OSDN Git Service
Elliott Hughes [Sat, 22 Dec 2012 01:50:09 +0000 (17:50 -0800)]
Merge "Support System.loadLibrary for libraries with transitive dependencies."
Elliott Hughes [Sat, 22 Dec 2012 00:11:32 +0000 (16:11 -0800)]
Merge "Fix a forecopy/forcecopy typo."
Elliott Hughes [Fri, 21 Dec 2012 23:19:31 +0000 (15:19 -0800)]
Fix a forecopy/forcecopy typo.
Change-Id: Ic8f8d6fdf278bfa41e54586bc187c0ba42248700
Elliott Hughes [Fri, 21 Dec 2012 20:31:43 +0000 (12:31 -0800)]
Merge "Add JNI statistics to the SIGQUIT output."
Elliott Hughes [Fri, 21 Dec 2012 19:43:46 +0000 (11:43 -0800)]
Add JNI statistics to the SIGQUIT output.
Bug:
7903975
Change-Id: Ibc011ef73c8632fec42b846fcdee20c970951757
Elliott Hughes [Thu, 20 Dec 2012 22:37:25 +0000 (14:37 -0800)]
Support System.loadLibrary for libraries with transitive dependencies.
Bug:
7896159
Bug: http://code.google.com/p/android/issues/detail?id=34416
Change-Id: Ic0522c4d7c3a3493e09b8dd4f68aabd86acb9b73
Elliott Hughes [Thu, 20 Dec 2012 02:05:01 +0000 (18:05 -0800)]
Merge "Change NewDirectByteBuffer to allow NULL if capacity == 0."
Elliott Hughes [Wed, 19 Dec 2012 23:45:18 +0000 (15:45 -0800)]
Change NewDirectByteBuffer to allow NULL if capacity == 0.
Also make the diagnostics more uniform.
Bug:
7892060
Change-Id: Ib3ca0d6241a5958d2c8681337b1883fc3d74cb54
Elliott Hughes [Wed, 19 Dec 2012 18:51:40 +0000 (10:51 -0800)]
Merge "Optimize dvmHash Loops."
You Kim [Wed, 19 Dec 2012 05:25:21 +0000 (14:25 +0900)]
Optimize dvmHash Loops.
pHashTable->tableSize is working as volatile variable & is evaluated on every iteration.
Help compiler to define tableSize as local(Register) variable.
Change-Id: Iecdea3ecb810319d461ff3924ef21fd6c2d69d1e
Elliott Hughes [Mon, 17 Dec 2012 21:45:41 +0000 (13:45 -0800)]
Merge "Optimize opcodes to avoid bubble in pipeline."
Elliott Hughes [Mon, 17 Dec 2012 18:59:43 +0000 (10:59 -0800)]
Merge "Remove possilbe leaks in dalvik."
You Kim [Sun, 16 Dec 2012 18:14:01 +0000 (03:14 +0900)]
Remove possilbe leaks in dalvik.
Add missing free() calls.
Change-Id: Id602b84ba8f15fef1851f37e3d5aac214ac67de1
Elliott Hughes [Mon, 17 Dec 2012 18:54:49 +0000 (10:54 -0800)]
Merge "Reclaiming attribute obj after pthread_create"
You Kim [Sun, 16 Dec 2012 17:36:44 +0000 (02:36 +0900)]
Reclaiming attribute obj after pthread_create
call pthread_attr_destroy in order to reclaim attribute object.
Change-Id: I2592903ef67101a7c656a3ab0ad243a9716298b5
Elliott Hughes [Mon, 17 Dec 2012 18:42:06 +0000 (10:42 -0800)]
Merge "Remove dead code."
Elliott Hughes [Mon, 17 Dec 2012 18:35:01 +0000 (10:35 -0800)]
Merge "Fail if the new index is out of range."
jwilson [Mon, 17 Dec 2012 18:16:33 +0000 (13:16 -0500)]
Fail if the new index is out of range.
We were silently truncating, which made an obvious problem
into a non-obvious one.
Bug: http://code.google.com/p/android/issues/detail?id=40409
Change-Id: I15576357c72ac0e98cf85c0a1d289fd5009468f9
Elliott Hughes [Mon, 17 Dec 2012 17:21:54 +0000 (09:21 -0800)]
Remove dead code.
