OSDN Git Service
Andreas Gampe [Mon, 14 Sep 2015 22:34:38 +0000 (15:34 -0700)]
ART: Decrease dex2oat watchdog timeout
Keep the dex2oat watchdog timeout lower than the package manager
timeout, so that dex2oat kills itself before the system server
watchdog kills the system because of the long installation.
Bug:
23629410
(cherry picked from commit
540138ae55ac1909606a436d7f52e20146c56657)
Change-Id: I5faa23a1715736a1c2ba3f302a6ee41130d3fbae
Brian Carlstrom [Tue, 29 Sep 2015 00:47:53 +0000 (17:47 -0700)]
Revert "Fix another source of undeterministic inlining." DO NOT MERGE
This reverts commit
bd7eef17de5f3d25f411702f9f260a96455b4cf7.
Bug:
24467386
Bug:
20037935
Change-Id: I04d1d0ffc67661370e4c17911d3b2b105d053bc1
Andreas Gampe [Sat, 29 Aug 2015 01:15:27 +0000 (18:15 -0700)]
ART: Weaken size check in hprof
Dumping the heap is a two-phase process. In the first phase, all
objects are visited to fill the header tables. In the second phase,
the tables are written out and the heap is visited again to write
the object records.
Deleting global references doesn't require the mutator lock. So it
is possible to delete a global ref in-between the first and the
second phase. This leads to a smaller set of object records. The
result is still safe, as the object's class etc. is still in the
header tables, and the object won't be deleted, so will still be
dumped by itself.
Bug:
23521263
Change-Id: I019c29b13ceb9f13e362c742662f1546d52c37a0
Andreas Gampe [Fri, 14 Aug 2015 21:07:43 +0000 (14:07 -0700)]
ART: Relax verifier aput checking
When checking on a null array, the cases of aput and aput-wide are
shared between integral and floating point types. Be careful to not
reject a valid program.
Bug:
21867457
Bug:
23201502
(cherry picked from commit
4bf4c78a6e8b7da7cf306e1dd17ff5a55d0c6c98)
Change-Id: I6c54a389c06e40a2dae00995aa16ff08a089e512
Andreas Gampe [Wed, 12 Aug 2015 17:48:12 +0000 (10:48 -0700)]
ART: Change UninitializedThis tracking in the verifier
Only relying on register types is error-prone. For example, we may
inadvertently reject correct code when the constructor terminates
abnormally.
Bug:
20843113
(cherry picked from commit
f10b6e109bfb595b6752d1b59db680694ac1684d)
(cherry picked from commit
af31802e5b74f5b9b8d3aadbaaf48cfde14ff7d1)
Change-Id: I8826cd167780df25a6166740f183d216483fa550
Goran Jakovljevic [Tue, 28 Jul 2015 12:26:12 +0000 (14:26 +0200)]
[MIPS] Avoid using odd numbered fp registers
Don't use odd numbered fp registers for single precision data on
MIPS32r6 (64-bit FPU).
Bug:
23050326
Change-Id: I35cc19df091149773411e2336b01c170929376bc
(cherry picked from commit
fc8156a3df88e259c892d50bf23f7c4f11531844)
Serguei Katkov [Mon, 10 Aug 2015 06:59:02 +0000 (12:59 +0600)]
ART: Fix Quick's DCE+GVN
DCE_GVN does not take into account the following case:
mov a, b
...
mov c, b
when optimization tries to replace a with c it must ensure that
for all uses of a there is no new definition of c before use.
Otherwise that use will incorrectly substituted with new c instead
of original b.
Bug:
23102860
Signed-off-by: Serguei Katkov <serguei.i.katkov@intel.com>
(cherry picked from commit
2f2f17399f6bdfc5ec94a875152c31ef79620520)
Change-Id: I1f08c99cedbe4fd1b96cad11f17d60ab551c7cf7
Andreas Gampe [Tue, 11 Aug 2015 00:15:32 +0000 (17:15 -0700)]
ART: Change UnresolvedMergedType merge
Change to use IsUnresolvedTypes.
Bug:
22881413
Change-Id: Ic27b97b475b28a5bdf994ea5178767e0efb0e752
Andreas Gampe [Fri, 7 Aug 2015 15:29:13 +0000 (08:29 -0700)]
Nicolas Geoffray [Sat, 8 Aug 2015 01:13:03 +0000 (18:13 -0700)]
Fix a bug in the register allocator around pair allocation.
