OSDN Git Service

android-x86/art.git
10 years agoam f2e92073: Merge "Add GcRoot to clean up and enforce read barriers."
Hiroshi Yamauchi [Tue, 29 Jul 2014 22:21:51 +0000 (22:21 +0000)]
am f2e92073: Merge "Add GcRoot to clean up and enforce read barriers."

* commit 'f2e920734dddbc0eee35c3ba27b88b732df4f82c':
  Add GcRoot to clean up and enforce read barriers.

10 years agoam 8df73882: Merge "Revert "Fix proxy tracing and enable tests that now work with...
Nicolas Geoffray [Tue, 29 Jul 2014 14:23:29 +0000 (14:23 +0000)]
am 8df73882: Merge "Revert "Fix proxy tracing and enable tests that now work with tracing.""

* commit '8df73882c60451e7f789bf9b1f3db2d7dc228640':
  Revert "Fix proxy tracing and enable tests that now work with tracing."

10 years agoam 01be6e3c: Merge "Fix braino in Makefile."
Nicolas Geoffray [Tue, 29 Jul 2014 14:23:28 +0000 (14:23 +0000)]
am 01be6e3c: Merge "Fix braino in Makefile."

* commit '01be6e3c4e4f083842ac9a6f8f9e6c2c937b8a15':
  Fix braino in Makefile.

10 years agoam 2dd6f729: Merge "Add a variable to disable PREBUILD in run-tests."
Nicolas Geoffray [Tue, 29 Jul 2014 14:23:27 +0000 (14:23 +0000)]
am 2dd6f729: Merge "Add a variable to disable PREBUILD in run-tests."

* commit '2dd6f729c05ff9a6c687888707c68e7894392c38':
  Add a variable to disable PREBUILD in run-tests.

10 years agoam 9d2f0100: Merge "AArch64: Clean up CalleeSaveMethod frame and the use of temp...
Andreas Gampe [Tue, 29 Jul 2014 14:23:26 +0000 (14:23 +0000)]
am 9d2f0100: Merge "AArch64: Clean up CalleeSaveMethod frame and the use of temp registers."

* commit '9d2f0100840a41ed6c6c95f92994db7975a8a211':
  AArch64: Clean up CalleeSaveMethod frame and the use of temp registers.

10 years agoam 2815f124: Merge "Fix proxy tracing and enable tests that now work with tracing."
Jeff Hao [Tue, 29 Jul 2014 01:40:37 +0000 (01:40 +0000)]
am 2815f124: Merge "Fix proxy tracing and enable tests that now work with tracing."

* commit '2815f1242c6c3ea1fc2df7bb5e4bd1924f4e75f7':
  Fix proxy tracing and enable tests that now work with tracing.

10 years agoam 4c6427a7: ART: Reject array-length on non-reference register
Andreas Gampe [Mon, 28 Jul 2014 23:16:22 +0000 (23:16 +0000)]
am 4c6427a7: ART: Reject array-length on non-reference register

* commit '4c6427a757cae2b1c79aafca71be780e54620e68':
  ART: Reject array-length on non-reference register

10 years agoam e5647289: Merge "ART: Fix missing initialization in stack_map_stream.h"
Andreas Gampe [Mon, 28 Jul 2014 22:47:58 +0000 (22:47 +0000)]
am e5647289: Merge "ART: Fix missing initialization in stack_map_stream.h"

* commit 'e5647289ccc7dafd8b91771a7bd1cd19416aed66':
  ART: Fix missing initialization in stack_map_stream.h

10 years agoMerge "Add GcRoot to clean up and enforce read barriers."
Hiroshi Yamauchi [Tue, 29 Jul 2014 22:00:56 +0000 (22:00 +0000)]
Merge "Add GcRoot to clean up and enforce read barriers."

10 years agoAdd GcRoot to clean up and enforce read barriers.
Hiroshi Yamauchi [Wed, 23 Jul 2014 01:08:23 +0000 (18:08 -0700)]
Add GcRoot to clean up and enforce read barriers.

Introduce a value-type wrapper around Object* for GC roots so that 1)
we won't have to directly add the read barrier code in many places and
2) we can avoid accidentally bypassing/missing read barriers on GC
roots (the GcRoot interface ensures that the read barrier is executed
on a read).

The jdwp test passed.

Bug: 12687968
Change-Id: Ib167c7c325b3c7e3900133578815f04d219972a1

10 years agoMerge "AArch64: Clean up CalleeSaveMethod frame and the use of temp registers."
Andreas Gampe [Tue, 29 Jul 2014 07:45:01 +0000 (07:45 +0000)]
Merge "AArch64: Clean up CalleeSaveMethod frame and the use of temp registers."

10 years agoAArch64: Clean up CalleeSaveMethod frame and the use of temp registers.
Zheng Xu [Fri, 25 Jul 2014 03:49:42 +0000 (11:49 +0800)]
AArch64: Clean up CalleeSaveMethod frame and the use of temp registers.

CalleeSaveMethod frame size changes :
SaveAll : 368 -> 176
RefOnly : 176 -> 96
RefsAndArgs : 304 -> 224

JNI register spill size changes :
160 -> 88

In the transition assembly, use registers following the rules:
1. x0-x7 as temp/argument registers.
2. IP0, IP1 as scratch registers.
3. After correct type of callee-save-frame has been setup, all registers
are scratch-able(probably except xSELF and xSUSPEND).
4. When restore callee-save-frame, IP0 and IP1 should be untouched.
5. From C to managed code, we assume all callee save register in AAPCS
will be restored by managed code except x19(SUSPEND).

