|
| 41b175aba41c9365a1c53b8a1afbd17129c87c14 |
|
19-May-2015 |
Vladimir Marko <vmarko@google.com> |
ART: Clean up arm64 kNumberOfXRegisters usage.
Avoid undefined behavior for arm64 stemming from 1u << 32 in
loops with upper bound kNumberOfXRegisters.
Create iterators for enumerating bits in an integer either
from high to low or from low to high and use them for
<arch>Context::FillCalleeSaves() on all architectures.
Refactor runtime/utils.{h,cc} by moving all bit-fiddling
functions to runtime/base/bit_utils.{h,cc} (together with
the new bit iterators) and all time-related functions to
runtime/base/time_utils.{h,cc}. Improve test coverage and
fix some corner cases for the bit-fiddling functions.
Bug: 13925192
(cherry picked from commit 80afd02024d20e60b197d3adfbb43cc303cf29e0)
Change-Id: I905257a21de90b5860ebe1e39563758f721eab82
|
| 80afd02024d20e60b197d3adfbb43cc303cf29e0 |
|
19-May-2015 |
Vladimir Marko <vmarko@google.com> |
ART: Clean up arm64 kNumberOfXRegisters usage.
Avoid undefined behavior for arm64 stemming from 1u << 32 in
loops with upper bound kNumberOfXRegisters.
Create iterators for enumerating bits in an integer either
from high to low or from low to high and use them for
<arch>Context::FillCalleeSaves() on all architectures.
Refactor runtime/utils.{h,cc} by moving all bit-fiddling
functions to runtime/base/bit_utils.{h,cc} (together with
the new bit iterators) and all time-related functions to
runtime/base/time_utils.{h,cc}. Improve test coverage and
fix some corner cases for the bit-fiddling functions.
Bug: 13925192
Change-Id: I704884dab15b41ecf7a1c47d397ab1c3fc7ee0f7
|
| 414369a2e3f23e1408fc1cbf4f623014bd95cb8f |
|
04-May-2015 |
Mathieu Chartier <mathieuc@google.com> |
Add some more DISALLOW_COPY_AND_ASSIGN
May help prevent bugs maybe.
(cherry picked from commit 3130cdf29eb203be0c38d1107a65d920ec39c106)
Change-Id: Ie73d469dfcd078492ecb3aa28682b42707221202
|
| 3130cdf29eb203be0c38d1107a65d920ec39c106 |
|
04-May-2015 |
Mathieu Chartier <mathieuc@google.com> |
Add some more DISALLOW_COPY_AND_ASSIGN
May help prevent bugs maybe.
Change-Id: Ie73d469dfcd078492ecb3aa28682b42707221202
|
| d45c84364db585f2e45a6264ae3116d004858677 |
|
02-May-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
Request full/partial GC type from VMRuntime.requestConcurrentGC().
(cherry pick commit 0ae98992d348b5bf3fc85a52d81b0e5f221f20d6)
Bug: 20554807
Change-Id: I572e846e9d0347b8b2692442e55f56c08b07d395
|
| 0ae98992d348b5bf3fc85a52d81b0e5f221f20d6 |
|
01-May-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
Request full/partial GC type from VMRuntime.requestConcurrentGC().
Bug: 20554807
Change-Id: I1cd855507c3f4e191035dbc2c429e1b8b344d81b
|
| 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
|
| 10d6886c9ce3ed87431cf10d376a69c23950fa61 |
|
15-Apr-2015 |
Mathieu Chartier <mathieuc@google.com> |
Disable parallel GC by default
Not using parallel GC seems to reduce avg pauses by ~0.1s on
EvaluateAndApplyChanges. Avoiding creating the thread pool should
help app launch slightly and reduce memory ussage.
Change-Id: Iebec2a17701c76e4145b41d7c0b4f6dd17806efa
|
| a1c9f013c034fbddb9337cc5c7ecf0e5a8b77547 |
|
02-Apr-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
getRuntimeStat() support (ART).
Export some runtime stats (currently GC stats) via
VMDebug.getRuntimeStat().
Added several new GC stats such as blocking GC counts and GC count
histograms.
Bug: 19825248
Change-Id: I8ece9ed241dc3982dfd983d7159090ba82940dce
|
| ddac42329314587f6f188bacf101b3cb15175b3c |
|
02-Apr-2015 |
Mathieu Chartier <mathieuc@google.com> |
Fix race with Heap::ClampGrowthLimit and GC
Aded logic for handling the temp bitmap if the GC is running and
the live bitmap is clamped to the mark bitmap. This fixes the SIGABRT
from ClampGrowthLimit if the GC clamped the bitmaps at this point.
Also added locking of the heap_bitmap_lock_ so that added a lock so
that the temp bitmap doesn't change from underneath us.
Bug: 20043461
Change-Id: Ib427e40bcdf149de0408b4b53e6524f51463f0af
|
| 4460a84be92b5a94ecfb5c650aef4945ab849c93 |
|
09-Mar-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
Rosalloc thread local allocation path without a cas.
Speedup on N4:
MemAllocTest 3044 -> 2396 (~21% reduction)
BinaryTrees 4101 -> 2929 (~26% reduction)
Bug: 9986565
Change-Id: Ia1d1a37b9e001f903c3c056e8ec68fc8c623a78b
|
| dfe3083054aa7be60adc24e3e58e0ae4773c41ff |
|
07-Mar-2015 |
Mathieu Chartier <mathieuc@google.com> |
Only allocate collectors if we may use them
I believe most of the savings come from not allocating the concurrent
copying 8MB MarkQueue.
Before AOSP hh native PSS:
72653 kB: Native
72998 kB: Native
72882 kB: Native
After AOSP hh native PSS:
64823 kB: Native
65986 kB: Native
64219 kB: Native
Bug: 17643507
Change-Id: Ic5a8e753beca36142c1bf36be6311051f7c78e47
|
| e4d99873eb1dbee956d76c1edef489103898b33f |
|
26-Feb-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
Add standard gc log for transition collections.
Factor out the gc log logic into a function.
Note we still don't log in background as before but do for things like
OOME compactions in foreground.
Change-Id: Ifd35a796cf26aba2fba83bf07e72902357d91f8b
|
| 446ba4baa35f9eed9ee598438f25fdf02e4a582c |
|
05-Feb-2015 |
Igor Murashkin <iam@google.com> |
Fix broken volantis build
Change-Id: Iedc1a94fd1ec4224b605b984c11cfcc8c1c510e3
|
| aaebaa0121be3b9d9f13630585304482cbcaeb4b |
|
26-Jan-2015 |
Igor Murashkin <iam@google.com> |
art: Refactor RuntimeOptions/ParsedOptions
Refactor the RuntimeOptions to be a
type-safe map (VariantMap, see runtime_options.h) and the ParsedOptions
to delegate the parsing to CmdlineParser (see cmdline/cmdline_parser.h).
This is the start of a command line parsing refactor, and may include
more in the future (dex2oat, patchoat, etc).
For more details of the command line parsing generator usage see cmdline/README.md
Change-Id: Ic67c6bca5e1f33bf2ec60e2e3ff8c366bab91563
|
| 4add3b4fa38ec42bb3c71d01cf70bce8e9a9fb4e |
|
15-Jan-2015 |
Lei Li <lei.l.li@intel.com> |
ART: clear dirty cards of alloc space for MS/CMS partial and full GCs
For MS/CMS partial and full GCs, we could clear the dirty cards of alloc
space when we process cards as we care about the dirty cards after GC
starts.
Change-Id: I1f9b32b20d75979387bc5d26b0cf9a256dcf20b6
Signed-off-by: Lei Li <lei.l.li@intel.com>
|
| cb535da36915f9d10bec3880b46f1de1f7a69f22 |
|
23-Jan-2015 |
Mathieu Chartier <mathieuc@google.com> |
Change AtomicStack to use StackReference
Previously used Object*, using StackReference saves memory on 64 bit
devices.
Bug: 12935052
Bug: 17643507
Change-Id: I035878690054eeeb24d655a900b8f26c837703ff
|
| 2cd334ae2d4287216523882f0d298cf3901b7ab1 |
|
09-Jan-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
More of the concurrent copying collector.
Bug: 12687968
Change-Id: I62f70274d47df6d6cab714df95c518b750ce3105
|
| 0c8c303c20cdaaf54d26e45cc17dc5afb820d8ef |
|
17-Jan-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
Clean up the locks around Heap::VisitObjects().
This is so that we could support suspending all threads when visiting
objects in the presence of a concurrent, moving collector.
Bug: 12687968
Change-Id: Icc8e60630465afde948ebc6ea91d4ebaff5d7837
|
| 68d1bfc0bb5a7be79354f6551dae6cd600c20e76 |
|
08-Jan-2015 |
Mathieu Chartier <mathieuc@google.com> |
Add clamp growth limit
Clamp growth limit shrinks the space memmaps to the current growth
limit. This reduces virtual memory usage for apps with small heaps.
Bug: 18387825
Bug: 17131630
(cherry picked from commit 379d09fe3c3feb7c2a2fb5a3623689b5ace7e79b)
Change-Id: I4f8e507fde1c8e10373615254aa0f1bf1b48a7ea
|
| 379d09fe3c3feb7c2a2fb5a3623689b5ace7e79b |
|
08-Jan-2015 |
Mathieu Chartier <mathieuc@google.com> |
Add clamp growth limit
Clamp growth limit shrinks the space memmaps to the current growth
limit. This reduces virtual memory usage for apps with small heaps.
Bug: 18387825
Bug: 17131630
Change-Id: I4a8fdc335d2c40492e991708adabcc46299efb7d
|
| a5eae69589ff562ad66c57665882cd16f237321c |
|
18-Dec-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add heap task processor
The heap task processor processes async tasks which may be delayed.
The motivation for this change is preventing deadlocks which
can occur when the daemon threads get suspended by GetThreadStack.
Other improvements, reduces daemon thread count by one.
Cleaner pending transition VS heap trimming logic.
Bug: 18739541
Change-Id: Idab52b2d9661a6385cada74b93ff297ddc55fc78
|
| e2c2f6e61df0bfa1516bfc8943e78ea610d4d878 |
|
17-Dec-2014 |
Mathieu Chartier <mathieuc@google.com> |
Fix concurrent GC ergonomics
Fixed a race with the gc_request_pending_ boolean which would cause
two concurrent GCs to start in a row in most cases. This broke sticky
CMS ergonomics since the second GC was a sticky CMS which started way
too early resulting in low throughput. Since the throughput was low,
it switch to partial / full for the next iteration.
The race happened as follows, allocating thread would request
concurrent GC which woke up the daemon thread. The daemon thread
cleared the gc_request_pending_ boolean, but before we set the
concurrent_start_bytes_ to max in to prevent more request, the
allocating thread would call RequestConcurrentGC again. This caused
the next WaitForConcurrentGCRequest to return right away and a
concurrent GC to occur earlier than necessary.
Changed the allocation rate ergonomics to use allocation rate
during the GC instead of allocation rate inbetween GCs, this is
better since the allocation rate may become slower if the GC steals
mutator time, resulting in concurrent GCs starting a bit earlier
than they need to.
Fixed a bug in GrowForUtilization where we didn't use the adjusted
max_free when we shrank down the heap, this caused the sticky CMS to
occasionally shrink the heap more than necessary.
EvaluateAndApplyChanges:
Before: ~12.6s GC time
After: ~7.75s GC time
Change-Id: I354bc825b3c44ccfbfe867af0d437b17fe1fe022
|
| 079101a17575114622f6e1d5be5c9ba643630e9a |
|
15-Dec-2014 |
Mathieu Chartier <mathieuc@google.com> |
Move GC daemon locking logic into heap
Fixes deadlock caused by acquirng the mutator lock while
synchronizing on the daemon thread.
Bug: 18739541
Change-Id: I925b8f0f3b58178da6eff17b9c073f655c39597b
|
| 446f9ee5031cf89b8964e29eba2c9f10a4d4aaf1 |
|
02-Dec-2014 |
Mathieu Chartier <mathieuc@google.com> |
Try normal allocation if large object allocation fails
If a large object allocation fails, we now try the normal allocators.
Bug: 18124612
(cherry picked from commit f1c4d0e3a27e9b39916750147ecdea1418fcc231)
Change-Id: Ib83ebe53fbdd83aa2d23fd10a8bb10e149f8918f
|
| f1c4d0e3a27e9b39916750147ecdea1418fcc231 |
|
02-Dec-2014 |
Mathieu Chartier <mathieuc@google.com> |
Try normal allocation if large object allocation fails
If a large object allocation fails, we now try the normal allocators.
