OSDN Git Service

android-x86/art.git
9 years agoam ba14a447: am e6caf046: Merge "Fix issue with Thread.setName hanging after Thread...
Brian Carlstrom [Thu, 28 Aug 2014 23:02:13 +0000 (23:02 +0000)]
am ba14a447: am e6caf046: Merge "Fix issue with Thread.setName hanging after Thread.start" into lmp-dev

* commit 'ba14a447405980211cc68203fa8845a75d97e929':
  Fix issue with Thread.setName hanging after Thread.start

9 years agoam e6caf046: Merge "Fix issue with Thread.setName hanging after Thread.start" into...
Brian Carlstrom [Thu, 28 Aug 2014 22:48:35 +0000 (22:48 +0000)]
am e6caf046: Merge "Fix issue with Thread.setName hanging after Thread.start" into lmp-dev

* commit 'e6caf04639aec134f7b668bf71cd6797a4759dd9':
  Fix issue with Thread.setName hanging after Thread.start

9 years agoMerge "Fix issue with Thread.setName hanging after Thread.start" into lmp-dev
Brian Carlstrom [Thu, 28 Aug 2014 22:23:09 +0000 (22:23 +0000)]
Merge "Fix issue with Thread.setName hanging after Thread.start" into lmp-dev

9 years agoam 5e894e1f: am 7654123b: Merge "Ensure proxy constructor is in dex cache" into lmp-dev
Brian Carlstrom [Thu, 28 Aug 2014 19:13:30 +0000 (19:13 +0000)]
am 5e894e1f: am 7654123b: Merge "Ensure proxy constructor is in dex cache" into lmp-dev

* commit '5e894e1f84c267f1a259f0831a5cdd5d2b9d6fa3':
  Ensure proxy constructor is in dex cache

9 years agoam 7654123b: Merge "Ensure proxy constructor is in dex cache" into lmp-dev
Brian Carlstrom [Thu, 28 Aug 2014 19:01:48 +0000 (19:01 +0000)]
am 7654123b: Merge "Ensure proxy constructor is in dex cache" into lmp-dev

* commit '7654123b2fd7abeb40522dac791fa26a1364d030':
  Ensure proxy constructor is in dex cache

9 years agoam 0bf01ef1: am 679c3ced: Merge "Prevent exception bugs in class linker" into lmp-dev
Mathieu Chartier [Thu, 28 Aug 2014 19:01:29 +0000 (19:01 +0000)]
am 0bf01ef1: am 679c3ced: Merge "Prevent exception bugs in class linker" into lmp-dev

* commit '0bf01ef1c5ac31ff35cb044b3fc48c3ea90b04be':
  Prevent exception bugs in class linker

9 years agoMerge "Ensure proxy constructor is in dex cache" into lmp-dev
Brian Carlstrom [Thu, 28 Aug 2014 18:54:15 +0000 (18:54 +0000)]
Merge "Ensure proxy constructor is in dex cache" into lmp-dev

9 years agoam 679c3ced: Merge "Prevent exception bugs in class linker" into lmp-dev
Mathieu Chartier [Thu, 28 Aug 2014 18:48:33 +0000 (18:48 +0000)]
am 679c3ced: Merge "Prevent exception bugs in class linker" into lmp-dev

* commit '679c3cedaf1550113104f5113b23f0fdec6175c0':
  Prevent exception bugs in class linker

9 years agoMerge "Prevent exception bugs in class linker" into lmp-dev
Mathieu Chartier [Thu, 28 Aug 2014 18:37:29 +0000 (18:37 +0000)]
Merge "Prevent exception bugs in class linker" into lmp-dev

9 years agoEnsure proxy constructor is in dex cache
Sebastien Hertz [Wed, 27 Aug 2014 13:32:56 +0000 (15:32 +0200)]
Ensure proxy constructor is in dex cache

Bug: 17262039

(cherry picked from commit ae94e350f100207359b8345d8d21e78e7cfb44c4)

Change-Id: I1cfc8dac1f63b9012fa098804135be8847b1daee

9 years agoFix issue with Thread.setName hanging after Thread.start
Brian Carlstrom [Thu, 28 Aug 2014 06:43:46 +0000 (23:43 -0700)]
Fix issue with Thread.setName hanging after Thread.start

When suspending a thread by peer the invariant that only attached threads are
suspended must be maintained. Add a ThreadList::Contains check which requires
making this method non-static.
Add some extra thread logging.

Bug: 17302037
Change-Id: I89af9840e398c2ceadeb736c431dfe3be8f3ba06

9 years agoPrevent exception bugs in class linker
Mathieu Chartier [Wed, 27 Aug 2014 22:28:28 +0000 (15:28 -0700)]
Prevent exception bugs in class linker

There were some places that could throw exceptions but still succeed.
This caused the allocation entrypoints to occasionally allocate a
heap object with a pending exception.
Also added some additional AssertNoExceptionPending.

Bug: 17164348

Change-Id: Ic6dd3b0cce9955349176503dd7f6c3da7ab0a6f1

9 years agoam aad269d7: Merge "Actually ask the pthread implementation for the stack guard size."
Elliott Hughes [Thu, 28 Aug 2014 17:21:32 +0000 (17:21 +0000)]
am aad269d7: Merge "Actually ask the pthread implementation for the stack guard size."