In quick compiler:
1. Use IP0, IP1 as scratch register.
2. Use IP1 as hidden argument register(IP0 will be scratched by
 trampoline.)

Change-Id: I05ed9d418b01b9e87218a7608536f57e7a286e4c

10 years agoMerge "Fix proxy tracing and enable tests that now work with tracing."
Jeff Hao [Tue, 29 Jul 2014 01:34:08 +0000 (01:34 +0000)]
Merge "Fix proxy tracing and enable tests that now work with tracing."

10 years agoFix proxy tracing and enable tests that now work with tracing.
Jeff Hao [Thu, 24 Jul 2014 23:26:09 +0000 (16:26 -0700)]
Fix proxy tracing and enable tests that now work with tracing.

Bug: 16386215
Change-Id: Iec2a372c921caceb050c6baf72d48b3d822899a4

10 years agoam daa31c58: Merge "ART: Reject array-length on non-reference register"
Andreas Gampe [Mon, 28 Jul 2014 20:40:31 +0000 (20:40 +0000)]
am daa31c58: Merge "ART: Reject array-length on non-reference register"

* commit 'daa31c58ec2fec209f8fdc383bf10dd2e2c64512':
  ART: Reject array-length on non-reference register

10 years agoART: Reject array-length on non-reference register
Andreas Gampe [Mon, 28 Jul 2014 20:14:34 +0000 (13:14 -0700)]
ART: Reject array-length on non-reference register

It should be a hard verifier failure if we detect an array-length
instruction over a non-reference-type register.

Bug: 16595896

(cherry picked from commit 65c9db8dfbeea3f708f95f058f4fed7c2af71052)

Change-Id: I56babcbe95d50fd66a132cf4434cf5fa780beb4e

10 years agoam 3f49507c: Merge "ART: Rework quick entrypoint code in Mir2Lir, cleanup"
Andreas Gampe [Mon, 28 Jul 2014 16:46:17 +0000 (16:46 +0000)]
am 3f49507c: Merge "ART: Rework quick entrypoint code in Mir2Lir, cleanup"

* commit '3f49507c333b7c8d36620870522a7a1055987ef8':
  ART: Rework quick entrypoint code in Mir2Lir, cleanup

10 years agoam 37f1c7c6: Merge "Implement array get and array put in optimizing."
Nicolas Geoffray [Mon, 28 Jul 2014 16:40:20 +0000 (16:40 +0000)]
am 37f1c7c6: Merge "Implement array get and array put in optimizing."

* commit '37f1c7c6e084c1160eb1b156ab63ac114a82625a':
  Implement array get and array put in optimizing.

10 years agoam 98cc1e55: Merge "ART: Fix target test paths"
Andreas Gampe [Sun, 27 Jul 2014 22:03:49 +0000 (22:03 +0000)]
am 98cc1e55: Merge "ART: Fix target test paths"

* commit '98cc1e552c2ccbe5d51bc81d49e79119280f5416':
  ART: Fix target test paths

10 years agoam b3bf7afc: Merge "Remove the per-project FDO support. Use the common FDO support...
Ian Rogers [Sun, 27 Jul 2014 18:37:29 +0000 (18:37 +0000)]
am b3bf7afc: Merge "Remove the per-project FDO support. Use the common FDO support from build/core. LOCAL_FDO_SUPPORT indicates if the module should enable FDO builds."

* commit 'b3bf7afc4288f97f9b591d19f03362e0577ed0ec':
  Remove the per-project FDO support. Use the common FDO support from build/core. LOCAL_FDO_SUPPORT indicates if the module should enable FDO builds.

10 years agoam 4483fbfd: Merge "Revert "Enable Load Store Elimination for ARM and ARM64""
Bill Buzbee [Sun, 27 Jul 2014 17:56:10 +0000 (17:56 +0000)]
am 4483fbfd: Merge "Revert "Enable Load Store Elimination for ARM and ARM64""

* commit '4483fbfdf17fd7230a52cc54bc5f7fbca82bbd75':
  Revert "Enable Load Store Elimination for ARM and ARM64"

10 years agoam 123c5743: Merge "Enable Load Store Elimination for ARM and ARM64"
Bill Buzbee [Sun, 27 Jul 2014 17:36:41 +0000 (17:36 +0000)]
am 123c5743: Merge "Enable Load Store Elimination for ARM and ARM64"

* commit '123c5743374137711163346e75aaad587d142317':
  Enable Load Store Elimination for ARM and ARM64

10 years agoam a65d63e4: Merge "SetConstantWide should mark both SSA regs as constant"
Bill Buzbee [Sun, 27 Jul 2014 02:53:49 +0000 (02:53 +0000)]
am a65d63e4: Merge "SetConstantWide should mark both SSA regs as constant"

* commit 'a65d63e4eb81605fd798f1ca038d651b0faa29eb':
  SetConstantWide should mark both SSA regs as constant

10 years agoam 5e3f45bd: Merge "ART: ReEnable check for a gap between method\'s arg "in" vregs...
Bill Buzbee [Sun, 27 Jul 2014 02:53:49 +0000 (02:53 +0000)]
am 5e3f45bd: Merge "ART: ReEnable check for a gap between method\'s arg "in" vregs and rest of its vregs"

* commit '5e3f45bdb987ba7fe7a4114cb5fe8483432015b6':
  ART: ReEnable check for a gap between method's arg "in" vregs and rest of its vregs

10 years agoam b97df048: Merge "ART: Native bridge command-line parameter"
Andreas Gampe [Sun, 27 Jul 2014 01:43:52 +0000 (01:43 +0000)]
am b97df048: Merge "ART: Native bridge command-line parameter"

* commit 'b97df0488b5391564b5ea195a0765c943787f88d':
  ART: Native bridge command-line parameter