Bug: 18504942
Change-Id: I18b9759d6af885556941542c57fec584f18197f1
|
| d582fa4ea62083a7598dded5b82dc2198b3daac7 |
|
06-Nov-2014 |
Ian Rogers <irogers@google.com> |
Instruction set features for ARM64, MIPS and X86.
Also, refactor how feature strings are handled so they are additive or
subtractive.
Make MIPS have features for FPU 32-bit and MIPS v2. Use in the quick compiler
rather than #ifdefs that wouldn't have worked in cross-compilation.
Add SIMD features for x86/x86-64 proposed in:
https://android-review.googlesource.com/#/c/112370/
Bug: 18056890
Change-Id: Ic88ff84a714926bd277beb74a430c5c7d5ed7666
|
| 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
|
| 2d2621a1463d2f3f03fa73503fa42e43657cdcfc |
|
24-Oct-2014 |
Mathieu Chartier <mathieuc@google.com> |
Optimize method linking
Added more inlining, removed imt array allocation and replaced it
with a handle scope. Removed some un-necessary handle scopes.
Added logic to base interface method tables from the superclass so
that we dont need to reconstruct for every interface (large win).
Facebook launch Dalvik KK MR2:
TotalTime: 3165
TotalTime: 3652
TotalTime: 3143
TotalTime: 3298
TotalTime: 3212
TotalTime: 3211
Facebook launch TOT before:
WaitTime: 3702
WaitTime: 3616
WaitTime: 3616
WaitTime: 3687
WaitTime: 3742
WaitTime: 3767
After optimizations:
WaitTime: 2903
WaitTime: 2953
WaitTime: 2918
WaitTime: 2940
WaitTime: 2879
WaitTime: 2792
LinkInterfaceMethods no longer one of the hottest methods, new list:
4.73% art::ClassLinker::LinkVirtualMethods(art::Thread*, art::Handle<art::mirror::Class>)
3.07% art::DexFile::FindClassDef(char const*) const
2.94% art::mirror::Class::FindDeclaredStaticField(art::mirror::DexCache const*, unsigned int)
2.90% art::DexFile::FindStringId(char const*) const
Bug: 18054905
Bug: 16828525
(cherry picked from commit 1fb463e42cf1d67595cff66d19c0f99e3046f4c4)
Change-Id: I27cc70178fd3655fbe5a3178887fcba189d21321
|
| 1fb463e42cf1d67595cff66d19c0f99e3046f4c4 |
|
24-Oct-2014 |
Mathieu Chartier <mathieuc@google.com> |
Optimize method linking
Added more inlining, removed imt array allocation and replaced it
with a handle scope. Removed some un-necessary handle scopes.
Added logic to base interface method tables from the superclass so
that we dont need to reconstruct for every interface (large win).
Facebook launch Dalvik KK MR2:
TotalTime: 3165
TotalTime: 3652
TotalTime: 3143
TotalTime: 3298
TotalTime: 3212
TotalTime: 3211
Facebook launch TOT before:
WaitTime: 3702
WaitTime: 3616
WaitTime: 3616
WaitTime: 3687
WaitTime: 3742
WaitTime: 3767
After optimizations:
WaitTime: 2903
WaitTime: 2953
WaitTime: 2918
WaitTime: 2940
WaitTime: 2879
WaitTime: 2792
LinkInterfaceMethods no longer one of the hottest methods, new list:
4.73% art::ClassLinker::LinkVirtualMethods(art::Thread*, art::Handle<art::mirror::Class>)
3.07% art::DexFile::FindClassDef(char const*) const
2.94% art::mirror::Class::FindDeclaredStaticField(art::mirror::DexCache const*, unsigned int)
2.90% art::DexFile::FindStringId(char const*) const
Bug: 18054905
Bug: 16828525
Change-Id: I27cc70178fd3655fbe5a3178887fcba189d21321
|
| cf7f19135f0e273f7b0136315633c2abfc715343 |
|
23-Oct-2014 |
Ian Rogers <irogers@google.com> |
C++11 related clean-up of DISALLOW_..
Move DISALLOW_COPY_AND_ASSIGN to delete functions. By no having declarations
with no definitions this prompts better warning messages so deal with these
by correcting the code.
Add a DISALLOW_ALLOCATION and use for ValueObject and mirror::Object.
Make X86 assembly operand types ValueObjects to fix compilation errors.
Tidy the use of iostream and ostream.
Avoid making cutils a dependency via mutex-inl.h for tests that link against
libart. Push tracing dependencies into appropriate files and mutex.cc.
x86 32-bit host symbols size is increased for libarttest, avoid copying this
in run-test 115 by using symlinks and remove this test's higher than normal
ulimit.
Fix the RunningOnValgrind test in RosAllocSpace to not use GetHeap as it
returns NULL when the heap is under construction by Runtime.
Change-Id: Ia246f7ac0c11f73072b30d70566a196e9b78472b
|
| 677cd61ad05d993c4d3b22656675874f06d6aabc |
|
15-Oct-2014 |
Ian Rogers <irogers@google.com> |
Make ART compile with GCC -O0 again.
Tidy up InstructionSetFeatures so that it has a type hierarchy dependent on
architecture.
Add to instruction_set_test to warn when InstructionSetFeatures don't agree
with ones from system properties, AT_HWCAP and /proc/cpuinfo.
Clean-up class linker entry point logic to not return entry points but to
test whether the passed code is the particular entrypoint. This works around
image trampolines that replicate entrypoints.
Bug: 17993736
(cherry picked from commit 6f3dbbadf4ce66982eb3d400e0a74cb73eb034f3)
Change-Id: I3e7595f437db4828072589d475a5453b7f31003e
|
| 6f3dbbadf4ce66982eb3d400e0a74cb73eb034f3 |
|
15-Oct-2014 |
Ian Rogers <irogers@google.com> |
Make ART compile with GCC -O0 again.
Tidy up InstructionSetFeatures so that it has a type hierarchy dependent on
architecture.
Add to instruction_set_test to warn when InstructionSetFeatures don't agree
with ones from system properties, AT_HWCAP and /proc/cpuinfo.
Clean-up class linker entry point logic to not return entry points but to
test whether the passed code is the particular entrypoint. This works around
image trampolines that replicate entrypoints.
Bug: 17993736
Change-Id: I5f4b49e88c3b02a79f9bee04f83395146ed7be23
|
| 13735955f39b3b304c37d2b2840663c131262c18 |
|
08-Oct-2014 |
Ian Rogers <irogers@google.com> |
stdint types all the way!
Change-Id: I4e4ef3a2002fc59ebd9097087f150eaf3f2a7e08
|
| 2dbe627954fd78a3659ab3cd42d2ead5b4529441 |
|
16-Sep-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add runtime options for large object space
Adds the two following options:
Option -XX:LargeObjectSpace={freelist, map, disabled}
Option -XX:LargeObjectThreshold=size specifies the size for which
primitive arrays are allocated in the large object space.
Added handling for large object space == null in collectors.
Fixed an error in the mem map space where we didn't use the page
aligned size LOS for bytes allocated, this resutled in heaps appearing
a bit smaller than they should be.
Change-Id: I6f17c3534b59e7dc68cd375153e7a846799b3da4
|
| 7b078e8c04f3e1451dbdd18543c8b9692b5b067e |
|
10-Sep-2014 |
Ian Rogers <irogers@google.com> |
Compile time performance improvements focusing on interpret-only.
Reduce virtual method dispatch in the method verifier and make more code
inline-able.
Add a StringPiece with const char* equality operator to avoid redundant
StringPieces and strlens.
Remove back link from register line to verifier and pass as argument to reduce
size of RegisterLine.
Remove instruction length from instruction flags and compute from the
instruction, again to reduce size.
Add suspend checks to resolve and verify to allow for more easy monitor
inflation and reduce contention on Locks::thread_list_suspend_thread_lock_.
Change ThrowEarlierClassFailure to throw pre-allocated exception.
Avoid calls to Thread::Current() by passing self.
Template specialize IsValidClassName.
Make ANR reporting with SIGQUIT run using checkpoints rather than suspending
all threads. This makes the stack/lock analysis less lock error prone.
Extra Barrier assertions and condition variable time out is now returned as a
boolean both from Barrier and ConditionVariable::Wait.
2 threaded host x86-64 interpret-only numbers from 341 samples:
Before change: Avg 176.137ms 99% CI 3.468ms to 1060.770ms
After change: Avg 139.163% 99% CI 3.027ms to 838.257ms
Reduction in average compile time after change is 20.9%.
Slow-down without change is 26.5%.
Bug: 17471626 - Fix bug where RegTypeCache::JavaLangObject/String/Class/Throwable
could return unresolved type when class loading is disabled.
Bug: 17398101
Change-Id: Id59ce3cc520701c6ecf612f7152498107bc40684
|
| af4edbd2b77fef0eee3827eb30cc1a0e8e6efeb1 |
|
09-Sep-2014 |
Mathieu Chartier <mathieuc@google.com> |
Fix and re-enable FreeList large object space for 64 bit
Not enabled on 32 bit due to virtual memory fragmentation concerns.
The new free list large object space ensures that allocations are
page aligned by using a side table for accounting data.
(cherry picked from commit 66e222aa48e6d2fe4c78a1df938364b82bc83e72)
Change-Id: Idbcbe75cb86b6d9b3d8b20f3048631a48c511458
|
| 66e222aa48e6d2fe4c78a1df938364b82bc83e72 |
|
09-Sep-2014 |
Mathieu Chartier <mathieuc@google.com> |
Fix and re-enable FreeList large object space for 64 bit
Not enabled on 32 bit due to virtual memory fragmentation concerns.
The new free list large object space ensures that allocations are
page aligned by using a side table for accounting data.
Bug: 17414549
Change-Id: Idbcbe75cb86b6d9b3d8b20f3048631a48c511458
|
| eb10cfe7849236ca9eb5a6d05ac0d06be05f19bf |
|
09-Sep-2014 |
Mathieu Chartier <mathieuc@google.com> |
Temporarily disable LOS on 64 bit
Problem is that MAP_32BIT simulation can take seconds on large object
allocations in the worst case.
Bug: 17414549
Change-Id: I463bb0b728eed14847a32a27a93640ec6080f6b3
|
| b27285583ca40253b76010948420a91b45f79257 |
|
08-Sep-2014 |
Mathieu Chartier <mathieuc@google.com> |
Revert "Fix heap trimmer daemon sleeping."
This reverts commit 698b44634396e8111e96e1116586fde89a838347.
(cherry picked from commit 4d3981c72a546b87c8f5d391ff8b5135b8ca99a7)
Change-Id: I7245bc08ebb95c12e7cc71b97edf9ca11ffbdf61
|
| 4d3981c72a546b87c8f5d391ff8b5135b8ca99a7 |
|
08-Sep-2014 |
Mathieu Chartier <mathieuc@google.com> |
Revert "Fix heap trimmer daemon sleeping."
This reverts commit 698b44634396e8111e96e1116586fde89a838347.
Change-Id: Ifa8c2d554b82289cc31803ad6027735d2c3c9be8
|
| 23d683ff957ed9967a045ad3da61f4a4e3c99f2a |
|
04-Sep-2014 |
Mathieu Chartier <mathieuc@google.com> |
Fix heap trimmer daemon sleeping.
Problem:
Heap trimmer daemon sleeping caused app launch occasionally to take a
while due stopping the heap trimming thread taking a long time.
The heap trimmer thread now never sleeps, we prevent issues caused
by frequent back and forth process state changes by only
transitioning to background if it has been kCollectorTransitionWait
time since the last transition to background. Similar logic for heap
trimming.
(cherry picked from commit 698b44634396e8111e96e1116586fde89a838347)
Change-Id: I5b2894e92eb79351eec35078d49c8457d2fd67a8
|
| 698b44634396e8111e96e1116586fde89a838347 |
|
04-Sep-2014 |
Mathieu Chartier <mathieuc@google.com> |
Fix heap trimmer daemon sleeping.
Problem:
Heap trimmer daemon sleeping caused app launch occasionally to take a
while due stopping the heap trimming thread taking a long time.
The heap trimmer thread now never sleeps, we prevent issues caused
by frequent back and forth process state changes by only
transitioning to background if it has been kCollectorTransitionWait
time since the last transition to background. Similar logic for heap
trimming.
Bug: 17310019
Change-Id: I23980421cc388f36b66a4c03ed15dd11d43f59b5
|
| 8ec31f9835a8d1513be9be7f9fa236fe4715750d |
|
03-Sep-2014 |
Mathieu Chartier <mathieuc@google.com> |
Fix native allocation watermark clamping.
