OSDN Git Service
Vladimir Marko [Mon, 18 Aug 2014 13:20:29 +0000 (13:20 +0000)]
am
dfa7a6c8: Disable load/store elimination. DO NOT MERGE
* commit '
dfa7a6c8711623bf20aad58f57ae980b497ef4d7':
Disable load/store elimination. DO NOT MERGE
Vladimir Marko [Mon, 18 Aug 2014 10:42:51 +0000 (11:42 +0100)]
Disable load/store elimination. DO NOT MERGE
The optimization has several issues and it has been
effectively disabled by
https://android-review.googlesource.com/103511
Disable it completely to avoid wasting compilation time.
Bug:
16862804
Change-Id: I42f69ae64170dc44f5fd6455b6a51c9bd3f93672
Andreas Gampe [Mon, 18 Aug 2014 05:12:55 +0000 (05:12 +0000)]
am
5e8a63ae: Merge "ART: Do not recursively abort when visiting locks in a bad state" into lmp-dev
* commit '
5e8a63ae39cb5db996229c6c4a1a21a4b8dec5a2':
ART: Do not recursively abort when visiting locks in a bad state
Andreas Gampe [Sun, 17 Aug 2014 09:39:32 +0000 (09:39 +0000)]
am
d84f916d: ART: Avoid expensive class lookup in compiler driver
* commit '
d84f916d35bb0dc74d7daa075b883e5ee1ce1109':
ART: Avoid expensive class lookup in compiler driver
Andreas Gampe [Sat, 16 Aug 2014 05:35:46 +0000 (22:35 -0700)]
ART: Avoid expensive class lookup in compiler driver
When we are not compiling multidex, avoid an expensive by-descriptor
lookup for classes when deciding whether to skip compilation.
Bug:
17072698
Change-Id: I19bcb2a2928207fcae0badd466cdc52945bd37c4
Andreas Gampe [Mon, 18 Aug 2014 05:06:15 +0000 (05:06 +0000)]
Merge "ART: Do not recursively abort when visiting locks in a bad state" into lmp-dev
Andreas Gampe [Sat, 16 Aug 2014 20:41:10 +0000 (13:41 -0700)]
ART: Do not recursively abort when visiting locks in a bad state
This avoids a nested abort in VisitLocks.
Bug:
17080621,
16382675
Change-Id: Id604976ac9dcac0e319fb25cab4d2cbc98d7ee24
Nicolas Geoffray [Fri, 15 Aug 2014 19:39:31 +0000 (19:39 +0000)]
am
ebee8de7: Fix bugs in the x86 and arm versions of generic JNI.
* commit '
ebee8de725d5fa83483642786b19ea453d865762':
Fix bugs in the x86 and arm versions of generic JNI.
Nicolas Geoffray [Fri, 15 Aug 2014 19:39:31 +0000 (19:39 +0000)]
am
9c290012: Execute an application even when dex2oat crashes.
* commit '
9c290012b7f505ae1943ab87236f775b97a46e2d':
Execute an application even when dex2oat crashes.
Nicolas Geoffray [Wed, 13 Aug 2014 02:40:45 +0000 (03:40 +0100)]
Fix bugs in the x86 and arm versions of generic JNI.
Also fix the run script of 116-nodex2oat to use the non-prebuild
script for target.
Bug:
17000769
(cherry-picked from commit
54accbca0b549b1b1ad3ef09655dad438bc1e104)
Change-Id: I439fcd710fb8bb408f3288ff8fb34fef23890adb
Nicolas Geoffray [Tue, 22 Jul 2014 09:48:00 +0000 (10:48 +0100)]
Execute an application even when dex2oat crashes.
Bug:
17000769
(cherry picked from commit
4fcdc94d22a4608e355aa8df36240181149d10e8)
Change-Id: Iccb1fec94fe64ce4c3097510952f275482b86aa9
Ian Rogers [Fri, 15 Aug 2014 18:42:34 +0000 (18:42 +0000)]
am
378fe3d3: Merge "Make Monitor::Wait more robust to spurious Inflate failures." into lmp-dev
* commit '
378fe3d39d3a55f61d82ea482b8ce3082aee0e21':
Make Monitor::Wait more robust to spurious Inflate failures.
Mathieu Chartier [Fri, 15 Aug 2014 17:51:28 +0000 (17:51 +0000)]
am
bb6eccca: Merge "Use handle in one case that spans a gc-point." into lmp-dev
* commit '
bb6eccca7dcd673e24eed2fe95915780aa38a5d6':
Use handle in one case that spans a gc-point.
