OSDN Git Service
Andreas Gampe [Mon, 29 Sep 2014 18:29:04 +0000 (18:29 +0000)]
Merge "ART: Do a pre-pass for monitor_enter dex pc search"
Ian Rogers [Mon, 29 Sep 2014 18:24:23 +0000 (18:24 +0000)]
Merge "Reduce scope of MIR analysis attributes."
Ian Rogers [Mon, 29 Sep 2014 17:49:11 +0000 (10:49 -0700)]
Reduce scope of MIR analysis attributes.
MIR analaysis attributes are only used in MIRGraph::AnalyzeBlock. Reduce the
scope to just the file that defines that function.
Reduce the size of the attributes from int32 to int16.
Remove the use of #defines in favor of enums.
Change-Id: I45ba3aa08beae67240507917b584d537a029838e
Andreas Gampe [Thu, 18 Sep 2014 20:16:38 +0000 (13:16 -0700)]
ART: Do a pre-pass for monitor_enter dex pc search
In case the method does not have any monitor_enter instructions,
it is unnecessary to run the full verifier. Speeds up stack dumps
and works around b/
17514582.
Bug:
17514582
Change-Id: I5201bfbb9fb6cad49596b4c72e71983b58d9f20c
Nicolas Geoffray [Mon, 29 Sep 2014 10:59:34 +0000 (10:59 +0000)]
Merge "Fix a bug in the insertion of parallel move."
Nicolas Geoffray [Mon, 29 Sep 2014 09:33:25 +0000 (10:33 +0100)]
Fix a bug in the insertion of parallel move.
To make sure we do not connect interval siblings in the
same parallel move, I added a new field in MoveOperands
that tells for which instruction this move is for.
A parallel move should not contains moves for the same instructions.
The checks revealed a bug when connecting siblings, where
we would choose the wrong parallel move.
Change-Id: I70f27ec120886745c187071453c78da4c47c1dd2
Nicolas Geoffray [Mon, 29 Sep 2014 08:47:31 +0000 (08:47 +0000)]
Merge "Fix test now that instructions can die at instruction entry."
Nicolas Geoffray [Mon, 29 Sep 2014 08:46:03 +0000 (09:46 +0100)]
Fix test now that instructions can die at instruction entry.
Change-Id: I816279c55d12de8e69ac0b6c88730bd676c03335
Nicolas Geoffray [Sat, 27 Sep 2014 11:56:12 +0000 (11:56 +0000)]
Merge "Register allocator: refine instructions liveness."
Nicolas Geoffray [Thu, 25 Sep 2014 13:39:26 +0000 (14:39 +0100)]
Register allocator: refine instructions liveness.
Add support for instructions that die at the beginning
of another instruction. Before, an instruction needed
to stay alive during the instruction, so the register
allocator was not able not reuse the register.
Change-Id: I5f11a80b0a20778227229eb797816edcc6365297
Mathieu Chartier [Sat, 27 Sep 2014 02:55:11 +0000 (02:55 +0000)]
Merge "Optimize get/set reflection performance"
Mathieu Chartier [Fri, 26 Sep 2014 21:32:37 +0000 (14:32 -0700)]
Optimize get/set reflection performance
Speedups mostly from reducing how often access checks are needed,
and adding more inlining, and adding templates.
Field_getInt from ~850ns -> 350ns.
Field_setInt from ~900ns -> 370ns.
Bug:
14063288
(cherry picked from commit
ffc788cb7b5b9f53734d7bb8af2d5e45d885546b)
Change-Id: I2441581ff3478c6ae43b6aa49939ff3f07555ec8
Mathieu Chartier [Fri, 26 Sep 2014 23:19:48 +0000 (23:19 +0000)]
Merge "Fix broken runtime SetStatsEnabled logic"
Mathieu Chartier [Fri, 26 Sep 2014 00:03:12 +0000 (17:03 -0700)]
Fix broken runtime SetStatsEnabled logic
Previously, Runtime::SetStatsEnabled wouldn't take stats_enabled_
into account when deciding whether or not to increment / decrement
teh stats enabled counter. This resulted in counter underflows and
other errors which caused some CTS tests to fail.
