OSDN Git Service
Nicolas Geoffray [Fri, 18 Mar 2016 11:36:20 +0000 (11:36 +0000)]
Check if the type of an instruction is erroneous before inlining.
We can get HInstructions typed with a class that is in an error state.
For such classes, we cannot look at the vtable or imt table as they
are not cleanly populated.
bug:
27683927
Change-Id: I0d64ca470e1cb6cf9b40e9f02bb9b0bb12c2bac1
Nicolas Geoffray [Thu, 17 Mar 2016 14:39:13 +0000 (14:39 +0000)]
Merge "Dump oat file information as part of DumpForSigQuit."
Nicolas Geoffray [Thu, 17 Mar 2016 14:20:58 +0000 (14:20 +0000)]
Merge "Fix load store elimination bug in the presence of null[i]."
Nicolas Geoffray [Thu, 17 Mar 2016 11:56:54 +0000 (11:56 +0000)]
Dump oat file information as part of DumpForSigQuit.
Change-Id: Ie8d256d8fa16565105cfe5c39cc96826983469da
Nicolas Geoffray [Thu, 17 Mar 2016 10:44:24 +0000 (10:44 +0000)]
Fix load store elimination bug in the presence of null[i].
Due to the dex specification, we can be in a state where
two array get with the same dex register inputs are typed
differently.
bug:
27683874
Change-Id: Ia821fd32e86c306093372249e7686332a7584263
Nicolas Geoffray [Thu, 17 Mar 2016 10:50:11 +0000 (10:50 +0000)]
Merge "Check GetDumpNativeStackOnSigQuit() for daemons."
Nicolas Geoffray [Thu, 17 Mar 2016 09:35:04 +0000 (09:35 +0000)]
Check GetDumpNativeStackOnSigQuit() for daemons.
The flag is passed by run-test to avoid libunwind crash flakiness,
so we should check it when dumping threads.
Change-Id: I3a3fcfd80ab254a315d58d629c81161fea9900dc
Andreas Gampe [Thu, 17 Mar 2016 01:03:54 +0000 (01:03 +0000)]
Merge "ART: Remember whether the cache was pruned"
Bill Buzbee [Wed, 16 Mar 2016 23:06:01 +0000 (23:06 +0000)]
Merge "ART: Mterp arm64 2-operand double rem fix"
Alex Light [Wed, 16 Mar 2016 23:02:59 +0000 (23:02 +0000)]
Merge "Revert "Fix issue with proxy invocation on default methods""
Alex Light [Wed, 16 Mar 2016 23:00:49 +0000 (23:00 +0000)]
Revert "Fix issue with proxy invocation on default methods"
This reverts commit
ec3b7ab8f008f36f1072d4ba03da204229b95976.
Change-Id: Idfb2c63c246cc5dae7670503e70ad3f62bb1e1eb
Hiroshi Yamauchi [Wed, 16 Mar 2016 22:55:50 +0000 (22:55 +0000)]
Merge "Fix valgrind-test-art-host to run both 32-bit and 64-bit tests."
Mathieu Chartier [Wed, 16 Mar 2016 22:02:20 +0000 (22:02 +0000)]
Merge "Add verify-art-runtime usage to dex2oat"
Mathieu Chartier [Wed, 16 Mar 2016 21:29:17 +0000 (14:29 -0700)]
Add verify-art-runtime usage to dex2oat
Bug:
27688793
(cherry picked from commit
c6068709a482065d6f5426640f1c9316797508e3)
Change-Id: Idcf6b8023a15c816168ce42456d2bd7b463d0951
buzbee [Wed, 16 Mar 2016 21:39:50 +0000 (14:39 -0700)]
ART: Mterp arm64 2-operand double rem fix
An instruction ordering bug caused 2-operand double-precision
rem operations to be performed incorrectly on the arm64 fast
interpreter. Also, fixes the existing omnibus-opcodes smoke test
to better catch 2-operand float and double operation problems
(the problem was masked in the existing test).