The main issue causing the test to fail is that
native_footprint_gc_watermark_ becoming > growth_limit_ due to no
clamping.
Temporary runFinalization fix is calling runFinalization 2x.
Bug: 17371542
(cherry picked from commit 4c7fc5950853b0c368e2148db77ced7c4d3c303c)
Change-Id: I05b85e95560c32c33d53bc96abf87d5262007395
|
| 4c7fc5950853b0c368e2148db77ced7c4d3c303c |
|
03-Sep-2014 |
Mathieu Chartier <mathieuc@google.com> |
Fix native allocation watermark clamping.
The main issue causing the test to fail is that
native_footprint_gc_watermark_ becoming > growth_limit_ due to no
clamping.
Temporary runFinalization fix is calling runFinalization 2x.
Bug: 17371542
Change-Id: I188cb530a44dd109e066a22091f12f8d2d4350c3
|
| 08487454218c2bfee6a9d669622c9ed935408783 |
|
03-Sep-2014 |
Mathieu Chartier <mathieuc@google.com> |
Change native allocations to use growth limit.
Previously native allocation tracking used a GC footprint limit
which would cause GC in the allocating thread. This prevented
excessive growth of the heap but could cause jank due to GC in
the allocating thread. The new behavior is using the growth_limit
instead of the native footprint limit.
(cherry picked from commit d9819ecc0bc21a2bb356a4de9b013e36fe618627)
Change-Id: Ia40ed830e8c674cc49d4c0a6fd773d6cb8ff97fa
|
| d9819ecc0bc21a2bb356a4de9b013e36fe618627 |
|
03-Sep-2014 |
Mathieu Chartier <mathieuc@google.com> |
Change native allocations to use growth limit.
Previously native allocation tracking used a GC footprint limit
which would cause GC in the allocating thread. This prevented
excessive growth of the heap but could cause jank due to GC in
the allocating thread. The new behavior is using the growth_limit
instead of the native footprint limit.
Bug: 17006948
Change-Id: I40f30af09bb25596a9f57fa50e2a155fb947b5fe
|
| bad0267eaab9d6a522d05469ff90501deefdb88b |
|
25-Aug-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add native memory accounting through custom allocator.
Added a custom allocator that lets you pass in a special tag which
specifices where the allocation came from. This is used when
dumping. The performance overhead is low since each allocation only
does a atomic add/sub for each allocation/free.
The measurements are dumped to traces.txt during SIGQUIT.
Example output:
I/art (27274): AllocatorTagHeap active=120 max=120 total=168
I/art (27274): AllocatorTagMonitorList active=1572 max=6240 total=11724
I/art (27274): AllocatorTagClassTable active=185208 max=185208 total=268608
I/art (27274): AllocatorTagInternTable active=430368 max=430368 total=436080
I/art (27274): AllocatorTagMaps active=5616 max=6168 total=34392
I/art (27274): AllocatorTagLOS active=1024 max=1536 total=2044
I/art (27274): AllocatorTagSafeMap active=0 max=51936 total=533688
I/art (27274): AllocatorTagLOSMaps active=144 max=1248 total=5760
I/art (27274): AllocatorTagReferenceTable active=10944 max=11840 total=19136
I/art (27274): AllocatorTagHeapBitmap active=32 max=40 total=56
I/art (27274): AllocatorTagHeapBitmapLOS active=8 max=8 total=8
I/art (27274): AllocatorTagVerifier active=0 max=18844 total=1073156
I/art (27274): AllocatorTagModUnionCardSet active=5300 max=5920 total=56020
I/art (27274): AllocatorTagModUnionReferenceArray active=24864 max=24864 total=24864
I/art (27274): AllocatorTagJNILibrarires active=320 max=320 total=320
I/art (27274): AllocatorTagOatFile active=1400 max=1400 total=5852
Change-Id: Ibb470ef2e9c9a24563bb46422d46a55799704d82
(cherry picked from commit 5369c40f75fdcb1be7a7c06db212ce965c83a164)
|
| 5369c40f75fdcb1be7a7c06db212ce965c83a164 |
|
25-Aug-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add native memory accounting through custom allocator.
Added a custom allocator that lets you pass in a special tag which
specifices where the allocation came from. This is used when
dumping. The performance overhead is low since each allocation only
does a atomic add/sub for each allocation/free.
The measurements are dumped to traces.txt during SIGQUIT.
Example output:
I/art (27274): AllocatorTagHeap active=120 max=120 total=168
I/art (27274): AllocatorTagMonitorList active=1572 max=6240 total=11724
I/art (27274): AllocatorTagClassTable active=185208 max=185208 total=268608
I/art (27274): AllocatorTagInternTable active=430368 max=430368 total=436080
I/art (27274): AllocatorTagMaps active=5616 max=6168 total=34392
I/art (27274): AllocatorTagLOS active=1024 max=1536 total=2044
I/art (27274): AllocatorTagSafeMap active=0 max=51936 total=533688
I/art (27274): AllocatorTagLOSMaps active=144 max=1248 total=5760
I/art (27274): AllocatorTagReferenceTable active=10944 max=11840 total=19136
I/art (27274): AllocatorTagHeapBitmap active=32 max=40 total=56
I/art (27274): AllocatorTagHeapBitmapLOS active=8 max=8 total=8
I/art (27274): AllocatorTagVerifier active=0 max=18844 total=1073156
I/art (27274): AllocatorTagModUnionCardSet active=5300 max=5920 total=56020
I/art (27274): AllocatorTagModUnionReferenceArray active=24864 max=24864 total=24864
I/art (27274): AllocatorTagJNILibrarires active=320 max=320 total=320
I/art (27274): AllocatorTagOatFile active=1400 max=1400 total=5852
Bug: 16238192
Change-Id: Ibb470ef2e9c9a24563bb46422d46a55799704d82
|
| 6a7824dc81aaab3cb09ced16affca72d1b1da649 |
|
22-Aug-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add fallback for non moving space being full.
When the non moving space becomes full, we disable moving GC and
make the main space the new non moving space.
Also added a runtime option for changing the non moving space size:
-XX:NonMovingSpaceCapacity.
Bug: 17189964
(cherry picked from commit 4c5a469683e433f126c9863cd393747d2e7c4a29)
Change-Id: If82e3c6f8a0f389e37e14a0b1e6d5126d571fd7a
|
| 4c5a469683e433f126c9863cd393747d2e7c4a29 |
|
22-Aug-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add fallback for non moving space being full.
When the non moving space becomes full, we disable moving GC and
make the main space the new non moving space.
Also added a runtime option for changing the non moving space size:
-XX:NonMovingSpaceCapacity.
Bug: 17189964
Change-Id: I04d00d0f457f6c1f5724bf32932b6b6ce4d341b1
|
| e4cab17634dc6d5809b40f2ed80d6459734ae2fe |
|
20-Aug-2014 |
Mathieu Chartier <mathieuc@google.com> |
Enable large object space for zygote.
We now enable the large object space before the zygote fork.
This reduces the size of the zygote and removes the need for
excessive explicit GCs during phone booting.
Changed the card set mod union table to support forgetting cards.
If a card has no non null references which are in another space
then it is removed from the set.
Added logging of the zygote size when you do a SIGQUIT.
Dalvik PSS is the same or slightly lower (1-3%).
Zygote space size:
Before: 15MB
After: 8MB (+ some large objects).
TODO: Combine remembered sets and mod union tables into a single
interface.
Bug: 16398684
Change-Id: Ie48cdf35004a0a37eedb1ccc1bf214b1fa9e0cca
|
| b225890e02bc8d5864217743eaef306d5387e0e9 |
|
07-Aug-2014 |
Mathieu Chartier <mathieuc@google.com> |
Change FreeMemory and TotalMemory behavior.
TotalMemory is now equal to the footprint limit (around when the
next GC will occur).
FreeMemory is now equal to TotalMemory() - bytes_allocated_.
Also added more memory dumping info to DumpGcPerformanceInfo.
Bug: 16520008
Bug: 15819878
(cherry picked from commit dd162fb5990cedf80a5093ecc0e77df82af5f754)
Change-Id: I68239a5295cc1372b3995eb781d67b9e25d6eaed
|
| dd162fb5990cedf80a5093ecc0e77df82af5f754 |
|
07-Aug-2014 |
Mathieu Chartier <mathieuc@google.com> |
Change FreeMemory and TotalMemory behavior.
TotalMemory is now equal to the footprint limit (around when the
next GC will occur).
FreeMemory is now equal to TotalMemory() - bytes_allocated_.
Also added more memory dumping info to DumpGcPerformanceInfo.
Bug: 16520008
Bug: 15819878
Change-Id: Ifa59a2f2136dd0704097411ee183925e6c387fae
|
| 68d8b42ddec39ec0174162d90d4abaa004d1983e |
|
17-Jul-2014 |
Ian Rogers <irogers@google.com> |
Wire up check JNI force copy mode.
Increase check JNI checks.
Break apart jni_internal.h in to jni_env_ext.h and java_vm_ext.h.
Fix the abuse of ScopedObjectAccess/annotalysis by ScopedCheck in the case
of VM routines.
Make class loader override and shared library class loader JNI global
references rather than mirror pointers.
Clean-ups to native bridge.
Change-Id: If7c6110b5aade7a402bfb67534af86a7b2cdeb55
|
| eb1e929c0d6e312beb313ec108e611f1e74ff45c |
|
06-Aug-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Implement TLAB fast paths in artAllocObjectFromCode.
GSS/TLAB GC speedup on N4 (ms):
MemAllocTest 2963 -> 2792
BinaryTrees 2205 -> 2113
Also, measured wth -XX:IgnoreMaxFootprint to invoke GC less often
(only when the bump pointer space is filled rather than based on the
target utilization):
MemAllocTest 2707 -> 2590
BinaryTrees 2023 -> 1906
TODO: implement fast paths for array allocations.
Bug: 9986565
Change-Id: I73ff6327b229704f8ae5924ae9b747443c229841
|
| 38c488bcd41ba632a646d7a1d790ec71a2fcf6fa |
|
16-Jul-2014 |
Mathieu Chartier <mathieuc@google.com> |
Recycle mem-maps for collector transitions.
We now create spaces when we need them for collector transitions or
homogeneous compaction by recycling mem maps. Change the bump
pointer space size to be as large as the heap capacity instead of
1/2 heap capacity like it used to be. For GSS, bump pointer spaces
are set to 32MB currently.
Changed GSS to have main space == non moving space since we don't
need to copy from the main space.
Fixes GC stress tests 074, 096.
Fixed test 080 oom throw with -Xmx2m for GC stress test, this was
broken since it was allocating a 4 MB array before starting the
OOM process.
Bug: 14059466
Bug: 16406852
(cherry picked from commit b363f666883860d40823d5528df3c98c897f74f4)
Change-Id: I62877cfa24ec944a6f34ffac30334f454a8002fd
|
| b76cac637691c29daa9c44e493b5bc26346ed116 |
|
23-Jul-2014 |
Mathieu Chartier <mathieuc@google.com> |
Revert "Revert "Disable adding main and non moving spaces to immune region in GSS""
This reverts commit f85c2fb317399ab540854cd7551ac47690366543.
|
| 5189e24fb6d42c04c48169ab2f15de56ecf3c828 |
|
24-Jul-2014 |
Mathieu Chartier <mathieuc@google.com> |
Fix memory leak in RemoveRememberedSet.
RemoveRememberedSet now deletes the remembered set.
Bug: 16532086
Change-Id: I01092931cc20cd0688dd42eed3dde9ad140889b2
|
| b363f666883860d40823d5528df3c98c897f74f4 |
|
16-Jul-2014 |
Mathieu Chartier <mathieuc@google.com> |
Recycle mem-maps for collector transitions.
We now create spaces when we need them for collector transitions or
homogeneous compaction by recycling mem maps. Change the bump
pointer space size to be as large as the heap capacity instead of
1/2 heap capacity like it used to be. For GSS, bump pointer spaces
are set to 32MB currently.
Changed GSS to have main space == non moving space since we don't
need to copy from the main space.
Fixes GC stress tests 074, 096.
Fixed test 080 oom throw with -Xmx2m for GC stress test, this was
broken since it was allocating a 4 MB array before starting the
OOM process.
Bug: 14059466
Bug: 16406852
Change-Id: I62877cfa24ec944a6f34ffac30334f454a8002fd
|
| f85c2fb317399ab540854cd7551ac47690366543 |
|
22-Jul-2014 |
Mathieu Chartier <mathieuc@google.com> |
Revert "Disable adding main and non moving spaces to immune region in GSS"
Bug: 16399257
This reverts commit be0562fb14e6754ee932b8d9c97e2a6df3a91119.
