| 2d857af288e020909a17dee1cb2b6e227a527d67 |
|
18-Jul-2016 |
Mathieu Chartier <mathieuc@google.com> |
Do allocation fence before pushing on allocation stack
Heap::VisitObjects relies on having valid classes for objects in
the allocation stack. If the writes reorder, the thread calling
VisitObjects could see the free list pointer instead of the class
of the object. I believe this is causing crashes in VisitObjects.
Bug: 28790624
Test: Volantis booted
(cherry picked from commit 011dc2c4b9f3a064cba801679aedd3251fe191e3)
Change-Id: Ib620acbcd641ccd59fea9a66f5587c5026110a22
|
| a048ca8e0ff6d0e79d070decf9fa2331e2d13824 |
|
08-Jun-2016 |
Nicolas Geoffray <ngeoffray@google.com> |
Fix FOUR_ARG_DOWNCALL assembly stubs on arm and x86.
They were creating a stack that the runtime did not understand.
bug:28348339
(cherry picked from commit 0f838aa279a296b5f14c231065bb2f96b02d9caf)
Change-Id: Iecaa274512b42ad76fccb48c11a0076d0292c240
|
| fc6898769ae1ef91ec3e41c0a273401213cb82cd |
|
26-Apr-2016 |
Andreas Gampe <agampe@google.com> |
ART: Log all monitor operations to systrace
Add a VLOG option ("-verbose:systrace-locks") to log all monitor
operations to systrace. This requires non-fastpath thread
entrypoints, and ATRACE tags for locking and unlocking.
Do a bit of cleanup to the entrypoint initialization to share
common setup.
Bug: 28423466
Change-Id: Ie67e4aa946ec15f8fcf8cb7134c5d3cff0119ab3
|
| 59028d90d51a800bcea8be354d77d7be924da3a0 |
|
29-Mar-2016 |
Goran Jakovljevic <Goran.Jakovljevic@imgtec.com> |
MIPS: Improving art_quick_imt_conflict_trampoline
This is fixing stub_test for MIPS32 and MIPS64. This is follow up
change for Ie74d1c77cf73d451a1142bdc5e3683f9f84bb4e7.
Change-Id: I3c53ef690aff49d7cf9ad3de3aaed9a3d2e1c6b9
|
| 796d63050a18f263b93ea34951a61deaecab3422 |
|
13-Mar-2016 |
Nicolas Geoffray <ngeoffray@google.com> |
Add an ImtConflictTable to better resolve IMT conflicts.
- Attach a ImtConflictTable to conflict runtime ArtMethod.
- Initially 0, a new one will be created at the first hit of
the conflict method.
- If the assembly code does not find a target method in the table,
we will create a new one again, copying the data from the previous
table and adding the new mapping.
Implemented for arm/arm64/x86/x64.
bug:27556801
bug:24769046
Change-Id: Ie74d1c77cf73d451a1142bdc5e3683f9f84bb4e7
|
| d72945c6b18831630af8755fcf041f5a59c82cae |
|
16-Mar-2016 |
Hiroshi Yamauchi <yamauchi@google.com> |
Revert "Revert "Assembly TLAB allocation fast path for arm64.""
This reverts commit 52fa2c698b995c21940f366cf3a44204ddf4f8e9.
Fix the mvn instructions.
Bug: 9986565
Change-Id: Ib7b2023cd54c57131442e1de85c64f40b818313d
|
| 52fa2c698b995c21940f366cf3a44204ddf4f8e9 |
|
16-Mar-2016 |
Nicolas Geoffray <ngeoffray@google.com> |
Revert "Assembly TLAB allocation fast path for arm64."
Assembly code does not compile
Bug: 9986565
This reverts commit b7e52b02a82c69e59c88f06945fb21672dfe9923.
Change-Id: I80de1bcd6270e2d76f79cb708811a41c039ea5bb
|
| 563cf332e4560601207b2d039b42f6b0fdff8ce8 |
|
16-Mar-2016 |
Nicolas Geoffray <ngeoffray@google.com> |
Revert "Tentative fix for build breakage with read barriers."
This reverts commit 448f21e2c1cb43b193736598719979ebd2aed0b0.
Change-Id: Id821effa06697b1464581bcd9c8f34cd0ea69896
|
| 448f21e2c1cb43b193736598719979ebd2aed0b0 |
|
16-Mar-2016 |
Nicolas Geoffray <ngeoffray@google.com> |
Tentative fix for build breakage with read barriers.
Change-Id: I490821357525bdeb19c330f350635bd645f8ab53
|
| b7e52b02a82c69e59c88f06945fb21672dfe9923 |
|
10-Mar-2016 |
Hiroshi Yamauchi <yamauchi@google.com> |
Assembly TLAB allocation fast path for arm64.
This is the arm64 version of CL 187537.
Speedup (GSS GC with TLAB on N9):
BinaryTrees: 591 -> 493 ms (-17%)
MemAllocTest: 792 -> 755 ms (-5%)
Bug: 9986565
Change-Id: Icdad28cab0fd835679c640b7eae59b33ac2d6654
|
| 10d4c08c0ea9df0a85a11e1c77974df24078c0ec |
|
24-Feb-2016 |
Hiroshi Yamauchi <yamauchi@google.com> |
Assembly region TLAB allocation fast path for arm.
This is for the CC collector.
Share the common fast path code with the tlab fast path code.
Speedup (on N5):
BinaryTrees: 2291 -> 902 ms (-60%)
MemAllocTest: 2137 -> 1845 ms (-14%)
Bug: 9986565
Bug: 12687968
Change-Id: Ica63094ec2f85eaa4fd04d202a20090399275d85
|
| f969a209c30e3af636342d2fb7851d82a2529bf7 |
|
09-Mar-2016 |
Roland Levillain <rpl@google.com> |
Fix and enable java.lang.StringFactory intrinsics.
The following intrinsics were not considered by the
intrinsics recognizer:
- StringNewStringFromBytes
- StringNewStringFromChars
- StringNewStringFromString
This CL enables them and add tests for them.
This CL also:
- Fixes the locations of the ARM64 & MIPS64
StringNewStringFromString intrinsics.
- Fixes the definitions of the FOUR_ARG_DOWNCALL macros on
ARM and x86, which are used to implement the
art_quick_alloc_string_from_bytes* runtime entry points.
- Fixes PC info (stack maps) recording in the
StringNewStringFromBytes, StringNewStringFromChars and
StringNewStringFromString ARM, ARM64 & MIPS64 intrinsics.
Bug: 27425743
Change-Id: I38c00d3f0b2e6b64f7d3fe9146743493bef9e45c
|
| 7e1ce285c3c2cbd8d97c744699c2f2347fa7487b |
|
12-Dec-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
Assembly TLAB allocation fast path for arm.
Speedup (GSS GC with TLAB on N5):
BinaryTrees: 1872 -> 796 ms (-57%)
MemAllocTest: 2522 -> 2219 ms (-12%)
Bug: 9986565
Change-Id: Icb9d1259461f3abe83a4a82c8aff911937eaf57d
|
| d9994f069dfeaa32ba929ca78816b5b83e2a4134 |
|
11-Feb-2016 |
Nicolas Geoffray <ngeoffray@google.com> |
Re-enable OSR.
