OSDN Git Service

android-x86/art.git
9 years agoMerge "ART: Refactor elf_writer_quick, add symbolizer"
Andreas Gampe [Sat, 6 Sep 2014 05:03:34 +0000 (05:03 +0000)]
Merge "ART: Refactor elf_writer_quick, add symbolizer"

9 years agoART: Refactor elf_writer_quick, add symbolizer
Andreas Gampe [Fri, 5 Sep 2014 04:47:42 +0000 (21:47 -0700)]
ART: Refactor elf_writer_quick, add symbolizer

Refactors some classes in elf_writer_quick.h to elf_builder.h to
be more friendly for re-use. Use this in oatdump to add a symtab
to an oat file.

Bug: 1718762117322125
Change-Id: I2333291334fd98bd09cc5717fb83cb18efe3a029

9 years agoMerge "Fix numerous issues with DdmVmInternal allocation tracking"
Brian Carlstrom [Fri, 5 Sep 2014 23:05:06 +0000 (23:05 +0000)]
Merge "Fix numerous issues with DdmVmInternal allocation tracking"

9 years agoFix numerous issues with DdmVmInternal allocation tracking
Brian Carlstrom [Fri, 5 Sep 2014 20:01:41 +0000 (13:01 -0700)]
Fix numerous issues with DdmVmInternal allocation tracking

Issues addressed:
- Using without JDWP attached caused native crash.
- When buffer is full (64k entries), number of entries reported was 0.
- Disabling tracking after disabling tracking caused native crash.
- Asking for allocations after disabled caused native crash.
- Lock ordering issues between mutator lock and alloc tracker lock.

Adding 098-ddmc test to cover these cases.

Bug: 17392248

(cherry picked from commit a5815065ac0877add9c0db3605d27b4d6c426e61)

Change-Id: Ib0bc18dfcdafcc050ab9dceed3d167dd878d1d7a

9 years agoMerge "Revert "Revert "x86: Fix art_quick_instrumentation_exit"""
Andreas Gampe [Fri, 5 Sep 2014 17:19:23 +0000 (17:19 +0000)]
Merge "Revert "Revert "x86: Fix art_quick_instrumentation_exit"""

9 years agoMerge "ART: X86: Improve multiply in div/rem by literal"
Ian Rogers [Fri, 5 Sep 2014 17:10:25 +0000 (17:10 +0000)]
Merge "ART: X86: Improve multiply in div/rem by literal"

9 years agoMerge "Avoid recomputing the dex checksum during class loading"
Calin Juravle [Fri, 5 Sep 2014 15:35:19 +0000 (15:35 +0000)]
Merge "Avoid recomputing the dex checksum during class loading"

9 years agoAvoid recomputing the dex checksum during class loading
Calin Juravle [Tue, 2 Sep 2014 14:53:55 +0000 (15:53 +0100)]
Avoid recomputing the dex checksum during class loading

Thread the already computed checksum to VerifyOatAndDexFileChecksums and
LoadMultiDexFilesFromOatFile to avoid recomputing it.

Bug:17346103

(cherry picked from commit ca3459398018360d9968a52eebf727df085caf83)

Change-Id: Ie244efaca55b9e7c814b35feec7fce11fab3856d

9 years agoMerge "Remove unreachable code from class_linker.cc"
Calin Juravle [Fri, 5 Sep 2014 15:26:00 +0000 (15:26 +0000)]
Merge "Remove unreachable code from class_linker.cc"

9 years agoRemove unreachable code from class_linker.cc
Calin Juravle [Fri, 5 Sep 2014 15:14:19 +0000 (16:14 +0100)]
Remove unreachable code from class_linker.cc

If the checksum doesn't match the oat_dex_file will be NULL, a case
which is handled in the first if.

Bug: 17402267

Change-Id: If8547ddc8fd7eff59a3b7dd36c07ac4ceb351361

9 years agoART: X86: Improve multiply in div/rem by literal
Mark Mendell [Tue, 2 Sep 2014 16:44:24 +0000 (12:44 -0400)]
ART: X86: Improve multiply in div/rem by literal

Rewrite the multiplication of the magic number by the dividend to save
one instruction by multiplying directly from the dividend, rather than
loading the dividend into EAX/RAX first.