Jeff Hao [Fri, 15 Aug 2014 17:38:50 +0000 (17:38 +0000)]
am
95b4c65d: Prevent stubs from being installed in java.lang.reflect.Proxy.<init>.
* commit '
95b4c65da93500cdbdcaa3e01010771ef3f466f1':
Prevent stubs from being installed in java.lang.reflect.Proxy.<init>.
Jeff Hao [Fri, 15 Aug 2014 17:38:49 +0000 (17:38 +0000)]
am
4bf8d11d: Fix proxy tracing and enable tests that now work with tracing.
* commit '
4bf8d11df5dccc1b276cd9c40a98e8a14d79a9c8':
Fix proxy tracing and enable tests that now work with tracing.
Jeff Hao [Fri, 15 Aug 2014 00:18:52 +0000 (17:18 -0700)]
Prevent stubs from being installed in java.lang.reflect.Proxy.<init>.
This CL is a better fix for proxy tracing and undoes the changes in
https://android-review.googlesource.com/#/c/103025/
Bug:
16386215
(cherry picked from commit
db8a664e0b68c7c4d36270cd21dce8de1912d7f9)
Change-Id: Ic9e0ea2af7cb2da5d90c56aa009de92dba14cc47
Jeff Hao [Thu, 24 Jul 2014 23:26:09 +0000 (16:26 -0700)]
Fix proxy tracing and enable tests that now work with tracing.
Also updates proxy_test to generate an image for GetQuickOatCodeFor.
Bug:
16386215
(cherry picked from commit
f0a3f09c3d54646166a55c05a6b39c7dd504129c)
Change-Id: I138edbad9e1646db8590f2b1b73f2788d9710e68
Calin Juravle [Fri, 15 Aug 2014 13:16:47 +0000 (13:16 +0000)]
am
93de4273: [ART] Move to system/core native bridge library
* commit '
93de4273d72a2558a7b3423547b5074cd76c5796':
[ART] Move to system/core native bridge library
Calin Juravle [Tue, 12 Aug 2014 19:55:20 +0000 (20:55 +0100)]
[ART] Move to system/core native bridge library
Some cleanups on the way:
- move NativeBridgeRuntimeCallbacks from a global to a field
- rename native_bridge_library_string to the more suggestive
navtive_bridge_library_path
- clean up imports
Bug:
16884833
(cherry picked from commit I73aab8e212860ba5aee9444d801806d3da326a41)
Change-Id: I67f037ea81d2a4ede4294afd8b84b7640e534a13
Narayan Kamath [Fri, 15 Aug 2014 09:12:08 +0000 (09:12 +0000)]
am
4f1d05f0: Merge "More efficient stack walk in exception throwing." into lmp-dev
* commit '
4f1d05f0531b980b16dbff18161f02160919ec31':
More efficient stack walk in exception throwing.
Ian Rogers [Fri, 15 Aug 2014 18:34:19 +0000 (18:34 +0000)]
Merge "Make Monitor::Wait more robust to spurious Inflate failures." into lmp-dev
Ian Rogers [Fri, 15 Aug 2014 18:09:28 +0000 (11:09 -0700)]
Make Monitor::Wait more robust to spurious Inflate failures.
Bug:
17062710
Change-Id: Ife5f6b335caacc70cab543cd568676d277d3beb6
Narayan Kamath [Fri, 15 Aug 2014 09:08:15 +0000 (09:08 +0000)]
Merge "More efficient stack walk in exception throwing." into lmp-dev
Mathieu Chartier [Fri, 15 Aug 2014 17:45:30 +0000 (17:45 +0000)]
Merge "Use handle in one case that spans a gc-point." into lmp-dev
Mingyao Yang [Wed, 13 Aug 2014 21:51:03 +0000 (14:51 -0700)]
Use handle in one case that spans a gc-point.
Bug:
16689428
(cherry picked from commit
38eecb0f4288a374c9b0b4b4df8793eb5fc6697c)
Change-Id: Id4a1bde9012e4dd333493e46997c200537c9cf55
Andreas Gampe [Thu, 14 Aug 2014 22:51:43 +0000 (22:51 +0000)]
am
b84d7a22: Revert "Revert "ART: Fix mac build""
* commit '
b84d7a226399dcc60bc0a421715ffc90ce9213f7':
Revert "Revert "ART: Fix mac build""
Andreas Gampe [Thu, 14 Aug 2014 22:40:19 +0000 (22:40 +0000)]
Revert "Revert "ART: Fix mac build""
This reverts commit
350b03bca65f87895275d1ed4bb895d5afeeeff1.