* commit 'aad269d708cb1dd18fe3a319a862f91f01696435':
  Actually ask the pthread implementation for the stack guard size.

9 years agoam 044489b4: am 5b41c2a8: Merge "Actually ask the pthread implementation for the...
Elliott Hughes [Thu, 28 Aug 2014 17:13:52 +0000 (17:13 +0000)]
am 044489b4: am 5b41c2a8: Merge "Actually ask the pthread implementation for the stack guard size." into lmp-dev

* commit '044489b463722e1377b49477b1c555f569389336':
  Actually ask the pthread implementation for the stack guard size.

9 years agoam 1b761a95: am d1e4e735: Merge "Revert "Avoid using NanoTime for compiler and verifi...
Ian Rogers [Thu, 28 Aug 2014 17:13:51 +0000 (17:13 +0000)]
am 1b761a95: am d1e4e735: Merge "Revert "Avoid using NanoTime for compiler and verifier"" into lmp-dev

* commit '1b761a957961329d845a55f2dac41200ac9b244f':
  Revert "Avoid using NanoTime for compiler and verifier"

9 years agoam dbc0ca70: Merge "Ensure proxy constructor is in dex cache"
Sebastien Hertz [Thu, 28 Aug 2014 17:07:16 +0000 (17:07 +0000)]
am dbc0ca70: Merge "Ensure proxy constructor is in dex cache"

* commit 'dbc0ca70bcee22561c2f0ab7129a7721369e9aea':
  Ensure proxy constructor is in dex cache

9 years agoam 5b41c2a8: Merge "Actually ask the pthread implementation for the stack guard size...
Elliott Hughes [Thu, 28 Aug 2014 17:00:01 +0000 (17:00 +0000)]
am 5b41c2a8: Merge "Actually ask the pthread implementation for the stack guard size." into lmp-dev

* commit '5b41c2a8f8e4a745b18685413b51c883cd993403':
  Actually ask the pthread implementation for the stack guard size.

9 years agoam d1e4e735: Merge "Revert "Avoid using NanoTime for compiler and verifier"" into...
Ian Rogers [Thu, 28 Aug 2014 17:00:00 +0000 (17:00 +0000)]
am d1e4e735: Merge "Revert "Avoid using NanoTime for compiler and verifier"" into lmp-dev

* commit 'd1e4e735cdf98e2ece7fa9a5dfeaad98d0c37cd1':
  Revert "Avoid using NanoTime for compiler and verifier"

9 years agoMerge "Actually ask the pthread implementation for the stack guard size."
Elliott Hughes [Thu, 28 Aug 2014 16:58:34 +0000 (16:58 +0000)]
Merge "Actually ask the pthread implementation for the stack guard size."

9 years agoActually ask the pthread implementation for the stack guard size.
Elliott Hughes [Wed, 27 Aug 2014 18:47:01 +0000 (11:47 -0700)]
Actually ask the pthread implementation for the stack guard size.

(cherry-pick 307091dc306c34ce9e4ee6cc3b467807b3a3bd12.)

Bug: 17111575
Change-Id: I23919b1e8aeff627a65daf57f1109bec60b196cc

9 years agoMerge "Actually ask the pthread implementation for the stack guard size." into lmp-dev
Elliott Hughes [Thu, 28 Aug 2014 16:50:40 +0000 (16:50 +0000)]
Merge "Actually ask the pthread implementation for the stack guard size." into lmp-dev

9 years agoMerge "Revert "Avoid using NanoTime for compiler and verifier"" into lmp-dev
Ian Rogers [Thu, 28 Aug 2014 16:50:38 +0000 (16:50 +0000)]
Merge "Revert "Avoid using NanoTime for compiler and verifier"" into lmp-dev

9 years agoMerge "Ensure proxy constructor is in dex cache"
Sebastien Hertz [Thu, 28 Aug 2014 16:50:17 +0000 (16:50 +0000)]
Merge "Ensure proxy constructor is in dex cache"

9 years agoEnsure proxy constructor is in dex cache
Sebastien Hertz [Wed, 27 Aug 2014 13:32:56 +0000 (15:32 +0200)]
Ensure proxy constructor is in dex cache

Bug: 17262039

Change-Id: I879341fa9c66f18ec64d6d4bf6a222a452c27739

9 years agoam 782e743e: am cd7a048c: Merge "Support booting without functioning boot.oat/art...
Brian Carlstrom [Wed, 27 Aug 2014 23:59:16 +0000 (23:59 +0000)]
am 782e743e: am cd7a048c: Merge "Support booting without functioning boot.oat/art patchoat." into lmp-dev

* commit '782e743e949240ffb97a523801f9dc5997e2b61b':
  Support booting without functioning boot.oat/art patchoat.

9 years agoam cd7a048c: Merge "Support booting without functioning boot.oat/art patchoat." into...
Brian Carlstrom [Wed, 27 Aug 2014 23:47:50 +0000 (23:47 +0000)]
am cd7a048c: Merge "Support booting without functioning boot.oat/art patchoat." into lmp-dev

* commit 'cd7a048cfac23e6ec3e1ba3ab4c1dddb322eaadf':
  Support booting without functioning boot.oat/art patchoat.