Change-Id: I7934243d90c0739a7f11a8d70f6fb2c000144587
Signed-off-by: Mark Mendell <mark.p.mendell@intel.com>
9 years agoRevert "Revert "x86: Fix art_quick_instrumentation_exit""
Andreas Gampe [Fri, 5 Sep 2014 05:43:34 +0000 (05:43 +0000)]
Revert "Revert "x86: Fix art_quick_instrumentation_exit""

Need to use GOT to avoid TEXTREL.

This reverts commit 62aa66f8532b8eb44e882596010b33e7781b0754.

Change-Id: Ife659c28d621596ad288999fa128dcf32fc0b1c7

9 years agoMerge "Revert "x86: Fix art_quick_instrumentation_exit""
Andreas Gampe [Fri, 5 Sep 2014 05:02:20 +0000 (05:02 +0000)]
Merge "Revert "x86: Fix art_quick_instrumentation_exit""

9 years agoRevert "x86: Fix art_quick_instrumentation_exit"
Andreas Gampe [Fri, 5 Sep 2014 05:02:08 +0000 (05:02 +0000)]
Revert "x86: Fix art_quick_instrumentation_exit"

GCC 4.8 can't handle it correctly.

This reverts commit 5324dc896b69fc14a24334f52169e2319b9b3842.

Change-Id: I2478a52f147b585658d0e5769c3fce7f2c75240e

9 years agoMerge "x86: Fix art_quick_instrumentation_exit"
Andreas Gampe [Fri, 5 Sep 2014 02:56:43 +0000 (02:56 +0000)]
Merge "x86: Fix art_quick_instrumentation_exit"

9 years agox86: Fix art_quick_instrumentation_exit
Dmitry Petrochenko [Wed, 27 Aug 2014 11:45:56 +0000 (18:45 +0700)]
x86: Fix art_quick_instrumentation_exit

Make sure that we use a move-immediate instead of a
move-from-address (previously this was a load, whereas now
we are interested in the address as a value).

Change-Id: I4048c4a97abcc6524fe976bdc35b834d8f3bc592
Signed-off-by: Dmitry Petrochenko <dmitry.petrochenko@intel.com>
9 years agoMerge "Add test case for runFinalization."
Mathieu Chartier [Thu, 4 Sep 2014 22:32:40 +0000 (22:32 +0000)]
Merge "Add test case for runFinalization."

9 years agoAdd test case for runFinalization.
Mathieu Chartier [Thu, 4 Sep 2014 21:15:35 +0000 (14:15 -0700)]
Add test case for runFinalization.

There was a bug causing runFinalization to return before recently
freed objects were finalized. This is a regression test for this bug.

Bug: 17381967
Change-Id: Ide6e2037685324423e83965fae3935f3e7f0aba6

9 years agoMerge "Fix arm64 build warning (promoted to error)."
Dave Allison [Thu, 4 Sep 2014 21:59:58 +0000 (21:59 +0000)]
Merge "Fix arm64 build warning (promoted to error)."

9 years agoMerge "ART: Change ART for new native bridge sequence"
Andreas Gampe [Thu, 4 Sep 2014 21:47:08 +0000 (21:47 +0000)]
Merge "ART: Change ART for new native bridge sequence"

9 years agoMerge "ART: Fix host run-test script"
Andreas Gampe [Thu, 4 Sep 2014 21:37:19 +0000 (21:37 +0000)]
Merge "ART: Fix host run-test script"

9 years agoFix arm64 build warning (promoted to error).
Dave Allison [Thu, 4 Sep 2014 21:33:42 +0000 (14:33 -0700)]
Fix arm64 build warning (promoted to error).

This refactors code that caused an warning on the arm64 build.

Change-Id: I1c0ddabb7061ea289d98b03554bf764c26256078

9 years agoART: Fix host run-test script
Andreas Gampe [Thu, 4 Sep 2014 21:19:51 +0000 (14:19 -0700)]
ART: Fix host run-test script

When using gdb, do not pipe and grep. GDB is interactive...

Change-Id: Idd944a8ff7a8c254f47c840bc5b2c1e352e8b105

9 years agoART: Change ART for new native bridge sequence
Andreas Gampe [Wed, 3 Sep 2014 04:22:18 +0000 (21:22 -0700)]
ART: Change ART for new native bridge sequence

Initialize or unload after a fork.