Change-Id: I56c27c510c8fc0ba2d8e691c788b433dce250f7a
Elliott Hughes [Mon, 17 Dec 2012 17:16:35 +0000 (09:16 -0800)]
Merge "Possible leak in dvmAllocRegion"
Elliott Hughes [Mon, 17 Dec 2012 17:09:01 +0000 (09:09 -0800)]
Merge "Fix handling of annotations whose classes are missing at runtime."
Elliott Hughes [Sat, 15 Dec 2012 01:27:24 +0000 (17:27 -0800)]
Fix handling of annotations whose classes are missing at runtime.
Bug: http://code.google.com/p/android/issues/detail?id=39854
Change-Id: I3089b693ff3084b093f991f991ac9273dea7356d
You Kim [Mon, 17 Dec 2012 15:14:07 +0000 (00:14 +0900)]
Possible leak in dvmAllocRegion
In error case mmapped ptr 'base' is not munmapped.
Change-Id: I9e43841371fee6b7c092fe9a8922ae82e78805ed
You Kim [Thu, 22 Nov 2012 14:51:38 +0000 (23:51 +0900)]
Optimize opcodes to avoid bubble in pipeline.
Possible bubbles between fetch(ldr) and extract(and)
Excute subsequent instruction(s) inbetween the bubble by reordering instruction.
Change-Id: I76d89ed6bc4ca44673231375cc529fdbfd5a8030
Elliott Hughes [Mon, 10 Dec 2012 23:05:11 +0000 (15:05 -0800)]
Merge "ExceptionInInitializerError should wrap only non Error exceptions."
Elliott Hughes [Mon, 10 Dec 2012 22:51:07 +0000 (14:51 -0800)]
ExceptionInInitializerError should wrap only non Error exceptions.
Backport from dalvik-dev, due to customer demand.
Bug: http://code.google.com/p/android/issues/detail?id=41196
Change-Id: I53ef57e54983f2173546463ac78e265ff5dd3284
Elliott Hughes [Fri, 7 Dec 2012 00:43:18 +0000 (16:43 -0800)]
Merge "Don't cache failures in the interface cache."
Elliott Hughes [Fri, 7 Dec 2012 00:33:38 +0000 (16:33 -0800)]
Don't cache failures in the interface cache.
If we try to reuse a failure, we'll be holding a NULL Method* without
the current thread having the appropriate exception pending, which will
lead to a SIGSEGV.
I've added a corresponding vm-test to cts/.
Bug: http://code.google.com/p/android/issues/detail?id=29358
Change-Id: I044e438c46fce256c8eff2dac5d86778734d5614
Elliott Hughes [Thu, 6 Dec 2012 23:37:19 +0000 (15:37 -0800)]
Merge "Remove dvmFindInterfaceMethodInCache2."
Elliott Hughes [Thu, 6 Dec 2012 23:24:32 +0000 (15:24 -0800)]
Remove dvmFindInterfaceMethodInCache2.
We can just use dvmFindInterfaceMethodInCache directly.
Change-Id: I2f3a660262ba7a39c05689df160ebdd2e7ec38a5
Elliott Hughes [Fri, 30 Nov 2012 22:52:23 +0000 (14:52 -0800)]
Merge "JIT: Performance fix for const doubles"
buzbee [Fri, 30 Nov 2012 14:46:45 +0000 (06:46 -0800)]
JIT: Performance fix for const doubles
Some recent Arm processors take a performance hit when
creating a floating point double by loading it as a pair of singles.
Legacy code to support soft floating point doubles as a pair of core
registers loaded double immediates in this way.
With the CL, we handle double immediates as a single unit.
(cherry-pick of
c8129911e598ad0ca8d7b31012444ab6ce8bce45.)
Change-Id: Ic1512e34bfd233a6f5ffd58ce843965adbbad875
Elliott Hughes [Fri, 30 Nov 2012 19:46:25 +0000 (11:46 -0800)]
Merge "Use ALOGE rather than SLOGE for mountEmulatedStorage failures."
Elliott Hughes [Fri, 30 Nov 2012 19:36:25 +0000 (11:36 -0800)]
Use ALOGE rather than SLOGE for mountEmulatedStorage failures.