Change-Id: I29e07a8fa1e972990e5bf1ddf8c9a3538ea5f9cf
|
| be0562fb14e6754ee932b8d9c97e2a6df3a91119 |
|
14-Jul-2014 |
Mathieu Chartier <mathieuc@google.com> |
Disable adding main and non moving spaces to immune region in GSS
Disabled adding the main and non moving space to the immune region.
This will enable us to recycle bump pointer spaces for malloc space
-> malloc space compaction as well as collector transitions.
Also added logic for falling back to the non moving space, we may
copy objects there.
Refactored mod union table logic into MarkReachableObjects.
No measurable performance benefit or regression.
Bug: 14059466
Bug: 16291259
(cherry picked from commit 4c13a3ff475f206c4d0a86ee2595c45392fd942f)
Change-Id: I858b4fbddca888e164052ad247565a0bdbea68b5
|
| 4c13a3ff475f206c4d0a86ee2595c45392fd942f |
|
14-Jul-2014 |
Mathieu Chartier <mathieuc@google.com> |
Disable adding main and non moving spaces to immune region in GSS
Disabled adding the main and non moving space to the immune region.
This will enable us to recycle bump pointer spaces for malloc space
-> malloc space compaction as well as collector transitions.
Also added logic for falling back to the non moving space, we may
copy objects there.
Refactored mod union table logic into MarkReachableObjects.
No measurable performance benefit or regression.
Bug: 14059466
Bug: 16291259
Change-Id: If663d9fdbde943b988173b7f6ac844e5f78a0327
|
| f37a88b8e6db6c587fa449a12e40cb46be1689fc |
|
10-Jul-2014 |
Zuo Wang <zuo.wang@intel.com> |
ART: Compacting ROS/DlMalloc spaces with semispace copy GC
Current semispace copy GC is mainly associated with bump pointer
spaces. Though it squeezes fragmentation most aggressively, an extra
copy is required to re-establish the data in the ROS/DlMalloc space to allow
CMS GCs to happen afterwards. As semispace copy GC is still stop-the-world,
this not only introduces unnecessary overheads but also longer response time.
Response time indicates the time duration between the start of transition
request and the start of transition animation, which may impact the user
experience.
Using semispace copy GC to compact the data in a ROS space to another ROS(or
DlMalloc space to another DlMalloc) space solves this problem. Although it
squeezes less fragmentation, CMS GCs can run immediately after the compaction.
We apply this algorithm in two cases:
1) Right before throwing an OOM if -XX:EnableHSpaceCompactForOOM is passed in
as true.
2) When app is switched to background if the -XX:BackgroundGC option has value
HSpaceCompact.
For case 1), OOMs are significantly delayed in the harmony GC stress test,
with compaction ratio up to 0.87. For case 2), compaction ratio around 0.5 is
observed in both built-in SMS and browser. Similar results have been obtained
on other apps as well.
Change-Id: Iad9eabc6d046659fda3535ae20f21bc31f89ded3
Signed-off-by: Wang, Zuo <zuo.wang@intel.com>
Signed-off-by: Chang, Yang <yang.chang@intel.com>
Signed-off-by: Lei Li <lei.l.li@intel.com>
Signed-off-by: Lin Zang <lin.zang@intel.com>
|
| 2f4f0a3d049a39687418087fc0165f7a21654b45 |
|
11-Jul-2014 |
Mathieu Chartier <mathieuc@google.com> |
Change default heap maximum size to be 256m.
Useful for command line benchmarks.
Change-Id: Ie525863cd8eff93c64ce76639b1108fbdad91633
|
| 654dd48e2230e16bfaa225decce72b52642e2f78 |
|
09-Jul-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Improve the OOME fragmentation message.
Change-Id: I390d3622f8d572ec7e34ea6dff9e1e0936e81ac1
|
| 7c88c6054dfc01f2ac36b19842d467626459fc45 |
|
09-Jul-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add easy way to ensure the next allocation does GC.
Added a class called ScopedHeapFill which changes the bytes allocated
counter to be equal to the growth limit. This causes the next
allocation to do a GC and possibly generate an OOM error. This is
useful for tests which need GC to happen at specific point.
Change-Id: Ibd8f3d5928b58534c5165ba7c296980002aa2c28
|
| a5b5c55c8585b7ce915f0c7e1f66d121a7f7a078 |
|
24-Jun-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add notion of released vs empty pages to ROSAlloc.
A notion of released vs empty pages helps get a more accurate view of
how much memory was released during heap trimming. Otherwise we get
that the same pages possibly get madvised multiple times without
getting dirtied.
Also enabled heap trimming of rosalloc spaces even when we care about
jank. This is safe to do since the trimming process only acquires
locks for short periods of time.
Dalvik PSS reduces from ~52M to ~50M after boot on N4.
Bug: 9969166
Change-Id: I4012e0a2554f413d18efe1a0371fe18d1edabaa9
|
| 146b1872a4b379e808e2738b4090d3293a34c2e8 |
|
27-Jun-2014 |
Mathieu Chartier <mathieuc@google.com> |
Don't allow negative GetFreeMemory.
The max allowed footprint is only updated after the GC. But we can
still allocate even if bytes_allocated > max_allowed_footprint_.
This means that we used to be able to get a negative value if
bytes_allocated > max_allowed_footprint_.
External bug:
https://code.google.com/p/android/issues/detail?id=72221
Change-Id: I4ef9a534e29211786e82cdcb2582c11ab37a348a
|
| 10fb83ad7442c8cf3356a89ec918e0786f110981 |
|
16-Jun-2014 |
Mathieu Chartier <mathieuc@google.com> |
Shared single GC iteration accounting for all GCs.
Previously, each garbage collector had data that was only used
during collection. Since only one collector can be running at any
given time, we can make this data be shared between all collectors.
This reduces memory usage since we don't need to have redundant
information for each GC types. Also reduced how much code is required
to sweep spaces.
Bug: 9969166
Change-Id: I31caf0ee4d572f75e0c66863fe7db12c08ae08e7
|
| 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
|
| d30e1d6b477650366c25822f956202fec041e1d5 |
|
09-Jun-2014 |
Mathieu Chartier <mathieuc@google.com> |
Fix GetFreeMemory to use fooprint limit instead of max memory.
Based on definitions in:
http://developer.android.com/reference/java/lang/Runtime.html
Bug: 15507122
Change-Id: I02f34682d7ac2d379a07631b5207b6cfb224da6b
|
| ffddfdf6fec0b9d98a692e27242eecb15af5ead2 |
|
03-Jun-2014 |
Tim Murray <timmurray@google.com> |
DO NOT MERGE
Merge ART from AOSP to lmp-preview-dev.
Change-Id: I0f578733a4b8756fd780d4a052ad69b746f687a9
|
| 8e1ebf467a9c8fa504d7a90e03df6053ecdceeb3 |
|
30-May-2014 |
Mathieu Chartier <mathieuc@google.com> |
Change GetFreeMemory to use the growth limit.
This makes more sense since it is what the allocator uses. Also fixed
somewhere where we were not properly passing in whether or not it was
a large object allocation.
Bug: 15327879
Change-Id: Ieab7af5427f5cdc2760390186b67e2c96d4bafa7
|
| c179016fe188bef09487e777aa0fd861f5cdf067 |
|
23-May-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add reserve area to allocation stacks.
This fixes an issue with heap verification which was caused when
the allocation stack overflowed. This resulted in heap verification
failures since we were storing the newly allocated object in a
handle scope without having it be live either in the live bitmap
or allocation stack. We now push the object in the reserve area
before we do a GC due to allocation stack overflow.
Change-Id: I83b42c4b3250d7eaab1b49e53066e21c8656a740
|
| 8e4a96d9a7ba7c555032ef583a089ebd2c53f5f7 |
|
21-May-2014 |
Mathieu Chartier <mathieuc@google.com> |
Change zygote_creation_lock_ to be member instead of static.
Static variables aren't thread safe and could cause the zygote to be
created twice.
Bug: 15133494
Change-Id: I65c8f089bed8de93f895b62b3dcff4c936931860
|
| 3e5cf305db800b2989ad57b7cde8fb3cc9fa1b9e |
|
21-May-2014 |
Ian Rogers <irogers@google.com> |
Begin migration of art::Atomic to std::atomic.
Change-Id: I4858d9cbed95e5ca560956b9dabd976cebe68333
|
| 8ab7e78be6835d63a2cee9f234334ed8fb409781 |
|
20-May-2014 |
Mathieu Chartier <mathieuc@google.com> |
Improve heap verification for invalid roots.
The new root verification prints the root type and owner thread id as
well as the type of the object.
Also a bit of work for planned multi-threaded verification.
Bug: 14289301
Change-Id: Ia73c517dc11ec6dd82f3d945604ee3836b3db536
|
| 700a402244a1a423da4f3ba8032459f4b65fa18f |
|
20-May-2014 |
Ian Rogers <irogers@google.com> |
Now we have a proper C++ library, use std::unique_ptr.
Also remove the Android.libcxx.mk and other bits of stlport compatibility
mechanics.
Change-Id: Icdf7188ba3c79cdf5617672c1cfd0a68ae596a61
|
| 2afe49450f2e018f18b5de45428b9174bfd6f196 |
|
19-May-2014 |
Brian Carlstrom <bdc@google.com> |
Follow up to "Add ISA directory to image and odex pathnames."
Change-Id: I7f08cc3052fbed93a56ccf1ab7675ae8bc129da9
|
| eb8167a4f4d27fce0530f6724ab8032610cd146b |
|
08-May-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add Handle/HandleScope and delete SirtRef.
Delete SirtRef and replaced it with Handle. Handles are value types
which wrap around StackReference*.
Renamed StackIndirectReferenceTable to HandleScope.
Added a scoped handle wrapper which wraps around an Object** and
restores it in its destructor.
Renamed Handle::get -> Get.
Bug: 8473721
Change-Id: Idbfebd4f35af629f0f43931b7c5184b334822c7a
|
| 78f7b4c04ab6e8b5581921bc95b67a9beee1c246 |
|
06-May-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add concurrent reference processing.
Concurrent reference processing currently works by going into native
code from java.lang.ref.Reference.get(). From there, we have a fast
path if the references aren't being processed which returns the
referent without needing to access any locks. In the slow path we
block until reference processing is complete. It may be possible to
improve the slow path if the referent is blackened.
TODO: Investigate doing the fast path in java code by using racy reads
of a static volatile boolean. This will work as long as there are no
suspend points inbetween the boolean read and referent read.
Bug: 14381653
Change-Id: I1546b55be4691fe4ff4aa6d857b234cce7187d87
|
| e76e70f424468f311c2061c291e8384263f3968c |
|
03-May-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add RecordFree to the GarbageCollector interface
RecordFree now calls the Heap::RecordFree as well as updates the
garbage collector's internal bytes freed accounting.
Change-Id: I8cb03748b0768e3c8c50ea709572960e6e4ad219
|
| 89a201e599da95d88590c7b4d069b76735e20fa8 |
|
02-May-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add GC cause to WaitForGcToComplete message.
Change-Id: I8fe107d90a84de065c407b8d29fd106267ac440d
|
| 8668c3cbdcf9471bd97e0da68a240051f2973074 |
|
25-Apr-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add finalizer references from the entrypoints.
We now have an invariant where we never allocate finalizable
objects with the Initialized or Resolved entrypoints. This speeds up
allocation by only doing the check in the slow path.
Before:
MemAllocTest: 3625, 3707, 3641
EvaluateAndApplyChanges: 3448, 3421, 3413
After:
MemAllocTest: 3164, 3109, 3135
EvaluateAndApplyChanges: 3272, 3299, 3353
Bug: 14078487
Change-Id: I2b0534af3e7c75ea5e5257cf3647744f7abfb74e
|
| 11d9f06a96a6909905c248ed684366190140095c |
|
23-Apr-2014 |
Narayan Kamath <narayan@google.com> |
Use instruction specific dalvik cache dirs.
- All oat & art files are now placed under /data/dalvik-cache/<isa>/.
- GetDalvikCacheOrDie now requires a mandatory subdirectory argument,
and is implicitly rooted under /data/.
- Added helper methods to convert InstructionSet enums into strings
and vice versa.
(cherry picked from commit 2974bc3d8a5d161d449dd66826d668d87bdc3cbe)
Change-Id: Ic7986938e6a7091a2af675ebafec768f7b5fb8cd
|
| 1b54f9cb38605046d772ba0e125d5c009f1de7d2 |
|
01-May-2014 |
Mathieu Chartier <mathieuc@google.com> |
Clean up Add/Remove space.