Also added some locking to prevent race conditions.
Bug:
17360878
(cherry picked from commit
a98ffd745bbecb2e84a492194950c0b94966546b)
Change-Id: I21d241a58d35bd6a607aa2305c6da81720bd0886
Ian Rogers [Fri, 26 Sep 2014 23:01:45 +0000 (23:01 +0000)]
Merge "Fix make test exclusion bugs."
Ian Rogers [Fri, 26 Sep 2014 22:55:25 +0000 (15:55 -0700)]
Fix make test exclusion bugs.
Change-Id: I439c3670a01096355c50774c628d6b0f6cb6c397
Ian Rogers [Fri, 26 Sep 2014 22:47:47 +0000 (22:47 +0000)]
Merge "Only include test targets if test-art* or build-art* is being built."
Ian Rogers [Fri, 26 Sep 2014 22:31:20 +0000 (15:31 -0700)]
Only include test targets if test-art* or build-art* is being built.
Change-Id: I229ba2a77d993b2c14acf9a76e102ef76b0aeba6
Mathieu Chartier [Fri, 26 Sep 2014 19:59:52 +0000 (19:59 +0000)]
Merge "Change heap base address"
Mathieu Chartier [Fri, 26 Sep 2014 18:37:42 +0000 (11:37 -0700)]
Change heap base address
Moved to 300 MB to allow room for other people who wish to use the
low 1 GB.
Bug:
17671634
(cherry picked from commit
23fb122b57c52ab1db98708a4e15765b46b26405)
Change-Id: Icb2784fc92cb5c46edb6e3a9de2d8383e012a07c
Hiroshi Yamauchi [Fri, 26 Sep 2014 18:36:01 +0000 (18:36 +0000)]
Merge "Fix verifier crash in GetQuickFieldAccess()."
Hiroshi Yamauchi [Fri, 26 Sep 2014 17:43:59 +0000 (10:43 -0700)]
Fix verifier crash in GetQuickFieldAccess().
Bug:
17671806
Change-Id: Ifb603f6f55a94826d656a4c73a7e328d74f6dac9
Ian Rogers [Fri, 26 Sep 2014 16:57:30 +0000 (16:57 +0000)]
Merge "ART Runtime: Add movw $xx,yy fault handling"
Ian Rogers [Fri, 26 Sep 2014 16:53:58 +0000 (16:53 +0000)]
Merge "Elimination of redundant instructuion"
Ian Rogers [Fri, 26 Sep 2014 16:43:28 +0000 (16:43 +0000)]
Merge "Fix CallStaticVoidMethodA for testEverything"
Ian Rogers [Fri, 26 Sep 2014 16:15:00 +0000 (16:15 +0000)]
Merge "Fix some more pedantic compiler warnings."
Marcus Oakland [Wed, 24 Sep 2014 14:23:06 +0000 (15:23 +0100)]
Fix CallStaticVoidMethodA for testEverything
The android.jni.cts.JniCTest#testEverything and
android.jni.cts.JniCppTest#testEverything CTS test were failing
because of a SIGABRT caused by a call to ScopedCheck::AbortF
from ScopedCheck::CheckInstance when CheckJNI::CallStaticVoidMethodA
was invoked. This was due to the CheckJNI::CallMethodA method being
called with the jobject obj parameter being passed jclass c, and the
jclass c parameter being passed nullptr. This problem was rectified
by swapping these two parameters, and the CTS tests then passed.
Change-Id: I025cfd85ab55eb3eadb287b56846d9d42f5b7e5e
Signed-off-by: Marcus Oakland <marcus.oakland@arm.com>
Andreas Gampe [Fri, 26 Sep 2014 07:17:36 +0000 (07:17 +0000)]
Merge "ART: Fix mac build"
Andreas Gampe [Fri, 26 Sep 2014 06:13:47 +0000 (23:13 -0700)]
ART: Fix mac build
Change-Id: I5821cf4d5b8f4142912c9cd3524a6aaa4b4a397c
Ian Rogers [Fri, 26 Sep 2014 00:36:39 +0000 (17:36 -0700)]
Fix some more pedantic compiler warnings.