Bug:
27604215
Change-Id: I9fac1e61d9defe7d623c611406c35921abb65b04
Mathieu Chartier [Wed, 16 Mar 2016 21:07:31 +0000 (21:07 +0000)]
Merge "Fix race in AllocEntrypointsInstrumented"
Alex Light [Wed, 16 Mar 2016 20:39:54 +0000 (20:39 +0000)]
Merge "Fix issue with proxy invocation on default methods"
Mathieu Chartier [Wed, 16 Mar 2016 18:25:29 +0000 (11:25 -0700)]
Fix race in AllocEntrypointsInstrumented
We were using the quick_alloc_entry_points_instrumentation_counter_,
this counter is updated before the threads are suspended. The
allocator could come out of a suspend point, see that threads are
supposedly still no instrumented, continue the allocation, then
suddenly quick_alloc_entry_points_instrumentation_counter_ becomes
1 and alloc_tracking_enabled_ becomes true resulting in a failing
DCHECK.
The fix is to add a boolean that is updated only when the threads
are suspended.
Bug:
27506909
(cherry picked from commit
77d993107773b7b9bd7f07ce08d0aaac1631bf84)
Change-Id: Id12983ef77c4fddb0394e4439ce0829777f1a70b
Hiroshi Yamauchi [Wed, 16 Mar 2016 18:39:33 +0000 (11:39 -0700)]
Fix valgrind-test-art-host to run both 32-bit and 64-bit tests.
Bug:
27156726
Change-Id: Ie93405d6238f7cc5eba351cfb6c5b0f5d8348908
David Srbecky [Wed, 16 Mar 2016 17:54:52 +0000 (17:54 +0000)]
Merge "Remove --native-debuggable compiler option."
Roland Levillain [Wed, 16 Mar 2016 17:54:34 +0000 (17:54 +0000)]
Merge "Make art::HCompare support boolean, byte, short and char inputs."
David Srbecky [Sun, 13 Mar 2016 22:00:07 +0000 (22:00 +0000)]
Remove --native-debuggable compiler option.
Check if both --debuggable and --generate-debug-info are set instead.
History: I have recently added the --native-debuggable flag to control
whether extra stack maps are generated in order to produce accurate
native line number mapping of all generated code. I didn't want to
include it into --debuggable since we compile boot.oat as debuggable
and so it would be affected.
On second thought, it would have been reasonable to generate the extra
stackmaps only if both --debuggable and --generate-debug-info are set,
instead of introducing another compiler flag. This means we do not
affect boot.oat but we still get the extra stack maps if DWARF debug
information is explicitly requested.
Change-Id: I7e5e557e4850a88b3b6f86178d2cb645fb1e1110
Nicolas Geoffray [Wed, 16 Mar 2016 17:10:40 +0000 (17:10 +0000)]
Merge "Forgot space."
Narayan Kamath [Wed, 16 Mar 2016 17:04:52 +0000 (17:04 +0000)]
Merge "runtime: delete native/java_lang_Runtime.h."
Nicolas Geoffray [Wed, 16 Mar 2016 17:01:39 +0000 (17:01 +0000)]
Forgot space.
Change-Id: If9eced6d5bd6d8d0248bad0e7b12095fb5a84d26
Nicolas Geoffray [Wed, 16 Mar 2016 16:58:22 +0000 (16:58 +0000)]
Merge "Fix inliner crash related to type propagation."
Roland Levillain [Tue, 15 Mar 2016 15:02:50 +0000 (15:02 +0000)]
Make art::HCompare support boolean, byte, short and char inputs.
Also extend tests covering the IntegerSignum, LongSignum,
IntegerCompare and LongCompare intrinsics and their
translation into an art::HCompare instruction.
Bug:
27629913
Change-Id: I0afc75ee6e82602b01ec348bbb36a08e8abb8bb8
Sebastien Hertz [Wed, 16 Mar 2016 16:16:41 +0000 (16:16 +0000)]
Merge "Re-enable checker tests 536 and 537"
Sebastien Hertz [Wed, 16 Mar 2016 15:20:10 +0000 (16:20 +0100)]
Re-enable checker tests 536 and 537
They were disabled due to a bug in Jack that is fixed now.