10 years agoam 22e59fd2: ART: Allow arrays with erroneous component type
Andreas Gampe [Sun, 27 Jul 2014 01:39:57 +0000 (01:39 +0000)]
am 22e59fd2: ART: Allow arrays with erroneous component type

* commit '22e59fd2861a4aab16adb2b80a1d5166382a6617':
  ART: Allow arrays with erroneous component type

10 years agoam 0db2c431: Merge "ART: Allow arrays with erroneous component type"
Andreas Gampe [Sun, 27 Jul 2014 01:38:09 +0000 (01:38 +0000)]
am 0db2c431: Merge "ART: Allow arrays with erroneous component type"

* commit '0db2c431a3971781a52629be62f572cc5b0d35df':
  ART: Allow arrays with erroneous component type

10 years agoART: Allow arrays with erroneous component type
Andreas Gampe [Thu, 24 Jul 2014 03:18:36 +0000 (20:18 -0700)]
ART: Allow arrays with erroneous component type

Array classes must tolerate having component type classes that are
erroneous. Change CreateArrayClass to use LookupClass when FindClass
failed.

Bug: 16019155

(cherry picked from commit dc13d7df5da49e93963035633a82699c68fa0971)

Change-Id: I506250949a1802898433e9099dcb8ef31dd89659

10 years agoam 0237ac84: ART: Special-case cb(n)z even for in-reg constant
Andreas Gampe [Sat, 26 Jul 2014 20:06:46 +0000 (20:06 +0000)]
am 0237ac84: ART: Special-case cb(n)z even for in-reg constant

* commit '0237ac84b1459cb1718dce23f3572ae2fe1bd77e':
  ART: Special-case cb(n)z even for in-reg constant

10 years agoam c4a5f588: Merge "ART: Special-case cb(n)z even for in-reg constant"
Andreas Gampe [Sat, 26 Jul 2014 20:05:21 +0000 (20:05 +0000)]
am c4a5f588: Merge "ART: Special-case cb(n)z even for in-reg constant"

* commit 'c4a5f588c5c17ae226b56d25539ea4672f0ea9a3':
  ART: Special-case cb(n)z even for in-reg constant

10 years agoART: Special-case cb(n)z even for in-reg constant
Andreas Gampe [Sat, 26 Jul 2014 08:40:39 +0000 (01:40 -0700)]
ART: Special-case cb(n)z even for in-reg constant

Call out to OpCmpImmBranch in GenCompareAndBranch if the constant
is zero and we are testing == or !=, even when zero has been loaded
to a register already. This avoids a register size mismatch on 64b
architectures when basically doing a null check, and generally
squashes a cmp + branch to a cbz or cbnz on Arm and Mips. X86 is
not degraded.

Bug: 16562601

(cherry picked from commit b07c1f9f4d6088ca2d4c1a10819e57b19acf7f22)

Change-Id: I42701e827feb848470aa991297755d808fa0a077

10 years agoam 9646172b: Merge "ART: Rewrite ParallelGC run-test"
Andreas Gampe [Sat, 26 Jul 2014 13:14:46 +0000 (13:14 +0000)]
am 9646172b: Merge "ART: Rewrite ParallelGC run-test"

* commit '9646172b4e04f981a3b1cd3af0e8c9e8a610bfd1':
  ART: Rewrite ParallelGC run-test

10 years agoam ff9d95d5: Merge "Tweaks to patchoat and other related things."
Andreas Gampe [Sat, 26 Jul 2014 13:14:45 +0000 (13:14 +0000)]
am ff9d95d5: Merge "Tweaks to patchoat and other related things."

* commit 'ff9d95d58ea9c36d26b96f52c9a7ac4bf92fed2d':
  Tweaks to patchoat and other related things.

10 years agoam f071e86c: Merge "ART: Use array-inl.h in reference_table_test" into lmp-dev
Andreas Gampe [Sat, 26 Jul 2014 04:18:01 +0000 (04:18 +0000)]
am f071e86c: Merge "ART: Use array-inl.h in reference_table_test" into lmp-dev

* commit 'f071e86c901a230f55e76bc6b55e19cca16a74cb':
  ART: Use array-inl.h in reference_table_test

10 years agoMerge "Remove the per-project FDO support. Use the common FDO support from build...
Ian Rogers [Sun, 27 Jul 2014 18:32:40 +0000 (18:32 +0000)]
Merge "Remove the per-project FDO support. Use the common FDO support from build/core. LOCAL_FDO_SUPPORT indicates if the module should enable FDO builds."

10 years agoRemove the per-project FDO support. Use the common FDO support from build/core.
Dehao Chen [Thu, 24 Jul 2014 23:12:59 +0000 (16:12 -0700)]
Remove the per-project FDO support. Use the common FDO support from build/core.
LOCAL_FDO_SUPPORT indicates if the module should enable FDO builds.

Change-Id: I1eab8bf418b7ceb060eea18bdd304a06597af3d3

10 years agoMerge "Revert "Enable Load Store Elimination for ARM and ARM64""
Bill Buzbee [Sun, 27 Jul 2014 17:49:57 +0000 (17:49 +0000)]
Merge "Revert "Enable Load Store Elimination for ARM and ARM64""

10 years agoRevert "Enable Load Store Elimination for ARM and ARM64"
Bill Buzbee [Sun, 27 Jul 2014 17:49:42 +0000 (17:49 +0000)]
Revert "Enable Load Store Elimination for ARM and ARM64"

On extended testing, I'm seeing a CHECK failure at utility_arm.cc:1201.

This reverts commit fcc36ba2a2b8fd10e6eebd21ecb6329606443ded.