We may get hints that do not work with the current implementation
of register pairs, which forces the allocation of (low + 1)
for the high register. For example, if the hint is EBX, we will
allocate ESP for the high register.
bug:
23043730
(cherry picked from commit
f29758111e71a7d14f3e52d78773561a5d59961f)
Change-Id: Ib395e36616017a87d3055218d72417f4e9ff6501
Roland Levillain [Fri, 7 Aug 2015 11:22:28 +0000 (12:22 +0100)]
Tighten default inlining settings when using the space filter.
(cherry picked from commit
a215b95d03cfe713018a245553b74d7eeee813df
and amended)
Bug:
21868508
Change-Id: Ic83813a966cef18e59447083926bf033aa587154
Serguei Katkov [Wed, 5 Aug 2015 11:03:30 +0000 (17:03 +0600)]
ART: Fix the simplifier for add/sub
Instruction simplifier for add/sub should not proceed with floats
because that might cause the incorrect behavior with signed zero.
Bug:
23001681
Signed-off-by: Serguei Katkov <serguei.i.katkov@intel.com>
(cherry picked from commit
115b53f609e74672fa93eea1845bb17340d5112a)
Change-Id: I9928724c4158b3961e32e376b9203fe01ba2e442
Pavel Vyssotski [Fri, 31 Jul 2015 07:03:17 +0000 (13:03 +0600)]
ART: x86_64 RoundDouble/Float intrinsics should initialize out value.
x86_64 RoundDouble intrinsic should initialize output register for the case of
"inPlusPointFive >= maxLong" as expected. The same for the RoundFloat intrinsic.
Fixed also the out register type in CreateSSE41FPToIntLocations provoking
a DCHECK failure.
Signed-off-by: Pavel Vyssotski <pavel.n.vyssotski@intel.com>
(cherry picked from commit
9ca257196b46fd7629bce0b338580e571e4113a8)
Bug:
22973442
Change-Id: If974e79d33311587d0b541a01ca8a4c9c11b9468
Mathieu Chartier [Thu, 30 Jul 2015 18:41:04 +0000 (11:41 -0700)]
Don't dump stack when deleting an invalid local ref
Bug:
21724724
Change-Id: Ifa99df9ee9f3265038a0c0f7944e834e42c498d4
Calin Juravle [Wed, 29 Jul 2015 14:58:48 +0000 (15:58 +0100)]
Allow for fine tuning the inliner.
Bug:
21868508
(cherry picked and squashed from commits
ec74835a7e4f2660250a2f3f9508cbbe5269e49a and
0941b9d48a9a8c6d80a1af7a0d0fc9f80fe2b9a1)
Change-Id: I1750e6bea20321d04680132281a6c2924531c5ae
Mingyao Yang [Wed, 22 Jul 2015 22:56:34 +0000 (15:56 -0700)]
HDeoptimize should hold values live in env.
Values that are not live in compiled code anymore may still be needed in
interpreter, due to code motion, etc.
(cherry-picked from commit
718493c6c3c8e380663cb8a94e57ce160a6c473f)
Bug:
22665511
Change-Id: I8b85833c5c462f8fe36f86d6026a51b07563995a
Fredrik Roubert [Mon, 27 Jul 2015 22:14:08 +0000 (00:14 +0200)]
Bugfix: Java char is 16 bits, can not be treated as boolean.
Using SetFieldBooleanVolatile() and SetFieldBoolean() happens to work
for char values that only use the lower 8 bits, but is a mistake that
was introduced by the "Add AccessibleObject and Field to mirror" commit:
https://android.googlesource.com/platform/art/+/
daaf326
(cherry picked from commit
3152c82b0d33e5fb0a4aa964ea58451c72734444)
Bug:
22772717
Change-Id: Iec02ba3084c992ea239ecef688d7d29c7e21ae16
Andreas Gampe [Fri, 24 Jul 2015 21:24:38 +0000 (21:24 +0000)]
Merge "ART: Change the stack dump format to be in line with debuggerd" into mnc-dev
Andreas Gampe [Thu, 23 Jul 2015 21:37:18 +0000 (14:37 -0700)]
ART: Change the stack dump format to be in line with debuggerd
Make offsets 16 digits on 64-bit platforms.
Add the ability to provide the build fingerprint, and print it in
the traces output.
Bug:
22693991
Change-Id: Ibd3d3b3bd65dce84acfb97a487807d6f280a9508
Vladimir Marko [Wed, 22 Jul 2015 16:50:37 +0000 (17:50 +0100)]
ART: Fix Quick/Optimizing suspend check assumption mismatch.