Bug:
25635944
Change-Id: Id68824a8b4abbb6710e61897ffb3ed5daf839413
Nicolas Geoffray [Wed, 16 Mar 2016 15:16:06 +0000 (15:16 +0000)]
Fix inliner crash related to type propagation.
- Do not agressively try to resolve a return type.
- Deal with unresolved return type.
bug:
25492619
Change-Id: Idc9c96a0b376fe5ee86b411c02ce7078c7f48c84
Narayan Kamath [Wed, 16 Mar 2016 15:11:15 +0000 (15:11 +0000)]
runtime: delete native/java_lang_Runtime.h.
Now unused, except for an unnecessary include.
bug:
27387202
Change-Id: Ie77dc0da156a64e8ea03af1878b44d20c16c5ae5
Calin Juravle [Wed, 16 Mar 2016 14:40:08 +0000 (14:40 +0000)]
Merge "Clean up NullCheck generation and record stats about it."
Calin Juravle [Wed, 16 Mar 2016 14:39:49 +0000 (14:39 +0000)]
Merge "Improve compiler stats"
Nicolas Geoffray [Wed, 16 Mar 2016 14:39:16 +0000 (14:39 +0000)]
Merge "Add more statistics for memory use of the JIT."
Nicolas Geoffray [Wed, 16 Mar 2016 14:20:06 +0000 (14:20 +0000)]
Add more statistics for memory use of the JIT.
Collect data for stack maps, profiling info, and compiled code.
bug:
27520994
Change-Id: Ic87361230c96ce0090027a37d750e948d806c597
David Srbecky [Wed, 16 Mar 2016 14:13:02 +0000 (14:13 +0000)]
Merge "Call __jit_debug_register_code() indirectly via global variable."
Calin Juravle [Wed, 16 Mar 2016 14:05:09 +0000 (14:05 +0000)]
Clean up NullCheck generation and record stats about it.
This removes redundant code from the generators and allows for easier
stat recording.
Change-Id: Iccd4368f9e9d87a6fecb863dee4e2145c97851c4
Calin Juravle [Wed, 16 Mar 2016 11:53:41 +0000 (11:53 +0000)]
Improve compiler stats
- report the max size of arena alloc
- report how many virtual or interface invokes were inlined
Change-Id: I82f154a8e25b5e3890181a1aa11346cdc3f93e37
Nicolas Geoffray [Wed, 16 Mar 2016 13:55:51 +0000 (13:55 +0000)]
Merge "Set USE_JACK to true."
Nicolas Geoffray [Wed, 16 Mar 2016 10:55:57 +0000 (10:55 +0000)]
Set USE_JACK to true.
This fixes running checker when just doing art/test/run-test.
Change-Id: Icfac5896cdfd64eb16b0127d5f16f552b93ad03e
Nicolas Geoffray [Wed, 16 Mar 2016 12:52:06 +0000 (12:52 +0000)]
Merge "Move specific build rule of 529 into the default script."
Nicolas Geoffray [Wed, 16 Mar 2016 12:39:17 +0000 (12:39 +0000)]
Move specific build rule of 529 into the default script.
Change-Id: I90ff4b4c64a4e077b360bbec7a8e21b52139f395
Nicolas Geoffray [Wed, 16 Mar 2016 11:19:18 +0000 (11:19 +0000)]
Merge "Collect memory use for the JIT."
Nicolas Geoffray [Tue, 8 Mar 2016 16:57:48 +0000 (16:57 +0000)]
Collect memory use for the JIT.
bug:
27520994
Change-Id: I67b0c5b822001bfde8738a988c1ade69f1a26e3f
Nicolas Geoffray [Wed, 16 Mar 2016 11:15:43 +0000 (11:15 +0000)]
Merge "Accept boolean as an input of HDivZeroCheck."
Nicolas Geoffray [Wed, 16 Mar 2016 11:03:54 +0000 (11:03 +0000)]
Accept boolean as an input of HDivZeroCheck.