Change-Id: Icae3d49cd7c8fcab09f2f989cbcb1d7e5c6d137a

10 years agoEnable Load Store Elimination for ARM and ARM64
Serban Constantinescu [Tue, 15 Jul 2014 16:44:21 +0000 (17:44 +0100)]
Enable Load Store Elimination for ARM and ARM64

This patch refactors the implementation of the LoadStoreElimination
optimisation pass. Please note that this pass was disabled and not
functional for any of the backends.

The current implementation tracks aliases and handles DalvikRegs as well
as Heap memory regions. It has been tested and it is known to optimise
out the following:
  * Load - Load
  * Store - Load
  * Store - Store
  * Load Literals

Change-Id: Iefae9b696f87f833ef35c451ed4d49c5a1b6fde0

10 years agoam 3bcac48f: Merge "ART: Fix run-test 114 ParallelGC to account for OOM"
Andreas Gampe [Fri, 25 Jul 2014 23:50:07 +0000 (23:50 +0000)]
am 3bcac48f: Merge "ART: Fix run-test 114 ParallelGC to account for OOM"

* commit '3bcac48f23094fa0f46315a080ec47fc368fd4c2':
  ART: Fix run-test 114 ParallelGC to account for OOM

10 years agoam 698d79d1: ART: Clean up API after change 102631
Andreas Gampe [Fri, 25 Jul 2014 23:04:22 +0000 (23:04 +0000)]
am 698d79d1: ART: Clean up API after change 102631

* commit '698d79d1e3f348102deaa03b5e0fc23641107239':
  ART: Clean up API after change 102631

10 years agoam 4dd37adc: Merge "ART: Clean up API after change 102631"
Andreas Gampe [Fri, 25 Jul 2014 23:04:03 +0000 (23:04 +0000)]
am 4dd37adc: Merge "ART: Clean up API after change 102631"

* commit '4dd37adcf68a72758236ca770272a5597cd84767':
  ART: Clean up API after change 102631

10 years agoART: Clean up API after change 102631
Andreas Gampe [Fri, 25 Jul 2014 20:36:56 +0000 (13:36 -0700)]
ART: Clean up API after change 102631

This adds an abort_on_error parameter to GetCurrentMethod, which is
by default true. This restores all previous behavior, except for
monitor installation, where it follows 101639.

Bug: 16556938

(cherry picked from commit 6ec8ebd178ed39aa09e4c7fad194900114c4121a)

Change-Id: I2dee19d30674f2ffda33b3dc963e12470f39edb9

10 years agoMerge "ART: Fix missing initialization in stack_map_stream.h"
Andreas Gampe [Mon, 28 Jul 2014 22:02:25 +0000 (22:02 +0000)]
Merge "ART: Fix missing initialization in stack_map_stream.h"

10 years agoART: Fix missing initialization in stack_map_stream.h
Andreas Gampe [Mon, 28 Jul 2014 21:53:22 +0000 (14:53 -0700)]
ART: Fix missing initialization in stack_map_stream.h

Fixes stack_map_test.

Change-Id: Ibd6659b589ec595a7af21ad3d727c0ab2dceb45c

10 years agoMerge "ART: Rework quick entrypoint code in Mir2Lir, cleanup"
Andreas Gampe [Mon, 28 Jul 2014 16:41:14 +0000 (16:41 +0000)]
Merge "ART: Rework quick entrypoint code in Mir2Lir, cleanup"

10 years agoART: Rework quick entrypoint code in Mir2Lir, cleanup
Andreas Gampe [Mon, 28 Jul 2014 02:44:50 +0000 (19:44 -0700)]
ART: Rework quick entrypoint code in Mir2Lir, cleanup

To reduce the complexity of calling trampolines in generic code,
introduce an enumeration for entrypoints. Introduce a header that lists
the entrypoint enum and exposes a templatized method that translates an
enum value to the corresponding thread offset value.

Call helpers are rewritten to have an enum parameter instead of the
thread offset. Also rewrite LoadHelper and GenConversionCall this way.
It is now LoadHelper's duty to select the right thread offset size.

Introduce InvokeTrampoline virtual method to Mir2Lir. This allows to
further simplify the call helpers, as well as make OpThreadMem specific
to X86 only (removed from Mir2Lir).

Make GenInlinedCharAt virtual, move a copy to X86 backend, and simplify
both copies. Remove LoadBaseIndexedDisp and OpRegMem from Mir2Lir, as they
are now specific to X86 only.

Remove StoreBaseIndexedDisp from Mir2Lir, as it was only ever used in the
X86 backend.

Remove OpTlsCmp from Mir2Lir, as it was only ever used in the X86 backend.

Remove OpLea from Mir2Lir, as it was only ever defined in the X86 backend.

Remove GenImmedCheck from Mir2Lir as it was neither used nor implemented.

Change-Id: If0a6182288c5d57653e3979bf547840a4c47626e

10 years agoMerge "SetConstantWide should mark both SSA regs as constant"
Bill Buzbee [Sun, 27 Jul 2014 02:46:58 +0000 (02:46 +0000)]
Merge "SetConstantWide should mark both SSA regs as constant"

10 years agoMerge "ART: Allow arrays with erroneous component type"
Andreas Gampe [Sun, 27 Jul 2014 01:31:16 +0000 (01:31 +0000)]
Merge "ART: Allow arrays with erroneous component type"

10 years agoART: Allow arrays with erroneous component type
Andreas Gampe [Thu, 24 Jul 2014 03:18:36 +0000 (20:18 -0700)]
ART: Allow arrays with erroneous component type

Array classes must tolerate having component type classes that are
erroneous. Change CreateArrayClass to use LookupClass when FindClass
failed.