Also fixes a fault_handler symbol.
Change-Id: I1315527400bffc8ded9f2a7e7fdb1c8a6c104b96
Dave Allison [Thu, 14 Aug 2014 22:07:57 +0000 (22:07 +0000)]
am
03c9785a: Revert "Revert "Reduce stack usage for overflow checks""
* commit '
03c9785a8a6d712775cf406c4371d0227c44148f':
Revert "Revert "Reduce stack usage for overflow checks""
Dave Allison [Thu, 14 Aug 2014 17:02:48 +0000 (17:02 +0000)]
Revert "Revert "Reduce stack usage for overflow checks""
Fixes stack protection issue.
Fixes mac build issue.
This reverts commit
83b1940e6482b9d8feba5c492507735686650ea5.
Change-Id: I7ba17252882b23a740bcda2ea94aacf398255406
Hiroshi Yamauchi [Wed, 13 Aug 2014 18:12:22 +0000 (11:12 -0700)]
More efficient stack walk in exception throwing.
In the exception handling code, we currently walk down the stack
twice, once to get the stack height which we use to compute frame IDs
(the bottom frame is zero), and once more to find the catch block to
jump to.
For a deep stack, this could result in very slow exception
handling. That is, if have a lot of finally or catch blocks that we
end up jumping to in a deep stack, we need to do a lot of
catch/rethrow chains. Since we'd need to walk down to the bottom each
time to compute frames IDs in each catch/rethrow, we'd need to walk
down O(N^2) frames at the worst case.
Instead of frames IDs ((the bottom frame is zero), we will use the
frame depth (the top frame is zero) and no longer need to walk down
the stack just to get the stack height. We walk down O(N) frames.
This was what was happening with
code.google.gson.functional.CircularReferenceTest. With this change,
the test run time went from ~120s down to ~3s on N5 and it no longer
crashes due to the thread suspension timeout.
(cherry pick commit
649278cec7119cdd1bea3d0b710dbb2aa7c650b6)
Bug:
16800209
Change-Id: Ie815df1e3e8fb9d82e40685d4cc2b8838fd8aa07
Mathieu Chartier [Thu, 14 Aug 2014 20:12:07 +0000 (20:12 +0000)]
am
b162bf5a: Use homogeneous space compaction if proper flag is set.
* commit '
b162bf5af5c2e508c6947471ceffaa98991794f4':
Use homogeneous space compaction if proper flag is set.
Mathieu Chartier [Thu, 14 Aug 2014 01:07:31 +0000 (18:07 -0700)]
Use homogeneous space compaction if proper flag is set.
If the flag ART_USE_HSPACE_COMPACT is set then we use
hspace compaction.
Bug:
16401001
Change-Id: I74e1db764f9ff16c007fa3bd16cb2c9b468ec83c
Andreas Gampe [Thu, 14 Aug 2014 18:40:01 +0000 (18:40 +0000)]
am
1d778570: Merge "ART: Fix run-test in case of relative TMPDIR" into lmp-dev
* commit '
1d7785704cafca20629c4bfb87c60d99dd69f72a':
ART: Fix run-test in case of relative TMPDIR
Andreas Gampe [Thu, 14 Aug 2014 18:08:53 +0000 (18:08 +0000)]
am
0b6952ca: Merge "Revert "ART: Fix mac build"" into lmp-dev
* commit '
0b6952cac86de294fb01898694b6a55f0a162a43':
Revert "ART: Fix mac build"
Vladimir Marko [Thu, 14 Aug 2014 17:59:50 +0000 (17:59 +0000)]
am
e470a027: Merge "Fix intrinsic Math.abs(double) for ARM." into lmp-dev
* commit '
e470a02792ce584fe0ff1b9a7dd6aebe62ca9be5':
Fix intrinsic Math.abs(double) for ARM.
Dave Allison [Thu, 14 Aug 2014 17:00:55 +0000 (17:00 +0000)]
am
b0bf089c: Merge "Revert "Reduce stack usage for overflow checks"" into lmp-dev
* commit '
b0bf089c2d80f861e4d8798e97828452b2e0b99b':
Revert "Reduce stack usage for overflow checks"
Mathieu Chartier [Thu, 14 Aug 2014 17:00:54 +0000 (17:00 +0000)]
am
a73a8b55: Move explicit GC after we are done initalizing all classes.
* commit '
a73a8b55415b544355438522524a640378698d85':
Move explicit GC after we are done initalizing all classes.