Fixes two bugs:
- Dealing with proxy methods, which the compiler and code cache
does not handle.
- Dealing with phi types, that may have been speculatively optimized
but do not hold once jumping to the compiled code.
Change-Id: I7dcd9976ef7b12128fff95d2b7ed3e69cc42e90a
|
| d9bc433a89c41a255d1b669d075f802597839bdc |
|
06-Feb-2016 |
Nicolas Geoffray <ngeoffray@google.com> |
Point fixes after OSR change.
- ldr -> ldrb in osr stub for arm32
- disable 570-checker-osr for tracing
- don't osr proxy methods.
Change-Id: I9c713c9b7eab86ca9beb75f228fb3b76185621ef
|
| b331febbab8e916680faba722cc84b66b84218a3 |
|
05-Feb-2016 |
Nicolas Geoffray <ngeoffray@google.com> |
Revert "Revert "Implement on-stack replacement for arm/arm64/x86/x86_64.""
This reverts commit bd89a5c556324062b7d841843b039392e84cfaf4.
Change-Id: I08d190431520baa7fcec8fbdb444519f25ac8d44
|
| bd89a5c556324062b7d841843b039392e84cfaf4 |
|
05-Feb-2016 |
David Brazdil <dbrazdil@google.com> |
Revert "Implement on-stack replacement for arm/arm64/x86/x86_64."
DCHECK whether loop headers are covered fails.
This reverts commit 891bc286963892ed96134ca1adb7822737af9710.
Change-Id: I0f9a90630b014b16d20ba1dfba31ce63e6648021
|
| 891bc286963892ed96134ca1adb7822737af9710 |
|
29-Jan-2016 |
Nicolas Geoffray <ngeoffray@google.com> |
Implement on-stack replacement for arm/arm64/x86/x86_64.
High-level overview:
- osr_method_threshold is used to know when to compile a method
in osr mode (-> treat all loops as irreducible).
- branch instructions in the compiler query whether they can
jump to an osr method.
- An osr entry point is found through the stack maps: if a stack
map is duplicated in the CodeInfo, it is an osr entry point.
Change-Id: Ifb39338cd281e2c7eccce67f4e18d46428be71e4
|
| a7a4759946d9f11c88dc108b2b6a9518ce9c1e18 |
|
24-Nov-2015 |
Nicolas Geoffray <ngeoffray@google.com> |
Revert "lambda: Add support for invoke-interface for boxed innate lambdas"
955-lambda is flaky
Bug: 24618608
Bug: 25107649
This reverts commit 457e874459ae638145cab6d572e34d48480e39d2.
(cherry picked from commit 3a0909248e04b22c3981cbf617bc2502ed5b6380)
Change-Id: I24884344d21d7a4262e53e3f5dba57032687ddb7
|
| 3a0909248e04b22c3981cbf617bc2502ed5b6380 |
|
24-Nov-2015 |
Nicolas Geoffray <ngeoffray@google.com> |
Revert "lambda: Add support for invoke-interface for boxed innate lambdas"
955-lambda is flaky
Bug: 24618608
Bug: 25107649
This reverts commit 457e874459ae638145cab6d572e34d48480e39d2.
Change-Id: I24884344d21d7a4262e53e3f5dba57032687ddb7
|
| 457e874459ae638145cab6d572e34d48480e39d2 |
|
23-Oct-2015 |
Igor Murashkin <iam@google.com> |
lambda: Add support for invoke-interface for boxed innate lambdas
Lambda closures created with the 'create-lambda' instruction
(termed "innate lambdas") can be turned into an object with 'box-lambda'.
This CL enables support for those kinds of lambdas to work with
'invoke-interface' by generating a proxy class for the lambda.
Note: MIPS32/64 support not included.
Bug: 24618608
Bug: 25107649
Change-Id: Ic8f1bb66ebeaed4097e758a50becf1cff6ccaefb
|
| 3c183cec1fb337ed0247ba500c6949063072f565 |
|
31-Jul-2015 |
Przemyslaw Szczepaniak <pszczepaniak@google.com> |
Fix asm for String#indexOf/compareTo intrinsics
Fixed code for arm arch, it's still broken
on remaining architectures (will be done soon in
following CLs).
OpenJdk string lacks offset and length fields and requires
user to fetch the length from array object.
Change-Id: I682911e4ee72d9597e8a9735ac414c837569eb8e
|
| dc412b6f49a65774b7af654f65cbff619cb7d85a |
|
15-Oct-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
Revert "Revert "Implement rosalloc fast path in assembly for 32 bit arm.""
With a heap poisoning fix.
This reverts commit cf91c7d973f3b2f491abc61d47c141782c96d46e.
Bug: 9986565
Change-Id: Ia72edbde65ef6119e1931a77cc4c595a0b80ce31
|
| cf91c7d973f3b2f491abc61d47c141782c96d46e |
|
15-Oct-2015 |
Nicolas Geoffray <ngeoffray@google.com> |
Revert "Implement rosalloc fast path in assembly for 32 bit arm."
Tentative. Looks like heap poisoning breaks with this change.
bug: 9986565
This reverts commit e6316940db61faead36f9642cce137d41fc8f606.
Change-Id: I5c63758221464fe319315f40ae79c656048faed0
|
| e6316940db61faead36f9642cce137d41fc8f606 |
|
08-Oct-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
Implement rosalloc fast path in assembly for 32 bit arm.
Measurements (N5, ms)
BinaryTrees: 1702 -> 987 (-42%)
MemAllocTest: 2480 -> 2270 (-8%)
Bug: 9986565
Change-Id: I460af3626ad724078463d27cf74a94b7ff7468c5
|
| e460d1df1f789c7c8bb97024a8efbd713ac175e9 |
|
29-Sep-2015 |
Calin Juravle <calin@google.com> |
Revert "Revert "Support unresolved fields in optimizing"
The CL also changes the calling convetion for 64bit static field set
to use kArg2 instead of kArg1. This allows optimizing to keep
the asumptions:
- arm pairs are always of form (even_reg, odd_reg)
- ecx_edx is not used as a register on x86.
This reverts commit e6f49b47b6a4dc9c7684e4483757872cfc7ff1a1.
Change-Id: I93159917565824084abc96775f31be1a4249f2f3
|
| 639bdd13993644a267f177f8f5936496bda65e2b |
|
03-Jun-2015 |
Andreas Gampe <agampe@google.com> |
ART: Single-frame deopt
Add deoptimization of a single frame. Works by removing the managed
code frame and jumping into the quick-to-interpreter bridge, and
the bridge understanding a stored ShadowFrame.
We need a separate fixup pass. For x86, we leave the return address
on the stack so we don't need to push it there.
Bug: 21611912
Change-Id: I06625685ced8b054244f8685ab50b238a705b9d2
|
| 0747466fca310eedea5fc49e37d54f240a0b3c0f |
|
25-Aug-2015 |
Sebastien Hertz <shertz@google.com> |
Revert "Revert "Fix deoptimization with pending exception""
This reverts commit 6e2d5747d00697a25251d25dd33b953e54709507.