Bug: 16019155
Change-Id: Id4868c5498431c85c199aa3cbecd23566dce3601

10 years agoMerge "ART: Special-case cb(n)z even for in-reg constant"
Andreas Gampe [Sat, 26 Jul 2014 19:57:14 +0000 (19:57 +0000)]
Merge "ART: Special-case cb(n)z even for in-reg constant"

10 years agoART: Special-case cb(n)z even for in-reg constant
Andreas Gampe [Sat, 26 Jul 2014 08:40:39 +0000 (01:40 -0700)]
ART: Special-case cb(n)z even for in-reg constant

Call out to OpCmpImmBranch in GenCompareAndBranch if the constant
is zero and we are testing == or !=, even when zero has been loaded
to a register already. This avoids a register size mismatch on 64b
architectures when basically doing a null check, and generally
squashes a cmp + branch to a cbz or cbnz on Arm and Mips. X86 is
not degraded.

Bug: 16562601
Change-Id: I1997760f43dc186a84247ad30ae91053f71d102d

10 years agoam 9c81c0cb: Merge "Fix main space memory leak and add checks."
Mathieu Chartier [Fri, 25 Jul 2014 19:39:37 +0000 (19:39 +0000)]
am 9c81c0cb: Merge "Fix main space memory leak and add checks."

* commit '9c81c0cb0dc8c8d8ae5dca3d2d82b0eec7af589f':
  Fix main space memory leak and add checks.

10 years agoam b2ac1ab4: Fix main space memory leak and add checks.
Mathieu Chartier [Fri, 25 Jul 2014 19:19:15 +0000 (19:19 +0000)]
am b2ac1ab4: Fix main space memory leak and add checks.

* commit 'b2ac1ab47f84bd989fb2c94a532f5dc7d177ebe4':
  Fix main space memory leak and add checks.

10 years agoam 5d790444: Merge "Fix dangling pointer bug when transitioning to background." into...
Mathieu Chartier [Fri, 25 Jul 2014 19:19:04 +0000 (19:19 +0000)]
am 5d790444: Merge "Fix dangling pointer bug when transitioning to background." into lmp-dev

* commit '5d79044420e2cb03c0cd27c47f834943983aeb4b':
  Fix dangling pointer bug when transitioning to background.

10 years agoMerge branch 'lmp-dev-plus-aosp' of https://googleplex-android.googlesource.com/_dire...
Will Leshner [Fri, 25 Jul 2014 19:12:13 +0000 (19:12 +0000)]
Merge branch 'lmp-dev-plus-aosp' of https://googleplex-android.googlesource.com/_direct/platform/art into lmp-dev-plus-aosp

10 years agoam d190d989: Merge "Fix dangling pointer bug when transitioning to background."
Mathieu Chartier [Fri, 25 Jul 2014 19:02:23 +0000 (19:02 +0000)]
am d190d989: Merge "Fix dangling pointer bug when transitioning to background."

* commit 'd190d989ac92d2a5b9a342692564f40bd2080895':
  Fix dangling pointer bug when transitioning to background.

10 years agoam 3847d663: Merge "Increase limit to 2600 on host clang builds for frame size."...
Stephen Hines [Fri, 25 Jul 2014 18:57:53 +0000 (18:57 +0000)]
am 3847d663: Merge "Increase limit to 2600 on host clang builds for frame size." into lmp-dev

* commit '3847d66393a9b14ae9b896543780dc02c038d3fa':
  Increase limit to 2600 on host clang builds for frame size.

10 years agoFix main space memory leak and add checks.
Mathieu Chartier [Fri, 25 Jul 2014 18:50:47 +0000 (11:50 -0700)]
Fix main space memory leak and add checks.

The hypothesis is that we were leaking the main space and its
bitmaps, then eventually we would run out of virtual address space,
which would cause a null bitmap (DCHECK). Finally when we tried
adding the space with a null bitmap to the heap bitmap it segfaulted.

Changed some non performance critical DCHECK -> CHECK.

Bug: 16563323

(cherry picked from commit 2796a1669ae0f3b96db8432fbd8be1b93bf335c4)

Change-Id: Ifa9d866c6c89eff22a547af4db70bc79a77690ed

10 years agoresolved conflicts for merge of 76fb6019 to lmp-dev-plus-aosp
Stephen Hines [Fri, 25 Jul 2014 18:53:48 +0000 (11:53 -0700)]
resolved conflicts for merge of 76fb6019 to lmp-dev-plus-aosp

Change-Id: Ifc9f9fa95dda1dbe60b3be35ef30d2ca707c72d5

10 years agoam bacabce9: Merge "Add back a deleted check related to verification." into lmp-dev
Stephen Hines [Fri, 25 Jul 2014 18:50:20 +0000 (18:50 +0000)]
am bacabce9: Merge "Add back a deleted check related to verification." into lmp-dev

* commit 'bacabce9abe54e67192f835ebd4e013728a2121e':
  Add back a deleted check related to verification.

10 years agoMerge "Fix main space memory leak and add checks."
Mathieu Chartier [Fri, 25 Jul 2014 19:06:45 +0000 (19:06 +0000)]
Merge "Fix main space memory leak and add checks."

10 years agoFix main space memory leak and add checks.
Mathieu Chartier [Fri, 25 Jul 2014 18:50:47 +0000 (11:50 -0700)]
Fix main space memory leak and add checks.

The hypothesis is that we were leaking the main space and its
bitmaps, then eventually we would run out of virtual address space,
which would cause a null bitmap (DCHECK). Finally when we tried
adding the space with a null bitmap to the heap bitmap it segfaulted.