Otherwise the actual failure isn't in logcat, and all you see there
is the rather confusing:
E dalvikvm: cannot mountExternalStorage(): Success
Change-Id: I10c96748b5ff5b6c3877312e7bdb2641026b4489
Elliott Hughes [Thu, 29 Nov 2012 16:57:40 +0000 (08:57 -0800)]
Merge "remove duplicated call to dvmJitCalleeSave()"
Rocky Zhang [Thu, 29 Nov 2012 00:02:29 +0000 (08:02 +0800)]
remove duplicated call to dvmJitCalleeSave()
Change-Id: Ib7b5f637a22fd5ba60bdb45cd34ac8c1781c7bd3
Signed-off-by: Rocky Zhang <zhangyan.hit@gmail.com>
Elliott Hughes [Tue, 27 Nov 2012 22:37:03 +0000 (14:37 -0800)]
Merge "[MIPS] Fix DDMS(profiling) bug."
Elliott Hughes [Tue, 27 Nov 2012 22:36:45 +0000 (14:36 -0800)]
Merge "[MIPS] Fix several profiling issues caused by incorrect register saves"
Elliott Hughes [Tue, 27 Nov 2012 22:35:39 +0000 (14:35 -0800)]
Merge "[MIPS] Fixing FP registers fs0 and fs1 corruption bug."
The Android Open Source Project [Tue, 27 Nov 2012 16:20:15 +0000 (08:20 -0800)]
Merge jb-mr1-dev into aosp/master - do not merge
Change-Id: Idf370f9acc8ecddd2d9c73a41691c65da457ffc0
Douglas Leung [Tue, 27 Nov 2012 02:57:38 +0000 (18:57 -0800)]
[MIPS] Fixing FP registers fs0 and fs1 corruption bug.
For performance reasons, we don't want to use "callee saved" FP registers.
The fix is to just use temporary FP registers instead.
Change-Id: Ifae20c707ff7022905a080892348bfbc151e4935
Signed-off-by: Douglas Leung <douglas@mips.com>
Douglas Leung [Thu, 22 Nov 2012 03:13:14 +0000 (19:13 -0800)]
[MIPS] Fix DDMS(profiling) bug.
Change-Id: I47e87d600dc6eb529bf1ae9c8cc2b81910b9513c
Signed-off-by: Douglas Leung <douglas@mips.com>
Chris Dearman [Thu, 22 Nov 2012 03:08:34 +0000 (19:08 -0800)]
[MIPS] Fix several profiling issues caused by incorrect register saves
rTEMP is used by the JAL and JALR macros so it needs to be preserved
if it is holding some other useful value. No need to save ra in rTEMP
if using JAL/JALR macros.
Change-Id: I1bf8c3b044faa944021b4331e0a71c1489b70241
Signed-off-by: Douglas Leung <douglas@mips.com>
Signed-off-by: Chris Dearman <chris@mips.com>
Yohann Roussel [Mon, 12 Nov 2012 20:07:44 +0000 (12:07 -0800)]
am
afe3a013: (-s ours) am
288f2280: Revert "Fix alignment when recompacting a DexMerger result. do not merge."
* commit '
afe3a0131e75367759e1d7a69cec0aac2a45524c':
Revert "Fix alignment when recompacting a DexMerger result. do not merge."
yroussel [Mon, 12 Nov 2012 20:07:43 +0000 (12:07 -0800)]
am
2d2f13c5: (-s ours) am
707ebe9f: Fix alignment when recompacting a DexMerger result. do not merge.
* commit '
2d2f13c50ba3c38dc4a8a8aaf483f6df155e0080':
Fix alignment when recompacting a DexMerger result. do not merge.
The Android Open Source Project [Mon, 12 Nov 2012 15:39:27 +0000 (07:39 -0800)]
am
d9de2f66: am
bb077798: Reconcile with jb-mr1-release - do not merge
* commit '
d9de2f6652a8a7f7d5b348bccd7b842ac3d5716d':
The Android Open Source Project [Mon, 12 Nov 2012 15:37:02 +0000 (07:37 -0800)]
am
bb077798: Reconcile with jb-mr1-release - do not merge
* commit '
bb07779818b583ec6d0b73c4062f485751332c81':
The Android Open Source Project [Mon, 12 Nov 2012 15:35:32 +0000 (07:35 -0800)]
am
d5c3cb5f: am
890e41c1: Merge
f870f2d - do not merge
* commit '
d5c3cb5f23429cf4e1c7b08c5e926174a64b4703':
Add dx option to always generate const-string/jumbo.do not merge.