Fixes the deoptimization path from compiled code (generated by the
Optimizing compiler) by adding wrapper artDeoptimizeFromCompiledCode.
This wrapper, called through the matching assembler stub
art_quick_deoptimize_from_compiled_code, pushes the deoptimization
context just before deoptimizing the stack.
Bug: 23371176
Bug: 19944235
Change-Id: Ia7082656998aebdd0157438f7e6504c120e10d3e
|
| 6306921722283d2b0f8aac01883ad83215d6e864 |
|
22-Aug-2015 |
Man Cao <manc@google.com> |
Add a missing reader barrier in entrypoint stub
Also refactored some comments.
Change-Id: I5c50f487bf9d71f1be5f6c8814bf039993fc1267
|
| 1aee900d5a0b3a8d78725a7551356bda0d8554e1 |
|
15-Jul-2015 |
Man Cao <manc@google.com> |
Add read barrier support to the entrypoints.
Also remove "THIS_LOAD_REQUIRES_READ_BARRIER" since reading
an ArtMethod* no longer needs read barrier.
stub_test should also work with read barriers now.
Change-Id: I3fba18042de2f867a18dbdc38519986212bd9769
|
| 76914b0fc848fe0deed566af2255db7c183deaf2 |
|
18-Jul-2015 |
Andreas Gampe <agampe@google.com> |
ART: Fix arm32 quick_invoke_stub CFI
Follow-up to 4808846b2a8647a448aaa05d561a4f60b190196b.
Change-Id: I534ba39a4f59eaea9cfec8e81c944675284fc427
|
| 3031c8da0c5009183f770b005c245f9bf2a4d01b |
|
14-Jul-2015 |
Andreas Gampe <agampe@google.com> |
ART: Remove art_quick_invoke_interface_trampoline
The function has only been used by the IMT conflict resolution
trampoline for a while. Merge the two, which saves a branch.
Change-Id: I2f8c9204adf839ddc5459cc04e70d98f858110a1
|
| 3d1e0c971a5022cf3f7d02a45fdfd3381d5a8f3b |
|
11-Jul-2015 |
Andreas Gampe <agampe@google.com> |
ART: Fix arm32 assembly stub CFI
Follow-up to 7ea6a170486d81b127e69673cd1020c4db628c93.
Change-Id: Ib6df28653374994c78cd5586f2381d36b74ca846
|
| c89b047ed85cfabbdc79a7b76be413fb366b501a |
|
24-Jun-2015 |
Andreas Gampe <agampe@google.com> |
ART: Fix arm32 instrumentation exit stub
Properly store, pass and restore floating point result.
Bug: 22047833
(cherry picked from commit 844fb67a8b5638c4e195dbbe510dc1626824a16c)
Change-Id: If5d9efbe63334c9b6353f76ce6e4fe519602482a
|
| 844fb67a8b5638c4e195dbbe510dc1626824a16c |
|
24-Jun-2015 |
Andreas Gampe <agampe@google.com> |
ART: Fix arm32 instrumentation exit stub
Properly store, pass and restore floating point result.
Bug: 22047833
Change-Id: If5d9efbe63334c9b6353f76ce6e4fe519602482a
|
| bfa5eb6e8d15ea73a36f8df449630f285a91e995 |
|
30-May-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
Add heap poisoning support to the entrypoints.
In preparation for full compiler/managed-code support.
Enable stub_test with heap poisoning.
Bug: 12687968
Change-Id: I79fc54ce6386c0a1eb9621759bb4cc23bc393a75
|
| 5b3ee56cd63ee9e3c70c0412d044b81ab9c94513 |
|
14-Apr-2015 |
Roland Levillain <rpl@google.com> |
Delegate long-to-float type conversions to the runtime on ARM.
On ARM, translate long-to-float type conversions (from both
Quick and Optimizing) as calls to the runtime routine
art_l2f, instead of generating ad hoc code, so as to improve
the precision of the conversions.
Bug: 20413424
Change-Id: I8c414ee1c6f4ff1f32ee78f75734cfd3cf579f71
|
| 3d21bdf8894e780d349c481e5c9e29fe1556051c |
|
22-Apr-2015 |
Mathieu Chartier <mathieuc@google.com> |
Move mirror::ArtMethod to native
Optimizing + quick tests are passing, devices boot.
TODO: Test and fix bugs in mips64.
Saves 16 bytes per most ArtMethod, 7.5MB reduction in system PSS.
Some of the savings are from removal of virtual methods and direct
methods object arrays.
Bug: 19264997
(cherry picked from commit e401d146407d61eeb99f8d6176b2ac13c4df1e33)
Change-Id: I622469a0cfa0e7082a2119f3d6a9491eb61e3f3d
Fix some ArtMethod related bugs
Added root visiting for runtime methods, not currently required
since the GcRoots in these methods are null.
Added missing GetInterfaceMethodIfProxy in GetMethodLine, fixes
--trace run-tests 005, 044.
Fixed optimizing compiler bug where we used a normal stack location
instead of double on ARM64, this fixes the debuggable tests.
TODO: Fix JDWP tests.
Bug: 19264997
Change-Id: I7c55f69c61d1b45351fd0dc7185ffe5efad82bd3
ART: Fix casts for 64-bit pointers on 32-bit compiler.
Bug: 19264997
Change-Id: Ief45cdd4bae5a43fc8bfdfa7cf744e2c57529457
Fix JDWP tests after ArtMethod change
Fixes Throwable::GetStackDepth for exception event detection after
internal stack trace representation change.
Adds missing ArtMethod::GetInterfaceMethodIfProxy call in case of
proxy method.
Bug: 19264997
Change-Id: I363e293796848c3ec491c963813f62d868da44d2
Fix accidental IMT and root marking regression
Was always using the conflict trampoline. Also included fix for
regression in GC time caused by extra roots. Most of the regression
was IMT.
Fixed bug in DumpGcPerformanceInfo where we would get SIGABRT due to
detached thread.
EvaluateAndApplyChanges:
From ~2500 -> ~1980
GC time: 8.2s -> 7.2s due to 1s less of MarkConcurrentRoots
Bug: 19264997
Change-Id: I4333e80a8268c2ed1284f87f25b9f113d4f2c7e0
Fix bogus image test assert
Previously we were comparing the size of the non moving space to
size of the image file.
Now we properly compare the size of the image space against the size
of the image file.
Bug: 19264997
Change-Id: I7359f1f73ae3df60c5147245935a24431c04808a
[MIPS64] Fix art_quick_invoke_stub argument offsets.
ArtMethod reference's size got bigger, so we need to move other args
and leave enough space for ArtMethod* and 'this' pointer.
This fixes mips64 boot.
Bug: 19264997
Change-Id: I47198d5f39a4caab30b3b77479d5eedaad5006ab
|
| e401d146407d61eeb99f8d6176b2ac13c4df1e33 |
|
22-Apr-2015 |
Mathieu Chartier <mathieuc@google.com> |
Move mirror::ArtMethod to native
Optimizing + quick tests are passing, devices boot.