All our arithmetic operations accept it.
bug:
27624718
Change-Id: I1f6bb95dc77ecb3fb2fcabb35a93b31c524bfa0a
Nicolas Geoffray [Wed, 16 Mar 2016 10:54:51 +0000 (10:54 +0000)]
Merge "Fix checker test."
Nicolas Geoffray [Wed, 16 Mar 2016 10:41:53 +0000 (10:41 +0000)]
Fix checker test.
Change-Id: I1388868ca69c9f964f2019ff79d5c7aa0f00d57c
Nicolas Geoffray [Wed, 16 Mar 2016 10:23:43 +0000 (10:23 +0000)]
Merge "Revert "Assembly TLAB allocation fast path for arm64.""
Nicolas Geoffray [Wed, 16 Mar 2016 10:23:03 +0000 (10:23 +0000)]
Revert "Assembly TLAB allocation fast path for arm64."
Assembly code does not compile
Bug:
9986565
This reverts commit
b7e52b02a82c69e59c88f06945fb21672dfe9923.
Change-Id: I80de1bcd6270e2d76f79cb708811a41c039ea5bb
Nicolas Geoffray [Wed, 16 Mar 2016 10:23:20 +0000 (10:23 +0000)]
Merge "Revert "Tentative fix for build breakage with read barriers.""
Nicolas Geoffray [Wed, 16 Mar 2016 10:22:09 +0000 (10:22 +0000)]
Revert "Tentative fix for build breakage with read barriers."
This reverts commit
448f21e2c1cb43b193736598719979ebd2aed0b0.
Change-Id: Id821effa06697b1464581bcd9c8f34cd0ea69896
Vladimir Marko [Wed, 16 Mar 2016 10:13:48 +0000 (10:13 +0000)]
Merge "Optimizing: Fix arm64 shifter operand test."
Roland Levillain [Wed, 16 Mar 2016 10:12:39 +0000 (10:12 +0000)]
Merge "Make art::HCompare side effect free."
Vladimir Marko [Wed, 16 Mar 2016 10:02:39 +0000 (10:02 +0000)]
Optimizing: Fix arm64 shifter operand test.
Generic simplifier removes 3 shifts from the test since
https://android-review.googlesource.com/208199
and then the arm64-specific shifter operands don't apply.
Bug:
27638111
Change-Id: Ie6c6ec1f40ec263f3e40e8129f29e0191fd9de50
Vladimir Marko [Wed, 16 Mar 2016 09:56:14 +0000 (09:56 +0000)]
Merge "ART: Fix ignoring duplicate fields."
Nicolas Geoffray [Wed, 16 Mar 2016 09:37:52 +0000 (09:37 +0000)]
Merge "Tentative fix for build breakage with read barriers."
Nicolas Geoffray [Wed, 16 Mar 2016 09:36:35 +0000 (09:36 +0000)]
Tentative fix for build breakage with read barriers.
Change-Id: I490821357525bdeb19c330f350635bd645f8ab53
Nicolas Geoffray [Wed, 16 Mar 2016 09:19:39 +0000 (09:19 +0000)]
Merge "Fix HDoubleConstant::IsZero and HFloatConstant::IsZero."
Nicolas Geoffray [Tue, 15 Mar 2016 23:02:48 +0000 (23:02 +0000)]
Merge "Disable test."
Nicolas Geoffray [Tue, 15 Mar 2016 23:01:33 +0000 (23:01 +0000)]
Disable test.
Change-Id: Id06de23646187967df3035ca423f4cd6c73cf0a7
Vladimir Marko [Tue, 15 Mar 2016 20:31:50 +0000 (20:31 +0000)]
ART: Fix ignoring duplicate fields.
Ignore duplicate fields when calculating class size without
embedded tables. We already ignored them when loading class
members. This fixes a check failure in debug build that the
final class is no smaller than the temporary class.