9 years agoMerge "Support booting without functioning boot.oat/art patchoat." into lmp-dev
Brian Carlstrom [Wed, 27 Aug 2014 23:40:53 +0000 (23:40 +0000)]
Merge "Support booting without functioning boot.oat/art patchoat." into lmp-dev

9 years agoActually ask the pthread implementation for the stack guard size.
Elliott Hughes [Wed, 27 Aug 2014 18:47:01 +0000 (11:47 -0700)]
Actually ask the pthread implementation for the stack guard size.

Bug: 17111575
Change-Id: I23919b1e8aeff627a65daf57f1109bec60b196cc

9 years agoam 9c188588: Merge "Fix gcstress and gcverify test targets"
Ian Rogers [Wed, 27 Aug 2014 23:30:28 +0000 (23:30 +0000)]
am 9c188588: Merge "Fix gcstress and gcverify test targets"

* commit '9c18858899738f79e75b5a31a96b91c8bfa8bdd7':
  Fix gcstress and gcverify test targets

9 years agoMerge "Fix gcstress and gcverify test targets"
Ian Rogers [Wed, 27 Aug 2014 23:12:35 +0000 (23:12 +0000)]
Merge "Fix gcstress and gcverify test targets"

9 years agoFix gcstress and gcverify test targets
Alex Light [Wed, 27 Aug 2014 23:08:57 +0000 (16:08 -0700)]
Fix gcstress and gcverify test targets

Make the build system look at the right variables to determine if these
should be on.

Change-Id: Iede993714eee773716cdc99181b10b280e96ca2a

9 years agoam 2602e205: am 074c8c6a: Merge "Avoid using NanoTime for compiler and verifier"...
Mathieu Chartier [Wed, 27 Aug 2014 22:06:02 +0000 (22:06 +0000)]
am 2602e205: am 074c8c6a: Merge "Avoid using NanoTime for compiler and verifier" into lmp-dev

* commit '2602e205729fe90cffff9adeb25d7f65a59bdf21':
  Avoid using NanoTime for compiler and verifier

9 years agoam 074c8c6a: Merge "Avoid using NanoTime for compiler and verifier" into lmp-dev
Mathieu Chartier [Wed, 27 Aug 2014 21:52:49 +0000 (21:52 +0000)]
am 074c8c6a: Merge "Avoid using NanoTime for compiler and verifier" into lmp-dev

* commit '074c8c6aac3bd42a173befd1a4eafd479ef12dc9':
  Avoid using NanoTime for compiler and verifier

9 years agoam 70644461: Merge "ART: Change temporaries to positive names"
Bill Buzbee [Wed, 27 Aug 2014 21:52:37 +0000 (21:52 +0000)]
am 70644461: Merge "ART: Change temporaries to positive names"

* commit '70644461909a68eb5cd032027ae0eb528143d99c':
  ART: Change temporaries to positive names

9 years agoRevert "Avoid using NanoTime for compiler and verifier"
Ian Rogers [Wed, 27 Aug 2014 21:46:57 +0000 (21:46 +0000)]
Revert "Avoid using NanoTime for compiler and verifier"

This reverts commit c89e6edbdce97c7d199b54bf63aed7ea01ebf1ff.

Change-Id: Id3b89121a19e64bf3c457c39b375f3d3c0fcc579

9 years agoMerge "Avoid using NanoTime for compiler and verifier" into lmp-dev
Mathieu Chartier [Wed, 27 Aug 2014 21:46:42 +0000 (21:46 +0000)]
Merge "Avoid using NanoTime for compiler and verifier" into lmp-dev

9 years agoam 03de1f07: am 52fb2a63: Merge "Handle nested signals" into lmp-dev
Dave Allison [Wed, 27 Aug 2014 21:36:07 +0000 (21:36 +0000)]
am 03de1f07: am 52fb2a63: Merge "Handle nested signals" into lmp-dev

* commit '03de1f07f5bdc2896d447415f2c3526556fa20f7':
  Handle nested signals

9 years agoam 90364c5f: Merge "Handle nested signals"
Dave Allison [Wed, 27 Aug 2014 21:35:22 +0000 (21:35 +0000)]
am 90364c5f: Merge "Handle nested signals"

* commit '90364c5f94069cf96e507b40076ea8eb2a6f303b':
  Handle nested signals

9 years agoMerge "ART: Change temporaries to positive names"
Bill Buzbee [Wed, 27 Aug 2014 21:33:14 +0000 (21:33 +0000)]
Merge "ART: Change temporaries to positive names"

9 years agoAvoid using NanoTime for compiler and verifier
Mathieu Chartier [Wed, 27 Aug 2014 21:10:58 +0000 (14:10 -0700)]
Avoid using NanoTime for compiler and verifier

Target timings before with -j2 --compiler-filter=interpret-only:
Approximate speedup 3%.