Mathieu Chartier [Mon, 11 Aug 2014 20:52:12 +0000 (13:52 -0700)]
Move explicit GC after we are done initalizing all classes.
Previously we had the GC happen for each dex file, this added a
few seconds of time on boot.oat creation.
Bug:
16853450
(cherry picked from commit
093ef212b7db4c17ed62df5cac0c1ac3a24e691d)
Change-Id: Ia469838eff1550c175c3465a478d8081023601bd
Andreas Gampe [Thu, 14 Aug 2014 16:17:10 +0000 (16:17 +0000)]
am
0c2a6d76: Merge "ART: Fix another potential N-th visitor issue" into lmp-dev
* commit '
0c2a6d76ccedb77e79adc68c723cd4ed8278fee9':
ART: Fix another potential N-th visitor issue
Dave Allison [Thu, 14 Aug 2014 16:59:07 +0000 (16:59 +0000)]
Merge "Revert "Reduce stack usage for overflow checks"" into lmp-dev
Dave Allison [Thu, 14 Aug 2014 16:54:09 +0000 (16:54 +0000)]
Revert "Reduce stack usage for overflow checks"
This reverts commit
63c051a540e6dfc806f656b88ac3a63e99395429.
Change-Id: I282a048994fcd130fe73842b16c21680053c592f
Andreas Gampe [Thu, 14 Aug 2014 16:11:58 +0000 (16:11 +0000)]
Merge "ART: Fix another potential N-th visitor issue" into lmp-dev
Andreas Gampe [Thu, 14 Aug 2014 16:08:46 +0000 (16:08 +0000)]
am
564f5830: ART: Fix mac build
* commit '
564f58305961986591ccb2fff04b9ccdd430a6db':
ART: Fix mac build
Andreas Gampe [Thu, 14 Aug 2014 02:54:09 +0000 (19:54 -0700)]
ART: Fix mac build
is_main_thread is unused in the case of a mac.
Bug:
15435566
(cherry picked from commit
7b1bf42dbc765524d54d45a70cb7eb1cccf0c617)
Change-Id: I9848d0156a4903ce73d20b5ce282b61ceb3039ff
Andreas Gampe [Thu, 14 Aug 2014 15:43:58 +0000 (15:43 +0000)]
am
c483a73c: Merge "ART: Fix class-linker handling" into lmp-dev
* commit '
c483a73cc2f9d63c1b6bccd1007117cf0d6dcd56':
ART: Fix class-linker handling
Andreas Gampe [Thu, 14 Aug 2014 12:20:48 +0000 (12:20 +0000)]
am
f8c4ed08: ART: Fix wrong command line parameter
* commit '
f8c4ed08832ec5302c9ea8441ec8cb39b87075ab':
ART: Fix wrong command line parameter
Calin Juravle [Thu, 14 Aug 2014 12:20:47 +0000 (12:20 +0000)]
am
4c37afb7: Get the native bridge library from the framework.
* commit '
4c37afb7b207424ff788ab8d8778ea4e8d4ac324':
Get the native bridge library from the framework.
Ian Rogers [Thu, 14 Aug 2014 12:20:46 +0000 (12:20 +0000)]
am
20ae1d77: Clean up native bridge
* commit '
20ae1d77463d1c3f206eac8d541124b4555464de':
Clean up native bridge
Andreas Gampe [Fri, 8 Aug 2014 19:05:10 +0000 (12:05 -0700)]
ART: Fix wrong command line parameter
Bug:
16843953
(cherry picked from commit
c4a7acf505944e7d937f0670b322e12302dd38dc)
Change-Id: Ib59e10a5d97a20ea1a013587645e0eeb204eda8d
Calin Juravle [Thu, 7 Aug 2014 13:53:41 +0000 (14:53 +0100)]
Get the native bridge library from the framework.
Framework now passes the name of the native bridge to the runtime so
there's no need to do the reading in libart.