Bug:
21868015
Bug:
27626707
Change-Id: Id72121724abc093d536c2329c61a24854a20122c
Alex Light [Tue, 15 Mar 2016 18:35:18 +0000 (11:35 -0700)]
Fix issue with proxy invocation on default methods
Bug:
27621360
Change-Id: I617eee243f475872235fd75b6a401e056111ea23
Hiroshi Yamauchi [Tue, 15 Mar 2016 19:56:24 +0000 (19:56 +0000)]
Merge "Assembly TLAB allocation fast path for arm64."
Hiroshi Yamauchi [Thu, 10 Mar 2016 05:14:41 +0000 (21:14 -0800)]
Assembly TLAB allocation fast path for arm64.
This is the arm64 version of CL 187537.
Speedup (GSS GC with TLAB on N9):
BinaryTrees: 591 -> 493 ms (-17%)
MemAllocTest: 792 -> 755 ms (-5%)
Bug:
9986565
Change-Id: Icdad28cab0fd835679c640b7eae59b33ac2d6654
Vladimir Marko [Tue, 15 Mar 2016 18:23:01 +0000 (18:23 +0000)]
Merge "ART: Fix shift simplification, x >>> 64."
Vladimir Marko [Tue, 15 Mar 2016 18:11:50 +0000 (18:11 +0000)]
ART: Fix shift simplification, x >>> 64.
Fix braino in
https://android-review.googlesource.com/208199
Bug:
27638111
Change-Id: I8f12008af8bba943664c8a9eac3f2d2f7c820e79
David Srbecky [Tue, 15 Mar 2016 17:02:41 +0000 (17:02 +0000)]
Call __jit_debug_register_code() indirectly via global variable.
This is speculative future optimization. It allows the debugger to
change the value of the global variable to point to its injected code.
This, in turn, allows it to handle JIT events in-process without
triggering breakpoint and without the inter-process cost of it.
Change-Id: I4961c8862603bfb43c84974a3a51110ee202bb1b
Vladimir Marko [Tue, 15 Mar 2016 17:03:18 +0000 (17:03 +0000)]
Merge "ART: Fix valgrind failure in image_test64."
Nicolas Geoffray [Tue, 15 Mar 2016 16:23:04 +0000 (16:23 +0000)]
Fix HDoubleConstant::IsZero and HFloatConstant::IsZero.
bug:
27639313
Change-Id: I2f30a65a07662dfce0a6d6f4ed356a8a0b3dcdef
Vladimir Marko [Tue, 15 Mar 2016 16:06:03 +0000 (16:06 +0000)]
Merge "Optimizing: Improve shift simplification, x >>> 64."
Narayan Kamath [Tue, 15 Mar 2016 15:39:10 +0000 (15:39 +0000)]
Merge "runtime: delete native/java_lang_Runtime.cc"
Vladimir Marko [Tue, 15 Mar 2016 14:57:32 +0000 (14:57 +0000)]
Optimizing: Improve shift simplification, x >>> 64.
Simplify shifts by a multiple of bit size, not just 0.
ARM codegen does not expect to see such shifts and it
is guarding against them with a DCHECK().
Bug:
27638111
Change-Id: I3ae8383d7edefa0facd375ce511e7a226d5468a1
Narayan Kamath [Tue, 15 Mar 2016 14:47:29 +0000 (14:47 +0000)]
runtime: delete native/java_lang_Runtime.cc
We're now using Runtime.c from ojluni, which does the same set
of things via the JVM interface. This removes unnecessary duplication
of code between Runtime.cc and OpenJdkJvm.cc.
Also includes a few changes to JVM_NativeLoad to bring it in sync
with the version in Runtime.cc.
bug:
27387202
Change-Id: I92aac81d131661aca588e352aee7a167593e17d4
Andreas Gampe [Sat, 12 Mar 2016 01:44:04 +0000 (17:44 -0800)]
ART: Remember whether the cache was pruned
Remember (and expose) whether the cache was pruned when initializing
the runtime.
Bug:
27618311
(cherry picked from commit
7ef1b4b20bd7cf485f5e443927c1e6fa797a0439)
Change-Id: I178d18d731a9dce5cb11ed0874c9c9c9d36d7d8e
Roland Levillain [Tue, 15 Mar 2016 14:57:31 +0000 (14:57 +0000)]
Make art::HCompare side effect free.