Before:
real    0m36.852s
real    0m36.050s
real    0m36.246s

After:
real    0m35.667s
real    0m35.779s
real    0m35.458s

Bug: 16853450

Change-Id: Id996172d8283639fe293ddfc26d01fc1a0beed0b

9 years agoresolved conflicts for merge of a42d23a7 to lmp-dev-plus-aosp
Mathieu Chartier [Wed, 27 Aug 2014 21:31:30 +0000 (14:31 -0700)]
resolved conflicts for merge of a42d23a7 to lmp-dev-plus-aosp

Change-Id: I0258f55e8199c06948878ba0e9bf7941bf60a435

9 years agoMerge "Handle nested signals"
Dave Allison [Wed, 27 Aug 2014 21:18:03 +0000 (21:18 +0000)]
Merge "Handle nested signals"

9 years agoam 52fb2a63: Merge "Handle nested signals" into lmp-dev
Dave Allison [Wed, 27 Aug 2014 19:54:05 +0000 (19:54 +0000)]
am 52fb2a63: Merge "Handle nested signals" into lmp-dev

* commit '52fb2a636a082cf5dc4c669a4b34085135709645':
  Handle nested signals

9 years agoSupport booting without functioning boot.oat/art patchoat.
Alex Light [Sat, 23 Aug 2014 00:49:35 +0000 (17:49 -0700)]
Support booting without functioning boot.oat/art patchoat.

Bug: 17000769

(cherry picked from commit 84d7605f93f1e6e86a16e02017e305c90e93117a)

Change-Id: I89c26a905af12ea288742368c2c038afd57a879a

9 years agoHandle nested signals
Dave Allison [Tue, 26 Aug 2014 18:07:58 +0000 (11:07 -0700)]
Handle nested signals

This allows for signals to be raised inside the ART signal handler.
This can occur when the JavaStackTraceHandler attempts to generate
a stack trace and something goes wrong.

It also fixes an issue where the fault manager was not being
correctly shut down inside the signal chaining code.  In this
case the signal handler was not restored to the original.

Bug: 17006816
Bug: 17133266

(cherry picked from commit fabe91e0d558936ac26b98d2b4ee1af08f58831d)

Change-Id: I10730ef52d5d8d34610a5293253b3be6caf4829e

9 years agoMerge "Handle nested signals" into lmp-dev
Dave Allison [Wed, 27 Aug 2014 19:42:39 +0000 (19:42 +0000)]
Merge "Handle nested signals" into lmp-dev

9 years agoam a0a0da29: Merge "Really fix tests."
Brian Carlstrom [Wed, 27 Aug 2014 19:25:08 +0000 (19:25 +0000)]
am a0a0da29: Merge "Really fix tests."

* commit 'a0a0da29e7d4d5c1bd471c49f1a4b6ec98fb767a':
  Really fix tests.

9 years agoMerge "Really fix tests."
Brian Carlstrom [Wed, 27 Aug 2014 19:01:17 +0000 (19:01 +0000)]
Merge "Really fix tests."

9 years agoam b68c6e57: Reduce interpret-only compile time.
Mathieu Chartier [Wed, 27 Aug 2014 18:58:37 +0000 (18:58 +0000)]
am b68c6e57: Reduce interpret-only compile time.

* commit 'b68c6e578a28a9717d78dfd522d9d9b8befaedf2':
  Reduce interpret-only compile time.

9 years agoReally fix tests.
Alex Light [Wed, 27 Aug 2014 18:13:47 +0000 (11:13 -0700)]
Really fix tests.

Remove extra line in Android.run-test.mk.
Update junit code in 082 and 021.
Set correct bootclasspath for --no-image.
Make host core.art depend on dex files being installed.
Make 118 pass in the correct bootclasspath.

Bug: 17290452

Change-Id: I415eddfa3632ec7eda927abe95925202de193749

9 years agoam 469f2ee9: Merge "Various run-test makefile fixes."
Ian Rogers [Wed, 27 Aug 2014 18:29:39 +0000 (18:29 +0000)]
am 469f2ee9: Merge "Various run-test makefile fixes."

* commit '469f2ee9a3c3d77f0352faed19d8c83c7397a638':
  Various run-test makefile fixes.

9 years agoMerge "Various run-test makefile fixes."
Ian Rogers [Wed, 27 Aug 2014 18:11:01 +0000 (18:11 +0000)]
Merge "Various run-test makefile fixes."

9 years agoVarious run-test makefile fixes.
Ian Rogers [Wed, 27 Aug 2014 17:54:11 +0000 (10:54 -0700)]
Various run-test makefile fixes.

Reduce number of test combinations based on global flags, rather than just
specific tests, to keep known broken list as small as possible.
Fix dist goal known broken definition.
Fix naming of rules in define-test-art-run-test to match the make target.
Undefine more variables after use.

Change-Id: I4b4eadd37fc1f816bc61d78c5866bfcee7b60244

9 years agoReduce interpret-only compile time.
Mathieu Chartier [Tue, 19 Aug 2014 21:29:46 +0000 (14:29 -0700)]
Reduce interpret-only compile time.

Before:
39.04user 5.18system 0:29.24elapsed 151%CPU (0avgtext+0avgdata 164176maxresident)k
38.87user 5.16system 0:29.14elapsed 151%CPU (0avgtext+0avgdata 164144maxresident)k

After:
36.26user 3.25system 0:27.00elapsed 146%CPU (0avgtext+0avgdata 162592maxresident)k
36.25user 3.28system 0:26.28elapsed 150%CPU (0avgtext+0avgdata 162688maxresident)k

Disabled implicit stack protection for the compiler, this reduces page faults.