Deleted the set_as_default parameter and added a new function
SetSpaceAsDefault instead.
Change-Id: Ic4c359854d08e64ac0d0df92f0105447adb9df36
|
| 2974bc3d8a5d161d449dd66826d668d87bdc3cbe |
|
23-Apr-2014 |
Narayan Kamath <narayan@google.com> |
Use instruction specific dalvik cache dirs.
- All oat & art files are now placed under /data/dalvik-cache/<isa>/.
- GetDalvikCacheOrDie now requires a mandatory subdirectory argument,
and is implicitly rooted under /data/.
- Added helper methods to convert InstructionSet enums into strings
and vice versa.
Change-Id: I9bff2e2ca534e0b93842a50d5b272ddf6d5745f3
|
| 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
|
| 6f365cc033654a5a3b45eaa1379d4b5f156b0cee |
|
23-Apr-2014 |
Mathieu Chartier <mathieuc@google.com> |
Enable concurrent sweeping for non-concurrent GC.
Refactored the GarbageCollector to let all of the phases be run by
the collector's RunPhases virtual method. This lets the GC decide
which phases should be concurrent and reduces how much baked in GC
logic resides in GarbageCollector.
Enabled concurrent sweeping in the semi space and non concurrent
mark sweep GCs. Changed the semi-space collector to have a swap semi
spaces boolean which can be changed with a setter.
Fixed tests to pass with GSS collector, there was an error related to
the large object space limit.
Before (EvaluateAndApplyChanges):
GSS paused GC time 7.81s/7.81s, score: 3920
After (EvaluateAndApplyChanges):
GSS paused GC time 6.94s/7.71s, score: 3900
Benchmark score doesn't go up since the GC happens in the allocating
thread. There is a slight reduction in pause times experienced by
other threads (0.8s total).
Added options for pre sweeping GC heap verification and pre sweeping
rosalloc verification.
Bug: 14226004
Bug: 14250892
Bug: 14386356
Change-Id: Ib557d0590c1ed82a639d0f0281ba67cf8cae938c
|
| 73d1e17b3afc7d5e56184f90bf819dc64956448a |
|
12-Apr-2014 |
Mathieu Chartier <mathieuc@google.com> |
Enable reading page map without lock in RosAlloc::BulkFree
Enabling this flag greatly reduces how much time was spent in the GC.
It was not done previously since it was regressing MemAllocTest. With
these RosAlloc changes, the benchmark score no longer regresses after
we enable the flag.
Changed Run::AllocSlot to only have one mode of allocation. The new
mode is finding the first free bit in the bitmap. This was
previously the slow path but is now the fast path. Some optimizations
which enabled this include always having the alloc bitmap bits which
correspond to invalid slots be set to 1. This prevents us from needing
a bound check since we will never end up allocating there.
Changed revoking thread local buffer to point to an invalid run. The
invalid run is just a run which always has all the allocation bits set
to 1. When a thread attempts to do a thread local allocation from here
it will always fail and go slow path. This eliminates the need for a
null check for revoked runs.
Changed zeroing of memory to happen during free, AllocPages should
always return zeroed memory. Added prefetching which happens when we
allocate a run.
Some refactoring to reduce duplicated code.
Ergonomics changes: Changed kStickyGcThroughputAdjustment to 1.0,
this helps reduce GC time.
Measurements (3 samples per benchmark):
Before: MemAllocTest scores: 3463, 3445, 3431
EvaluateAndApplyChanges score | total GC time
Iter 1: 3485, 23.602436s
Iter 2: 3434, 22.499882s
Iter 3: 3483, 23.253274s
After: MemAllocTest scores: 3495, 3417, 3409
EvaluateAndApplyChanges score | total GC time:
Iter 1: 3375, 17.463462s
Iter 2: 3358, 16.185188s
Iter 3: 3367, 15.822312s
Bug: 8788501
Bug: 11790317
Bug: 9986565
Change-Id: Ifd273a054824028dabed27c07c081dde1816f93c
|
| 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
|
| 2f8da3e9ff60e5cb2a3fdf57dbcb67f513b9c2c2 |
|
16-Apr-2014 |
Mathieu Chartier <mathieuc@google.com> |
Decrease target utilization for foreground apps.
GC time in FormulaEvaluationActions.EvaluateAndApplyChanges goes from
26.1s to 23.2s. Benchmark score goes down ~50 in
FormulaEvaluationActions.EvaluateAndApplyChanges, and up ~50 in
GenericCalcActions.MemAllocTest.
Bug: 8788501
Change-Id: I412af1205f8b67e70a12237c990231ea62167bc0
|
| a8e8f9c0a8e259a807d7b99a148d14104c24209d |
|
09-Apr-2014 |
Mathieu Chartier <mathieuc@google.com> |
Refactor space bitmap to support different alignments.
Required for:
Using space bitmaps instead of std::set in mod union table +
remembered set.
Using a bitmap instead of set for large object marking.
Bug: 13571028
Change-Id: Id024e9563d4ca4278f79607cdb2f81895121b113
|
| 31f441464c0c8f840aba37e236ad133f30308d70 |
|
08-Apr-2014 |
Mathieu Chartier <mathieuc@google.com> |
Clean-up heap spaces.
We now use the CMS collector instead of the semispace collector when
the phone is booting. We still perform compaction during the zygote
space creation. This reduces time spent in GC by ~2s during boot
and doesn't affect zygote space size.
Changed the space creation logic to create the temp space when a
background transition occurs.
Added a flag to each space which is true if you are allowed to
move objects that are within this space.
Removed SwapSemiSpaces call from the semi space collector, it is now
the job of the caller to do this with threads suspended. This
simplifies the logic in the zygote compaction / heap transition code
since these do not copy from one semispace to another.
Added Space::Clear to RosAllocSpace and DlMallocSpace. This greatly
simplifies the code used for collector transitions.
Time spent in GC creating zygote space:
Before: 3.4s, After: 1.28s
No change in zygote space size.
Bug: 13878055
Change-Id: I700348ab7d5bf3aa537c0cd70c0fed09aa4b0623
|
| 7bf9f190cd33a7e2f8584299eb889e9df66e0323 |
|
04-Apr-2014 |
Mathieu Chartier <mathieuc@google.com> |
Improve invalid root dumping.
The invalid root dumping now attempts to print the root type.
Change-Id: Ie821296d569f34909ba6e2705f5c347cd2143a3a
|
| 4cd662e54440f76fc920cb2c67acab3bba8b33dd |
|
04-Apr-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Fix Object::Clone()'s pre-fence barrier.
Pass in a pre-fence barrier object that sets in the array length
instead of setting it after returning from AllocObject().
Fix another potential bug due to the wrong default pre-fence barrier
parameter value. Since this appears error-prone, removed the default
parameter value and make it an explicit parameter.
Fix another potential moving GC bug due to a lack of a SirtRef.
Bug: 13097759
Change-Id: I466aa0e50f9e1a5dbf20be5a195edee619c7514e
|
| 4aeec176eaf11fe03f342aadcbb79142230270ed |
|
28-Mar-2014 |
Mathieu Chartier <mathieuc@google.com> |
Refactor some GC code.
Reduced amount of code in mark sweep / semi space by moving
common logic to garbage_collector.cc. Cleaned up mod union tables
and deleted an unused implementation.
Change-Id: I4bcc6ba41afd96d230cfbaf4d6636f37c52e37ea
|
| d5307ec41c8344be0c32273ec4f574064036187d |
|
28-Mar-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
An empty collector skeleton for a read barrier-based collector.
Bug: 12687968
Change-Id: Ic2a3a7b9943ca64e7f60f4d6ed552a316ea4a6f3
|
| afe4998fc15b8de093d6b282c9782d7182829e36 |
|
27-Mar-2014 |
Mathieu Chartier <mathieuc@google.com> |
Change sticky GC ergonomics to use GC throughput.
The old sticky ergonomics used partial/full GC when the bytes until
the footprint limit was < min free. This was suboptimal. The new
sticky GC ergonomics do partial/full GC when the throughput
of the current sticky GC iteration is <= mean throughput of the
partial/full GC.
Total GC time on FormulaEvaluationActions.EvaluateAndApplyChanges.
Before: 26.4s
After: 24.8s
No benchmark score change measured.
Bug: 8788501
Change-Id: I90000305e93fd492a8ef5a06ec9620d830eaf90d
|
| c93c530efc175954160c3834c93961a1a946a35a |
|
21-Mar-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Revoke rosalloc thread-local buffers at the checkpoint.
In the mark sweep collector, rosalloc thread-local buffers were
revoked during the pause. Now, they are revoked at the thread
checkpoint, as opposed to during the pause, which appears to help
reduce the pause time.
In Ritz MemAllocTest, the average sticky pause time went down ~20%
(925 us -> 724 us).
Bug: 13394464
Bug: 9986565
Change-Id: I104992a11b46d59264c0b9aa2db82b1ccf2826bc
|
| 36bf21665b43adc2485167cbb3a044dbdf4931ee |
|
20-Mar-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add Heap::RunningOnValgrind and call it from the spaces.
Makes it easier to disable valgrind support.
Change-Id: I1bde792f1b76a2dd968fa03c6142e92fcc3670b0
|
| 601276abdb746b03675ff945745aa936694d3439 |
|
20-Mar-2014 |
Mathieu Chartier <mathieuc@google.com> |
Fix RecordFree to take signed parameters.
RecordFree can get negative bytes allocated when background
compaction foreground transitions occur. This caused a DCHECK to
fail on debug builds. Also did some refactoring in
PreProcessReferences.
Bug: 13568814
Change-Id: I57543f1c78544a94f1d241459698b736dba8cfa8
|
| 3e41780cb3bcade3b724908e00443a9caf6977ef |
|
20-Mar-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Refactor the garbage collector driver (GarbageCollector::Run).
Bug: 12687968
Change-Id: Ifc9ee86249f7938f51495ea1498cf0f7853a27e8
|
| 1ad2784ad9f311ebf9fe0677d33818648f423f9c |
|
20-Mar-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add soft reference pre processing.
Soft reference pre-processing does soft reference preservation with
mutators running. After this is done, it does another pass with
mutators paused in the ProcessReference code. This helps lower pauses
since most preserved soft references have their referents recursive
marked outside the pause.
Changed ergonomics to have non sticky collectors always clear the
soft references.
Maps pauses ~10ms -> ~3ms on Nexus 4.
Bug: 13421927
Change-Id: I1370f7bb6934034869aa5afca0c377876267aa8e
|
| 38e68e9978236db87c9008bbe47db80525d2fa16 |
|
07-Mar-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Use the card table to speed up the GSS collector.
Scan only dirty cards, as opposed to the whole space, to find
references from the non-moving spaces to the bump pointer spaces at
bump pointer space only collections.
With this change, the Ritz MemAllocTest speeds up by 8-10% on host and
2-3% on N4. The Ritz EvaluateFibonacci speeds up by 8% and its average
pause time is reduced by 43% on N4.
Bug: 11650816
Change-Id: I1eefe75776bc37e24673b301ffa65a25f9bd4cde
|
| 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
|
| 7bf52d28978a6a747795cc0c78b04a3e83f0cf16 |
|
13-Mar-2014 |
Mathieu Chartier <mathieuc@google.com> |
Fix heap trimming logic.
The new heap trimming logic introduced in the c/84301 was not very
good since the heap trim would get avoided since the daemon thread
was coming to the heap trimming code before the scheduled time.
The new logic is to do the heap trim if the last heap trim occurred
more than kHeapTrimWait ns ago.
Change-Id: I9d0e6766bf0c68e5f7fb15fb059140e1f1264216
|
| 5ccd498d4aa450b0381344724b072a932709a59a |
|
11-Mar-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Put the post zygote non-moving space next to the malloc space.
This change fixes the following problem with the GSS collector: if the
post zygote non-moving space happens to be located after the bump
pointer spaces, the bump pointer space is treated wrongfully as an
immune space. The reasons are 1) the immune space is represented by a
simple address range and 2) the GSS collector treats the post zygote
non-moving space as an immune space at a bump pointer space only
collection.
In addition, this change makes it a reality that all the non-moving
spaces are adjacent, which we tend to assume in the code (eg. the
notion of the immune space represented by a simple address range.)
This should help avoid potential confusions in the future.
Fix a DCHECK failure where usable_size isn't set properly when
-XX:UseTLAB is used.
Change-Id: I585920e433744a390f87e9a25bef6114b2a4623f
|
| c645f1ddb7c40bea6a38eda4b3f83f6b6dec405b |
|
07-Mar-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add more VerifyObject calls.