Change-Id: I5a20de1cb68dd1802937b369b14c50c9c1031c67

9 years agoMerge "Make nested signal more generic"
Dave Allison [Thu, 4 Sep 2014 18:06:25 +0000 (18:06 +0000)]
Merge "Make nested signal more generic"

9 years agoMerge "Fix heap trimmer daemon sleeping."
Mathieu Chartier [Thu, 4 Sep 2014 17:51:02 +0000 (17:51 +0000)]
Merge "Fix heap trimmer daemon sleeping."

9 years agoFix heap trimmer daemon sleeping.
Mathieu Chartier [Thu, 4 Sep 2014 04:26:36 +0000 (21:26 -0700)]
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

9 years agoMake nested signal more generic
Dave Allison [Fri, 29 Aug 2014 19:06:16 +0000 (12:06 -0700)]
Make nested signal more generic

This moves the nested signal catching code up to a slightly higher
level in the fault manager.  Originally it was down in the DumpJavaStack
function but it needs to be up in the caller of that to allow for other
handlers like the stack dumper.

Change-Id: I31029f555dd9b501f4b5c0ed5c95777004873126

9 years agoMerge "VisitClassesWithoutClassesLock isn't safe if classes move."
Ian Rogers [Thu, 4 Sep 2014 17:23:53 +0000 (17:23 +0000)]
Merge "VisitClassesWithoutClassesLock isn't safe if classes move."

9 years agoMerge "Fix stack overflow and duplicate methods while tracing."
Ian Rogers [Thu, 4 Sep 2014 17:09:14 +0000 (17:09 +0000)]
Merge "Fix stack overflow and duplicate methods while tracing."

9 years agoFix stack overflow and duplicate methods while tracing.
Jeff Hao [Wed, 3 Sep 2014 20:48:16 +0000 (13:48 -0700)]
Fix stack overflow and duplicate methods while tracing.

Bug: 16386215
Change-Id: I0d0ae0113a3a00013ce84a1f5a110e2c52f19b86
(cherry picked from commit 6b28a456b075fe53dfb7e924a44dbf35d0d41eb3)

9 years agoMerge "ART: Fix x86_64 GenSelect case when destination is Ref"
Ian Rogers [Thu, 4 Sep 2014 16:14:37 +0000 (16:14 +0000)]
Merge "ART: Fix x86_64 GenSelect case when destination is Ref"

9 years agoART: Fix x86_64 GenSelect case when destination is Ref
Jean Christophe Beyler [Thu, 4 Sep 2014 15:34:28 +0000 (08:34 -0700)]
ART: Fix x86_64 GenSelect case when destination is Ref

Reference in x86_64 is a 64-bit solo register. As a result, the invocation
  of OpRegImm results in an error when Select opcode of the kind:
    ref = boolean ? null : null;
because opRegImm does not support 64-bit destination for OpMov.

The case above is only possible for ref because no one other constant except
  null is possible.

Bug: 17327895
Change-Id: I7541e744ec1c8619711712fd17be72764efcf3a8
Signed-off-by: Serguei Katkov <serguei.i.katkov@intel.com>
9 years agoMerge "ART: Reduce LockCallTemps usage"
Ian Rogers [Thu, 4 Sep 2014 15:48:22 +0000 (15:48 +0000)]
Merge "ART: Reduce LockCallTemps usage"

9 years agoART: Reduce LockCallTemps usage
Maxim Kazantsev [Mon, 18 Aug 2014 11:43:55 +0000 (18:43 +0700)]
ART: Reduce LockCallTemps usage

Using FlushAllRegs/LockCallTemps in integer arithmetics causes
excess register flushing and clobbering. This patch adds API that
allows to flush, clobber and lock only those registers we really
need for calculations.

Change-Id: Idabaa4fff4d18a33e5040a80f66f2df6432f8be0
Signed-off-by: Max Kazantsev <maxim.kazantsev@intel.com>
9 years agoMerge "ART: Allow oatdump to print vr stack locations"
Ian Rogers [Thu, 4 Sep 2014 15:36:20 +0000 (15:36 +0000)]
Merge "ART: Allow oatdump to print vr stack locations"

9 years agoART: Allow oatdump to print vr stack locations
Razvan A Lupusoru [Sat, 30 Aug 2014 00:56:46 +0000 (17:56 -0700)]
ART: Allow oatdump to print vr stack locations

For both debugging and performance analysis, it is necessary to understand
stack layout. This patch adds capability to oatdump to print out the offsets
of the locals, ins, method*, and out VRs.