Argument name shadows field name.
Lack of case statements for kMips64.
Change-Id: Ib3ca16f79a4d44ae80ef32f22cc79cbe9527f4f1
Ian Rogers [Thu, 25 Sep 2014 23:53:21 +0000 (23:53 +0000)]
Merge "Fix Class::IsInSamePackage to not read beyond the end of a StringPiece."
Ian Rogers [Thu, 25 Sep 2014 22:35:37 +0000 (15:35 -0700)]
Fix Class::IsInSamePackage to not read beyond the end of a StringPiece.
Fix length of string piece to be size_type rather than int because expecting
negatively sized strings is a bad idea and we should use unsigned types to
defensively guard against mistakes.
Remove max_size and capacity since the return type is inconsistent between
Google and the STL, and we don't need the functions.
Add a bound check in libartd to operator[] accesses.
Change-Id: I1b87a03d8fbd95e7dbb106745e304d1083898075
Hiroshi Yamauchi [Thu, 25 Sep 2014 22:59:28 +0000 (22:59 +0000)]
Merge "Some optimizations for the array alloc path."
Hiroshi Yamauchi [Thu, 25 Sep 2014 18:46:46 +0000 (11:46 -0700)]
Some optimizations for the array alloc path.
- Force Array::Alloc() to be inlined.
- Simplify the array size overflow check.
- Turn fill_usable into a template parameter.
- Remove a branch in Array::DataOffset() and avoid
Primitive::ComponentSize(), which has a switch, in the array alloc
path.
- Strength reductions in the array size computation by using component
size shifts instead of component sizes. Store component size shift
in the upper 16 bits of primitive_type field.
- Speedup: ~4% (3435->3284) in MemAllocTest on N4.
Bug:
9986565
Change-Id: I4b142ffac4ab8b5b915836f1660a949d6442344c
Mathieu Chartier [Thu, 25 Sep 2014 22:30:08 +0000 (22:30 +0000)]
Merge "Reduce IndirectReferenceTable memory usage"
Mathieu Chartier [Thu, 25 Sep 2014 22:27:43 +0000 (15:27 -0700)]
Reduce IndirectReferenceTable memory usage
Changed the slot side table and the main table to be a single
table, reduced number of slots per reference from 4 to 3 to make
the IrtEntry a power of 2 size.
Before:
20848 kB: Dalvik Other
After:
16760 kB: Dalvik Other
Bug:
17643507
(cherry picked from commit
a1de6b93426cfc66a64eb1b57303348aab5e766d)
Change-Id: I362475235a887c60eff6870bb10051a6be3d5814
Mark Mendell [Thu, 25 Sep 2014 19:13:39 +0000 (15:13 -0400)]
ART Runtime: Add movw $xx,yy fault handling
Handle a movw $xx,yy instruction in the segmentation violation fault
handler. This can be detected by a 0x66 prefix byte for the move
immediate instruction, using this to set the size of the immediate
value.
This test case shows up in 122-npe using my private backend changes.
Change-Id: If224a9dd0084f50c91da418808cbe2d491308600
Category: Device Enablement
Domain: AOSP.ART-Quick
Origin: Internal
Upstream-Candidate: yes
Signed-off-by: Mark Mendell <mark.p.mendell@intel.com>
Vladimir Marko [Thu, 25 Sep 2014 18:08:03 +0000 (18:08 +0000)]
Merge "Implement method calls using relative BL on ARM."
Vladimir Marko [Fri, 1 Aug 2014 18:04:18 +0000 (19:04 +0100)]
Implement method calls using relative BL on ARM.
Store the linker patches with each CompiledMethod instead of
keeping them in CompilerDriver. Reorganize oat file creation
to apply the patches as we're writing the method code. Add
framework for platform-specific relative call patches in the
OatWriter. Implement relative call patches for ARM.