Bug:
16843953
(cherry picked from commit
421b6466977d4b1ccd453f23b7b492f219099702)
Change-Id: I9a0a8b8c64949e05c6f90bc846d68f0797640f6d
Ian Rogers [Thu, 17 Jul 2014 18:09:10 +0000 (11:09 -0700)]
Clean up native bridge
(partially cherry picked from commit
68d8b42ddec39ec0174162d90d4abaa004d1983e)
Change-Id: I85813e97d73b1b331646dd0e880108dfbfe80f69
Andreas Gampe [Thu, 14 Aug 2014 06:05:38 +0000 (23:05 -0700)]
ART: Fix another potential N-th visitor issue
Bug:
16867274
Change-Id: I5466af563032041ef4a547894f40f303871302c6
Andreas Gampe [Thu, 14 Aug 2014 05:21:55 +0000 (05:21 +0000)]
am
7e9cdbbc: Merge "ART: Tolerate shallow call stack in VMStack_getCallingClassLoader" into lmp-dev
* commit '
7e9cdbbc14c015028d4b239a09c5af33ffc2087c':
ART: Tolerate shallow call stack in VMStack_getCallingClassLoader
Dan Albert [Thu, 14 Aug 2014 05:21:54 +0000 (05:21 +0000)]
am
c85a899b: Revert "Revert "Remove GCC atomic workarounds.""
* commit '
c85a899b23327c3d5693c8dabf5b482c3fb5ded8':
Revert "Revert "Remove GCC atomic workarounds.""
Dan Albert [Thu, 14 Aug 2014 05:14:18 +0000 (05:14 +0000)]
am
f0a37663: Initialize art::Atomic\'s value to 0.
* commit '
f0a37663309e468f11858bffa3221e7223cf577d':
Initialize art::Atomic's value to 0.
Dan Albert [Mon, 11 Aug 2014 23:38:02 +0000 (16:38 -0700)]
Revert "Revert "Remove GCC atomic workarounds.""
This reverts commit
626a2468e4e4f39db7b0f35a4fee87293e360e92.
(cherry picked from commit
aab0f86e3b079598d41c3a00bfa765a7589c5110)
Change-Id: I2333b54dcfd5ee8227a6479d2e3814b25377bdfd
Dan Albert [Tue, 12 Aug 2014 18:48:34 +0000 (11:48 -0700)]
Initialize art::Atomic's value to 0.
art::Mutex and art::ConditionVariable were DCHECKing that the value was
zero on entering the constructor without ever initializing the value.
Since 0 is a sensible default, make it the default for art::Atomic
rather than initializing for each use.
Bug:
16301104
(cherry picked from commit
6a3f8d93ddf09b5f6667820089e488958cba8361)
Change-Id: Iab69bccac9f904111a72eff8162c7f0ca979a08a
Dave Allison [Wed, 13 Aug 2014 23:00:21 +0000 (23:00 +0000)]
am
63c051a5: Reduce stack usage for overflow checks
* commit '
63c051a540e6dfc806f656b88ac3a63e99395429':
Reduce stack usage for overflow checks
Dave Allison [Fri, 25 Jul 2014 23:15:27 +0000 (16:15 -0700)]
Reduce stack usage for overflow checks
This reduces the stack space reserved for overflow checks to 12K, split
into an 8K gap and a 4K protected region. GC needs over 8K when running
in a stack overflow situation.
Also prevents signal runaway by detecting a signal inside code that
resulted from a signal handler invokation. And adds a max signal count to
the SignalTest to prevent it running forever.
Also reduces the number of iterations for the InterfaceTest as this was
taking (almost) forever with the --trace option on run-test.
Bug:
15435566
Change-Id: Id4fd46f22d52d42a9eb431ca07948673e8fda694
Conflicts:
compiler/optimizing/code_generator_x86_64.cc
runtime/arch/x86/fault_handler_x86.cc
runtime/arch/x86_64/quick_entrypoints_x86_64.S
Ian Rogers [Wed, 13 Aug 2014 21:38:41 +0000 (21:38 +0000)]
am
c739de01: Merge "Avoid use of std::string where we have const char*." into lmp-dev
* commit '
c739de01b905729b8587003a52a55b62358e979d':
Avoid use of std::string where we have const char*.
Mathieu Chartier [Wed, 13 Aug 2014 19:01:33 +0000 (19:01 +0000)]
am
83715081: Add allocation rate tracking to systrace.
* commit '
837150818af0eee993f7d93c5a50c3c7b46f1dab':
Add allocation rate tracking to systrace.
Mathieu Chartier [Wed, 13 Aug 2014 17:08:05 +0000 (10:08 -0700)]
Add allocation rate tracking to systrace.
Bug:
16956980
Change-Id: I7074f4ae0f75d1bb58e9860ded4e17ce6978d74e
Dave Allison [Wed, 13 Aug 2014 18:08:06 +0000 (18:08 +0000)]
am
b0f05b96: Add implicit checks for x86_64 architecture.
* commit '
b0f05b9654eb005bc8c8e15f615a7f5a312f640c':
Add implicit checks for x86_64 architecture.