Change-Id: I73512f59e4fd2d2b12725a6c76d602182c46ff78
Signed-off-by: Razvan A Lupusoru <razvan.a.lupusoru@intel.com>
9 years agoMerge "ART: Vectorization opcode implementation fixes"
Ian Rogers [Thu, 4 Sep 2014 15:27:29 +0000 (15:27 +0000)]
Merge "ART: Vectorization opcode implementation fixes"

9 years agoMerge "Add numerator check for integer divide and modulo"
Ian Rogers [Thu, 4 Sep 2014 15:20:55 +0000 (15:20 +0000)]
Merge "Add numerator check for integer divide and modulo"

9 years agoMerge "ART: Only call CalculateBasicBlockInformation once"
Ian Rogers [Thu, 4 Sep 2014 15:14:44 +0000 (15:14 +0000)]
Merge "ART: Only call CalculateBasicBlockInformation once"

9 years agoART: Only call CalculateBasicBlockInformation once
Jean Christophe Beyler [Tue, 2 Sep 2014 21:22:17 +0000 (14:22 -0700)]
ART: Only call CalculateBasicBlockInformation once

CalculateBasicBlockInformation gets called twice now when dirty is true.

Change-Id: I8aa0569e4c00ea884a45ecea957a4ae06a328969
Signed-off-by: Jean Christophe Beyler <jean.christophe.beyler@intel.com>
9 years agoMerge "ART: Fix computation of frame size for direct proxy methods"
Ian Rogers [Thu, 4 Sep 2014 14:22:44 +0000 (14:22 +0000)]
Merge "ART: Fix computation of frame size for direct proxy methods"

9 years agoART: Fix computation of frame size for direct proxy methods
Serguei Katkov [Fri, 29 Aug 2014 11:20:15 +0000 (18:20 +0700)]
ART: Fix computation of frame size for direct proxy methods

Proxy method has only one direct method and it is a constructor which
is cloned from java.lang.reflect.Proxy class together with code.
As a result its body is a compiled quick code and frame size should be
computed accordingly in contrast with other virtual methods which are
invoked through stub.

Change-Id: I0ac99b2f567b281c9342152bad1149dd6cd39480
Signed-off-by: Serguei Katkov <serguei.i.katkov@intel.com>
9 years agoMerge "Fix native allocation watermark clamping."
Mathieu Chartier [Thu, 4 Sep 2014 04:36:08 +0000 (04:36 +0000)]
Merge "Fix native allocation watermark clamping."

9 years agoFix native allocation watermark clamping.
Mathieu Chartier [Wed, 3 Sep 2014 17:30:11 +0000 (10:30 -0700)]
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

9 years agoMerge "Add exception check to AllocateInternalWithGc"
Mathieu Chartier [Thu, 4 Sep 2014 03:32:56 +0000 (03:32 +0000)]
Merge "Add exception check to AllocateInternalWithGc"

9 years agoAdd exception check to AllocateInternalWithGc
Mathieu Chartier [Wed, 3 Sep 2014 18:21:08 +0000 (11:21 -0700)]
Add exception check to AllocateInternalWithGc

Trying to figure out if the pending exception comes before or during
AllocateInternalWithGc.

Bug: 17164348

(cherry picked from commit 98d43cb0da3c34d1888d1e31d953c2940fd0f897)

Change-Id: I2aac4b67f0c3f1c0fe3c9dc321316f76494c4074

9 years agoVisitClassesWithoutClassesLock isn't safe if classes move.
Ian Rogers [Fri, 29 Aug 2014 22:40:08 +0000 (15:40 -0700)]
VisitClassesWithoutClassesLock isn't safe if classes move.

Which they do, so avoid by doing an array allocation.
Also, tidy member variables to the end of ClassLinker.
Remove unnecessary mutable. Tidy and fix a locks required/excluded.

Change-Id: I2404a9e7a1ea997d68ab1206f97d2a20dffbda06

9 years agoMerge "Remove abuse of mirror::Object* to reference special values."
Ian Rogers [Thu, 4 Sep 2014 00:20:48 +0000 (00:20 +0000)]
Merge "Remove abuse of mirror::Object* to reference special values."

9 years agoRemove abuse of mirror::Object* to reference special values.
Ian Rogers [Wed, 3 Sep 2014 23:16:56 +0000 (16:16 -0700)]
Remove abuse of mirror::Object* to reference special values.