Quick's SuspendCheckElimination (SCE) expects that every
method contains a suspend check and it eliminates suspend
checks in loops containing an invoke. Optimizing eliminates
the suspend check from leaf methods, so the combination of
a Quick-compiled loop calling an Optimizing-compiled leaf
method can lead to missing suspend checks and potentially
leading to ANRs.
Enable Quick's kLeafOptimization flag to remove suspend
checks from leaf methods and disable Quick's SCE. This
aligns the suspend check placement for the two backends
and avoids the broken combination.
Currently, all methods containing a try-catch are compiled
with Quick, so it's relatively easy to create a regression
test. However, this test will not be valid when Optimizing
starts supporting try-catch.
Bug:
22657404
(cherry picked from commit
d29e8487ff1774b6eb5f0e18d854415c1ee8f6b0)
Change-Id: I733c38bf68bfc2c618f2f2e6b59f8b0e015d7be1
Mathieu Chartier [Tue, 21 Jul 2015 20:34:45 +0000 (20:34 +0000)]
Merge "Don't check code pointer for proxies in ArtMethod::GetQuickFrameInfo" into mnc-dev
Mathieu Chartier [Sat, 18 Jul 2015 02:53:47 +0000 (19:53 -0700)]
Don't check code pointer for proxies in ArtMethod::GetQuickFrameInfo
(cherry picked from commit
7c0fe5e99d1d150b3072a3ae4cd0f0362fe32834)
Bug:
22541058
Change-Id: Id084a69ccd9835147ec71df6bb759e0f1b5e6fbf
Mathieu Chartier [Sat, 18 Jul 2015 01:46:32 +0000 (18:46 -0700)]
Get non proxy method for StackDumpVisitor
Fixes failing DCHECK.
(cherry picked from commit
ef76c3fa57ddf3f7ca32e118f599da8cb4eb1560)
Bug:
22553459
Change-Id: I6fe1012f9a5f8d29cb98bf25e3a279da51c5e1c0
Nicolas Geoffray [Fri, 17 Jul 2015 10:11:09 +0000 (11:11 +0100)]
Reduce space filter threshold to 128.
In order to get closer to our code increase limit.
Bug:
21868508
(cherry picked from commit
432bf3d525dd0736c91238d69e29b4db6a2f7b80)
Change-Id: Iab19fe99a8a869b0fb841bac317016e4a3dc00e2
Nicolas Geoffray [Fri, 17 Jul 2015 08:54:20 +0000 (08:54 +0000)]
Merge "ART: Boolean simplifier fix" into mnc-dev
Serguei Katkov [Wed, 15 Jul 2015 09:32:09 +0000 (15:32 +0600)]
ART: Boolean simplifier fix
Boolean simplifier should not remove condition instruction
if it has side effects.
(cherry picked from commit
108ceb48edea654c544b483c302212e43d79a472)
Change-Id: I00dd595da1bba0f027f5c8647035e9fd8ba62ee1
Serguei Katkov [Tue, 14 Jul 2015 13:04:48 +0000 (19:04 +0600)]
ART: DCE should know that array-length can throw NPE
array-length can throw NPE so it should be taking into account.
Bug:
22521944
Signed-off-by: Serguei Katkov <serguei.i.katkov@intel.com>
(cherry picked from commit
b016c6dd3c30b04104a0a43dc294ce93e5f63874)
Change-Id: I6232430a02f9e6d156ad2aae0eb5a630118c0e79
Christopher Ferris [Mon, 13 Jul 2015 20:19:56 +0000 (20:19 +0000)]
Merge "Fix cfi information for x86." into mnc-dev
Christopher Ferris [Mon, 13 Jul 2015 19:21:00 +0000 (12:21 -0700)]
Fix cfi information for x86.
With this change, I can now enable debug malloc fully.
Bug:
19071092
Change-Id: Iabfb18505feffb480504a8bbfbac60dd219ed48a
Roland Levillain [Mon, 13 Jul 2015 12:33:04 +0000 (13:33 +0100)]
Re-enable run-test 449-checker-bce on MIPS64.