TODO: Test and fix bugs in mips64.
Saves 16 bytes per most ArtMethod, 7.5MB reduction in system PSS.
Some of the savings are from removal of virtual methods and direct
methods object arrays.
Bug: 19264997
Change-Id: I622469a0cfa0e7082a2119f3d6a9491eb61e3f3d
|
| 8ea18d0f066f63fa4e5d154f14327468bf288e2b |
|
26-May-2015 |
Nicolas Geoffray <ngeoffray@google.com> |
Pass the dex method index directly to interface trampoline.
This avoids computing the dex pc and re-finding the method
index again. I have kept the code for kDebugBuild.
Change-Id: Icd60e0deade755e32b54021c0875b1af592b8c3e
|
| 7ea6a170486d81b127e69673cd1020c4db628c93 |
|
19-May-2015 |
Nicolas Geoffray <ngeoffray@google.com> |
Don't hardcode the location of the caller.
This is to avoid shooting ourselves in the foot when
dealing with inlined frames. Instead, use common methods
for fetching the caller and its dex pc.
Change-Id: I3467a7b50cf163022d332e80356f0aab747de252
|
| 848f70a3d73833fc1bf3032a9ff6812e429661d9 |
|
15-Jan-2014 |
Jeff Hao <jeffhao@google.com> |
Replace String CharArray with internal uint16_t array.
Summary of high level changes:
- Adds compiler inliner support to identify string init methods
- Adds compiler support (quick & optimizing) with new invoke code path
that calls method off the thread pointer
- Adds thread entrypoints for all string init methods
- Adds map to verifier to log when receiver of string init has been
copied to other registers. used by compiler and interpreter
Change-Id: I797b992a8feb566f9ad73060011ab6f51eb7ce01
|
| 5ea536aa4a6414db01beaf6f8bd8cb9adc5cfc92 |
|
20-Apr-2015 |
Vladimir Marko <vmarko@google.com> |
Remove ArtMethod* parameter from dex cache entry points.
Load the ArtMethod* using an optimized stack walk instead.
This reduces the size of the generated code.
Three of the entry points are called only from a slow-path
and the fourth (InitializeTypeAndVerifyAccess) is rare and
already slow enough that the one or two extra loads
(depending on whether we already have the ArtMethod* in a
register) are insignificant. And as we're starting to use
PC-relative addressing of the dex cache arrays (already
done by Quick for the boot image), having the ArtMethod* in
a register becomes less likely anyway.
Change-Id: Ib19b9d204e355e13bf386662a8b158178bf8ad28
|
| 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
|
| e15ea086439b41a805d164d2beb07b4ba96aaa97 |
|
10-Feb-2015 |
Hiroshi Yamauchi <yamauchi@google.com> |
Reserve bits in the lock word for read barriers.
This prepares for the CC collector to use the standard object header
model by storing the read barrier state in the lock word.
Bug: 19355854
Bug: 12687968
Change-Id: Ia7585662dd2cebf0479a3e74f734afe5059fb70f
|
| 126d65952a03b3e44d5021208673c01920a982a4 |
|
03-Mar-2015 |
Nicolas Geoffray <ngeoffray@google.com> |
Fix generic JNI stubs to not discard the Java native frame.
Change-Id: Ic856b442fdde5ce91673fc5856eb0dfc84c75d28
|
| 1cc7dbabd03e0a6c09d68161417a21bd6f9df371 |
|
18-Dec-2014 |
Andreas Gampe <agampe@google.com> |
ART: Reorder entrypoint argument order
Shuffle the ArtMethod* referrer backwards for easier removal.
Clean up ARM & MIPS assembly code.
Change some macros to make future changes easier.
Change-Id: Ie2862b68bd6e519438e83eecd9e1611df51d7945
|
| 4808846b2a8647a448aaa05d561a4f60b190196b |
|
12-Dec-2014 |
Nicolas Geoffray <ngeoffray@google.com> |
Save all registers in native to Java stubs.
This will make things more friendly when experimenting with the
number of callee saves in optimizing.
Change-Id: Iefd9a2da329a420eb69fc2fa9e91c06bbda30cdb
|
| d0a0b3e189e6f46d410d61dab4d0cce113be41d5 |
|
31-Oct-2014 |
Christopher Ferris <cferris@google.com> |
Fix arm unwind data for art code.
Bug: 17392751
Change-Id: Ic85d4468b0f931aa9e38dee12d466a1e04ebf944
|
| 2d7210188805292e463be4bcf7a133b654d7e0ea |
|
10-Nov-2014 |
Mathieu Chartier <mathieuc@google.com> |
Change 64 bit ArtMethod fields to be pointer sized
Changed the 64 bit entrypoint and gc map fields in ArtMethod to be
pointer sized. This saves a large amount of memory on 32 bit systems.
Reduces ArtMethod size by 16 bytes on 32 bit.
Total number of ArtMethod on low memory mako: 169957
Image size: 49203 methods -> 787248 image size reduction.
Zygote space size: 1070 methods -> 17120 size reduction.
App methods: ~120k -> 2 MB savings.
Savings per app on low memory mako: 125K+ per app
(less active apps -> more image methods per app).
Savings depend on how often the shared methods are on dirty pages vs
shared.
TODO in another CL, delete gc map field from ArtMethod since we
should be able to get it from the Oat method header.
Bug: 17643507
Change-Id: Ie9508f05907a9f693882d4d32a564460bf273ee8
(cherry picked from commit e832e64a7e82d7f72aedbd7d798fb929d458ee8f)
|
| e832e64a7e82d7f72aedbd7d798fb929d458ee8f |
|
10-Nov-2014 |
Mathieu Chartier <mathieuc@google.com> |
Change 64 bit ArtMethod fields to be pointer sized
Changed the 64 bit entrypoint and gc map fields in ArtMethod to be
pointer sized. This saves a large amount of memory on 32 bit systems.
Reduces ArtMethod size by 16 bytes on 32 bit.
Total number of ArtMethod on low memory mako: 169957
Image size: 49203 methods -> 787248 image size reduction.
Zygote space size: 1070 methods -> 17120 size reduction.
App methods: ~120k -> 2 MB savings.
Savings per app on low memory mako: 125K+ per app
(less active apps -> more image methods per app).
Savings depend on how often the shared methods are on dirty pages vs
shared.
TODO in another CL, delete gc map field from ArtMethod since we
should be able to get it from the Oat method header.
Bug: 17643507
Change-Id: Ie9508f05907a9f693882d4d32a564460bf273ee8
|
| 5667fdbb6e441dee7534ade18b628ed396daf593 |
|
23-Oct-2014 |
Zheng Xu <zheng.xu@arm.com> |
ARM: Use hardfp calling convention between java to java call.
This patch default to use hardfp calling convention. Softfp can be enabled
by setting kArm32QuickCodeUseSoftFloat to true.
We get about -1 ~ +5% performance improvement with different benchmark
tests. Hopefully, we should be able to get more performance by address the left
TODOs, as some part of the code takes the original assumption which is not
optimal.