Remove kInvalidIndirectRefObject, kClearedJniWeakGlobal and
ObjectRegistry::kInvalidObject. Handle error conditions by passing in or
returning an error value.
GetObjectRefType is simplified to be faster and not return invalid references
that are not expected according to the spec. Adjust check JNI and
jni_internal_test appropriately.
Fix cases in the debugger/JDWP of out arguments being passed by reference.
Bug: 17376993

Change-Id: I3ce8a28c01827e163f4dc288449959464da788b1

9 years agoART: Vectorization opcode implementation fixes
Lupusoru, Razvan A [Mon, 28 Jul 2014 21:11:01 +0000 (14:11 -0700)]
ART: Vectorization opcode implementation fixes

This patch fixes the implementation of the x86 vectorization opcodes.

Change-Id: I0028d54a9fa6edce791b7e3a053002d076798748
Signed-off-by: Razvan A Lupusoru <razvan.a.lupusoru@intel.com>
Signed-off-by: Udayan Banerji <udayan.banerji@intel.com>
Signed-off-by: Philbert Lin <philbert.lin@intel.com>
9 years agoMerge "Fix SetupResourceMasks to pass correct parameters"
Vladimir Marko [Wed, 3 Sep 2014 11:48:42 +0000 (11:48 +0000)]
Merge "Fix SetupResourceMasks to pass correct parameters"

9 years agoMerge "ART fix oat debug source map operations"
Vladimir Marko [Wed, 3 Sep 2014 08:53:49 +0000 (08:53 +0000)]
Merge "ART fix oat debug source map operations"

9 years agoMerge "Reduce lock contention when debugging"
Sebastien Hertz [Wed, 3 Sep 2014 07:17:29 +0000 (07:17 +0000)]
Merge "Reduce lock contention when debugging"

9 years agoMerge "Quick compiler, aarch64: Insane sanity checker"
buzbee [Wed, 3 Sep 2014 01:41:16 +0000 (01:41 +0000)]
Merge "Quick compiler, aarch64: Insane sanity checker"

9 years agoMerge "Fix native allocation test."
Mathieu Chartier [Wed, 3 Sep 2014 01:14:14 +0000 (01:14 +0000)]
Merge "Fix native allocation test."

9 years agoMerge "Change native allocations to use growth limit."
Mathieu Chartier [Wed, 3 Sep 2014 01:13:57 +0000 (01:13 +0000)]
Merge "Change native allocations to use growth limit."

9 years agoMerge "ART: Fix StubTest Interface Trampoline test"
Andreas Gampe [Wed, 3 Sep 2014 01:00:21 +0000 (01:00 +0000)]
Merge "ART: Fix StubTest Interface Trampoline test"

9 years agoART: Fix StubTest Interface Trampoline test
Andreas Gampe [Wed, 21 May 2014 22:37:53 +0000 (15:37 -0700)]
ART: Fix StubTest Interface Trampoline test

The imt conflict test does not actually need a proper conflict,
as we do not invoke through a lookup.

Added simple check for art_quick_invoke_interface_trampoline_with_access_check.
Only non-failure testing, as failure requires a properly setup frame etc.

Change-Id: I173dac04db950bd23f62eaf13341a7b19802e5e8

9 years agoFix native allocation test.
Mathieu Chartier [Wed, 3 Sep 2014 00:36:08 +0000 (17:36 -0700)]
Fix native allocation test.

Forgot to update with last CL.

(cherry picked from commit c60bdf667b5c6fa767548942f0115547523593b4)

Change-Id: I41ad168a5906ca07938bdd212a55f0e995ee5df7

9 years agoChange native allocations to use growth limit.
Mathieu Chartier [Tue, 2 Sep 2014 23:21:01 +0000 (16:21 -0700)]
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

9 years agoMerge "Add missing read barriers to intern table."
Mathieu Chartier [Wed, 3 Sep 2014 00:47:15 +0000 (00:47 +0000)]
Merge "Add missing read barriers to intern table."

9 years agoMerge "ART: Fix read-out-of-bounds in the compiler"
Andreas Gampe [Wed, 3 Sep 2014 00:31:30 +0000 (00:31 +0000)]
Merge "ART: Fix read-out-of-bounds in the compiler"

9 years agoART: Fix read-out-of-bounds in the compiler
Andreas Gampe [Wed, 27 Aug 2014 21:24:42 +0000 (14:24 -0700)]
ART: Fix read-out-of-bounds in the compiler

In case of a wide dalvik register, asking for the constant value
can lead to a read out of bounds.