(cherry picked from commit
c3f73f788a2f46e4007433bd3ed3dce64e2f5f33)
Bug:
21555893
Change-Id: I92e11637cc1f8f694e4d4138512cbcf47e22a249
Roland Levillain [Mon, 13 Jul 2015 12:32:04 +0000 (12:32 +0000)]
Merge "Initialize pDeoptimize quick entry pointer for Mips." into mnc-dev
Douglas Leung [Mon, 13 Jul 2015 09:55:45 +0000 (10:55 +0100)]
Initialize pDeoptimize quick entry pointer for Mips.
This fixes art test 449-checker-bce.
(cherry picked from commit
2e8bf554574516e0e6a9b10ccd9f13bf56189795)
Bug:
21555893
Change-Id: I9a87c430168ea35984289b0b4a417aa0c659b6d5
Nicolas Geoffray [Mon, 13 Jul 2015 09:27:05 +0000 (09:27 +0000)]
Merge "Return an invalid StackMap when none can be found." into mnc-dev
Nicolas Geoffray [Fri, 10 Jul 2015 09:56:40 +0000 (10:56 +0100)]
Return an invalid StackMap when none can be found.
bug:
22389275
Partial cherry-pick of:
https://android-review.googlesource.com/#/c/151853
(commit
e12997fbce8e22431be58cac9db2535f7b4a7ac3)
Change-Id: Ia30b817be1b50d97243ba32967eeee359ed679c4
Mathieu Chartier [Sat, 11 Jul 2015 01:26:41 +0000 (18:26 -0700)]
Fix proxy handling in FindDeclaredVirtualMethod
Added missing GetInterfaceMethodIfProxy and test.
Fixed formatting.
Bug:
22411819
https://code.google.com/p/android-developer-preview/issues/detail?id=2635
(cherry picked from commit
72156e28fd6bc72ac965b29446f8801b2e82f2fd)
Change-Id: I3eece9c72091bb9d0262aacf0a75ec6908b5f4d2
Andreas Gampe [Thu, 9 Jul 2015 18:30:14 +0000 (11:30 -0700)]
ART: Change merges with Undefined to Undefined
The result of a merge with an Undefined type should be Undefined.
Conflicts are allowed to be copied around, but Undefined registers
should not be touched at all, except to be written into.
Add a success test case (the register isn't used) and a fail test
case (the register is tried to be copied).
Bug:
22331663
Bug:
22371999
(cherry picked from commit
97a1ff353f254b6e46c7501fe3f0e3254c2517b4)
Change-Id: I9697ce31c1d2ab5aee0433dcf1253bcca79c2983
Andreas Gampe [Thu, 9 Jul 2015 18:01:07 +0000 (11:01 -0700)]
ART: Fix test
Fixes a breakage from
38536287f61c9c0fc3bab8c1950cf8c74881482a. Ensure
that v4 is not undefined when an exception may be thrown (at the
new-instance).
Bug:
22331663
Bug:
22371999
(cherry picked from commit
7135ff0fdfd02796d40e7ef3c7527ff127512bd3)
Change-Id: Ice36ddda5ca92ea32fa88523dd616d4f2a61027e
Andreas Gampe [Thu, 9 Jul 2015 21:44:01 +0000 (21:44 +0000)]
Merge "ART: Improve the symbolize script" into mnc-dev
Andreas Gampe [Thu, 9 Jul 2015 21:39:56 +0000 (14:39 -0700)]
ART: Improve the symbolize script
Allow a list of files to pull. This improves turnaround time when
tracking down perf hotspots.
Bug:
21760614
Change-Id: I824e875053940d59eb00f9d796ef402632905c39
Sebastien Hertz [Tue, 24 Feb 2015 16:39:50 +0000 (17:39 +0100)]
Support compiling run-tests with jack
This CL adds support to compile run-test source files with jack. When
a test needs to rely on class files, we use jill to convert them to a
jack library.
We need to pass the full classpath to jack containing at least core
classes (like java.lang.Object). This means the Android tree must
have been compiled with jack first so we find all the necessary
classes.jack files.
Some tests still rely on dex files generated with the old toolchain.
We keep building them this way for the moment and will update them
later, when they get ready for Jack.
Also updates a few tests dealing with garbage collection to avoid a
situation where a reference can be retained by a local DEX register.
Bug:
19467889
(cherry picked from commit
19ac0276208f0afef6ba8a4ab34b74a59b8d11d7)
Change-Id: Ia5a989b83430ffe8298a869a1da970b756721bb0
Andreas Gampe [Thu, 9 Jul 2015 00:22:57 +0000 (17:22 -0700)]
ART: Allow to set and copy conflicts in the verifier
As long as conflicts are not actively being used, it is OK to set
them in a register line or copy them around.