Change-Id: Ie2effb3d92b61ac8f356140eba09dc37d62290f8
Nicolas Geoffray [Thu, 25 Sep 2014 16:37:24 +0000 (16:37 +0000)]
Merge "Make the script bash because it uses PIPESTATUS."
Nicolas Geoffray [Thu, 25 Sep 2014 15:40:07 +0000 (16:40 +0100)]
Make the script bash because it uses PIPESTATUS.
Change-Id: Idc85935e0d9eebc96faed5112b5a0af8cf53192c
Nicolas Geoffray [Thu, 25 Sep 2014 15:37:26 +0000 (15:37 +0000)]
Merge "Make suspend checks note have side effects."
Nicolas Geoffray [Thu, 25 Sep 2014 15:33:42 +0000 (16:33 +0100)]
Make suspend checks note have side effects.
Also adjust gtests.
Change-Id: I5e1a3e53115812b45ec7f4b6f50ba468fa7ac6b1
Nicolas Geoffray [Thu, 25 Sep 2014 13:37:06 +0000 (13:37 +0000)]
Merge "Optimize suspend checks in optimizing compiler."
Roland Levillain [Thu, 25 Sep 2014 13:28:27 +0000 (13:28 +0000)]
Merge "Optimizing compiler: remove unnecessary `explicit' keywords."
Roland Levillain [Thu, 25 Sep 2014 12:41:09 +0000 (12:41 +0000)]
Merge "Optimizing compiler: check inputs & uses definitions in CFG."
Roland Levillain [Thu, 25 Sep 2014 11:15:20 +0000 (12:15 +0100)]
Optimizing compiler: remove unnecessary `explicit' keywords.
Change-Id: I5927fd92d53308c81e14edbd6e7d1c943bfa085b
Nicolas Geoffray [Wed, 24 Sep 2014 17:10:46 +0000 (18:10 +0100)]
Optimize suspend checks in optimizing compiler.
- Remove the ones added during graph build (they were added
for the baseline code generator).
- Emit them at loop back edges after phi moves, so that the test
can directly jump to the loop header.
- Fix x86 and x86_64 suspend check by using cmpw instead of cmpl.
Change-Id: I6fad5795a55705d86c9e1cb85bf5d63dadfafa2a
Vladimir Marko [Thu, 25 Sep 2014 11:23:21 +0000 (11:23 +0000)]
Merge "Deprecate GrowableArray, use ArenaVector instead."
Roland Levillain [Thu, 25 Sep 2014 09:10:38 +0000 (10:10 +0100)]
Optimizing compiler: check inputs & uses definitions in CFG.
Ensure each input and each use of an instruction is defined
in a block of the control-flow graph.
Change-Id: If4a83b02825230329b0b4fd84255dcb7c3219684
Ian Rogers [Thu, 25 Sep 2014 01:55:35 +0000 (01:55 +0000)]
Merge "ART: Reject branch/switch/throw to move-result."
Ian Rogers [Thu, 25 Sep 2014 01:20:05 +0000 (01:20 +0000)]
Merge "Fix "const-class v0, I" for libartd.so"
Andreas Gampe [Wed, 24 Sep 2014 22:00:11 +0000 (22:00 +0000)]
Merge "ART: Update for split native bridge initialization"
Stephen Kyle [Mon, 22 Sep 2014 12:53:15 +0000 (13:53 +0100)]
ART: Reject branch/switch/throw to move-result.
move-result* instructions must occur after invoke instructions,
however it was still possible to branch or switch to a move-result*
instruction with creative DEX manipulation. The verifier now rejects
this situation, as well as having a move-result* instruction as the
first instruction in a try block's handler. This now ensures that
move-result* must happen dynamically after an invoke, not just
statically.