Bug: 17302671

(cherry picked from commit ade731854d18839823e57fb2d3d67238c5467d15)

Change-Id: Ie1849cd67cc418c97cbd7a8524f027f9b66e4c96

9 years agoMerge "ART: Tighten verifier list reading and offsets"
Andreas Gampe [Wed, 3 Sep 2014 00:06:22 +0000 (00:06 +0000)]
Merge "ART: Tighten verifier list reading and offsets"

9 years agoART: Tighten verifier list reading and offsets
Andreas Gampe [Tue, 2 Sep 2014 18:17:34 +0000 (11:17 -0700)]
ART: Tighten verifier list reading and offsets

Check offsets and sizes for header entries of a dex file.

Bug: 17347459

(cherry picked from commit 78d639ef4be3ad7314846e1e6c1261d7d30f83fa)

Change-Id: Ia1727c33dea51f7a8e345f3799f1ba414708239c

9 years agoAdd missing read barriers to intern table.
Mathieu Chartier [Tue, 2 Sep 2014 23:20:45 +0000 (16:20 -0700)]
Add missing read barriers to intern table.

Also deleted GcRoot::Assign.

Change-Id: Ib1ea739cf79c72fc92a8628cd9095c11b660e018

9 years agoMerge "Change intern table to unordered set."
Mathieu Chartier [Tue, 2 Sep 2014 23:42:02 +0000 (23:42 +0000)]
Merge "Change intern table to unordered set."

9 years agoMerge "Disable SignalTest with GC stress."
Ian Rogers [Tue, 2 Sep 2014 23:32:25 +0000 (23:32 +0000)]
Merge "Disable SignalTest with GC stress."

9 years agoFix SetupResourceMasks to pass correct parameters
Junmo Park [Sat, 30 Aug 2014 11:13:02 +0000 (20:13 +0900)]
Fix SetupResourceMasks to pass correct parameters

Calling SetupTargetResourceMasks function with correct parameter.
SetupTargetResourceMasks(lir, flags, use_mask, def_mask) is correct order.

Change-Id: I49118c3f2d3d06fac83fa3d5014b7fdaff6a2b00
Signed-off-by: Junmo Park <junmoz.park@samsung.com>
9 years agoDisable SignalTest with GC stress.
Ian Rogers [Tue, 2 Sep 2014 22:59:28 +0000 (15:59 -0700)]
Disable SignalTest with GC stress.

Also, only add to broken tests list run-tests that will be generated.

Change-Id: Ibd363c059c799110da7f3f5ec8ceba55277d5e0e

9 years agoQuick compiler, aarch64: Insane sanity checker
buzbee [Tue, 2 Sep 2014 22:39:57 +0000 (15:39 -0700)]
Quick compiler, aarch64: Insane sanity checker

Fix bit-mask generation used in the sanity checking code
for long division by small constants via multiplication.  The
failing case was a 64-bit wide mask.  The macro was using the
desired width of the result as a shift count - and a left
shift of a 64-bit integer by 64 bits is undefined.

Note that the generated code is correct - it's just the sanity
checking code that was wrong.

Change-Id: Ic6654662f848d3caab2c3b1ef0d92fe2eb9673f4

9 years agoMerge "Pre-allocate the NoClassDefFoundError to be thrown for boot classes."
Ian Rogers [Tue, 2 Sep 2014 22:49:45 +0000 (22:49 +0000)]
Merge "Pre-allocate the NoClassDefFoundError to be thrown for boot classes."

9 years agoPre-allocate the NoClassDefFoundError to be thrown for boot classes.
Ian Rogers [Wed, 4 Jun 2014 23:57:15 +0000 (16:57 -0700)]
Pre-allocate the NoClassDefFoundError to be thrown for boot classes.

Bring over a Dalvik "optimization".
Bug: 12804658
Bug: 16853450
Change-Id: I6419de7bd2ba18d91479cb52489104954f5c4524

9 years agoChange intern table to unordered set.
Mathieu Chartier [Sat, 30 Aug 2014 01:16:58 +0000 (18:16 -0700)]
Change intern table to unordered set.