DONE:
1. Interpreter to quick code
2. Quick code to interpreter
3. Transition assembly and callee-saves
4. Trampoline(generic jni, resolution, invoke with access check and etc.)
5. Pass fp arg reg following aapcs(gpr and stack do not follow aapcs)
6. Quick helper assembly routines to handle ABI differences
7. Quick code method entry
8. Quick code method invocation
9. JNI compiler
TODO:
10. Rework ArgMap, FlushIn, GenDalvikArgs and affected common code.
11. Rework CallRuntimeHelperXXX().
Change-Id: I9965d8a007f4829f2560b63bcbbde271bdcf6ec2
|
| 1d8cdbc5202378a5f1a4b3a1fba610675ed4dcd5 |
|
23-Sep-2014 |
Ian Rogers <irogers@google.com> |
Refactor quick entrypoints
Remove FinishCalleeSaveFrameSetup.
Assembly routines write down anchor into TLS as well as placing runtime
method in callee save frame.
Simplify artSet64InstanceFromCode by not computing the referrer from the
stack in the C++ code.
Move assembly offset tests next to constant declaration and tidy arch_test.
Change-Id: Iededeebc05e54a1e2bb7bb3572b8ba012cffa1c8
|
| 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
|
| 832336b3c9eb892045a8de1bb12c9361112ca3c5 |
|
09-Oct-2014 |
Ian Rogers <irogers@google.com> |
Don't copy fill array data to quick literal pool.
Currently quick copies the fill array data from the dex file to the literal
pool. It then has to go through hoops to pass this PC relative address down
to out-of-line code. Instead, pass the offset of the table to the out-of-line
code and use the CodeItem data associated with the ArtMethod. This reduces
the size of oat code while greatly simplifying it.
Unify the FillArrayData implementation in quick, portable and the interpreters.
Change-Id: I9c6971cf46285fbf197856627368c0185fdc98ca
|
| 37f05ef45e0393de812d51261dc293240c17294d |
|
17-Jul-2014 |
Fred Shih <ffred@google.com> |
Reduced memory usage of primitive fields smaller than 4-bytes
Reduced memory used by byte and boolean fields from 4 bytes down to a
single byte and shorts and chars down to two bytes. Fields are now
arranged as Reference followed by decreasing component sizes, with
fields shuffled forward as needed.
Bug: 8135266
Change-Id: I65eaf31ed27e5bd5ba0c7d4606454b720b074752
|
| 38bd1e1b8aa39bdc6e67c5dc205cc8d294d985c5 |
|
19-Aug-2014 |
Andreas Gampe <agampe@google.com> |
ART: Fix ARM invoke stub, run Generic JNI in jni_compiler_test
On devices which support it, run all jni compiler tests also with
the Generic JNI infrastructure.
Fix art_quick_invoke_stub for ARM32 to enforce 16B stack alignment.
As the native convention does not require as much, the previous
code would keep it unaligned.
Bug: 17107929
(cherry picked from commit 6e498695b60f1532d2264ec6badb1cd6e10ecaa9)
Change-Id: I1481500ff7c627ab4e5bfbfd7245099a639aa049
|
| 6e498695b60f1532d2264ec6badb1cd6e10ecaa9 |
|
19-Aug-2014 |
Andreas Gampe <agampe@google.com> |
ART: Fix ARM invoke stub, run Generic JNI in jni_compiler_test
On devices which support it, run all jni compiler tests also with
the Generic JNI infrastructure.
Fix art_quick_invoke_stub for ARM32 to enforce 16B stack alignment.
As the native convention does not require as much, the previous
code would keep it unaligned.
Bug: 17107929
Change-Id: Ibce423949846e447d2b0246d07973b6b717691ac
|
| 29b3841ad8c1c18ee7ddd2d8cab85806b3d62eaa |
|
13-Aug-2014 |
Andreas Gampe <agampe@google.com> |
ART: Set default visibility to protected
Set default visibility of symbols to protected. This allows the
linker to optimize internal calls and helps avoid plt calls.
Make almost all assembly stubs hidden, as -fvisibility does not
seem to apply to them. Change the assembly tests accordingly. Also
allows to clean up previous hacks to avoid plt calls.
Bug: 16974467
(cherry picked from commit 235e77bd9f19e4faefda109be40f8744f3a66f40)
Change-Id: I9030dcf6116251f434f94a2b08e56e12085af652
|
| ebee8de725d5fa83483642786b19ea453d865762 |
|
13-Aug-2014 |
Nicolas Geoffray <ngeoffray@google.com> |
Fix bugs in the x86 and arm versions of generic JNI.
Also fix the run script of 116-nodex2oat to use the non-prebuild
script for target.
Bug: 17000769
(cherry-picked from commit 54accbca0b549b1b1ad3ef09655dad438bc1e104)
Change-Id: I439fcd710fb8bb408f3288ff8fb34fef23890adb
|
| b038ba66a166fb264ca121632f447712e0973b5b |
|
14-Aug-2014 |
Dave Allison <dallison@google.com> |
Revert "Revert "Reduce stack usage for overflow checks""
Fixes stack protection issue.
Fixes mac build issue.
This reverts commit 83b1940e6482b9d8feba5c492507735686650ea5.
Change-Id: I7ba17252882b23a740bcda2ea94aacf398255406
|
| 4cf00ba324f5f6884059796a6ba41937f32e1844 |
|
14-Aug-2014 |
Dave Allison <dallison@google.com> |
Revert "Reduce stack usage for overflow checks"
This reverts commit 63c051a540e6dfc806f656b88ac3a63e99395429.
Change-Id: I282a048994fcd130fe73842b16c21680053c592f
|
| 03c9785a8a6d712775cf406c4371d0227c44148f |
|
14-Aug-2014 |
Dave Allison <dallison@google.com> |
Revert "Revert "Reduce stack usage for overflow checks""
Fixes stack protection issue.
Fixes mac build issue.
This reverts commit 83b1940e6482b9d8feba5c492507735686650ea5.
Change-Id: I7ba17252882b23a740bcda2ea94aacf398255406
|
| 83b1940e6482b9d8feba5c492507735686650ea5 |
|
14-Aug-2014 |
Dave Allison <dallison@google.com> |
Revert "Reduce stack usage for overflow checks"
This reverts commit 63c051a540e6dfc806f656b88ac3a63e99395429.
Change-Id: I282a048994fcd130fe73842b16c21680053c592f
|
| 235e77bd9f19e4faefda109be40f8744f3a66f40 |
|
13-Aug-2014 |
Andreas Gampe <agampe@google.com> |
ART: Set default visibility to protected
Set default visibility of symbols to protected. This allows the
linker to optimize internal calls and helps avoid plt calls.
Make almost all assembly stubs hidden, as -fvisibility does not
seem to apply to them. Change the assembly tests accordingly. Also
allows to clean up previous hacks to avoid plt calls.