Added verify object calls to SirtRef, IndirectReferenceTable,
ReferenceTable.
Removed un-needed verify object in ScopedObjectAccess / DecodeJObject
since object sources are handled.
Bug: 12934910
Change-Id: I55a46a8ea61fed2a77526eda27fd2cce97a9b125
|
| 719d1a33f6569864f529e5a3fff59e7bca97aad0 |
|
06-Mar-2014 |
Ian Rogers <irogers@google.com> |
Enable annotalysis on clang ART builds.
Fix clang build errors aswell as restructure locking/mutex code for correct
thread safety analysis support.
Reorder make dependencies so that host builds build first as they should
provide better compilation errors than target.
Remove host's use of -fno-omit-frame-pointer as it has no value with correct
use of CFI, which we should have.
Change-Id: I72cea8da9a3757b1a0b3acb4081feccb7c6cef90
|
| f517f1a994fab72ba484bbbac6911e315f59f6cd |
|
07-Mar-2014 |
Mathieu Chartier <mathieuc@google.com> |
Restore obj after RequestConcurrentGC.
RequestConcurrentGC can cause thread suspension, this means that
another thread could transition the heap or cause moving GC.
Bug: 12934910
Change-Id: I5c07161e2e849d7acbdf939f1c24e1ba361a1d6a
|
| a5f9de040a7a740b7b4979bc8d47f95eb05718ae |
|
01-Mar-2014 |
Mathieu Chartier <mathieuc@google.com> |
Defer heap transitions to heap trimming daemon.
This fixes the case where quick back and forth process state changes
would cause a lot of heap transitions.
Change-Id: Ia5be792edcf26079b2aa23b9c115f6b0a9a39a1c
|
| 6dda898d47b3e8931e4404330e81b7110108e34f |
|
06-Mar-2014 |
Mathieu Chartier <mathieuc@google.com> |
Disable compaction for jni workarounds.
Compaction can't work when jni workarounds is enabled. Also some
other refactoring.
Change-Id: Ia7b0f2b39c79f5a0a5f50874d823b950ab02a0c3
|
| d889178ec78930538d9d6a66c3df9ee9afaffbb4 |
|
02-Mar-2014 |
Mathieu Chartier <mathieuc@google.com> |
Guard entrypoint changing by runtime shutdown lock.
There was a race when we changed the allocation entrypoints where a
new thread would be starting (Thread::Init) and initialize to the
wrong entrypoints. Guarding allocation entrypoint changing
with the runtime shutdown lock fixes this race condition since
Thread::Init is only called with the runtime shutdown lock held.
Bug: 13250963
Change-Id: I8eb209c124b6bf17020de874e1b0083f158b8200
|
| 15d3402bbf8265eb1165694da2e4117eb128f3bc |
|
27-Feb-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add custom SIGSEGV handler to help find heap corruption.
The new signal handler prints heap diagnostics when you get a SIGSEGV.
Added a fault message member in runtime which is modifiable by
Runtime::SetFaultMessage. When you get a SIGSEGV it will print out
whatever is stored in this string as well as the normal information.
This is useful for debugging heap corruption since it lets you see
which threads were in which methods when the last GC occured.
Added some smarter object dumping logic when the faulting address is
in the heap.
Bug: 12934910
Change-Id: Ia72be2c39f70ad711cbd746d66fad2b617d5d29f
|
| bd0a65339a08dc28c6b56d2673f1f13b6bddd7aa |
|
27-Feb-2014 |
Mathieu Chartier <mathieuc@google.com> |
Enable large object space for command line runs.
Added a dynamic large_object_threshold_ variable which is max int
when the large object space is disabled.
No longer clear timing logger timings after zygote creation since
it is not required and removes the boot semispace timings.
Change-Id: I693865f4699cc32381199377239854c6ec42f37e
|
| 6fac447555dc94a935b78198479cce645c837b89 |
|
26-Feb-2014 |
Ian Rogers <irogers@google.com> |
Make allocations report usable size.
Work-in-progress to allow arrays to fill usable size. Bug: 13028925.
Use C++11's override keyword on GCC >= 2.7 to ensure that we override GC and
allocator methods.
Move initial mirror::Class set up into a Functor so that all allocated objects
have non-zero sizes. Use this property to assert that all objects are never
larger than their usable size.
Other bits of GC related clean-up, missing initialization, missing use of
const, hot methods in .cc files, "unimplemented" functions that fail at
runtime in header files, reducing header file includes, move valgrind's space
into its own files, reduce number of array allocation routines.
Change-Id: Id5760041a2d7f94dcaf17ec760f6095ec75dadaa
|
| c22c59ef8513b4cbbfd25073d1afbf58196b522a |
|
25-Feb-2014 |
Mathieu Chartier <mathieuc@google.com> |
Remove started runtime check in RevokeAllThreadLocalAllocationStacks
This check occasionally caused some thread local allocation stacks
to incorrectly not get revoked when multiple threads were allocating
without a started runtime. This showed up in image_test with
compaction enabled when we were initializing classes in the compiler
driver.
Change-Id: I7f28d072feea333c2503e35265ba25c51a6308fe
|
| 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
|
| 90d706849a09590f150748ff97256ef718aca441 |
|
21-Feb-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Fix oatdumpd.
Bug: 13063529
Change-Id: I777cd7621ed522b8ab84c5bb1a9d2e5583817a08
|
| 3bb57c7b41bf5419fe895e7aa664d8d430205ba8 |
|
18-Feb-2014 |
Mathieu Chartier <mathieuc@google.com> |
Change ProcessReferences to not use RecursiveMarkObject.
Calling ProcessMarkStack in RecursiveMarkObject caused a lot of
overhead due to timing logger splits. Changed the logic to be the
same as prior to the reference queue refactoring which involves
calling process mark stack after preserving soft references and
enqueueing finalizer references.
FinalizingGC longest pause is reduced by around 1/2 down to ~300ms.
Benchmark score ~400000 -> ~600000.
Also changed the timing logger splits in the GC to have (Paused) if
the split is a paused part of the GC.
Bug: 12129382
Change-Id: I7476d4f23670b19d70738e2fd48e37ec2f57e9f4
|
| f5b0e20b5b31f5f5465784adcf2a204dcd69c7fd |
|
12-Feb-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Thread-local allocation stack.
With this change, Ritz MemAllocTest gets ~14% faster on N4.
Bug: 9986565
Change-Id: I2fb7d6f7c5daa63dd4fc73ba739e6ae4ed820617
|
| d68ac700820f3e4253c8b4bcf718daf452f6da4c |
|
11-Feb-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add more checking to ReleasePrimitiveArray.
When we ReleasePrimitiveArray, we now check that the elements pointer
is not a heap address if it is not equal to the java array's data.
Bug: 12845603
Change-Id: I458862f4dc586ba1c414647c7eb81b978c4ccb7e
|
| 83c8ee000d525017ead8753fce6bc1020249b96a |
|
28-Jan-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add root types and thread id to root visiting.
Enables us to pass the root type and thread id to hprof.
Bug: 12680863
Change-Id: I6a0f1f9e3aa8f9b4033d695818ae7ca3460d67cb
|
| 412c7fced915fc8d4d5e4166e977d55c809168a6 |
|
07-Feb-2014 |
Mathieu Chartier <mathieuc@google.com> |
Make debugger / jdwp compaction safe.
Fixed GetInstances, GetReferringObjects, CountInstances to use
VisitObjects instead of the live bitmap.
We now treat the object registry as system weaks and update the
objects when/if they move. Also added the recent_allocation_records_
as roots.
Bug: 12936165
Change-Id: I615c289efbf2977ceab5c4ffa73d216d799e6e33
|
| a4adbfd44032d70e166e6f18096bbbed05a990ba |
|
05-Feb-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
RosAlloc verification.
If enabled, RosAlloc verification checks the allocator internal
metadata and invariants to detect bugs, heap corruptions, and race
conditions. Added runtime options for enabling and disabling
it. Enable it for the debug build.
Bug: 9986565
Bug: 12592026
Change-Id: I923742b87805ae839f1549d78d0d492733da6a58
|
| 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
|
| 816021e1ef6a9b1751cb455b51f53f98bcac1756 |
|
06-Feb-2014 |
Mathieu Chartier <mathieuc@google.com> |
Remove unused code in heap.h
Change-Id: I77268e92deeade14315e704dcfe77618382c46b8
|
| d5a89ee74c00bd24a34b6e795828a0b86842630f |
|
31-Jan-2014 |
Mathieu Chartier <mathieuc@google.com> |
Fix race conditions caused by StartGC.
Race1: Heap trimming could happen when we were transitioning the heap.
This caused the space to get deleted in the middle of the trim.
Race2: IncrementDisableCompactingGC needed to WaitForConcurrentGC if
we were running a moving GC or about to starting a moving GC.
Race3: The logic for whether or not we had a compacting GC was
calculated before StartGC in CollectGarbageInternal. This could cause
us to get blocked waiting for the GC to complete and come out of the
wait with a new collector_type_ due to a heap transition.
Change-Id: I07c36ae5df1820e9cca70cf239e46175c1eb9575
|
| a1602f28c0e3127ad511712d4b08db89737ae901 |
|
14-Jan-2014 |
Mathieu Chartier <mathieuc@google.com> |
Add zygote space as its own space type.
Helps prevent errors caused from doing invalid operations on the
old alloc space.
Removed some duplicated code in mark_sweep.cc and semi_space.cc.
Change-Id: I67a772cab30d698744c918aad581053f282a4a99
|
| 1d27b34d3b18a5a0c832dae9768366dc08ef8d1c |
|
28-Jan-2014 |
Mathieu Chartier <mathieuc@google.com> |
Change DisableGC to DisableMovingGC.
Also removed the WaitForConcurrentGC in IncrementDisableMovingGC
since we do not currently support any type of concurrent moving
collectors.
This fixes the performance regression introduced by waiting for the
concurrent GC which manifested itself in framework perf benchmarks
as a result of background compaction.
Change-Id: I524f9ab52e1992419626a27649f232ca6967b03d
|
| 9be9a7a344096e040f2edfbdf8e479634058a2d7 |
|
24-Jan-2014 |
Mathieu Chartier <mathieuc@google.com> |
Improve IsMovableObject logic.
The new logic says the main space is non movable iff neither the
background collector or foreground collector are compacting.
Temporarily fixes performance regressions and stability issues
caused by buggy JNI code.
Also addressed a few review comments from previous CLs.
Change-Id: Id5f91765b5fbdae387b566c0bda3392557218ba1
|
| 938a03b3f196895e1fce2b862ae01039deca4c34 |
|
17-Jan-2014 |
Mathieu Chartier <mathieuc@google.com> |
Fix heap verification and add runtime options.
Post GC heap verification is now enabled by default for debug builds.
Added four Xgc runtime options: preverify, nopreverify, postverify,
nopostverify.
Fixed another bug where the non-moving space could theoretically end
up outside of the card table address range.
Bug: 10672951
Change-Id: I5ba3cd96dbfb900263dba795242df74c563deaa3
|
| 6f4ffe41649f1e6381e8cda087ad3749206806e5 |
|
13-Jan-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Improve the generational mode.
- Turn the compile-time flags for generational mode into a command
line flag.
- In the generational mode, always collect the whole heap, as opposed
to the bump pointer space only, if a collection is an explicit,
native allocation-triggered or last attempt one.
Change-Id: I7a14a707cc47e6e3aa4a3292db62533409f17563
|
| 05e713a3a009a0825826ce6d494582fcd6dd6a8d |
|
09-Jan-2014 |
Hiroshi Yamauchi <yamauchi@google.com> |
Bump pointer space only collection.
Add a mode of collection that collects only the bump pointer spaces,
as opposed to the whole heap. It's disabled behind a flag.
TODO: It still scans the entire non-moving space to look for
inter-space references. Use a card table like technique to limit the
scope of scanning and speed it up.
Bug: 11650816
Change-Id: I56cb11e78e47a578bff644e6e6c63d978cfedf39
|
| db7f37d57b6ac83abe6815d0cd5c50701b6be821 |
|
10-Jan-2014 |
Mathieu Chartier <mathieuc@google.com> |
Refactor large object sweeping.
Moved basic sweeping logic into large_object_space.cc.
Renamed SpaceSetMap -> ObjectSet.
Change-Id: I938c1f29f69b0682350347da2bd5de021c0e0224
|
| fc5b528fc46ccff655cfb3578847350f74064e8d |
|
10-Jan-2014 |
Mathieu Chartier <mathieuc@google.com> |
Fix IsMovableObject to handle background compaction.