Bug:
22331663
(cherry picked from commit
38536287f61c9c0fc3bab8c1950cf8c74881482a)
Change-Id: I61999e2d9c92f9bdedcb0a5dea54df3c5bb130ca
Andreas Gampe [Tue, 7 Jul 2015 00:20:18 +0000 (17:20 -0700)]
ART: Avoid soft- after hard-fail in verifier
Follow-up to
414000ec4d728b5c85f8c6dee4f867fecde59b01.
Bug:
21886894
Bug:
22245124
(cherry picked from commit
890da29d279c44923f9271450b8a4423b1267655)
Change-Id: I3728187bd8526cdf84f95b123e6fad5d6c5a8b3c
Andreas Gampe [Wed, 8 Jul 2015 20:12:25 +0000 (20:12 +0000)]
Merge "ART: Release inputs in Long.reverse intrinsic in x86" into mnc-dev
Andreas Gampe [Tue, 7 Jul 2015 20:25:58 +0000 (13:25 -0700)]
ART: Release inputs in Long.reverse intrinsic in x86
In the worst case we are using two temps each for input and output.
Then we do not have a temp left over for the swap operations. The
input is dead, however, after the first swap. So try to release it
(a no-op if it isn't actually a temp).
Bug:
22324327
(cherry picked from commit
575422fa5be7389bdaff5e2d25dd87b1d2d4de85)
Change-Id: I1fc50159afdad14160e34abeaf4670958171d6b2
Andreas Gampe [Mon, 6 Jul 2015 21:00:39 +0000 (14:00 -0700)]
ART: Fix opsize in LoadArgDirect
If the destination register is a reference, use kReference for the
op size.
Bug:
22244733
(cherry picked from commit
185a5586c8b796e770e9b4b7ac2befa8ccdaca7e)
Change-Id: Idf52f2ee4c65b5dc41cb66257d95281dc6f32255
Douglas Leung [Thu, 2 Jul 2015 23:42:08 +0000 (16:42 -0700)]
Add implicit null pointer and stack overflow checks for Mips.
(cherry picked from commit
22bb5a2ebc1e2724179faf4660b2735dcb185f21)
Bug:
21555893
Change-Id: I2a995be128a5603d08753c14956dd8c8240ac63c
Nicolas Geoffray [Thu, 2 Jul 2015 14:48:27 +0000 (15:48 +0100)]
Do not create a HBoundType when the instruction is non-null.
We don't need to refine the type after a null check, if the
instruction is known non null or null. As a side effect, this
avoids replacing HLoadClass instructions with HBoundType instructions.
bug:
22116987
Change-Id: I565ae95db5a64faec30e026674636e398e0bf445
Andreas Gampe [Wed, 1 Jul 2015 17:50:21 +0000 (17:50 +0000)]
Merge "ART: Symbolize all oat files in /data folder." into mnc-dev
Andreas Gampe [Wed, 1 Jul 2015 17:50:07 +0000 (17:50 +0000)]
Merge "ART: Fix data loss when symbolizing oat files." into mnc-dev
Zheng Xu [Tue, 21 Apr 2015 09:35:25 +0000 (17:35 +0800)]
ART: Symbolize all oat files in /data folder.
Now we store oat files not only in /data/dalvik-cache, but also in
other places. This patch tries to pull all oat files from device
/data folder.
Bug:
21760614
(cherry picked from commit
468bcf63207420f18c0c8a8621aa2d774393c155)
Change-Id: Icf81cf28c29da2e248e4fbd84f5920f46ddc4cd6
Zheng Xu [Fri, 19 Jun 2015 06:12:45 +0000 (14:12 +0800)]
ART: Fix data loss when symbolizing oat files.
Bug:
21760614
(cherry picked from commit
98088c424efd45ea3c172d4bc16e191f26a007cc)
Change-Id: I1b40b8ee313a4d187439d969d083d12fb3997cda
Nicolas Geoffray [Wed, 1 Jul 2015 11:36:37 +0000 (11:36 +0000)]
Merge "Do not replace a live phi with a dead phi." into mnc-dev
Nicolas Geoffray [Wed, 1 Jul 2015 08:54:33 +0000 (08:54 +0000)]
Merge "Revert "Revert "Do not update the type of something we already know.""" into mnc-dev
Andreas Gampe [Tue, 30 Jun 2015 18:23:44 +0000 (11:23 -0700)]
ART: Remove bad FindSymbol call
The removed call is superfluous under normal operation and harmful
when running a native-bridged library.