The Android Open Source Project [Mon, 12 Nov 2012 15:35:23 +0000 (07:35 -0800)]
Reconcile with jb-mr1-release - do not merge
Change-Id: Ic0fb72b0939908c2e072e4ae5d60c9d24de9dc66
The Android Open Source Project [Mon, 12 Nov 2012 15:33:04 +0000 (07:33 -0800)]
am
890e41c1: Merge
f870f2d - do not merge
* commit '
890e41c1c4289b97476bc90d2e1e443fffec4b13':
Add dx option to always generate const-string/jumbo.do not merge.
The Android Open Source Project [Mon, 12 Nov 2012 15:29:38 +0000 (07:29 -0800)]
Merge
f870f2d - do not merge
Change-Id: Ib8ea0f0d579f78fcab1deca05ff1beac9ba9bc03
The Android Automerger [Sat, 10 Nov 2012 00:25:17 +0000 (16:25 -0800)]
merge in jb-mr1-release history after reset to jb-mr1-dev
jeffhao [Tue, 16 Oct 2012 00:04:46 +0000 (17:04 -0700)]
Add dx option to always generate const-string/jumbo.do not merge.
This allows large dex files with many strings to be merged properly.
(cherry picked from commit
266f45ff7da18022faf5f77df76c69f8cdad313f)
Change-Id: I5fe4c55d84a91101a4f89f590117aa6dc0bfc0f2
Yohann Roussel [Wed, 7 Nov 2012 20:13:16 +0000 (12:13 -0800)]
am
288f2280: Revert "Fix alignment when recompacting a DexMerger result. do not merge."
* commit '
288f22800c8fe1eb44afff78ec1fff815379183e':
Revert "Fix alignment when recompacting a DexMerger result. do not merge."
Elliott Hughes [Wed, 7 Nov 2012 19:19:49 +0000 (11:19 -0800)]
am
6bc15e50: Merge "Fix x86 JIT log spam."
* commit '
6bc15e501214d1e5544661438c1447d1cc4f98a5':
Fix x86 JIT log spam.
Elliott Hughes [Wed, 7 Nov 2012 19:04:56 +0000 (11:04 -0800)]
Merge "Fix x86 JIT log spam."
Elliott Hughes [Wed, 7 Nov 2012 18:56:59 +0000 (10:56 -0800)]
Fix x86 JIT log spam.
Removes all the E and I logging in cases like these:
E( 2901) JIT couldn't compile Ljava/lang/Number;<init> dex_pc=0
I( 2901) codeGenBasicBlockJit returns negative number
E( 2901) JIT couldn't compile Ljava/lang/String;<init> dex_pc=0
I( 2901) codeGenBasicBlockJit returns negative number
E( 2901) JIT couldn't compile Ljava/util/Hashtable$HashtableEntry;<init> dex_pc=0
I( 2901) codeGenBasicBlockJit returns negative number
E( 2901) JIT couldn't compile Ljava/lang/AbstractStringBuilder;<init> dex_pc=0
I( 2901) codeGenBasicBlockJit returns negative number
E( 2901) JIT couldn't compile Ljava/util/HashMap$HashMapEntry;<init> dex_pc=0
I( 2901) codeGenBasicBlockJit returns negative number
Change-Id: I020c01c11a3840e700bbeb39237da1a6d508be8a
Yohann Roussel [Wed, 7 Nov 2012 10:03:30 +0000 (02:03 -0800)]
Revert "Fix alignment when recompacting a DexMerger result. do not merge."
This reverts commit
707ebe9f1984f7b1ed2e3465012c7c0eb3ef2bd6
Looks like jb-dev is missing some other patches before it possible to cherry-pick this one.
Change-Id: I8cbdd649d03696cb80be0dacc8640275468ef162
yroussel [Wed, 7 Nov 2012 03:16:39 +0000 (19:16 -0800)]
am
707ebe9f: Fix alignment when recompacting a DexMerger result. do not merge.