Before we assumed that only the bump pointer space contained
movable objects. This is incorrect since the background compaction
foreground -> background transition copies objects from the
RosAlloc / DlMalloc space (now called main space) to the bump pointer
space.
Bug: 8981901
Change-Id: Ie7bb9a5d82012988573e944ede2f6a8a77d8872f
|
| e6da9af8dfe0a3e3fbc2be700554f6478380e7b9 |
|
16-Dec-2013 |
Mathieu Chartier <mathieuc@google.com> |
Background compaction support.
When the process state changes to a state which does not perceives
jank, we copy from the main free-list backed allocation space to
the bump pointer space and enable the semispace allocator.
When we transition back to foreground, we copy back to a free-list
backed space.
Create a seperate non-moving space which only holds non-movable
objects. This enables us to quickly wipe the current alloc space
(DlMalloc / RosAlloc) when we transition to background.
Added multiple alloc space support to the sticky mark sweep GC.
Added a -XX:BackgroundGC option which lets you specify
which GC to use for background apps. Passing in
-XX:BackgroundGC=SS makes the heap compact the heap for apps which
do not perceive jank.
Results:
Simple background foreground test:
0. Reboot phone, unlock.
1. Open browser, click on home.
2. Open calculator, click on home.
3. Open calendar, click on home.
4. Open camera, click on home.
5. Open clock, click on home.
6. adb shell dumpsys meminfo
PSS Normal ART:
Sample 1:
88468 kB: Dalvik
3188 kB: Dalvik Other
Sample 2:
81125 kB: Dalvik
3080 kB: Dalvik Other
PSS Dalvik:
Total PSS by category:
Sample 1:
81033 kB: Dalvik
27787 kB: Dalvik Other
Sample 2:
81901 kB: Dalvik
28869 kB: Dalvik Other
PSS ART + Background Compaction:
Sample 1:
71014 kB: Dalvik
1412 kB: Dalvik Other
Sample 2:
73859 kB: Dalvik
1400 kB: Dalvik Other
Dalvik other reduction can be explained by less deep allocation
stacks / less live bitmaps / less dirty cards.
TODO improvements: Recycle mem-maps which are unused in the current
state. Not hardcode 64 MB capacity of non movable space (avoid
returning linear alloc nightmares). Figure out ways to deal with low
virtual address memory problems.
Bug: 8981901
Change-Id: Ib235d03f45548ffc08a06b8ae57bf5bada49d6f3
|
| 692fafd9778141fa6ef0048c9569abd7ee0253bf |
|
30-Nov-2013 |
Mathieu Chartier <mathieuc@google.com> |
Thread local bump pointer allocator.
Added a thread local allocator to the heap, each thread has three
pointers which specify the thread local buffer: start, cur, and
end. When the remaining space in the thread local buffer isn't large
enough for the allocation, the allocator allocates a new thread
local buffer using the bump pointer allocator.
The bump pointer space had to be modified to accomodate thread
local buffers. These buffers are called "blocks", where a block
is a buffer which contains a set of adjacent objects. Blocks
aren't necessarily full and may have wasted memory towards the
end. Blocks have an 8 byte header which specifies their size and is
required for traversing bump pointer spaces.
Memory usage is in between full bump pointer and ROSAlloc since
madvised memory limits wasted ram to an average of 1/2 page per
block.
Added a runtime option -XX:UseTLAB which specifies whether or
not to use the thread local allocator. Its a NOP if the garbage
collector is not the semispace collector.
TODO: Smarter block accounting to prevent us reading objects until
we either hit the end of the block or GetClass() == null which
signifies that the block isn't 100% full. This would provide a
slight speedup to BumpPointerSpace::Walk.
Timings: -XX:HeapMinFree=4m -XX:HeapMaxFree=8m -Xmx48m
ritzperf memalloc:
Dalvik -Xgc:concurrent: 11678
Dalvik -Xgc:noconcurrent: 6697
-Xgc:MS: 5978
-Xgc:SS: 4271
-Xgc:CMS: 4150
-Xgc:SS -XX:UseTLAB: 3255
Bug: 9986565
Bug: 12042213
Change-Id: Ib7e1d4b199a8199f3b1de94b0a7b6e1730689cad
|
| c528dba35b5faece51ca658fc008b688f8b690ad |
|
26-Nov-2013 |
Mathieu Chartier <mathieuc@google.com> |
Enable moving classes.
Slight reduction in Zygote size, memory savings are in the noise.
Before: Zygote size: 8739224
After: Zygote size: 8733568
Fixed a bug where we didn't set the concurrent start bytes after
switching the allocator from bump pointer to ROSAlloc in the
zygote. This caused excessive memory usage.
Added the method verifiers as roots to fix an issue caused by
RegTypes holding a Class*.
Added logic to clear card table in the SemiSpace collector, this
reduces DalvikOther from ~2400k -> ~1760k when using the SemiSpace
collector.
Added a missing lock to the timing loggers which caused a rare
one time crash in std::set.
Bug: 11771255
Bug: 8499494
Bug: 10802951
Change-Id: I99d2b528cd51c1c5ed7012e3220b3aefded680ae
|
| 7bf82af01ec250a4ed2cee03a0e51d179fa820f9 |
|
07-Dec-2013 |
Mathieu Chartier <mathieuc@google.com> |
Fix memory usage regression and clean up collector changing code.
Memory usage regressed since we didn't properly update
concurrent_start_bytes_ when changing collectors.
Bug: 12034247
Change-Id: I1c69e71cd2919e0d3bf75485a4ac0b0aeca59278
|
| ca2a24da53869a04e1947aa46d06ccce5247d6f4 |
|
26-Nov-2013 |
Mathieu Chartier <mathieuc@google.com> |
Add interface for updating process state.
Called from activity manager to let the heap know when it should
perform compaction and trimming.
Bug: 8981901
Change-Id: Ib8ea48d2dc9d6901c3f2e0554391721d6691e726
|
| 0de9f73afe3e835b63f2ee0c1416930656449f3f |
|
23-Nov-2013 |
Mathieu Chartier <mathieuc@google.com> |
Add -xGc: MS, CMS, SS options to specify which GC to use.
Can be used for running tests or benchmarks with semispace,
marksweep or concurrent marksweep.
Change-Id: Ic9ab1220150f2c7c9c30df4ffee45b9d303094b3
|
| 2e899a92439dc6bdaaa67b8230933006284aa600 |
|
23-Nov-2013 |
Hiroshi Yamauchi <yamauchi@google.com> |
Fix a crash with -XX:DumpGCPerformanceOnShutdown.
DumpGcPerformanceInfo() could call RosAllocSpace::InspectAllRosAlloc()
which needs the thread list to be still alive. Fix by moving the
DumpGcPerformanceInfo() call from the Heap destructor up to the
beginning of the Runtime destructor so that the thread list is still
alive when it's called.
Bug: 11830901
Change-Id: Ib094d60916943c8cb1d4b769d805b4ca03269f90
|
| ff3b24aa929a9db79daeef7c0b0522da099700a9 |
|
23-Nov-2013 |
Mathieu Chartier <mathieuc@google.com> |
Add developer option for dumping GC cumulative timings on shutdown.
The option is "-XX:DumpGCPerformanceOnShutdown".
Bug: 9986416
Change-Id: If6ebb26b3e611a9dead197740dbfc64e548dc388
|
| 50482234bed852766498321f71d2ff5e46e4fec2 |
|
21-Nov-2013 |
Mathieu Chartier <mathieuc@google.com> |
Compact zygote.
We now start out using the bump pointer allocator and switch to
the free list collector (ROSAlloc) after the zygote forks.
Before compaction:
Zygote size: 9060352
After compaction
Zygote size: 8425864
The main reason the size doesn't reduce more is that most of the
zygote space is non-movable objects allocated by
VMRuntime.newNonMovableObject. The objects which are non-movable
but could be movable include around 10000 classes and some number
of fields and methods.
Bug: 8981901
Change-Id: Iea21b70fb7af27cb7e92d72070d278a5cd4026ac
|
| 4b5553031edf6ca5212f4eb183073e43abac2b0d |
|
21-Nov-2013 |
Mathieu Chartier <mathieuc@google.com> |
Fix memory leak caused by not adding lage objects to allocation stack.
Large objects weren't being added to allocation stack. This was
causing them to never be marked as live and therefore never freed.
Change-Id: Ie84e3cd7a417a89870752cb21bd7d562c3427284
|
| 1febddf359ae500ef1bb01ab4883b076fcb56440 |
|
20-Nov-2013 |
Mathieu Chartier <mathieuc@google.com> |
Set array length before fence in allocation code path.
Could not delete SetLength since it is required by
space_test.
Bug: 11747779
Change-Id: Icf1ead216b6ff1b519240ab0d0ca30d68429d5b6
|
| cbb2d20bea2861f244da2e2318d8c088300a3710 |
|
15-Nov-2013 |
Mathieu Chartier <mathieuc@google.com> |
Refactor allocation entrypoints.
Adds support for switching entrypoints during runtime. Enables
addition of new allocators with out requiring significant copy
paste. Slight speedup on ritzperf probably due to more inlining.
TODO: Ensuring that the entire allocation path is inlined so
that the switch statement in the allocation code is optimized
out.
Rosalloc measurements:
4583
4453
4439
4434
4751
After change:
4184
4287
4131
4335
4097
Change-Id: I1352a3cbcdf6dae93921582726324d91312df5c9
|
| d31fb9718a6180304cd951619dc36be8e090a641 |
|
19-Nov-2013 |
Hiroshi Yamauchi <yamauchi@google.com> |
Enable rosalloc by default.
Bug: 9986565
Change-Id: I485341d30a21704ddbc45d8f531ef5593a358bb5
|
| cf58d4adf461eb9b8e84baa8019054c88cd8acc6 |
|
26-Sep-2013 |
Hiroshi Yamauchi <yamauchi@google.com> |
A custom 'runs-of-slots' memory allocator.
Bug: 9986565
Change-Id: I0eb73b9458752113f519483616536d219d5f798b
|
| 39e3261168e7761fea6d873494d7c5d191285791 |
|
13-Nov-2013 |
Mathieu Chartier <mathieuc@google.com> |
Refactor reference queues.
Refactored the reference queue processing to reside in the heap code.
This removes significant code duplication in the semispace and
marksweep garbage collectors.
Changed the soft reference behaviour to preserve all soft references
unless the GC requires them to be cleared to avoid an out of memory
error. It may be worth investigating a better heuristic in the
future to preserve soft references by LRU order.
Change-Id: I1f3ff5bd4b3c5149271f4bb4fc94ba199e2f9bc2
|
| 5fe9af720048673e62ee29597a30bb9e54c903c5 |
|
14-Nov-2013 |
Ian Rogers <irogers@google.com> |
Fix memory leaks relating to timing logger.
Bug: 11670287.
We use pointers to uninitialized values for control-flow in the timing logger
code, add TODO comments to clean this up later.
Remove base namespace and other bits of tidying.
Change-Id: I1e6600a1e92f974c8f58f3a405a4e4abb4d9f80f
|
| 590fee9e8972f872301c2d16a575d579ee564bee |
|
13-Sep-2013 |
Mathieu Chartier <mathieuc@google.com> |
Compacting collector.
The compacting collector is currently similar to semispace. It works by
copying objects back and forth between two bump pointer spaces. There
are types of objects which are "non-movable" due to current runtime
limitations. These are Classes, Methods, and Fields.
Bump pointer spaces are a new type of continuous alloc space which have
no lock in the allocation code path. When you allocate from these it uses
atomic operations to increase an index. Traversing the objects in the bump
pointer space relies on Object::SizeOf matching the allocated size exactly.
Runtime changes:
JNI::GetArrayElements returns copies objects if you attempt to get the
backing data of a movable array. For GetArrayElementsCritical, we return
direct backing storage for any types of arrays, but temporarily disable
the GC until the critical region is completed.
Added a new runtime call called VisitObjects, this is used in place of
the old pattern which was flushing the allocation stack and walking
the bitmaps.
Changed image writer to be compaction safe and use object monitor word
for forwarding addresses.
Added a bunch of added SIRTs to ClassLinker, MethodLinker, etc..
TODO: Enable switching allocators, compacting on background, etc..
Bug: 8981901
Change-Id: I3c886fd322a6eef2b99388d19a765042ec26ab99
|
| 0732d59d42523b8c2d807de3a380d2fbfa781364 |
|
06-Nov-2013 |
Mathieu Chartier <mathieuc@google.com> |
Improve object clone performance and make compaction proof.