Added support for not timing every method compilation and verification. NanoTime is
slow and adds ~2 seconds of real time. This is currently enabled since people want
to know which methods are slow to compile.

Bug: 16853450

(cherry picked from commit 8e219ae27624116b6d23e858fb21e93342f81d66)

Change-Id: I349ffb3f36db8c437137387aa6914dc17d743f09

9 years agoHandle nested signals
Dave Allison [Tue, 26 Aug 2014 18:07:58 +0000 (11:07 -0700)]
Handle nested signals

This allows for signals to be raised inside the ART signal handler.
This can occur when the JavaStackTraceHandler attempts to generate
a stack trace and something goes wrong.

It also fixes an issue where the fault manager was not being
correctly shut down inside the signal chaining code.  In this
case the signal handler was not restored to the original.

Bug: 17006816
Bug: 17133266
Change-Id: I9c25bf4f6921e6a107aefbdf47d2c0db9f41508f

9 years agoam 270c3e6f: Merge "Fix builds where dist_goal is defined."
Ian Rogers [Wed, 27 Aug 2014 04:03:23 +0000 (04:03 +0000)]
am 270c3e6f: Merge "Fix builds where dist_goal is defined."

* commit '270c3e6f95f047595a37ac15eb0734573377347d':
  Fix builds where dist_goal is defined.

9 years agoMerge "Fix builds where dist_goal is defined."
Ian Rogers [Wed, 27 Aug 2014 03:40:19 +0000 (03:40 +0000)]
Merge "Fix builds where dist_goal is defined."

9 years agoFix builds where dist_goal is defined.
Ian Rogers [Wed, 27 Aug 2014 03:39:04 +0000 (20:39 -0700)]
Fix builds where dist_goal is defined.

Change-Id: I93b3fb468cf760aa84d792b2651cdde366fc6033

9 years agoam 3d159704: Merge "Fix test build targets."
Ian Rogers [Wed, 27 Aug 2014 03:25:09 +0000 (03:25 +0000)]
am 3d159704: Merge "Fix test build targets."

* commit '3d159704d1f03d120cef8ad67a7a65f436fe8d00':
  Fix test build targets.

9 years agoMerge "Fix test build targets."
Ian Rogers [Wed, 27 Aug 2014 03:01:34 +0000 (03:01 +0000)]
Merge "Fix test build targets."

9 years agoFix test build targets.
Ian Rogers [Wed, 27 Aug 2014 02:59:52 +0000 (19:59 -0700)]
Fix test build targets.

Missing includes for libarttest.
Missed merge for adding IMAGE_TYPES to known broken tests.

Change-Id: I64227b3410bac9a1de8e0efd3b89eab78cd3cf5b

9 years agoam ff092c2d: Merge changes Id3113019,I89c26a90
Ian Rogers [Wed, 27 Aug 2014 02:41:22 +0000 (02:41 +0000)]
am ff092c2d: Merge changes Id3113019,I89c26a90

* commit 'ff092c2d5f672979c97f4c3b9d165e609ccf7ce0':
  Add run-test support for running without a working patchoat/dex2oat
  Support booting without functioning boot.oat/art patchoat.

9 years agoam 709368e6: Merge "Reduce interpret-only compile time."
Mathieu Chartier [Wed, 27 Aug 2014 02:41:15 +0000 (02:41 +0000)]
am 709368e6: Merge "Reduce interpret-only compile time."

* commit '709368e616791209b02d39adb6da5e55782cb45f':
  Reduce interpret-only compile time.

9 years agoam 2cdd2f2f: Merge "ART: Added dirty flag to optimizations"
Ian Rogers [Wed, 27 Aug 2014 02:41:13 +0000 (02:41 +0000)]
am 2cdd2f2f: Merge "ART: Added dirty flag to optimizations"

* commit '2cdd2f2f4c18de88c0daeddec362c0471d823b0b':
  ART: Added dirty flag to optimizations

9 years agoam ab66df8b: Merge "ART: Add capability for a pass to have options"
Ian Rogers [Wed, 27 Aug 2014 02:41:11 +0000 (02:41 +0000)]
am ab66df8b: Merge "ART: Add capability for a pass to have options"

* commit 'ab66df8bc9e32133cec45c56f2c17d6d705bfb21':
  ART: Add capability for a pass to have options

9 years agoART: Change temporaries to positive names
Razvan A Lupusoru [Sat, 7 Jun 2014 00:04:52 +0000 (17:04 -0700)]
ART: Change temporaries to positive names

Changes compiler temporaries to have positive names. The numbering now
puts them above the code VRs (locals + ins, in that order). The patch also
introduces APIs to query the number of temporaries, locals and ins.