* commit '
707ebe9f1984f7b1ed2e3465012c7c0eb3ef2bd6':
Fix alignment when recompacting a DexMerger result. do not merge.
Elliott Hughes [Wed, 7 Nov 2012 01:39:48 +0000 (17:39 -0800)]
am
11ef8f2b: am
9a8b82ae: Merge "Always build WITH_JIT for the target."
* commit '
11ef8f2b8b74b1f0556e0e55f76e4984189c0803':
Always build WITH_JIT for the target.
Elliott Hughes [Wed, 7 Nov 2012 01:37:44 +0000 (17:37 -0800)]
am
9a8b82ae: Merge "Always build WITH_JIT for the target."
* commit '
9a8b82ae06ef2e0fbd6b0a342a2394a6830afd95':
Always build WITH_JIT for the target.
Elliott Hughes [Wed, 7 Nov 2012 01:17:32 +0000 (17:17 -0800)]
Merge "Always build WITH_JIT for the target."
yroussel [Wed, 31 Oct 2012 15:25:42 +0000 (16:25 +0100)]
Fix alignment when recompacting a DexMerger result. do not merge.
(cherry picked from commit
e0e7b076e427a327e27247aa409d0b0fa8b0e0aa)
Change-Id: Ib8092bd9ce9ba5490040a41c5d604e6a138acc7d
Elliott Hughes [Wed, 7 Nov 2012 00:34:28 +0000 (16:34 -0800)]
Always build WITH_JIT for the target.
Without this, by default MIPS and x86 builds build without a JIT.
Change-Id: Ieff5c25fd4e40401ed511e952a4f189615131327
The Android Open Source Project [Tue, 6 Nov 2012 19:26:04 +0000 (11:26 -0800)]
am
6fbc2f19: am
e7e77c4b: Reconcile with jb-mr1-factory-release jb-mr1-release - do not merge
* commit '
6fbc2f19af5adee4b423a203fe2e00c5afb5803f':
The Android Open Source Project [Tue, 6 Nov 2012 19:22:55 +0000 (11:22 -0800)]
am
e7e77c4b: Reconcile with jb-mr1-factory-release jb-mr1-release - do not merge
* commit '
e7e77c4bd64c87798344d690e13418b481ab2d05':
Revert the fix for incorrect package-private overriding.
One more errno for encrypted devices.
Revert "Upgrade to dlmalloc 2.8.5."
Revert "Ensure we fence the writing of clazz to an Object post allocation."
The Android Open Source Project [Tue, 6 Nov 2012 19:20:15 +0000 (11:20 -0800)]
Reconcile with jb-mr1-factory-release jb-mr1-release - do not merge
Change-Id: Ia24bfd680910be16fa6f89fbd48db251b076d8f6
The Android Open Source Project [Wed, 31 Oct 2012 16:15:35 +0000 (09:15 -0700)]
Reconcile with jb-mr1-factory-release jb-mr1-release - do not merge
Change-Id: Idcb8619166f794d334655aaf35232bb6a09e1c9e
Elliott Hughes [Tue, 30 Oct 2012 00:14:58 +0000 (17:14 -0700)]
am
2988c4f2: Merge "Fix dalvik\'s use of readdir_r(3)."
* commit '
2988c4f272f62af2e96f1e6161d4e99bc1dc1b16':
Fix dalvik's use of readdir_r(3).
jeffhao [Mon, 29 Oct 2012 20:52:18 +0000 (13:52 -0700)]
am
f6650edf: am
d454d96b: am
a0475a59: (-s ours) am
d5cc3680: (-s ours) am
9db54b1e: Add dx option to always generate const-string/jumbo. do not merge.
* commit '
f6650edf4225b489e39bf560fed406c749ee0c5b':
jeffhao [Mon, 29 Oct 2012 20:52:17 +0000 (13:52 -0700)]
am
3daa7a27: am
07df9514: am
69dd411c: (-s ours) Add dx option to always generate const-string/jumbo. do not merge.