All our back ends implement all comparisons without making a
runtime call, so we can mark art::HCompare as a side effect
free instruction unconditionally.
Change-Id: I9a9e7c09156c642edb6af1fe84408f887e762f2e
Vladimir Marko [Tue, 15 Mar 2016 12:41:27 +0000 (12:41 +0000)]
ART: Fix valgrind failure in image_test64.
Explicitly clear LengthPrefixedArray<ArtMethod> padding
in ImageWriter. This also removes a potential source of
non-deterministic data in compiled boot image.
Bug:
26687569
Bug:
27552451
Change-Id: Icf4bf34c9ea5b3159462760fbf556d287841191d
Vladimir Marko [Tue, 15 Mar 2016 13:23:38 +0000 (13:23 +0000)]
Merge "Optimizing: Prevent potential valgrind error."
Vladimir Marko [Tue, 15 Mar 2016 10:40:33 +0000 (10:40 +0000)]
Optimizing: Prevent potential valgrind error.
This CL preemptively extends the workaround from
https://android-review.googlesource.com/208230
to an almost identical bit of code.
Bug:
27651442
Change-Id: I7683d42b46b16f2293916defc6ef1d871dc9af6c
Vladimir Marko [Tue, 15 Mar 2016 09:51:10 +0000 (09:51 +0000)]
Merge "Optimizing: Fix valgrind error in image_test64."
Jeff Hao [Tue, 15 Mar 2016 02:41:14 +0000 (02:41 +0000)]
Merge "Fix zygote handling of when secondary image is out of date."
Aart Bik [Tue, 15 Mar 2016 02:40:15 +0000 (02:40 +0000)]
Merge "Fixed bug in BCE, with regression test."
Jeff Hao [Tue, 15 Mar 2016 01:50:49 +0000 (18:50 -0700)]
Fix zygote handling of when secondary image is out of date.
Simple fix for case when primary image is current, but a secondary image
is out of date. Will want to do a cleaner fix in the future.
Bug:
27340451
(cherry-pick of commit
b8510d2311964489342d8875ef7f5b681ba874b4)
Change-Id: Ia6db863e3d3c72120630bcbae8d399a78dda2710
Mathieu Chartier [Tue, 15 Mar 2016 00:11:27 +0000 (00:11 +0000)]
Merge "Add oatdump support for app images"
Mathieu Chartier [Wed, 9 Mar 2016 00:49:58 +0000 (16:49 -0800)]
Add oatdump support for app images
Example usage on host:
oatdumpd --app-oat=art/plus32.odex --app-image=art/plus32.art
--image=art/oats/system@framework@boot.art --instruction-set=arm
TODO: Add to oatdump test.
Bug:
27408512
Bug:
22858531
(cherry picked from commit
bcb6a72569a1401b36a3ad3b6aa4d13e29966cf0)
Change-Id: I9d1aa7eaa16795e5fbabc6974d245849e16b1d03
Andreas Gampe [Mon, 14 Mar 2016 23:37:15 +0000 (23:37 +0000)]
Merge "ART: Update dalvik-cache for zygote after OTA"
Andreas Gampe [Sat, 21 Nov 2015 03:26:51 +0000 (19:26 -0800)]
ART: Update dalvik-cache for zygote after OTA
Detect OTA artifacts and move them over to the regular dalvik-cache
on boot.
Bug:
25612095
(cherry picked from commit
754479f7aa056602de8282f4f03718b42c20b89b)
Change-Id: I9a1a22641cc7833435098627a5cc6d954521c4ea
Aart Bik [Mon, 14 Mar 2016 21:11:26 +0000 (14:11 -0700)]
Fixed bug in BCE, with regression test.
The fix is twofold:
(1) Ensure that bound checks are never eliminated more than once
to guard against any conceivable situation in which the same
bounds check appears multiple times in array length use list.
(2) Specially reject BoundsCheck(x,x) since that always goes OOB.