Change-Id: Ida97852f4051310fdaf38bed1d6e1c5a541c85c5
Signed-off-by: Stephen Kyle <stephen.kyle@arm.com>
Brian Carlstrom [Wed, 24 Sep 2014 16:35:50 +0000 (16:35 +0000)]
Merge "Restore kReasonLogging for kDexoptNeeded and kPatchoatNeeded"
Brian Carlstrom [Tue, 23 Sep 2014 20:53:28 +0000 (13:53 -0700)]
Restore kReasonLogging for kDexoptNeeded and kPatchoatNeeded
Bug:
17565501
(cherry picked from commit
63582f211b7123d02f18dfa7075f6758ccb2b07c)
Change-Id: Ie1196191d92286e543ed099c7d1ea4712802912d
Logan Chien [Tue, 23 Sep 2014 15:01:47 +0000 (23:01 +0800)]
Fix "const-class v0, I" for libartd.so
This CL fixes the verification of "const-class v0, I"
by adding HasClassVirtual() to PrimitiveType.
Without this CL, the DCHECK for HasClass() in RegType::GetClass()
will raise an assertion failure, since the verifier is trying to get
the class object for the dex cache.
This issue can be reproduced by running the CTS
dot.junit.opcodes.const_class.Main_testN2 with libartd.so.
Change-Id: I605a3fcf7dd9d63e1f739fbb3153d0367911bb3e
Andreas Gampe [Wed, 24 Sep 2014 07:34:13 +0000 (07:34 +0000)]
Merge "ART: Refactor elf_writer_quick for elf section order"
Andreas Gampe [Wed, 24 Sep 2014 05:58:07 +0000 (22:58 -0700)]
ART: Refactor elf_writer_quick for elf section order
This writes ELF sections in approximate order (debug sections are
currently out-of-order) to avoid incompletely written files
appearing OK to ART loading code.
Bug:
17622827
(cherry picked from commit
3c5ec99e156632a734c2359642ae928102850597)
Change-Id: I812fd42fcf2823dbfaf5891bf101d3b59406f263
Brian Carlstrom [Wed, 24 Sep 2014 07:17:45 +0000 (07:17 +0000)]
Merge "dex2oat should truncate files opened from file descriptors"
Brian Carlstrom [Wed, 24 Sep 2014 04:13:28 +0000 (21:13 -0700)]
dex2oat should truncate files opened from file descriptors
Bug:
17622827
(cherry picked from commit
aa93129ca54cc897fdc2c8ddd0a7136f7b5632ed)
Change-Id: I5467d488b4a57f7be35c72c16775df5bd5d69a52
jgu21 [Wed, 10 Sep 2014 10:57:17 +0000 (06:57 -0400)]
ART: Update for split native bridge initialization
Change-Id: I0b93da93251c6b4638de786bf98cf99df07c3fc2
Roland Levillain [Tue, 23 Sep 2014 17:13:54 +0000 (17:13 +0000)]
Merge "Optimizing compiler: ensure loop header dominates loop's blocks."
Sebastien Hertz [Tue, 23 Sep 2014 14:27:33 +0000 (14:27 +0000)]
Merge "Enable computed goto interpreter on the host"
Vladimir Marko [Mon, 22 Sep 2014 13:50:02 +0000 (14:50 +0100)]
Deprecate GrowableArray, use ArenaVector instead.
Purge GrowableArray from Quick and Portable.
Remove GrowableArray<T>::Iterator.
Change-Id: I92157d3a6ea5975f295662809585b2dc15caa1c6
Nicolas Geoffray [Tue, 23 Sep 2014 13:36:28 +0000 (13:36 +0000)]
Merge "Support for saving and restoring live registers in a slow path."
Nicolas Geoffray [Fri, 19 Sep 2014 10:01:00 +0000 (11:01 +0100)]
Support for saving and restoring live registers in a slow path.
And use it in suspend check slow paths.
Change-Id: I79caf28f334c145a36180c79a6e2fceae3990c31
Sebastien Hertz [Thu, 28 Aug 2014 16:25:09 +0000 (18:25 +0200)]
Enable computed goto interpreter on the host
We disabled the computed goto interpreter when compiling with clang because of
compilation issues. Since we compile with clang on the host, we used the
switch-based interpreter.
Now clang is able to compile the computed goto interpreter, we can use it by
default on the host.
Change-Id: Ifa2a2d3c171b1116ecd3b3f213382c3c85e93f66
Nicolas Geoffray [Tue, 23 Sep 2014 11:25:56 +0000 (11:25 +0000)]
Merge "Fix code generation with materialized conditions."