* commit '
3daa7a27d4320c799a4ee8f04dc1e6dfa733734f':
jeffhao [Mon, 29 Oct 2012 20:42:09 +0000 (13:42 -0700)]
am
d454d96b: am
a0475a59: (-s ours) am
d5cc3680: (-s ours) am
9db54b1e: Add dx option to always generate const-string/jumbo. do not merge.
* commit '
d454d96bf19438d2fc4301714abf55f7fecd0001':
jeffhao [Mon, 29 Oct 2012 20:42:08 +0000 (13:42 -0700)]
am
07df9514: am
69dd411c: (-s ours) Add dx option to always generate const-string/jumbo. do not merge.
* commit '
07df951456410a66353455108d86958985a15115':
Elliott Hughes [Fri, 26 Oct 2012 23:45:29 +0000 (16:45 -0700)]
Merge "Fix dalvik's use of readdir_r(3)."
The Android Automerger [Fri, 26 Oct 2012 23:22:35 +0000 (16:22 -0700)]
merge in jb-mr1-release history after reset to jb-mr1-dev
Elliott Hughes [Fri, 26 Oct 2012 22:27:20 +0000 (15:27 -0700)]
Fix dalvik's use of readdir_r(3).
Change-Id: Ic2c9b88b2b3204b4c02f4838766dc4d9ebad06a2
Nick Kralevich [Fri, 19 Oct 2012 22:00:49 +0000 (15:00 -0700)]
resolved conflicts for merge of
e206e7c9 to jb-mr1-dev-plus-aosp
Change-Id: I52fcfbdeb6bbb39c5330ce2be589ec546805aa1f
Nick Kralevich [Fri, 19 Oct 2012 21:38:11 +0000 (14:38 -0700)]
Merge "Push to AOSP"
Nick Kralevich [Fri, 19 Oct 2012 20:45:06 +0000 (13:45 -0700)]
jeffhao [Wed, 17 Oct 2012 18:10:42 +0000 (11:10 -0700)]
am
a0475a59: (-s ours) am
d5cc3680: (-s ours) am
9db54b1e: Add dx option to always generate const-string/jumbo. do not merge.
* commit '
a0475a59b58c39c9a3babddc371815c6d51e3f39':
Add dx option to always generate const-string/jumbo. do not merge.
jeffhao [Wed, 17 Oct 2012 18:08:33 +0000 (11:08 -0700)]
am
69dd411c: (-s ours) Add dx option to always generate const-string/jumbo. do not merge.
* commit '
69dd411c7ef47e3ea3f6ff8be9a4d903f545f0d5':
Add dx option to always generate const-string/jumbo. do not merge.
jeffhao [Wed, 17 Oct 2012 17:59:15 +0000 (10:59 -0700)]
am
13b52dd6: am
a0475a59: am
d5cc3680: (-s ours) am
9db54b1e: Add dx option to always generate const-string/jumbo. do not merge.
* commit '
13b52dd62109b7ce81eb284e159c02cefbf820a3':
Add dx option to always generate const-string/jumbo. do not merge.
jeffhao [Wed, 17 Oct 2012 17:56:37 +0000 (10:56 -0700)]
am
a0475a59: am
d5cc3680: (-s ours) am
9db54b1e: Add dx option to always generate const-string/jumbo. do not merge.
* commit '
a0475a59b58c39c9a3babddc371815c6d51e3f39':
Add dx option to always generate const-string/jumbo. do not merge.
jeffhao [Wed, 17 Oct 2012 17:53:43 +0000 (10:53 -0700)]
am
d5cc3680: (-s ours) am
9db54b1e: Add dx option to always generate const-string/jumbo. do not merge.
* commit '
d5cc36801bb86ba4e8b3db6efc67282f41b5b784':
Add dx option to always generate const-string/jumbo. do not merge.
jeffhao [Wed, 17 Oct 2012 17:50:43 +0000 (10:50 -0700)]
am
9db54b1e: Add dx option to always generate const-string/jumbo. do not merge.
* commit '
9db54b1e21b8e994658890328e57caa822e444a7':
Add dx option to always generate const-string/jumbo. do not merge.