Andreas Gampe [Thu, 14 Aug 2014 15:40:19 +0000 (15:40 +0000)]
Merge "ART: Fix class-linker handling" into lmp-dev
Andreas Gampe [Thu, 14 Aug 2014 04:49:37 +0000 (21:49 -0700)]
ART: Fix class-linker handling
ResolveMethod did not account correctly for the mutual exclusivity
of direct and static methods. In such a case we threw a NoSuchMethodError,
while the correct behavior is to throw an IncompatibleClassChangeError.
Bug:
16956477
Change-Id: Id014affe0b8a43dbd75570b123b921d5853ab135
Andreas Gampe [Thu, 14 Aug 2014 05:18:47 +0000 (05:18 +0000)]
Merge "ART: Tolerate shallow call stack in VMStack_getCallingClassLoader" into lmp-dev
Andreas Gampe [Tue, 12 Aug 2014 01:51:53 +0000 (18:51 -0700)]
ART: Tolerate shallow call stack in VMStack_getCallingClassLoader
When the call stack does not have the three methods we expect,
the visitor will return a nullptr value.
Add a test to JniTest (and refactor the test a little for code reuse).
Bug:
16867274
Change-Id: I5fb8f91f372a41e0bc11ef9f70640834591afa53
Dave Allison [Wed, 16 Jul 2014 23:04:32 +0000 (16:04 -0700)]
Add implicit checks for x86_64 architecture.
This combines the x86 and x86_64 fault handlers into one. It also
merges in the change to the entrypoints for X86_64.
Replaces generic instruction length calculator with one that only
works with the specific instructions we use.
Bug:
16256184
Change-Id: I1e8ab5ad43f46060de9597615b423c89a836035c
Signed-off-by: Chao-ying Fu <chao-ying.fu@intel.com>
Andreas Gampe [Wed, 13 Aug 2014 07:50:56 +0000 (07:50 +0000)]
am
4527f61c: ART: Don\'t leave rex_ & length_ uninitialized
* commit '
4527f61cbc1a3b01ab249181900b8ea547edaf90':
ART: Don't leave rex_ & length_ uninitialized
Andreas Gampe [Wed, 13 Aug 2014 06:24:16 +0000 (23:24 -0700)]
ART: Don't leave rex_ & length_ uninitialized
Don't leave fields uninitialized in the x86-64 utils assembler.
Bug:
16993295
(cherry picked from commit
181211de89df5615395a55114fe551b08e86bc4d)
Change-Id: I2d648991fc2a6a4d5f518eb9ef0bb10857f6318a
Sebastien Hertz [Wed, 13 Aug 2014 06:14:50 +0000 (06:14 +0000)]
am
13cccfa3: Merge "Fix crash in VirtualMachine.AllThreads JDWP command" into lmp-dev
* commit '
13cccfa3ce61a41a176358ff2aa53e60d0acacd5':
Fix crash in VirtualMachine.AllThreads JDWP command
Vladimir Marko [Thu, 14 Aug 2014 17:53:07 +0000 (17:53 +0000)]
Merge "Fix intrinsic Math.abs(double) for ARM." into lmp-dev
Vladimir Marko [Thu, 14 Aug 2014 14:23:32 +0000 (15:23 +0100)]
Fix intrinsic Math.abs(double) for ARM.
Bug:
16930909
(cherry picked from commit
a5f90b6f0f5b33487e71eaeb05508555f17dcf30)
Change-Id: I3ee40ffbeeabf53c72ea5781d02cda1b92b13289
Sebastien Hertz [Wed, 13 Aug 2014 06:09:49 +0000 (06:09 +0000)]
Merge "Fix crash in VirtualMachine.AllThreads JDWP command" into lmp-dev
Andreas Gampe [Thu, 14 Aug 2014 18:37:56 +0000 (18:37 +0000)]
Merge "ART: Fix run-test in case of relative TMPDIR" into lmp-dev
Andreas Gampe [Wed, 13 Aug 2014 22:40:22 +0000 (15:40 -0700)]
ART: Fix run-test in case of relative TMPDIR
Relative paths should be resolved to absolute paths, as we are
chdir-ing in the test.
Bug:
16499668
Change-Id: Ic7db35e21f519ced77c4ac468d8e909bcb40b24d
Andreas Gampe [Thu, 14 Aug 2014 18:05:50 +0000 (18:05 +0000)]
Merge "Revert "ART: Fix mac build"" into lmp-dev
Andreas Gampe [Thu, 14 Aug 2014 18:04:12 +0000 (18:04 +0000)]
Revert "ART: Fix mac build"
The underlying CL was reverted.