BUG=
27628526
Change-Id: I399ec4254323e0cfcd0a68898f403cfab7b35135
Vladimir Marko [Mon, 14 Mar 2016 21:21:29 +0000 (21:21 +0000)]
Optimizing: Fix valgrind error in image_test64.
Bug:
27651442
Change-Id: Id9b80c6015dbc3b82966766ca4ad010be770f116
Mathieu Chartier [Mon, 14 Mar 2016 20:50:00 +0000 (20:50 +0000)]
Merge "Do not include image header in decompressed size"
Mathieu Chartier [Mon, 14 Mar 2016 18:02:59 +0000 (11:02 -0700)]
Do not include image header in decompressed size
Could cause a buffer overflow since we told LZ4 to decompress more
bytes than necessary.
Bug:
27561308
(cherry picked from commit
324eb2c6b049f1677133f0b708730e904c5e73ab)
Change-Id: I524c03b8f0e8a84814dbd8678285184e9d4da9f1
Mathieu Chartier [Mon, 14 Mar 2016 20:14:30 +0000 (20:14 +0000)]
Merge "Fix cases where we miss instrumentation changes"
Mathieu Chartier [Sat, 12 Mar 2016 01:22:23 +0000 (17:22 -0800)]
Fix cases where we miss instrumentation changes
Moved allocation stack push to after we record the allocation since
it can cause thread suspension.
Added handling in entrypoint utils for thread suspension cases.
Keep the AllocRecordObjectMap around since we do not want to delete
it if there are any threads waiting on new_record_condition_. The
condition guards adding stack traces while the GC is running. If we
delete the map and there are still waiters that did not resume, it
caused a CHECK failure. This could happen in cases where one thread
disables allocation tracking while other threads are about to
resume from the condition.
Bug:
27506909
(cherry picked from commit
0f394021d6abfdf9ebea6c7f405ec936a812ea62)
Change-Id: I1dc51e9f000684b4032b57beab59d317ece26f06
Vladimir Marko [Mon, 14 Mar 2016 19:13:14 +0000 (19:13 +0000)]
Merge "Optimizing: Fix TypeConversion(And(x, const)) simplification."
Vladimir Marko [Mon, 14 Mar 2016 18:00:05 +0000 (18:00 +0000)]
Optimizing: Fix TypeConversion(And(x, const)) simplification.
Avoid introducing implicit conversions when simplifying the
expression TypeConversion(And(x, const)). Previously, when
we dropped the And, we could end up with a TypeConversion to
the same type which should be eliminated on subsequent pass
of the block's instructions; however, a subsequent dependent
TypeConversion in the same block would be processed earlier
and we would unexpectedly see its input as the conversion to
the same type, failing a DCHECK().
Bug:
27626509
Change-Id: I5874a9ceafbf635cf3391beea807ede8468ab5c3
Hiroshi Yamauchi [Mon, 14 Mar 2016 18:47:50 +0000 (18:47 +0000)]
Merge "Clean up the valgrind test targets."
Hiroshi Yamauchi [Mon, 14 Mar 2016 18:18:36 +0000 (11:18 -0700)]
Clean up the valgrind test targets.
Allow the 32-bit or the 64-bit valgrind tests to run separately or both.
Bug:
27156726
Change-Id: Ib439eb2f57d3302be152cb5af1a8960a41725f31
Vladimir Marko [Mon, 14 Mar 2016 17:42:19 +0000 (17:42 +0000)]
Merge "Optimizing: Do not re-record standby checks for dynamic BCE."
Nicolas Geoffray [Mon, 14 Mar 2016 15:54:54 +0000 (15:54 +0000)]
Merge "Revert "Enable compilation of secondary dex files""
Nicolas Geoffray [Mon, 14 Mar 2016 15:54:39 +0000 (15:54 +0000)]
Revert "Enable compilation of secondary dex files"
Reverting until we fix the buildbot failures on jdwp and libcore.
Bug:
26719109
This reverts commit
770b39eb564ec16c672ace83515535b055edb2b1.
Change-Id: Ifba366d2ddc9311737b6ea29c007475a40cc803f