Kenny Root [Wed, 17 Oct 2012 16:46:36 +0000 (09:46 -0700)]
resolved conflicts for merge of
ab84215a to jb-mr1-dev-plus-aosp
Change-Id: I0420c747fbe23b24e6ba2a5865587fe732eff387
Kenny Root [Wed, 17 Oct 2012 16:35:52 +0000 (09:35 -0700)]
Merge "Remove HAVE_SELINUX guards"
Kenny Root [Wed, 17 Oct 2012 06:38:15 +0000 (23:38 -0700)]
Remove HAVE_SELINUX guards
Change-Id: I1cb99cd25857a70049220d4368f21d97c3b80ebf
The Android Automerger [Wed, 17 Oct 2012 13:59:13 +0000 (06:59 -0700)]
merge in jb-mr1-release history after reset to jb-mr1-dev
jeffhao [Wed, 17 Oct 2012 01:57:28 +0000 (18:57 -0700)]
am
3b53278d: am
69dd411c: Add dx option to always generate const-string/jumbo. do not merge.
* commit '
3b53278d274cbf4398917ee76f024cbb71ea28f6':
Add dx option to always generate const-string/jumbo. do not merge.
jeffhao [Wed, 17 Oct 2012 01:54:58 +0000 (18:54 -0700)]
am
69dd411c: Add dx option to always generate const-string/jumbo. do not merge.
* commit '
69dd411c7ef47e3ea3f6ff8be9a4d903f545f0d5':
Add dx option to always generate const-string/jumbo. do not merge.
jeffhao [Tue, 16 Oct 2012 00:04:46 +0000 (17:04 -0700)]
Add dx option to always generate const-string/jumbo. do not merge.
This allows large dex files with many strings to be merged properly.
(cherry picked from commit
266f45ff7da18022faf5f77df76c69f8cdad313f)
Change-Id: I0277b1a0a81b3a77092a9e9f8f648ff190e29723
jeffhao [Tue, 16 Oct 2012 00:04:46 +0000 (17:04 -0700)]
Add dx option to always generate const-string/jumbo. do not merge.
This allows large dex files with many strings to be merged properly.
(cherry picked from commit
266f45ff7da18022faf5f77df76c69f8cdad313f)
Change-Id: I5e85346610c4c5280ecdd9e97c63eaef4420642e
The Android Automerger [Tue, 16 Oct 2012 13:59:12 +0000 (06:59 -0700)]
merge in jb-mr1-release history after reset to jb-mr1-dev
Elliott Hughes [Tue, 16 Oct 2012 03:20:52 +0000 (20:20 -0700)]
am
cda473c1: Merge "Handle MAX_INT case in packed-switch"
* commit '
cda473c171f967db7472d1832cd4cf2111665264':
Handle MAX_INT case in packed-switch
Elliott Hughes [Tue, 16 Oct 2012 03:20:51 +0000 (20:20 -0700)]
am
20b6e422: Merge "Remove Jumbo opcode support"
* commit '
20b6e422412a2d004e3a92e5bfbba77d3de5e6c3':
Remove Jumbo opcode support
Elliott Hughes [Tue, 16 Oct 2012 01:07:08 +0000 (18:07 -0700)]
Merge "Handle MAX_INT case in packed-switch"
Chris Dearman [Thu, 12 Jul 2012 00:10:36 +0000 (17:10 -0700)]
Handle MAX_INT case in packed-switch
Allows regression 095-switch-MAX_INT to pass
Change-Id: Iec97fb456d43dbb14ee096fe708c74aff5ba1f5a
Signed-off-by: Chris Dearman <chris@mips.com>
Elliott Hughes [Tue, 16 Oct 2012 00:36:47 +0000 (17:36 -0700)]
Merge "Remove Jumbo opcode support"
Elliott Hughes [Mon, 15 Oct 2012 21:29:23 +0000 (14:29 -0700)]
Revert the fix for incorrect package-private overriding.
This fixes FIFA 2012, the download portion of which relies on the
old bug, without breaking instrumentation tests. The problem is that
dexopt tries to quicken method calls, and gets confused by this case
where the apparent static method resolution differs from the actual
one at runtime, depending on the targetSdkVersion of the specific app.
dexopt can't make an ahead-of-time decision in a world where the rules
might change at runtime.
Bug:
7301030
Bug:
7343420
Change-Id: Iaa15611f099546b7e54279cfd6abc9b4cdcb9812