Changed some non performance critical DCHECK -> CHECK.

Bug: 16563323
Change-Id: I08a1f873752e28ebcf63ebbd90f92d994d7ca96b

10 years agoMerge "Fix dangling pointer bug when transitioning to background."
Mathieu Chartier [Fri, 25 Jul 2014 18:42:53 +0000 (18:42 +0000)]
Merge "Fix dangling pointer bug when transitioning to background."

10 years agoFix dangling pointer bug when transitioning to background.
Mathieu Chartier [Fri, 25 Jul 2014 17:13:51 +0000 (10:13 -0700)]
Fix dangling pointer bug when transitioning to background.

Dangling pointer left behind from the old rosalloc / dlmalloc
spaces. We now avoid using this pointer by using main_space_
and non_moving_space_ as well as clear the pointer when we remove
the space.

Bug: 16567203

Change-Id: Ida9ff30783e89cd4a4d86a4d0e912701692101f1

10 years agoMerge "Fix dangling pointer bug when transitioning to background." into lmp-dev
Mathieu Chartier [Fri, 25 Jul 2014 18:50:34 +0000 (18:50 +0000)]
Merge "Fix dangling pointer bug when transitioning to background." into lmp-dev

10 years agoFix dangling pointer bug when transitioning to background.
Mathieu Chartier [Fri, 25 Jul 2014 17:13:51 +0000 (10:13 -0700)]
Fix dangling pointer bug when transitioning to background.

Dangling pointer left behind from the old rosalloc / dlmalloc
spaces. We now avoid using this pointer by using main_space_
and non_moving_space_ as well as clear the pointer when we remove
the space.

Bug: 16567203

(cherry picked from commit 00b5915828f89daaefd9e8fb215658360f76762c)

Change-Id: I5962929e4a0bb4db6f531d25ee322da7ab3f5dd4

10 years agoMerge "resolved conflicts for merge of 9ef2ec9b to lmp-dev-plus-aosp" into lmp-dev...
Stephen Hines [Fri, 25 Jul 2014 18:48:31 +0000 (18:48 +0000)]
Merge "resolved conflicts for merge of 9ef2ec9b to lmp-dev-plus-aosp" into lmp-dev-plus-aosp

10 years agoam 9fbe946f: Merge "Fix JNI compiler frame size adjustments."
Vladimir Marko [Fri, 25 Jul 2014 18:45:05 +0000 (18:45 +0000)]
am 9fbe946f: Merge "Fix JNI compiler frame size adjustments."

* commit '9fbe946f49724ff7ae30da5d366a002d1e6addfa':
  Fix JNI compiler frame size adjustments.

10 years agoam 1ad334e6: Merge "Rename openDexFileNative to openDexFile."
Narayan Kamath [Fri, 25 Jul 2014 18:45:04 +0000 (18:45 +0000)]
am 1ad334e6: Merge "Rename openDexFileNative to openDexFile."

* commit '1ad334e6519c274b504c4d5d222c8e40920d02ee':
  Rename openDexFileNative to openDexFile.

10 years agoam 86507ed4: Merge "Add a new stack map scheme that encodes compilation info per...
Nicolas Geoffray [Fri, 25 Jul 2014 18:45:03 +0000 (18:45 +0000)]
am 86507ed4: Merge "Add a new stack map scheme that encodes compilation info per pc."

* commit '86507ed48af6046ee9e5bc84daeb795fc8acc3bf':
  Add a new stack map scheme that encodes compilation info per pc.

10 years agoam e2f654a4: Merge "ART: Fix wrong CHECK in GetCurrentLocationForThrow"
Andreas Gampe [Fri, 25 Jul 2014 18:45:00 +0000 (18:45 +0000)]
am e2f654a4: Merge "ART: Fix wrong CHECK in GetCurrentLocationForThrow"

* commit 'e2f654a463976f811c5358fc0de68c0492601274':
  ART: Fix wrong CHECK in GetCurrentLocationForThrow

10 years agoam b7563b64: Merge "ART: Turn on ART_TEST_KEEP_GOING by default, clean up unused...
Andreas Gampe [Fri, 25 Jul 2014 18:44:59 +0000 (18:44 +0000)]
am b7563b64: Merge "ART: Turn on ART_TEST_KEEP_GOING by default, clean up unused vars"

* commit 'b7563b641059bcff82eb4624edb806a7ce7f39c0':
  ART: Turn on ART_TEST_KEEP_GOING by default, clean up unused vars

10 years agoresolved conflicts for merge of 9ef2ec9b to lmp-dev-plus-aosp
Stephen Hines [Fri, 25 Jul 2014 18:44:37 +0000 (11:44 -0700)]
resolved conflicts for merge of 9ef2ec9b to lmp-dev-plus-aosp

Change-Id: I22a24179518bf5de94a22b21cb4c86fce3e012e6

10 years agoresolved conflicts for merge of 307f6508 to lmp-dev-plus-aosp
Andreas Gampe [Fri, 25 Jul 2014 18:38:28 +0000 (11:38 -0700)]
resolved conflicts for merge of 307f6508 to lmp-dev-plus-aosp

Change-Id: If49761765f3346993fb67068e79dad65e161c6a5

10 years agoam ee2d1ada: Fix JNI compiler frame size adjustments.
Vladimir Marko [Fri, 25 Jul 2014 15:51:39 +0000 (15:51 +0000)]
am ee2d1ada: Fix JNI compiler frame size adjustments.

* commit 'ee2d1ada2b9d0400e56c63656f9abfbccc880377':
  Fix JNI compiler frame size adjustments.