This reverts commit
564f58305961986591ccb2fff04b9ccdd430a6db.
Change-Id: Ic9fe93a95c4357d0045a430713be7da517f626b5
Ian Rogers [Wed, 13 Aug 2014 21:34:53 +0000 (21:34 +0000)]
Merge "Avoid use of std::string where we have const char*." into lmp-dev
Ian Rogers [Tue, 12 Aug 2014 09:30:58 +0000 (02:30 -0700)]
Avoid use of std::string where we have const char*.
Removing the ClassHelper caused std::string creation for all calls to
Class::GetDescriptor and a significant performance regression. Make the
std::string an out argument so the caller can maintain it and its life time
while allowing GetDescriptor to return the common const char* case.
Don't generate GC maps when compilation is disabled.
Remove other uses of std::string that are occuring on critical paths.
Use the cheaper SkipClass in CompileMethod in CompilerDriver.
Specialize the utf8 as utf16 comparison code for the common shorter byte
encoding.
Force a bit of inlining, remove some UNLIKELYs (they are prone to pessimizing
code), add some LIKELYs.
x86-64 host 1-thread interpret-only of 57 apks:
Before: 29.539s
After: 23.467s
Regular compile:
Before: 1m35.347s
After: 1m20.056s
Bug:
16853450
Change-Id: Ic705ea24784bee24ab80084d06174cbf87d557ad
Conflicts:
runtime/utils.cc
Mathieu Chartier [Tue, 12 Aug 2014 17:45:00 +0000 (17:45 +0000)]
am
5f1dcee2: Merge "Change FreeMemory and TotalMemory behavior." into lmp-dev
* commit '
5f1dcee2b3cdaf0f745958ff366a84279940c65e':
Change FreeMemory and TotalMemory behavior.
Jeff Hao [Tue, 12 Aug 2014 17:23:26 +0000 (17:23 +0000)]
am
b21f1336: Make method tracing deoptimize on startup.
* commit '
b21f1336be9dd0c2607b2ff53834d62a218d1c83':
Make method tracing deoptimize on startup.
Jeff Hao [Tue, 12 Aug 2014 01:00:29 +0000 (18:00 -0700)]
Make method tracing deoptimize on startup.
Bug:
16848366
(cherry picked from commit
bc678bba8ec45dfcb1865676dd6d8188da9e56fc)
Change-Id: Idf75288026a837b95f4619acadec57e38bacbd7b
Mathieu Chartier [Tue, 12 Aug 2014 17:39:22 +0000 (17:39 +0000)]
Merge "Change FreeMemory and TotalMemory behavior." into lmp-dev
Mathieu Chartier [Thu, 7 Aug 2014 00:06:33 +0000 (17:06 -0700)]
Change FreeMemory and TotalMemory behavior.
TotalMemory is now equal to the footprint limit (around when the
next GC will occur).
FreeMemory is now equal to TotalMemory() - bytes_allocated_.
Also added more memory dumping info to DumpGcPerformanceInfo.
Bug:
16520008
Bug:
15819878
(cherry picked from commit
dd162fb5990cedf80a5093ecc0e77df82af5f754)
Change-Id: I68239a5295cc1372b3995eb781d67b9e25d6eaed
Sebastien Hertz [Tue, 12 Aug 2014 07:09:37 +0000 (09:09 +0200)]
Fix crash in VirtualMachine.AllThreads JDWP command
When collecting all the started threads, we may encounter a thread in the
process of being started from Java code (Thread.start) in the threads list. The
crash occurs when we attempt to access the java.lang.Thread peer (calling
Thread::GetPeer) but the JNI global reference to it (Thread::jpeer) has not
been destroyed yet (which is verified with a check). This only happens when the
thread is starting up.
We now check the thread finished starting up before accessing the Java peer.
This also prevents from returning non-started threads in the reply.
Bug:
16660101
Change-Id: I9e445e5da3a6951143d6c3c9a0d4f136398fde2f
Zheng Xu [Tue, 12 Aug 2014 15:49:08 +0000 (15:49 +0000)]
am
d74824bd: AArch64: Fix art_quick_string_compareto.
* commit '
d74824bdd01b2a76b310e1275a1114d39833a708':
AArch64: Fix art_quick_string_compareto.
Zheng Xu [Tue, 12 Aug 2014 09:19:12 +0000 (17:19 +0800)]
AArch64: Fix art_quick_string_compareto.