Bug: 16974467
Change-Id: I9030dcf6116251f434f94a2b08e56e12085af652
|
| 63c051a540e6dfc806f656b88ac3a63e99395429 |
|
26-Jul-2014 |
Dave Allison <dallison@google.com> |
Reduce stack usage for overflow checks
This reduces the stack space reserved for overflow checks to 12K, split
into an 8K gap and a 4K protected region. GC needs over 8K when running
in a stack overflow situation.
Also prevents signal runaway by detecting a signal inside code that
resulted from a signal handler invokation. And adds a max signal count to
the SignalTest to prevent it running forever.
Also reduces the number of iterations for the InterfaceTest as this was
taking (almost) forever with the --trace option on run-test.
Bug: 15435566
Change-Id: Id4fd46f22d52d42a9eb431ca07948673e8fda694
Conflicts:
compiler/optimizing/code_generator_x86_64.cc
runtime/arch/x86/fault_handler_x86.cc
runtime/arch/x86_64/quick_entrypoints_x86_64.S
|
| 648d7112609dd19c38131b3e71c37bcbbd19d11e |
|
26-Jul-2014 |
Dave Allison <dallison@google.com> |
Reduce stack usage for overflow checks
This reduces the stack space reserved for overflow checks to 12K, split
into an 8K gap and a 4K protected region. GC needs over 8K when running
in a stack overflow situation.
Also prevents signal runaway by detecting a signal inside code that
resulted from a signal handler invokation. And adds a max signal count to
the SignalTest to prevent it running forever.
Also reduces the number of iterations for the InterfaceTest as this was
taking (almost) forever with the --trace option on run-test.
Bug: 15435566
Change-Id: Id4fd46f22d52d42a9eb431ca07948673e8fda694
|
| 54accbca0b549b1b1ad3ef09655dad438bc1e104 |
|
13-Aug-2014 |
Nicolas Geoffray <ngeoffray@google.com> |
Fix bugs in the x86 and arm versions of generic JNI.
Also fix the run script of 116-nodex2oat to use the non-prebuild
script for target.
Bug: 17000769
Change-Id: I439fcd710fb8bb408f3288ff8fb34fef23890adb
|
| 6a763f149f0c6f705cab8fb1984aa6f9198b2841 |
|
10-Aug-2014 |
Dan Albert <danalbert@google.com> |
Proper fix after -Bsymbolic change.
Marking the offending symbols as hidden allows the linker to resolve
them at build time.
Bug: 16853291
Change-Id: I1480c1520c23f81e42831a66737bba7bdf3433c2
|
| 58bea4a6114b9ea4b00e36186357e5b431430123 |
|
10-Aug-2014 |
Dan Albert <danalbert@google.com> |
Proper fix after -Bsymbolic change.
Marking the offending symbols as hidden allows the linker to resolve
them at build time.
Bug: 16853291
Change-Id: I1480c1520c23f81e42831a66737bba7bdf3433c2
|
| d842648ee0ae8df1a899c53e96a98ccc6cca9d4c |
|
08-Aug-2014 |
Dan Albert <danalbert@google.com> |
Revert "Fix build without -Bsymbolic."
This reverts commit bae21bb7f6ca917e0954c970f0bfd2bfa3dcc0a9.
|
| bae21bb7f6ca917e0954c970f0bfd2bfa3dcc0a9 |
|
07-Aug-2014 |
Dan Albert <danalbert@google.com> |
Fix build without -Bsymbolic.
For ARM: Gold is unable to create PLT based relocations for conditional
branches.
For x86: Some uses of SYMBOL() should have been PLT_SYMBOL().
Bug: 16853291
Change-Id: Id70488b077256a70137c4417f21be2c2d1d4341c
|
| 76a0723959c2185ace249e705774af9a679a0e0d |
|
06-Aug-2014 |
Jeff Hao <jeffhao@google.com> |
When performing a long jump, update SP last.
Bug: 16571371
Change-Id: I719d9880a64e277330a2da698d5f6f7d320014d0
|
| 467f816c3c4483fb7766db3fc50e2fff10a92554 |
|
06-Aug-2014 |
Jeff Hao <jeffhao@google.com> |
When performing a long jump, update SP last.
Bug: 16571371
(cherry picked from commit 76a0723959c2185ace249e705774af9a679a0e0d)
Change-Id: Ib3a7bcda8aecdf7e92ea40a76de32aed83a42ee0
|
| c200a4abeca91e19969f5b35543f17f812ba32b9 |
|
17-Jun-2014 |
Andreas Gampe <agampe@google.com> |
ART: Rework Generic JNI, add ARM version
Refactors and optimizes Generic JNI. This version uses TwoWordReturn
to avoid writing to / loading from the bottom of the alloca.
Change-Id: I3287007c976f79c9fd32d3b3a43f2d1371bf4cd3
|
| 04f4d8abe45d6e79eca983e057de76aea24b7df9 |
|
30-May-2014 |
Wei Jin <wejin@google.com> |
Add an optimization for removing redundant suspend tests in ART
This CL:
(1) eliminates redundant suspend checks (dominated by another check),
(2) removes the special treatment of the R4 register, which got
reset on every native call, possibly yielding long execution
sequences without any suspend checks, and
(3) fixes the absence of suspend checks in leaf methods.
(2) and (3) increase the frequency of suspend checks, which improves
the performance of GC and the accuracy of profile data. To
compensate for the increased number of checks, we implemented an
optimization that leverages dominance information to remove
redundant suspend checks on back edges. Based on the results of
running the Caffeine benchmark on Nexus 7, the patch performs
roughly 30% more useful suspend checks, spreading them much more
evenly along the execution trace, while incurring less than 1%
overhead. For flexibility consideration, this CL defines two flags
to control the enabling of optimizations. The original
implementation is the default.
Change-Id: I31e81a5b3c53030444dbe0434157274c9ab8640f
Signed-off-by: Wei Jin <wejin@google.com>
|
| 675967d981a3d17aaedf4ca6e07cc3a76e066921 |
|
15-May-2014 |
Andreas Gampe <agampe@google.com> |
ART: Fix comments for dmb
Update the comments to mention correct barrier type.
Bug: 14680557
Change-Id: I26df43078c189cf16cc65e99b851981cbd063017
|
| 4fc046e78efbc98541388cdda986b5d8a2b951ad |
|
07-May-2014 |
Andreas Gampe <agampe@google.com> |
ART: Add lock and unlock stubs for ARM64, fix for X86-64
Basic translation of ARM stubs using dmb memory barrier.
Fix placement of dmb in unlock_object of ARM and ARM64.
Update lock and unlock tests in stub_test to force fat locks.
Fix X86-64 unlock stub.
Change-Id: Ie2e4328d9631e06843115888644e75fde8b319ee
|
| 5cd33753b96d92c03e3cb10cb802e68fb6ef2f21 |
|
16-Apr-2014 |
Dave Allison <dallison@google.com> |
Handle implicit stack overflow without affecting stack walks
This changes the way in which implicit stack overflows are handled
to satisfy concerns about changes to the stack walk code.
Instead of creating a gap in the stack and checking for it in
the stack walker, use the ManagedStack infrastructure to concoct
an invisible gap that will never be seen by a stack walk.