10 years agoFix JNI compiler frame size adjustments.
Vladimir Marko [Thu, 24 Jul 2014 16:01:58 +0000 (17:01 +0100)]
Fix JNI compiler frame size adjustments.

Bug: 16321952

(cherry picked from 4e24b9da7c5ea6e517ecdbc09aa14a0995100b7e)

Change-Id: I486b7ad7ebefc7db9b56fe92809ef495c316292b

10 years agoam 195bb33e: Merge "ART: Account for multidex location strings in VMClassLoader"
Andreas Gampe [Fri, 25 Jul 2014 12:41:18 +0000 (12:41 +0000)]
am 195bb33e: Merge "ART: Account for multidex location strings in VMClassLoader"

* commit '195bb33e08f262bde1e4367f14690f8328653ab6':
  ART: Account for multidex location strings in VMClassLoader

10 years agoam 6d5e82f7: ART: Fix wrong CHECK in GetCurrentLocationForThrow
Andreas Gampe [Fri, 25 Jul 2014 12:25:59 +0000 (12:25 +0000)]
am 6d5e82f7: ART: Fix wrong CHECK in GetCurrentLocationForThrow

* commit '6d5e82f78d7b121a33f935521e0a7f5335cab213':
  ART: Fix wrong CHECK in GetCurrentLocationForThrow

10 years agoam b3ec51e6: ART: Relax CurrentMethodVisitor requirements on GetDexPC
Andreas Gampe [Fri, 25 Jul 2014 12:25:58 +0000 (12:25 +0000)]
am b3ec51e6: ART: Relax CurrentMethodVisitor requirements on GetDexPC

* commit 'b3ec51e63c5041f77389ee001b47a8897b2a7acc':
  ART: Relax CurrentMethodVisitor requirements on GetDexPC

10 years agoam 9157c0ca: ART: Account for multidex location strings in VMClassLoader
Andreas Gampe [Fri, 25 Jul 2014 12:25:53 +0000 (12:25 +0000)]
am 9157c0ca: ART: Account for multidex location strings in VMClassLoader

* commit '9157c0ca6a57fa9f6378aebcfb2e3ff76622499a':
  ART: Account for multidex location strings in VMClassLoader

10 years agoART: Fix wrong CHECK in GetCurrentLocationForThrow
Andreas Gampe [Fri, 25 Jul 2014 10:07:06 +0000 (03:07 -0700)]
ART: Fix wrong CHECK in GetCurrentLocationForThrow

Under certain circumstances kDexNoIndex is actually an encoded
dex pc. Change the setup of CurrentMethodVisitor to account for that.

Bug: 16556938

(cherry picked from commit f9df5c1639a9418fcdf70476556a4c30b210701e)

Change-Id: Icb1d42ccd22296e2e0c629ebbd7e4d8c7d87f07a

10 years agoART: Relax CurrentMethodVisitor requirements on GetDexPC
Andreas Gampe [Thu, 17 Jul 2014 05:20:31 +0000 (22:20 -0700)]
ART: Relax CurrentMethodVisitor requirements on GetDexPC

In case we want to dump a Java stack after an unhandled fault, in
case we hold a thinlocked monitor, that monitor might get inflated.
That can cause an abort as we may not have enough/correct information
for the state at the bottom-most call.

Relax GetDexPc in the CurrentMethodVisitor to not abort when it cannot
find a dex pc. Instead, let the caller handle such a case. This CL
allows the locking_dex_pc_ in Monitor to be DexFile::kDexNoIndex,
which avoids the above abort.

Bug: 1635280216556938

(cherry picked from commit 4a8c3fa4bd8d95fac5671ab778dd00b6dc3ec0e4)

Change-Id: I60d3d89603de65752d9230ceac6abb0242388c18

10 years agoART: Account for multidex location strings in VMClassLoader
Andreas Gampe [Thu, 24 Jul 2014 22:35:50 +0000 (15:35 -0700)]
ART: Account for multidex location strings in VMClassLoader

To look up resources, look in the unadorned location.

Bug: 16530747

(cherry picked from commit cb8f9e8a2941971c049b26745ea713c859342d9b)

Change-Id: I5dce30c5e3c9f38fb5436c43a3f4a51fe4062c24

10 years agoam 4857b868: Merge "Integrate ART with NativeBridge interfaces"
Andreas Gampe [Fri, 25 Jul 2014 04:09:24 +0000 (04:09 +0000)]
am 4857b868: Merge "Integrate ART with NativeBridge interfaces"

* commit '4857b868f7d5941255afc847c07f49ae496d0e1c':
  Integrate ART with NativeBridge interfaces

10 years agoam c0dcf291: Fix stale remembered sets error.
Mathieu Chartier [Thu, 24 Jul 2014 23:33:21 +0000 (23:33 +0000)]
am c0dcf291: Fix stale remembered sets error.

* commit 'c0dcf29143051cf34d6c032a7b385ac4579bd4ec':
  Fix stale remembered sets error.

10 years agoam 5dc158e9: Fix access to long/double stack values from debugger
Sebastien Hertz [Thu, 24 Jul 2014 23:30:54 +0000 (23:30 +0000)]
am 5dc158e9: Fix access to long/double stack values from debugger

* commit '5dc158e9b6635b6c5d3916ae4094e8886feb4580':
  Fix access to long/double stack values from debugger

10 years agoMerge "Remove errors/warnings from tautological comparisons." into lmp-dev
Stephen Hines [Fri, 25 Jul 2014 17:27:58 +0000 (17:27 +0000)]
Merge "Remove errors/warnings from tautological comparisons." into lmp-dev

10 years agoRemove errors/warnings from tautological comparisons.
Stephen Hines [Wed, 16 Jul 2014 04:41:06 +0000 (21:41 -0700)]
Remove errors/warnings from tautological comparisons.