Intern table active used bytes goes from 430k to 317k on system
server. Similar %wise savings on other apps.

Bug: 16238192

(cherry picked from commit d910fcef539e12ab181e56ec80684f39c4e95733)

Change-Id: Ic70395124435c6f420a77e6d8639404a160f395a

9 years agoMerge "Reduce and speed-up class def searches."
Ian Rogers [Tue, 2 Sep 2014 21:22:20 +0000 (21:22 +0000)]
Merge "Reduce and speed-up class def searches."

9 years agoMerge "Change heap transitions + hspace compation to use VLOG(heap)"
Mathieu Chartier [Tue, 2 Sep 2014 19:56:48 +0000 (19:56 +0000)]
Merge "Change heap transitions + hspace compation to use VLOG(heap)"

9 years agoChange heap transitions + hspace compation to use VLOG(heap)
Mathieu Chartier [Tue, 2 Sep 2014 19:33:25 +0000 (12:33 -0700)]
Change heap transitions + hspace compation to use VLOG(heap)

VLOG(gc) is spammy since it is enabled.

(cherry picked from commit 8bd2961b385c870953e650c253a04ee24b76809b)

Change-Id: Id75256cdc6a9802818bb56f77533cca88ace42a6

9 years agoMerge "ART: Fix typo"
Andreas Gampe [Tue, 2 Sep 2014 17:43:07 +0000 (17:43 +0000)]
Merge "ART: Fix typo"

9 years agoART: Fix typo
Andreas Gampe [Tue, 2 Sep 2014 17:22:20 +0000 (10:22 -0700)]
ART: Fix typo

Bug: 17327877

(cherry picked from commit b373ba1bf7e67d4dd8286df08065ac76aa8ac1ee)

Change-Id: I6132a8afb4b69c8d22a01b477f6f324c789ed666

9 years agoReduce lock contention when debugging
Sebastien Hertz [Tue, 19 Aug 2014 13:33:43 +0000 (15:33 +0200)]
Reduce lock contention when debugging

Uses a ReaderWriterMutex for the breakpoint lock to reduce contention during
debugging session.

Also adds missing thread safety annotations on fields and methods related to
instrumentation and debugging.

Bug: 16814665
Bug: 11667502

(cherry picked from commit 59d9d668d4f4286813afe2b4e7c6db839222ce96)

Change-Id: I5f1156da8c6cc8316d6db16e6cfb2470fe289ad3

9 years agoMerge "ART: Fix dex file verifier type-list handling"
Andreas Gampe [Tue, 2 Sep 2014 16:23:23 +0000 (16:23 +0000)]
Merge "ART: Fix dex file verifier type-list handling"

9 years agoART: Fix dex file verifier type-list handling
Andreas Gampe [Fri, 29 Aug 2014 23:07:49 +0000 (16:07 -0700)]
ART: Fix dex file verifier type-list handling

It is rare, but valid, to have an empty type list.

Bug: 17327877

(cherry picked from commit 277a7c7b4bb9c421380592fd3998d2e79e4035b3)

Change-Id: Ib3a8ff3e5ccd8fe7c04b1e97485bf3e6de72aa4d

9 years agoReduce and speed-up class def searches.
Ian Rogers [Sat, 30 Aug 2014 03:19:11 +0000 (20:19 -0700)]
Reduce and speed-up class def searches.

Use the class linker for descriptor lookups from the compile driver so that
dex caches are populated.
Reduce the scope of functions for scanning class paths to just the class
linker where they are performed.
If we see more than a threshold number of find class def misses on a dex file
lazily compute an index, so that future lookups are constant time (part of the
collection code is taken from
https://android-review.googlesource.com/#/c/103865/3). Note that we take a lazy
approach so that we don't serialize on loading dex files, this avoids the
reason the index was removed in 8b2c0b9abc3f520495f4387ea040132ba85cae69.
Remove an implicit and unnecessary std::string creation for PrintableString.

Single threaded interpret-only dex2oat performance is improved by roughly 10%.

Bug: 16853450

Change-Id: Icf72df76b0a4328f2a24075e81f4ff267b9401f4

9 years agoMerge "Revert "Revert "ART: Prepare for ELF64."""
Nicolas Geoffray [Tue, 2 Sep 2014 08:12:44 +0000 (08:12 +0000)]
Merge "Revert "Revert "ART: Prepare for ELF64."""