Also, this uses madvise to tell the kernel that the main stack's
protected region will probably never be accessed, and instead
of using memset to map the pages in, use memcpy to read from
them. This will save 32K on the main stack.
Also adds a 'signals' verbosity level as per a review request.
Bug: 14066862
Change-Id: I5257305feeaea241d11e6aa6f021d2a81da20b81
|
| 5c1e4352614d61fed6868567e58b96682828cb4d |
|
22-Apr-2014 |
Andreas Gampe <agampe@google.com> |
Add "arch_test" gtest for assembly stub constants, add some ARM64 assembly code
Add a test that (1) checks all callee-save method frame sizes for
all architectures, (2) checks thread offsets for the runtime
architecture and (3) checks callee-save method offsets for the
runtime architecture.
The "asm_support_XXX.h" files now only contain definitions that are
common between all architectures. Architecture-specific definitions
(i.e., special registers names) have been pushed into the corresponding
.S file. This change was required to be able to undefine definitions
in the test, so that multiple tests can be written in one file.
Test (1) above is in a sense two-stage. The arch_test gtest compares
constants (if it finds them) against the frame size as reported by
the ArtMethods created by the Runtime. This works for all architectures
as we can provide the instruction-set to CreateCalleeSaveMethod. The
second stage of the "test" are preprocessor tests with "#error" in the
case that the constants are not the expected value.
Optimally I'd like to change that to an actual runtime test exercising
the assembly code, which would also allow to check whether the right
registers are stored.
Also added missing assembly code for ARM64 for the callee-save macros.
Also fix X86_64 compilation for Clang 3.5.
Change-Id: I018e6433dffd3d31ba3bfcd75661653f4c7b6552
|
| 525cde2dd7cc7ad4212765ad3975cf260a934d3e |
|
23-Apr-2014 |
Andreas Gampe <agampe@google.com> |
Add a GTest for Assembly Stubs, Add some ARM64 and X86-64 Stubs
This GTest adds some runtime testing for the stubs that does not
rely on the compiler. This should allow to add or update the stubs
and do testing, especially on architectures without working compiler.
This test is a bit dangerous: if it doesn't know how to handle an
architecture, it will only log a warning. This is so that testing
does not break at the moment. The warning is forced to stdout, too,
so that it is always visible.
Add art_quick_check_cast to ARM64 and X86-64. Add art_quick_memcpy
to X86-64. The latter should be removed in a good compiler, as it is
practically only overhead. Add minor CFI information in ARM.
Change-Id: Ia9c6d0f4035eb1527c12b5f6067dece59e25528d
|
| 8325296769a77ecf3ab647b5ab516f439f5b3206 |
|
04-Apr-2014 |
Dave Allison <dallison@google.com> |
Disable use of R4 as a promotable register
When we are using implicit suspend checks we can potentially
use r4 as a register into which variables can be promoted. However
the runtime doesn't save this and thus will corrupt it. Not good.
This disables the promotion of r4 until we can figure out how
to make the runtime save it properly.
Change-Id: Ib95ce93579e1c364de5ecc8e728f2cb7990da77a
|
| 2da882315a61072664f7ce3c212307342e907207 |
|
27-Feb-2014 |
Andreas Gampe <agampe@google.com> |
Initial changes towards Generic JNI option
Some initial changes that lead to an UNIMPLEMENTED. Works
by not compiling for JNI right now and tracking native methods
which have neither quick nor portable code. Uses new trampoline.
Change-Id: I5448654044eb2717752fd7359f4ef8bd5c17be6e
|
| 5d885c82522f9f3f0775a6255b64bcd08c2a4c9e |
|
22-Feb-2014 |
Ian Rogers <irogers@google.com> |
Make ARM assembly labels local.
Avoids the symbols being part of the symbol file and confusing tools like gdb.
Change-Id: I90f9ce4a058cd4b69757e32df58fc88ab8d81b68
|
| 0177e53ea521ad58b70c305700dab32f1ac773b7 |
|
12-Feb-2014 |
Ian Rogers <irogers@google.com> |
Work in the direction of hard float quick ABIs.
Pass a shorty to ArtMethod::Invoke so that register setup can use it.
Document x86-64 ABI.
Add extra debug output for one JNI native method registration fails, namely a
dump of the Class and its dex file's location.
Add hack to get testing of OatMethod's without GC maps working in 64bit.
Change-Id: Ic06b68e18eac33637df2caf5e7e775ff95ae70f3
|
| ef7d42fca18c16fbaf103822ad16f23246e2905d |
|
06-Jan-2014 |
Ian Rogers <irogers@google.com> |
Object model changes to support 64bit.
Modify mirror objects so that references between them use an ObjectReference
value type rather than an Object* so that functionality to compress larger
references can be captured in the ObjectRefererence implementation.
ObjectReferences are 32bit and all other aspects of object layout remain as
they are currently.
Expand fields in objects holding pointers so they can hold 64bit pointers. Its
expected the size of these will come down by improving where we hold compiler
meta-data.
Stub out x86_64 architecture specific runtime implementation.
Modify OutputStream so that reads and writes are of unsigned quantities.
Make the use of portable or quick code more explicit.
Templatize AtomicInteger to support more than just int32_t as a type.
Add missing, and fix issues relating to, missing annotalysis information on the
mutator lock.
Refactor and share implementations for array copy between System and uses
elsewhere in the runtime.
Fix numerous 64bit build issues.
Change-Id: I1a5694c251a42c9eff71084dfdd4b51fff716822
|
| b122a4bbed34ab22b4c1541ee25e5cf22f12a926 |
|
20-Nov-2013 |
Ian Rogers <irogers@google.com> |
Tidy up memory barriers.
Change-Id: I937ea93e6df1835ecfe2d4bb7d84c24fe7fc097b
|
| 7410f29b4dae223befac036ea567d7f33351dad1 |
|
24-Nov-2013 |
Mathieu Chartier <mathieuc@google.com> |
Fix dumpsys meminfo <pid>.
Added a case for BumpPointerSpaces. Confirmed working non-debug.
Should also work in debug builds.
Bug: 11830794
Change-Id: I12053ff16eec403dcd4a780e13095e3212a77132
|
| 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
|
| 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
|
| bbb32c277b800a8ee378c16a3645ab6d4d19aef1 |
|
06-Nov-2013 |
Dave Allison <dallison@google.com> |
Add .cfi_restore for assembly functions
Bug: 11256318
This adds .cfi_restore directives to the assembly language functions
for ARM, MIPS and x86.
Change-Id: Ic5a80e190aa6cf2d2fcfaecec14de03221ff386b
|
| 4e6a31eb97f22f4480827474b30b9e64f396eace |
|
31-Oct-2013 |
Mathieu Chartier <mathieuc@google.com> |
Lazily compute object identity hash codes.
Before, we computed identity hashcodes whenever we inflated a monitor.
This caused issues since it meant that we would have all of these
hash codes in the image, causing locks to excessively inflate during
application run time.
This change makes it so that we lazily compute hash codes. When a
thin lock gets inflated, we assign a hash code of 0 assigned to it.