The compiler temp infrastructure suffered from several issues
which are also addressed by this patch:
-There is no longer a queue of compiler temps. This would be polluted
with Method* when post opts were called multiple times.
-Sanity checks have been added to allow requesting of temps from BE
and to prevent temps after frame is committed.
-None of the structures holding temps can overflow because they are
allocated to allow holding maximum temps. Thus temps can be requested
by BE with no problem.
-Since the queue of compiler temps is no longer maintained, it is no
longer possible to refer to a temp that has invalid ssa (because it
was requested before ssa was run).
-The BE can now request temps after all ME allocations and it is guaranteed
to actually receive them.
-ME temps are now treated like normal VRs in all cases with no special
handling. Only the BE temps are handled specially because there are no
references to them from MIRs.
-Deprecated and removed several fields in CompilationUnit that saved
register information and updated callsites to call the new interface from
MIRGraph.

Change-Id: Ia8b1fec9384a1a83017800a59e5b0498dfb2698c
Signed-off-by: Razvan A Lupusoru <razvan.a.lupusoru@intel.com>
Signed-off-by: Udayan Banerji <udayan.banerji@intel.com>
9 years agoMerge changes Id3113019,I89c26a90
Ian Rogers [Wed, 27 Aug 2014 01:05:55 +0000 (01:05 +0000)]
Merge changes Id3113019,I89c26a90

* changes:
  Add run-test support for running without a working patchoat/dex2oat
  Support booting without functioning boot.oat/art patchoat.

9 years agoMerge "Reduce interpret-only compile time."
Mathieu Chartier [Wed, 27 Aug 2014 00:37:27 +0000 (00:37 +0000)]
Merge "Reduce interpret-only compile time."

9 years agoMerge "ART: Added dirty flag to optimizations"
Ian Rogers [Wed, 27 Aug 2014 00:36:31 +0000 (00:36 +0000)]
Merge "ART: Added dirty flag to optimizations"

9 years agoAdd run-test support for running without a working patchoat/dex2oat
Alex Light [Mon, 25 Aug 2014 20:25:56 +0000 (13:25 -0700)]
Add run-test support for running without a working patchoat/dex2oat

Also add it to the test-art target.

Bug: 17262039

Change-Id: Id31130194c46df66fc48852d6f8884f14cb08db4

9 years agoMerge "ART: Add capability for a pass to have options"
Ian Rogers [Wed, 27 Aug 2014 00:26:00 +0000 (00:26 +0000)]
Merge "ART: Add capability for a pass to have options"

9 years agoART: Add capability for a pass to have options
Razvan A Lupusoru [Thu, 3 Jul 2014 01:16:51 +0000 (18:16 -0700)]
ART: Add capability for a pass to have options

This patch adds capability to have pass options. These are needed when a pass
has multiple flags that can be tweaked. The user is now allowed to pass those
options via command line.

Since passes are treated as singletons and they are immutable, the overridden
options provided by user are set on the compilation unit. Doing this way also
allows a selectivity system to tweak the option per compilation instead of
doing it globally (due to the single pass existing).

The following command line flags have been added:
--print-pass-options - This prints all passes that have options along with
their defaults.
--pass-options= - This is used to pass the overridden options in format of
PassName:PassOption:PassOptionSetting

Change-Id: Ib5156f5d2ff51a0c64c4ea0fa050bd2170663417
Signed-off-by: Razvan A Lupusoru <razvan.a.lupusoru@intel.com>
Signed-off-by: Jean Christophe Beyler <jean.christophe.beyler@intel.com>
9 years agoam a9ffc181: Merge "Use loops to build combinations of run-tests."
Ian Rogers [Wed, 27 Aug 2014 00:12:08 +0000 (00:12 +0000)]
am a9ffc181: Merge "Use loops to build combinations of run-tests."

* commit 'a9ffc181c61f5634753bf1039ed50bf22c800fde':
  Use loops to build combinations of run-tests.

9 years agoART: Added dirty flag to optimizations
Jean Christophe Beyler [Fri, 18 Jul 2014 22:33:57 +0000 (15:33 -0700)]
ART: Added dirty flag to optimizations

- Added a dirty flag to the PassMEDataHolder and use it to determine
    if we want to perform the post-optimization passes or not.

Change-Id: I59a85f5198f9224d5fe8893397bfaf68a978d8f1
Signed-off-by: Jean Christophe Beyler <jean.christophe.beyler@intel.com>
9 years agoSupport booting without functioning boot.oat/art patchoat.
Alex Light [Sat, 23 Aug 2014 00:49:35 +0000 (17:49 -0700)]
Support booting without functioning boot.oat/art patchoat.

Bug: 17000769

Change-Id: I89c26a905af12ea288742368c2c038afd57a879a

9 years agoMerge "Use loops to build combinations of run-tests."
Ian Rogers [Tue, 26 Aug 2014 23:37:01 +0000 (23:37 +0000)]
Merge "Use loops to build combinations of run-tests."

9 years agoam 061d3103: Merge "ART: Add non-temporal store support"
Ian Rogers [Tue, 26 Aug 2014 23:36:08 +0000 (23:36 +0000)]
am 061d3103: Merge "ART: Add non-temporal store support"

* commit '061d3103991ab4c9a98a14228cbd4c8654da1ac8':
  ART: Add non-temporal store support

9 years agoUse loops to build combinations of run-tests.
Ian Rogers [Tue, 19 Aug 2014 23:52:36 +0000 (16:52 -0700)]
Use loops to build combinations of run-tests.