Though __memcmp16() is implemented in the same module as the caller, it
is still possible that the toolchain would put __memcmp16() into plt. In
that case, IP registers can be trashed when loading the function address.
Use x14/x15 to replace IP0/IP1.
Bug:
16974467
(cherry picked from commit
62ddb328860e907eb76ccd3abed63ba75438fea8)
Change-Id: I40e39d075860bc78624ce6ef8b4f8e33e57fc58c
Calin Juravle [Tue, 12 Aug 2014 11:40:34 +0000 (11:40 +0000)]
am
8c6c5211: Revert "Rename openDexFileNative to openDexFile."
* commit '
8c6c52110c5afd438741694520203a6d7e81a2a4':
Revert "Rename openDexFileNative to openDexFile."
Calin Juravle [Wed, 6 Aug 2014 19:07:41 +0000 (19:07 +0000)]
Revert "Rename openDexFileNative to openDexFile."
https://android-review.googlesource.com/#/c/103383/ introduces
absolute paths so we need back the old method name
Bug:
16644204
Bug:
16930747
This reverts commit
350b626eb1cfbaee7619e9453a4c018c07aed003.
(cherry picked from commit
a2069c7152eb46e8c64b4c8d7e19e5fab5de6df4)
Change-Id: I55c8cb31bb79b1ed8ea51f63ec504d77f76a582a
Andreas Gampe [Mon, 11 Aug 2014 15:07:05 +0000 (15:07 +0000)]
am
1c98286d: Merge "ART: Skip compiling redefined classes in apps" into lmp-dev
* commit '
1c98286d8ceaf3715e06f50acd6f3b0aa5a06f2e':
ART: Skip compiling redefined classes in apps
Andreas Gampe [Mon, 11 Aug 2014 15:03:15 +0000 (15:03 +0000)]
Merge "ART: Skip compiling redefined classes in apps" into lmp-dev
Dan Albert [Sun, 10 Aug 2014 17:20:16 +0000 (17:20 +0000)]
am
a0b530dd: Fix the Mac build.
* commit '
a0b530dd5e76662156b40c8a84ebb446b6332598':
Fix the Mac build.
Dan Albert [Sun, 10 Aug 2014 08:16:40 +0000 (08:16 +0000)]
am
6a763f14: Proper fix after -Bsymbolic change.
* commit '
6a763f149f0c6f705cab8fb1984aa6f9198b2841':
Proper fix after -Bsymbolic change.
Dan Albert [Sun, 10 Aug 2014 06:43:10 +0000 (23:43 -0700)]
Fix the Mac build.
Apparently Mac uses a different assembler directive for .hidden.
Change-Id: Ic2adb6bef82e376c3e5f4a29258a85a0dac00cd5
Dan Albert [Sun, 10 Aug 2014 01:12:10 +0000 (18:12 -0700)]
Proper fix after -Bsymbolic change.
Marking the offending symbols as hidden allows the linker to resolve
them at build time.
Bug:
16853291
Change-Id: I1480c1520c23f81e42831a66737bba7bdf3433c2
Dan Albert [Sat, 9 Aug 2014 06:51:23 +0000 (06:51 +0000)]
am
679ec1bf: Merge "Workaround for test failures after Bsymbolic fix." into lmp-dev
* commit '
679ec1bf0fb5cfd2c74cdf0747b5b8b8f4a5bc06':
Workaround for test failures after Bsymbolic fix.
Andreas Gampe [Sat, 9 Aug 2014 06:33:01 +0000 (06:33 +0000)]
am
55a39dbf: Merge "ART: Turn on dex2oat watchdog on target" into lmp-dev
* commit '
55a39dbfae0935a60e0de3a05573b3b1ee25f579':
ART: Turn on dex2oat watchdog on target
Andreas Gampe [Sat, 9 Aug 2014 06:15:57 +0000 (06:15 +0000)]
am
8c18c2aa: ART: Generate chained compare-and-branch for short switches
* commit '
8c18c2aaedb171f9b03ec49c94b0e33449dc411b':
ART: Generate chained compare-and-branch for short switches
Andreas Gampe [Wed, 6 Aug 2014 17:09:01 +0000 (10:09 -0700)]
ART: Generate chained compare-and-branch for short switches
Refactor Mir2Lir to generate chained compare-and-branch sequences
for short switches on all architectures.
Bug:
16241558
(cherry picked from commit
48971b3242e5126bcd800cc9c68df64596b43d13)
Change-Id: I0bb3071b8676523e90e0258e9b0e3fd69c1237f4