Bug: 11551604
Bug: 8981901
Change-Id: I60646d838dbb51e125303d1a8fe869191aa63e78
|
| 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
|
| 1eb512d33f94d1dd7ea38263307ba0f7a0dfa653 |
|
19-Oct-2013 |
Ian Rogers <irogers@google.com> |
Fast JNI support.
Use a modifier to signal a native method is a fast JNI method. If the
modifier is set then don't perform runnable transitions.
Change-Id: I7835b4d837bfdd1cb8e2d54b919c0d5e6cf90499
|
| 3b4c18933c24b8a33f38573c2ebcdb9aa16efeb5 |
|
13-Sep-2013 |
Hiroshi Yamauchi <yamauchi@google.com> |
Split the allocation path into 'instrumented' and 'uninstrumented'
ones.
The instrumented path is equivalent to the existing allocation path
that checks for three instrumentation mechanisms (the debugger
allocation tracking, the runtime allocation stats collection, and
valgrind) for every allocation. The uinstrumented path does not
perform these checks. We use the uninstrumented path by default and
enable the instrumented path only when any of the three mechanisms is
enabled. The uninstrumented version of Heap::AllocObject() is inlined.
This change improves the Ritz MemAllocTest by ~4% on Nexus 4 and ~3%
on Host/x86.
Bug: 9986565
Change-Id: I3e68dfff6789d77bbdcea98457b694e1b5fcef5f
|
| 11409ae81a3eaf84d7fd2b3c85b8b06d2bae27f0 |
|
23-Sep-2013 |
Mathieu Chartier <mathieuc@google.com> |
Refactor and improve mod-union tables.
Allow support for adding more mod union tables, reduces the amount
of baked in logic. Adds support for updating mod union table references
from compaction (not for ReferenceCache table yet).
Change-Id: I1beeda00839ed86ef0e853beff5ce10d0ab2b9d1
|
| 0f72e4136aecaf6976fdb55916bbd7b6d5c9c77b |
|
07-Sep-2013 |
Mathieu Chartier <mathieuc@google.com> |
Improve heap verification.
Re-enabled checking the allocation stack for heap verification.
Added tracking of recent frees in DlMallocSpace if debug spaces is
enabled. This is useful when you have heap corruption caused by a
live object referencing a recently freed object.
Added various other sanity checks in the GC.
Bug: 10626133
Change-Id: I5ada11966336ae9a06615b16f4b933f05b5d0c32
|
| b4ea4de2d6b63a3855968f2748878018a27af106 |
|
18-Sep-2013 |
Mathieu Chartier <mathieuc@google.com> |
Fix soft reference clearing issue.
There was a bug where we would check that the pending next field was
non null before enqueueing up cleared references. This was causing
references to not get queued up during ProcessReferences.
Bug: 10626133
Change-Id: Ic1e00e42045092280b4abb3d41f1c58f7adbc3de
|
| 2775ee4f82dff260663ca16adddc0b15327aaa42 |
|
21-Aug-2013 |
Mathieu Chartier <mathieuc@google.com> |
Add more runtime options.
Changed HeapGCThreads to be split into two different options:
-XX:ParallelGCThreads: Which specifies how many threads the GC may
use when the mutators are suspended.
-XX:ConcGCThreads: Which specifies how many threads the GC may use
when the mutators are running.
Added runtime options to specify long pause / long GC thresholds:
-XX:LongPauseThreshold (default 5ms)
-XX:LongGCThreshold (default 100ms)
These thresholds were previously constants, but are now runtime
options. If we exceed either of the thresholds, we print the GC
message.
Added a new runtime option: -XX:IgnoreMaxFootprint which makes it
that the GC only does GC when the number of bytes allocated hits
the growth limit. This causes GC to occur much less frequently and
can be useful to measure how much of an impact GC has on performance.
Changed the GC behaviour to use only one thread when we do not care
about pauses to prevent jank that can be caused by 2 simultaneous GC
on different processes fighting for CPU time.
Added thread pool functionality for changing the maximum number of
active workers.
Fixed an accounting error where we didn't count large objects in the
total freed.
Bug: 9986416
Change-Id: I86afa358d93dcd3780e18ac5d85bdb1a130cb7e7
|
| 720ef7680573c1afd12f99f02eee3045daee5168 |
|
17-Aug-2013 |
Mathieu Chartier <mathieuc@google.com> |
Fix non concurrent GC ergonomics.
If we dont have concurrent GC enabled, we need to force GC for alloc
when we hit the maximum allowed footprint so that our heap doesn't
keep growing until it hits the growth limit.
Refactored a bit of stuff.
Change-Id: I8eceac4ef01e969fd286ebde3a735a09d0a6dfc1
|
| 94c32c5f01c7d44781317bf23933ed0a5bc4b796 |
|
09-Aug-2013 |
Mathieu Chartier <mathieuc@google.com> |
More parallel GC, rewritten parallel mark stack processing.
Card scanning may now be done in parallel. This speeds up sticky and
reduces pause times for all GC types.
Speedup on my mako (ritz perf):
Average pause time for sticky GC (~250 samples):
Without parallel cards scanning enabled: 2.524904215ms
Parallel card scanning (num_gc_threads_): 1.552123552ms
Throughput (~250 samples):
Sticky GC throughput with parallel card scanning: 69MB/s
Sticky GC throughput without parallel card scanning: 51MB/s
Rewrote the mark stack processing to be LIFO and use a prefetch queue
like the non parallel version.
Cleaned up some of the logcat printing for the activity manager
process state listening.
Added unlikely hints to object scanning since arrays and classes are
scanned much less often than normal objects.
Fixed a bug where the number of GC threads was clamped to 1 due to a
bool instead of a size_t.
Fixed a race condition when we added references to the reference
queues. Sharded the reference queue lock into one lock for each reference
type (weak, soft, phatom, finalizer).
Changed timing splits to be different for processing gray objects with
and without mutators paused since sticky GC does both.
Mask out the class bit when visiting fields as an optimization, this is
valid since classes are held live by the class linker.
Partially completed: Parallel recursive mark + finger.
Bug: 10245302
Bug: 9969166
Bug: 9986532
Bug: 9961698
Change-Id: I142d09718c4609b7c2387cb28f517a6983c73288
|
| f981da1d60864a730f744ef2cc3a19391c8303f2 |
|
08-Aug-2013 |
Mathieu Chartier <mathieuc@google.com> |
Use JNI to access activity manager process state.
After a GC occurs, we use JNI to access activity manager process
state to decide whether or not we will trim.
Change-Id: Iad981e3a7cdc694729d8792cc0f19e0262154388
(cherry picked from commit c39e342317d77e701b4cd01cd5b05902e6512f4b)
|
| c39e342317d77e701b4cd01cd5b05902e6512f4b |
|
08-Aug-2013 |
Mathieu Chartier <mathieuc@google.com> |
Use JNI to access activity manager process state.
After a GC occurs, we use JNI to access activity manager process
state to decide whether or not we will trim.
Change-Id: Iad981e3a7cdc694729d8792cc0f19e0262154388
|
| 9642c96bd5a1ccc4e221de9c0af4a545af8182d2 |
|
06-Aug-2013 |
Mathieu Chartier <mathieuc@google.com> |
Move allocation stack unmarking outside of pause.
Reduces pause time by moving the allocation stack unmarking outside
of the pause. This is especially helpful for devices which have longer
GC times since these times result in having more things to be
unmarked in the allocation stack.
Bug: 9969166
Change-Id: I570f2213cbdda9d90545b64538e2cbeb0dc32d16
|
| e0a53e99e2a01f8668d6616c3cec7e2f5a711286 |
|
05-Aug-2013 |
Mathieu Chartier <mathieuc@google.com> |
Add low memory mode option to ART.
Useful so that we match the option I added here:
https://googleplex-android-review.googlesource.com/#/c/328940/
In ART low memory mode reduces the maximum number of histogram
buckets. We also trim no matter the utilization.
Change-Id: I655ba63312c0a6574569cdd5171ca81ea338c2aa
|
| 50b2928501fe489c108472e7648ec98cdca62e10 |
|
30-Jul-2013 |
Hiroshi Yamauchi <yamauchi@google.com> |
Improve the allocation speed.
- Improves the Ritz MemAllocTest benchmark result by ~500 ms (or ~5%) on Nexus 4.
- Move the memset() call that zeroes the allocated memory out of the lock region.
- De-virtualize/Inline the allocation call chains into Heap::AllocObject().
- Turn Heap::measure_allocation_time_ into a static const variable.
- Surround the VerifyObject() call with kIsDebugBuild.
Bug: 9986565
Change-Id: Ib70b6d051a80ec329788b30256565561f031da2a
|
| 834b394ee759ed31c5371d8093d7cd8cd90014a8 |
|
31-Jul-2013 |
Brian Carlstrom <bdc@google.com> |
Merge remote-tracking branch 'goog/dalvik-dev' into merge-art-to-dalvik-dev
Change-Id: I323e9e8c29c3e39d50d9aba93121b26266c52a46
|
| 58b42b986742565cccfde4366495ad23d0ba57d1 |
|
26-Jul-2013 |
Mathieu Chartier <mathieuc@google.com> |
Fix portable build.
Move dlmalloc include inside of cc file.
Change-Id: Ic36bdbf9fa964b93a23d1a45f603bab1bf9583d7
|
| 6f28d91aab952e3244fbb4e707fa38f85538f374 |
|
25-Jul-2013 |
Anwar Ghuloum <anwarg@google.com> |
Add systrace support to NewTimingLogger, migrate compiler timing logging to NewTimingLogger
Rpleaced old TimingLogger by NewTimingLogger, renamed NewTimingLogger to TimingLogger, added systrace support to TimingLogger.
Tests passing, phone booting, systrace working.
Change-Id: I2aeffb8bcb7f0fd979d8a2a3a8bcfbaa02413679
|
| 0a9dc05e704bfd033bac2aa38a4fc6f6b8e6cf93 |
|
25-Jul-2013 |
Mathieu Chartier <mathieuc@google.com> |
GC data structures allocation tracking
Adds a new stl compatible allocator that is used in most GC data
structures. When the data structures allocate and free memory, it
lets the heap know of how much memory was allocated or freed. Using
this info, we dump the approximated stl data structures memory usage
when a sigquit occurs.
The allocation tracking can be disabled with a compile time boolean
flag to remove performance impact.
Change-Id: Idddb6713169e07be913bceeb50f305c8573e4392
|
| 63a54345598861030178e033ffbd72c0e231a4c9 |
|
23-Jul-2013 |
Mathieu Chartier <mathieuc@google.com> |
Add option for changing number of GC threads.
The number of threads was previously set to 1 by an accidential
checkin. This hurts on all devices which aren't dual core. We now
properly use sysconf to determine how many threads we should create.
Also added a -XX:HeapGCThreads heap option which lets us change
how many GC threads we create. The default value is equal to the
number of processors on the device minus one.
Change-Id: If65065ef09174a3813b8741efdd5ea7bbe82a4e2
|
| 09b07a96094086e205948717666025909a75163b |
|
15-Jul-2013 |
Hiroshi Yamauchi <yamauchi@google.com> |
Show size/alloc/free per Dalvik heap space in dumpsys
Add the heap size/alloc/free stats to the Dalvik heap space breakdown section in dumpsys meminfo.
Also, now the zygote heap has a distict ashmem region name.
Bug: 9532137
Bug: 8266259
Change-Id: Ieeb02f5f5ebf7ffe35d4b55ad81d46989af2827b
|
| 82353313bbe00d033b6733535d0dcd6f56c36d8d |
|
18-Jul-2013 |
Mathieu Chartier <mathieuc@google.com> |
Add interface for updating the heap's process state
In the future, the heap's process state will be used
to determine GC behavior.
Change-Id: Iba4f038d28dbf483b6573d8feb25e4246ead1d50
|
| df62950e7a32031b82360c407d46a37b94188fbb |
|
18-Jul-2013 |
Brian Carlstrom <bdc@google.com> |
Fix cpplint whitespace/parens issues
Change-Id: Ifc678d59a8bed24ffddde5a0e543620b17b0aba9
|
| fc0e3219edc9a5bf81b166e82fd5db2796eb6a0d |
|
17-Jul-2013 |
Brian Carlstrom <bdc@google.com> |
Fix multiple inclusion guards to match new pathnames
Change-Id: Id7735be1d75bc315733b1773fba45c1deb8ace43
|
| 56d947fbc9bc2992e2f93112fafb73e50d2aaa7a |
|
15-Jul-2013 |
Brian Carlstrom <bdc@google.com> |
Add verification of boot.oat generated on device
Change-Id: I069586205a9a92fc7375ccf5cdde136bbbcfc800
|
| 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
|