Add a JNI force copy test.
Split tracing from GC, relocation from prebuilding.
As the combinations of tests is sufficiently large to cause make serious work,
only build rules for tests that will be executed. Default to small number of
tests rather than full. Use ART_TEST_FULL=true to override.

Change-Id: Ic25431f29071b840d2e2295be6853511c0b52cfa

9 years agoMerge "ART: Add non-temporal store support"
Ian Rogers [Tue, 26 Aug 2014 22:55:27 +0000 (22:55 +0000)]
Merge "ART: Add non-temporal store support"

9 years agoReduce interpret-only compile time.
Mathieu Chartier [Tue, 19 Aug 2014 21:29:46 +0000 (14:29 -0700)]
Reduce interpret-only compile time.

Before:
39.04user 5.18system 0:29.24elapsed 151%CPU (0avgtext+0avgdata 164176maxresident)k
38.87user 5.16system 0:29.14elapsed 151%CPU (0avgtext+0avgdata 164144maxresident)k

After:
36.26user 3.25system 0:27.00elapsed 146%CPU (0avgtext+0avgdata 162592maxresident)k
36.25user 3.28system 0:26.28elapsed 150%CPU (0avgtext+0avgdata 162688maxresident)k

Disabled implicit stack protection for the compiler, this reduces page faults.

Added support for not timing every method compilation and verification. NanoTime is
slow and adds ~2 seconds of real time. This is currently enabled since people want
to know which methods are slow to compile.

Bug: 16853450

Change-Id: I349ffb3f36db8c437137387aa6914dc17d743f09

9 years agoART: Add non-temporal store support
Jean Christophe Beyler [Fri, 25 Jul 2014 19:32:18 +0000 (12:32 -0700)]
ART: Add non-temporal store support

Added non-temporal store support as a hint from the ME.
Added the implementation of the memory barrier
extended instruction that supports non-temporal stores
by explicitly serializing all previous store-to-memory instructions.

Change-Id: I8205a92083f9725253d8ce893671a133a0b6849d
Signed-off-by: Jean Christophe Beyler <jean.christophe.beyler@intel.com>
Signed-off-by: Chao-ying Fu <chao-ying.fu@intel.com>
9 years agoam 0038fcaa: Merge "Fix race condition in TransitionCollector."
Mathieu Chartier [Tue, 26 Aug 2014 18:24:56 +0000 (18:24 +0000)]
am 0038fcaa: Merge "Fix race condition in TransitionCollector."

* commit '0038fcaa6f2457f5de5544fc93cd57a305e466de':
  Fix race condition in TransitionCollector.

9 years agoam 652e3a16: am f59b605f: Merge "Fix race condition in TransitionCollector." into...
Mathieu Chartier [Tue, 26 Aug 2014 18:24:00 +0000 (18:24 +0000)]
am 652e3a16: am f59b605f: Merge "Fix race condition in TransitionCollector." into lmp-dev

* commit '652e3a164fe1348de573dcdb4d79926592117bcd':
  Fix race condition in TransitionCollector.

9 years agoam d6a4d7d8: am ccb1a5d4: ART: Reject field accesses to non-reference vregs.
Stephen Kyle [Tue, 26 Aug 2014 18:23:59 +0000 (18:23 +0000)]
am d6a4d7d8: am ccb1a5d4: ART: Reject field accesses to non-reference vregs.

* commit 'd6a4d7d8856785c79ba24426abcc0c684dc901f1':
  ART: Reject field accesses to non-reference vregs.

9 years agoam 0b05ec67: Merge "ART: Improve error message on oat writer failure"
Andreas Gampe [Tue, 26 Aug 2014 18:14:02 +0000 (18:14 +0000)]
am 0b05ec67: Merge "ART: Improve error message on oat writer failure"

* commit '0b05ec674139c2b2e0914b0bb0e1898f93a9c852':
  ART: Improve error message on oat writer failure

9 years agoam f59b605f: Merge "Fix race condition in TransitionCollector." into lmp-dev
Mathieu Chartier [Tue, 26 Aug 2014 18:12:29 +0000 (18:12 +0000)]
am f59b605f: Merge "Fix race condition in TransitionCollector." into lmp-dev

* commit 'f59b605f82c0e2b7bfc6019ee7bd650628f50d3a':
  Fix race condition in TransitionCollector.

9 years agoam ccb1a5d4: ART: Reject field accesses to non-reference vregs.
Stephen Kyle [Tue, 26 Aug 2014 18:12:28 +0000 (18:12 +0000)]
am ccb1a5d4: ART: Reject field accesses to non-reference vregs.

* commit 'ccb1a5d46bc1b4ee888e74775f5c1291d600ff54':
  ART: Reject field accesses to non-reference vregs.

9 years agoMerge "Fix race condition in TransitionCollector."
Mathieu Chartier [Tue, 26 Aug 2014 18:03:03 +0000 (18:03 +0000)]
Merge "Fix race condition in TransitionCollector."

9 years agoFix race condition in TransitionCollector.
Mathieu Chartier [Sat, 23 Aug 2014 20:56:03 +0000 (13:56 -0700)]
Fix race condition in TransitionCollector.