This value signifies no hash code. When we try to get the identity
hash code of an object with an inflated monitor, it gets computed if
it is 0.
Change-Id: Iae6acd1960515a36e74644e5b1323ff336731806
|
| 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
|
| 88474b416eb257078e590bf9bc7957cee604a186 |
|
24-Oct-2013 |
Jeff Hao <jeffhao@google.com> |
Implement Interface Method Tables (IMT).
Change-Id: Idf7fe85e1293453a8ad862ff2380dcd5db4e3a39
|
| 1984152ac92aad244ae15184d12f9ceade686b7b |
|
22-Oct-2013 |
Mathieu Chartier <mathieuc@google.com> |
Add missing callee save restore to quick entrypoints.
Added callee save restoration to art_quick_proxy_invoke_handler,
art_quick_resolution_trampoline, art_quick_to_interpreter_bridge.
Bug: 8981901
Change-Id: Iaa5bf862834b49cc48bce4f4ffa34a6797024e6d
|
| 2a6c7b7c21adcd4493542604305585b852ccf554 |
|
16-Oct-2013 |
Mathieu Chartier <mathieuc@google.com> |
Restore callee save registers.
The callee saves weren't all being restored as an optimiation, but
this breaks compaction if register promotion is enabled. The reason
for this is that these registers may contain references which the
GC will update.
Change-Id: I810f56b4ed1f92c632155e30c0838269cb95f3c5
|
| a9a8254c920ce8e22210abfc16c9842ce0aea28f |
|
04-Oct-2013 |
Ian Rogers <irogers@google.com> |
Improve quick codegen for aput-object.
1) don't type check known null.
2) if we know types in verify don't check at runtime.
3) if we're runtime checking then move all the code out-of-line.
Also, don't set up a callee-save frame for check-cast, do an instance-of test
then throw an exception if that fails.
Tidy quick entry point of Ldivmod to Lmod which it is on x86 and mips.
Fix monitor-enter/exit NPE for MIPS.
Fix benign bug in mirror::Class::CannotBeAssignedFromOtherTypes, a byte[]
cannot be assigned to from other types.
Change-Id: I9cb3859ec70cca71ed79331ec8df5bec969d6745
|
| d9c4fc94fa618617f94e1de9af5f034549100753 |
|
02-Oct-2013 |
Ian Rogers <irogers@google.com> |
Inflate contended lock word by suspending owner.
Bug 6961405.
Don't inflate monitors for Notify and NotifyAll.
Tidy lock word, handle recursive lock case alongside unlocked case and move
assembly out of line (except for ARM quick). Also handle null in out-of-line
assembly as the test is quick and the enter/exit code is already a safepoint.
To gain ownership of a monitor on behalf of another thread, monitor contenders
must not hold the monitor_lock_, so they wait on a condition variable.
Reduce size of per mutex contention log.
Be consistent in calling thin lock thread ids just thread ids.
Fix potential thread death races caused by the use of FindThreadByThreadId,
make it invariant that returned threads are either self or suspended now.
Code size reduction on ARM boot.oat 0.2%.
Old nexus 7 speedup 0.25%, new nexus 7 speedup 1.4%, nexus 10 speedup 2.24%,
nexus 4 speedup 2.09% on DeltaBlue.
Change-Id: Id52558b914f160d9c8578fdd7fc8199a9598576a
|
| 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
|
| 637859cef76e7d41f7fdfd1f1aea7d60b4a315dc |
|
27-Aug-2013 |
Ian Rogers <irogers@google.com> |
Make entrypoints Thumb2.
Save the 64bit shifts that are more efficient as ARM code.
Move the standard .S set up code to asm_support_arm.S.
Change-Id: I38b95a88a3658e311020b59abfbe16f843b509ff
|
| d8cf66be7fa0803b26508005cefb27e1498d2963 |
|
03-Aug-2013 |
Ian Rogers <irogers@google.com> |
Place CFA in the regular .debug_frame section.
Change-Id: Id161fe9dbad3244e102337c80b0ac2d7da0c3aa9
(cherry picked from commit 1a2f1bc64974254ab3246ca5c0682c5bb4317bad)
|
| 468532ea115657709bc32ee498e701a4c71762d4 |
|
05-Aug-2013 |
Ian Rogers <irogers@google.com> |
Entry point clean up.
Create set of entry points needed for image methods to avoid fix-up at load time:
- interpreter - bridge to interpreter, bridge to compiled code
- jni - dlsym lookup
- quick - resolution and bridge to interpreter
- portable - resolution and bridge to interpreter
Fix JNI work around to use JNI work around argument rewriting code that'd been
accidentally disabled.
Remove abstact method error stub, use interpreter bridge instead.
Consolidate trampoline (previously stub) generation in generic helper.
Simplify trampolines to jump directly into assembly code, keeps stack crawlable.
Dex: replace use of int with ThreadOffset for values that are thread offsets.
Tidy entry point routines between interpreter, jni, quick and portable.
Change-Id: I52a7c2bbb1b7e0ff8a3c3100b774212309d0828e
(cherry picked from commit 848871b4d8481229c32e0d048a9856e5a9a17ef9)
|
| 848871b4d8481229c32e0d048a9856e5a9a17ef9 |
|
05-Aug-2013 |
Ian Rogers <irogers@google.com> |
Entry point clean up.
Create set of entry points needed for image methods to avoid fix-up at load time:
- interpreter - bridge to interpreter, bridge to compiled code
- jni - dlsym lookup
- quick - resolution and bridge to interpreter
- portable - resolution and bridge to interpreter
Fix JNI work around to use JNI work around argument rewriting code that'd been
accidentally disabled.
Remove abstact method error stub, use interpreter bridge instead.
Consolidate trampoline (previously stub) generation in generic helper.
Simplify trampolines to jump directly into assembly code, keeps stack crawlable.
Dex: replace use of int with ThreadOffset for values that are thread offsets.
Tidy entry point routines between interpreter, jni, quick and portable.
Change-Id: I52a7c2bbb1b7e0ff8a3c3100b774212309d0828e
|
| 1a2f1bc64974254ab3246ca5c0682c5bb4317bad |
|
03-Aug-2013 |
Ian Rogers <irogers@google.com> |
Place CFA in the regular .debug_frame section.
Change-Id: Id161fe9dbad3244e102337c80b0ac2d7da0c3aa9
|
| 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
|
| 7655f29fabc0a12765de828914a18314382e5a35 |
|
29-Jul-2013 |
Ian Rogers <irogers@google.com> |
Portable refactorings.
Separate quick from portable entrypoints.
Move architectural dependencies into arch.
Change-Id: I9adbc0a9782e2959fdc3308215f01e3107632b7c
|
| 166db04e259ca51838c311891598664deeed85ad |
|
26-Jul-2013 |
Ian Rogers <irogers@google.com> |
Move assembler out of runtime into compiler/utils.
Other directory layout bits of clean up. There is still work to separate quick
and portable in some files (e.g. argument visitor, proxy..).
Change-Id: If8fecffda8ba5c4c47a035f0c622c538c6b58351
|