Nicolas Geoffray [Mon, 22 Sep 2014 14:51:11 +0000 (15:51 +0100)]
Fix code generation with materialized conditions.
Change-Id: I8630af3c13fc1950d3fa718d7488407b00898796
Roland Levillain [Tue, 23 Sep 2014 09:50:22 +0000 (10:50 +0100)]
Optimizing compiler: ensure loop header dominates loop's blocks.
Change-Id: I6b2f1fdaac9f91dc5d9901cc2ad4c83745e90e70
Roland Levillain [Tue, 23 Sep 2014 09:02:32 +0000 (09:02 +0000)]
Merge "Add loop- and phi-related checks in the optimizing compiler."
Mathieu Chartier [Mon, 22 Sep 2014 21:40:06 +0000 (21:40 +0000)]
Merge "Add ScopedAssertNoThreadSuspension"
Roland Levillain [Mon, 22 Sep 2014 16:13:44 +0000 (17:13 +0100)]
Add loop- and phi-related checks in the optimizing compiler.
- Ensure the pre-header block is first in the list of
predecessors of a loop header.
- Ensure the loop header has only two predecessors and that
only the second one is the back edge.
- Ensure there is only one back edge per loop.
- Ensure the first input of a phi is not itself.
- Ensure the number of phi inputs is the same as the number
of its predecessors.
- Ensure phi input at index I either comes from the Ith
predecessor or from a block that dominates this
predecessor.
Change-Id: I4db5c68cfbc9b74d2d03125753d0143ece625378
Nicolas Geoffray [Mon, 22 Sep 2014 14:22:08 +0000 (14:22 +0000)]
Merge "Also remove environment links to removed instructions."
Nicolas Geoffray [Mon, 22 Sep 2014 11:27:27 +0000 (12:27 +0100)]
Also remove environment links to removed instructions.
Change-Id: I505163fb8683269c7d3fe21b34df92337d244552
Nicolas Geoffray [Mon, 22 Sep 2014 10:56:06 +0000 (10:56 +0000)]
Merge "Fix a bug in the handling of moves in register allocator."
Nicolas Geoffray [Mon, 22 Sep 2014 10:51:51 +0000 (11:51 +0100)]
Fix a bug in the handling of moves in register allocator.
Change-Id: Iaf1f34b0bece4f252290a97c3b73cc06e365985a
Sebastien Hertz [Mon, 22 Sep 2014 10:30:36 +0000 (10:30 +0000)]
Merge "Optimize JDWP stack local values access"
Anton Shamin [Mon, 22 Sep 2014 09:43:51 +0000 (16:43 +0700)]
Elimination of redundant instructuion
Elimination of storing offset in phys register (while it's already
stored in ECX by default) from two address shift operation.
Change-Id: Iaf9abde65f505ab855f73f9da7bd13fa46c64fbb
Signed-off-by: Anton Shamin <anton.shamin@intel.com>
Sebastien Hertz [Mon, 1 Sep 2014 15:07:11 +0000 (17:07 +0200)]
Optimize JDWP stack local values access
The StackFrame.GetValues and StackFrame.SetValues JDWP commands can refer to
multiple variables at the same time in a given frame. However we used to walk
the stack until getting to the requested frame for each variable.
Now, we walk the stack only once until getting to the frame so the context is
initialized. Then we read/write value for each variable from this context.
Bug:
17343501
Bug:
15680615
Change-Id: I2a4128f29a3c5856b994e280037c0a09eb48c5c8
Sebastien Hertz [Mon, 22 Sep 2014 09:10:17 +0000 (09:10 +0000)]
Merge "Do not hold breakpoint lock when running the verifier"
Nicolas Geoffray [Mon, 22 Sep 2014 08:31:34 +0000 (08:31 +0000)]
Merge "Add adb root to test-art-target, required by adb remount."