There was a minor race condition that could happen if multiple
threads called TransitionCollector at the same time. Starting out
in CMS:

Thread 1: TransitionCollector(SS).
Thread 2: TransitionCollector(CMS) sees that it is already CMS so
has copying_transition == false. But then thread 2 changes to SS.
Thread 1 resumes but incorrectly doesn't check the
disable_moving_gc_count_ possibly resulting in errors if
disable_moving_gc_count_ != 0 due to JNI since we are going from
SS -> CMS and are goign to move objects.

(cherry picked from commit de2233bbf64c7dde96513a729f6a97aebc2f0424)

Change-Id: I2fdda85f7cb7cbb0df3d1466a3c74853c5304c1b

9 years agoMerge "Fix race condition in TransitionCollector." into lmp-dev
Mathieu Chartier [Tue, 26 Aug 2014 18:01:37 +0000 (18:01 +0000)]
Merge "Fix race condition in TransitionCollector." into lmp-dev

9 years agoMerge "ART: Improve error message on oat writer failure"
Andreas Gampe [Tue, 26 Aug 2014 17:55:56 +0000 (17:55 +0000)]
Merge "ART: Improve error message on oat writer failure"

9 years agoART: Improve error message on oat writer failure
Andreas Gampe [Tue, 22 Jul 2014 05:56:08 +0000 (22:56 -0700)]
ART: Improve error message on oat writer failure

Try to log the pending exception to help resolve issues.

Bug: 16406811
Change-Id: I035ae9e59a5ee02b9e90c35c0644ec088d3c7d12

9 years agoam 3b6711fa: Merge "Fix usage of literals within a macro."
Ian Rogers [Tue, 26 Aug 2014 17:49:14 +0000 (17:49 +0000)]
am 3b6711fa: Merge "Fix usage of literals within a macro."

* commit '3b6711faf7b0b10eaa6c48ba854160bcecd00166':
  Fix usage of literals within a macro.

9 years agoMerge "Fix usage of literals within a macro."
Ian Rogers [Tue, 26 Aug 2014 17:36:09 +0000 (17:36 +0000)]
Merge "Fix usage of literals within a macro."

9 years agoFix race condition in TransitionCollector.
Mathieu Chartier [Sat, 23 Aug 2014 20:56:03 +0000 (13:56 -0700)]
Fix race condition in TransitionCollector.

There was a minor race condition that could happen if multiple
threads called TransitionCollector at the same time. Starting out
in CMS:

Thread 1: TransitionCollector(SS).
Thread 2: TransitionCollector(CMS) sees that it is already CMS so
has copying_transition == false. But then thread 2 changes to SS.
Thread 1 resumes but incorrectly doesn't check the
disable_moving_gc_count_ possibly resulting in errors if
disable_moving_gc_count_ != 0 due to JNI since we are going from
SS -> CMS and are goign to move objects.

Bug: 17189964

Change-Id: I4f0d002717516b81355d66f26e0e8ebe3958348d

9 years agoART: Reject field accesses to non-reference vregs.
Stephen Kyle [Fri, 22 Aug 2014 14:03:07 +0000 (15:03 +0100)]
ART: Reject field accesses to non-reference vregs.

The verifier will now reject any iget/iput insts
that do not have a reference object in vB.

Bug: 17207857
Signed-off-by: Stuart Monteith <stuart.monteith@arm.com>
(cherry picked from commit 695c5982eeddee18c643cc2f94c9d78fa75599ab)

Change-Id: I3790b8687673643bb142549d74f36a3e2978bf13

9 years agoFix usage of literals within a macro.
Ian Rogers [Tue, 26 Aug 2014 17:34:09 +0000 (10:34 -0700)]
Fix usage of literals within a macro.

Macs handle arguments in macros as $num meaning literals must be $$num.

Change-Id: Ia30c7ef4e089561e9f23f353d7080e800b117d70

9 years agoam 2f1f3b26: Merge "Map heap maps at a lower address."
Mathieu Chartier [Tue, 26 Aug 2014 17:29:19 +0000 (17:29 +0000)]
am 2f1f3b26: Merge "Map heap maps at a lower address."

* commit '2f1f3b26e360ed5362a3e6655ed0acdccea3a6a7':
  Map heap maps at a lower address.

9 years agoam b7fbafda: Merge "ART: kSuper also has IncompatibleClassChangeError"
Andreas Gampe [Tue, 26 Aug 2014 17:29:18 +0000 (17:29 +0000)]
am b7fbafda: Merge "ART: kSuper also has IncompatibleClassChangeError"

* commit 'b7fbafdac8cf0ff8f860e6e85646d6c88d3914ed':
  ART: kSuper also has IncompatibleClassChangeError

9 years agoam 654881ff: Merge "ART: Fix IncompatibleClassChange check for ArtMethod"
Andreas Gampe [Tue, 26 Aug 2014 17:29:17 +0000 (17:29 +0000)]
am 654881ff: Merge "ART: Fix IncompatibleClassChange check for ArtMethod"

* commit '654881ffd780066fe1160320059ae629416b4652':
  ART: Fix IncompatibleClassChange check for ArtMethod