9 years agoRevert "Revert "ART: Prepare for ELF64.""
Nicolas Geoffray [Tue, 2 Sep 2014 08:12:09 +0000 (08:12 +0000)]
Revert "Revert "ART: Prepare for ELF64.""

Fix was commited in "Fix frame size issue with 64-bit"

This reverts commit 7eebd95bddc7c6866ba29010d28b4f8251b43242.

Change-Id: I799b7fcab14ae24471d2f9d6d3a5e5cedd4aa7b8

9 years agoMerge "Fix frame size issue with 64-bit"
Narayan Kamath [Mon, 1 Sep 2014 15:55:16 +0000 (15:55 +0000)]
Merge "Fix frame size issue with 64-bit"

9 years agoMerge "Revert "ART: Prepare for ELF64.""
Nicolas Geoffray [Mon, 1 Sep 2014 15:14:16 +0000 (15:14 +0000)]
Merge "Revert "ART: Prepare for ELF64.""

9 years agoRevert "ART: Prepare for ELF64."
Nicolas Geoffray [Mon, 1 Sep 2014 15:13:00 +0000 (15:13 +0000)]
Revert "ART: Prepare for ELF64."

Build fails on arm64 with error message:
art/compiler/elf_writer_quick.cc:1083:1: error: the frame size of 2016 bytes is larger than 1728 bytes [-Werror=frame-larger-than=]

This reverts commit 45b8f1ab2782d02830c77751f883456de799dfb6.

Change-Id: I61b2e5f27ba4b2b32eacaec51fb5e5ed12fd8066

9 years agoMerge "Add native memory accounting through custom allocator."
Mathieu Chartier [Sat, 30 Aug 2014 01:58:53 +0000 (01:58 +0000)]
Merge "Add native memory accounting through custom allocator."

9 years agoAdd native memory accounting through custom allocator.
Mathieu Chartier [Mon, 25 Aug 2014 20:08:22 +0000 (13:08 -0700)]
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)

9 years agoMerge "ART: Fix accidental "1""
Andreas Gampe [Sat, 30 Aug 2014 01:16:09 +0000 (01:16 +0000)]
Merge "ART: Fix accidental "1""

9 years agoART: Fix accidental "1"
Andreas Gampe [Sat, 30 Aug 2014 01:15:04 +0000 (18:15 -0700)]
ART: Fix accidental "1"

Should have been "0".

Change-Id: I9ad3cf1e6be6d0fcf4c8f3700ddeec4e6d5ce167

9 years agoMerge "ART: Initialize fields of ParsedOptions"
Andreas Gampe [Sat, 30 Aug 2014 00:28:46 +0000 (00:28 +0000)]
Merge "ART: Initialize fields of ParsedOptions"

9 years agoFix frame size issue with 64-bit
Brian Carlstrom [Fri, 29 Aug 2014 23:20:48 +0000 (16:20 -0700)]
Fix frame size issue with 64-bit

Change-Id: Ib2f33dfb79952ba39f2d7e5303a2a2e6c4b0a0f6

9 years agoMerge "ART: Address issues with kIntrinsicMinMaxDouble for x86"
Ian Rogers [Fri, 29 Aug 2014 23:09:47 +0000 (23:09 +0000)]
Merge "ART: Address issues with kIntrinsicMinMaxDouble for x86"

9 years agoMerge "ART: Update CFG dumper to handle extended better"
Ian Rogers [Fri, 29 Aug 2014 23:04:06 +0000 (23:04 +0000)]
Merge "ART: Update CFG dumper to handle extended better"

9 years agoART: Initialize fields of ParsedOptions
Andreas Gampe [Fri, 29 Aug 2014 23:01:25 +0000 (16:01 -0700)]
ART: Initialize fields of ParsedOptions

Avoids certain paths where we touched uninitialized fields.

Bug: 17330858
Change-Id: Ia68161b12eefefacad6fc524f487f8e2caf6342a

9 years agoMerge "Fix master-art-host build by removing flakey test (temporarily)"
Dave Allison [Fri, 29 Aug 2014 21:20:05 +0000 (21:20 +0000)]
Merge "Fix master-art-host build by removing flakey test (temporarily)"

9 years agoMerge "ART: Prepare for ELF64."
Brian Carlstrom [Fri, 29 Aug 2014 21:17:50 +0000 (21:17 +0000)]
Merge "ART: Prepare for ELF64."