Sebastien Hertz [Mon, 22 Sep 2014 08:15:52 +0000 (08:15 +0000)]
Merge "Fix JDWP crash when reporting exception"
Sebastien Hertz [Fri, 19 Sep 2014 10:07:51 +0000 (12:07 +0200)]
Fix JDWP crash when reporting exception
The exception's throw location may be null so we need to handle that
case. Also fixes a memset issue.
Bug:
17571297
(cherry picked from commit
bbb63897d7f2d99219cb50721fe530521e08ddff)
Change-Id: Iedebb58f9460c5f04913c269200e51161bda1ba9
Sebastien Hertz [Thu, 18 Sep 2014 14:03:34 +0000 (16:03 +0200)]
Do not hold breakpoint lock when running the verifier
When setting a breakpoint, we need to know whether the method may be
inlined. We run the method verifier but that may cause thread
suspension. Therefore we must not hold any lock at this time. The
issue is we do hold the breakpoint lock so we fails a check in debug
mode.
This CL ensures we don't hold the breakpoint lock when running the
method verifier to detect inlining.
Bug:
17562442
Change-Id: Ia6b128fc8917ce00025b68ae4ac62fb2a1f154e6
Christopher Ferris [Sun, 21 Sep 2014 23:19:37 +0000 (23:19 +0000)]
Merge "Dump kernel/native stacks of pure native threads."
Christopher Ferris [Sat, 20 Sep 2014 00:23:59 +0000 (17:23 -0700)]
Dump kernel/native stacks of pure native threads.
Bug:
16463406
(cherry picked from commit
7da37cca5a5f71a0cd2442b2a119f76fa62e84b5)
Change-Id: I22a26d4ec4e3048ddcfd668eee4fccd8b2b3b727
Mathieu Chartier [Sat, 20 Sep 2014 00:52:37 +0000 (17:52 -0700)]
Add ScopedAssertNoThreadSuspension
Added a new class, ScopedAssertNoThreadSuspension.
Deleted some unnecessary ScopedAssertNoThreadSuspension since
VisitObjects already has a ScopedAssertNoThreadSuspension.
Change-Id: I29ec0006120c39a27184d30e2d1d0c179e203776
Ian Rogers [Sat, 20 Sep 2014 00:34:21 +0000 (00:34 +0000)]
Merge "Fix sput-wide verification flag."
Ian Rogers [Fri, 19 Sep 2014 23:23:01 +0000 (23:23 +0000)]
Merge "ART: Address three issues with x86 assembler before enabling load store elimination."
Andreas Gampe [Fri, 19 Sep 2014 16:54:18 +0000 (16:54 +0000)]
Merge "ART: Allow oatdump to verify oat file"
Ian Rogers [Fri, 19 Sep 2014 16:51:13 +0000 (16:51 +0000)]
Merge "ART: Fix host test failure (assembler_thumb_test)."
Ian Rogers [Fri, 19 Sep 2014 16:48:19 +0000 (16:48 +0000)]
Merge "ART: Overflow of bound check in ArrayCopy intrinsic"
Bill Buzbee [Fri, 19 Sep 2014 16:11:08 +0000 (16:11 +0000)]
Merge "Slow path should break def tracking"
Bill Buzbee [Fri, 19 Sep 2014 15:52:49 +0000 (15:52 +0000)]
Merge "AddIntrinsicSlowPath with resume requires clobbering"
Ian Rogers [Fri, 19 Sep 2014 15:44:24 +0000 (15:44 +0000)]
Merge "ART: Fix instruction flags of kMirOpCheck and kMirOpCheckPart2"
Ian Rogers [Fri, 19 Sep 2014 15:37:26 +0000 (15:37 +0000)]
Merge "X86 QBE: Make some X86 routines virtual"
Nicolas Geoffray [Fri, 19 Sep 2014 13:21:14 +0000 (13:21 +0000)]
Merge "First optimization in new compiler: simple GVN."
Nicolas Geoffray [Mon, 8 Sep 2014 16:30:24 +0000 (17:30 +0100)]
First optimization in new compiler: simple GVN.
Change-Id: Ibe0efa4e84fd020a53ded310a92e0b4363f91b12