Change-Id: I2728b08d55abccd14c111c6c9da93068d4e14275

10 years agoMerge "ART: Reject array-length on non-reference register"
Andreas Gampe [Mon, 28 Jul 2014 20:30:00 +0000 (20:30 +0000)]
Merge "ART: Reject array-length on non-reference register"

10 years agoART: Reject array-length on non-reference register
Andreas Gampe [Mon, 28 Jul 2014 20:14:34 +0000 (13:14 -0700)]
ART: Reject array-length on non-reference register

It should be a hard verifier failure if we detect an array-length
instruction over a non-reference-type register.

Bug: 16595896
Change-Id: I059510b15c846551b2f8ded86cfd8127543324f4

10 years agoMerge "ART: Fix target test paths"
Andreas Gampe [Sun, 27 Jul 2014 21:56:42 +0000 (21:56 +0000)]
Merge "ART: Fix target test paths"

10 years agoART: Fix target test paths
Andreas Gampe [Sat, 12 Jul 2014 04:14:35 +0000 (21:14 -0700)]
ART: Fix target test paths

Nativetest is now nativetest64 on 64b targets.

Change-Id: I4c4c6bba1a56525df2993708caaae3e6f7f5f2f6

10 years agoMerge "Enable Load Store Elimination for ARM and ARM64"
Bill Buzbee [Sun, 27 Jul 2014 17:30:43 +0000 (17:30 +0000)]
Merge "Enable Load Store Elimination for ARM and ARM64"

10 years agoMerge "ART: ReEnable check for a gap between method's arg "in" vregs and rest of...
Bill Buzbee [Sun, 27 Jul 2014 02:46:40 +0000 (02:46 +0000)]
Merge "ART: ReEnable check for a gap between method's arg "in" vregs and rest of its vregs"

10 years agoMerge "ART: Native bridge command-line parameter"
Andreas Gampe [Sun, 27 Jul 2014 01:36:55 +0000 (01:36 +0000)]
Merge "ART: Native bridge command-line parameter"

10 years agoART: Native bridge command-line parameter
Andreas Gampe [Fri, 25 Jul 2014 09:32:19 +0000 (02:32 -0700)]
ART: Native bridge command-line parameter

Add a command-line parameter for the native bridge library, slight
refactor/cleanup.

Add run-test 115 to test the native bridge interface. Currently the
tests are black-listed for the target, as the setup for the test
is too complicated in the current infrastructure.

Change-Id: I6ccf19485e8c30b96e9f2fd5425278cb1ebd403f

10 years agoMerge "Integrate ART with NativeBridge interfaces"
Andreas Gampe [Fri, 25 Jul 2014 04:01:32 +0000 (04:01 +0000)]
Merge "Integrate ART with NativeBridge interfaces"

10 years agoIntegrate ART with NativeBridge interfaces
Yong WU [Thu, 24 Jul 2014 13:32:15 +0000 (21:32 +0800)]
Integrate ART with NativeBridge interfaces

Native-bridge will provide the following interfaces to ART:

struct NativeBridgeCallbacks {
  bool  (*initialize   )(NativeBridgeArtCallbacks* vm_itf);
  void* (*loadLibrary  )(const char* libpath, int flag);
  void* (*getTrampoline)(void* handle, const char* name, const char* shorty,
                         uint32_t len);
  bool  (*isSupported  )(const char* libpath);
};

Native-bridge will expose a symbol NativeBridgeItf with the
type of NativeBridgeCallbacks to ART.

And ART will provide the interfaces below to native-bridge:

struct NativeBridgeArtCallbacks {
  int   (*logger               )(int prio, const char* tag, const char* fmt, ...);
  const char* (*getMethodShorty)(JNIEnv* env, jmethodID mid);
  int   (*getNativeMethodCount )(JNIEnv* env, jclass clazz);
  int   (*getNativeMethods     )(JNIEnv* env, jclass clazz, JNINativeMethod* methods,
                                 uint32_t method_count);
};

Based on the interfaces, if an ART call to dlopen fails to open a native library,
it queries the native bridge by using NativeBridgeCallbacks::isSupported(). If the
native library is supported by native-bridge, ART can load the native library
using NativeBridgeCallbacks::loadLibrary() and get a trampoline for a specific
native method using NativeBridgeCallbacks::getTrampoline(). ART can then call
the native method using the normal signature and the address of the trampoline.

On the other side, in the case of a native method calling JNI native function
CallXXXXMethodY(), native-bridge calls back to Art for the shorty of the method
using NativeBridgeArtCallbacks::getMethodShorty() so that it can prepare based
on host calling convention.

In case of JNI function RegisterNatives()/UnregisterNatives(), native bridge can
call back to NativeBridgeArtCallbacks::getNativeMethodCount() and NativeBridgeArtCallbacks
::getNativeMethods() to get all native methods of specified class so that all
corresponding trampolines can be prepared/destroyed.

Class NativeBridge is created to encapsulate the function pointers of
NativeBridgeCallbacks and provides better abstraction to ART.

Note: functionality is turned off in native_bridge.cc at the moment.

Change-Id: I652755044957a7960254648652b538cce70dd011

10 years agoam 54f3c041: Merge "Fix memory leak in RemoveRememberedSet."
Mathieu Chartier [Thu, 24 Jul 2014 20:55:16 +0000 (20:55 +0000)]
am 54f3c041: Merge "Fix memory leak in RemoveRememberedSet."

* commit '54f3c041365647fc79260b65e70b4290d0b50f1c':
  Fix memory leak in RemoveRememberedSet.