Bug:
22194296
Bug: https://code.google.com/p/android-developer-preview/issues/detail?id=2367
(cherry picked from commit
8fec90b421732b3eb278555f57aaf48fda4fa4ba)
Change-Id: I3eb8f435a6a974a3e87a370b254ba8752f2a59b3
Andreas Gampe [Tue, 30 Jun 2015 16:01:09 +0000 (16:01 +0000)]
Merge "ART: Unlink target oat file before compiling" into mnc-dev
Andreas Gampe [Sat, 27 Jun 2015 01:58:42 +0000 (18:58 -0700)]
ART: Unlink target oat file before compiling
Dex2oat must create a new file (new inode) when writing to a target
given by name, as the existing file may be in use. So unlink any
existing file first.
Bug:
22047255
Change-Id: Ief942c71564076e39c1e8340d5c4c55286f75896
Nicolas Geoffray [Mon, 29 Jun 2015 13:34:46 +0000 (14:34 +0100)]
Do not replace a live phi with a dead phi.
A dead phi is not properly typed. Therefore, always use the live phi
equivalent instead.
bug:
21865466
(cherry picked from commit
4230e1895b915a22363452823b0e51eabe92cb60)
Change-Id: If9a3313fc953c2eb4be85b625676e48112fcb1f6
Nicolas Geoffray [Mon, 29 Jun 2015 12:43:54 +0000 (13:43 +0100)]
Revert "Revert "Do not update the type of something we already know.""
This reverts commit
63107a804ce17db9789051e1fe310d99d1dae1cb.
bug:
22116987
(cherry picked from commit
f9a199571417b5a5a62d94d05a064077e14dd2c4)
Change-Id: Ia516f2cbce6d22df37f3a0854abdd0b54d3ea72d
Mathieu Chartier [Sat, 27 Jun 2015 22:42:27 +0000 (15:42 -0700)]
ART: Fix CFI annotation for art_quick_aput_obj
Fix the CFI state after an early return.
Bug:
22014525
(cherry picked from commit
2738639bcd30b908d825725169b7497ed047debb)
Change-Id: I56b9ba8cf8c47d70a642f064e59c7e04a476dd2f
Roland Levillain [Mon, 29 Jun 2015 08:18:30 +0000 (08:18 +0000)]
Merge "MIPS: Initial version of optimizing compiler for MIPS64R6." into mnc-dev
Vladimir Marko [Sat, 27 Jun 2015 12:32:11 +0000 (13:32 +0100)]
Quick: Disable GVN, DCE and LVN for type conflicts.
(cherry picked from commit
f7e7327d7a873d21aaae01471f46a4d64d65433b)
Bug:
22136903
Change-Id: I657d4b5e623696e1264c2c6c9ad4e7fd466d759c
Andreas Gampe [Sat, 27 Jun 2015 00:33:47 +0000 (17:33 -0700)]
ART: Fix invalid access and DCHECK in verifier
If we get a throwing failure when setting types from the signature,
the work instruction index is still invalid. Do not try to copy the
line then.
As a throwing failure might happen in the above instance, but the
flow analysis expects to have a cleared failure flag before processing
each instruction, clear the flag.
Bug:
21645819
Bug:
22080519
Change-Id: I224c4dad98fa5bb50e62210f0ee30c0dd020e3a6
Mathieu Chartier [Fri, 26 Jun 2015 20:57:03 +0000 (20:57 +0000)]
Merge "Require mutator lock for DeleteLocalRef" into mnc-dev
Mathieu Chartier [Fri, 26 Jun 2015 17:47:08 +0000 (10:47 -0700)]
Require mutator lock for DeleteLocalRef
There was a race condition where suspended threads could call
DeleteLocalRef while the GC was marking their roots. This could
cause the GC to attempt to mark a null object.
(cherry picked from commit
dd06afecd67fc6a60856436ab150a4d1b9a8ad29)
Bug:
22119403
Change-Id: I962c717bb87b2acb2a4710a2d7ab16793e031401
Andreas Gampe [Tue, 23 Jun 2015 05:53:45 +0000 (22:53 -0700)]
ART: Fix streaming tracing issues
Fix a lock ordering issue in streaming-mode tracing.
Fix a moving-GC issue in streaming-mode tracing. DexCache
objects are not good keys for a map.
Expose streaming mode for testing in run-tests.
Bug:
21760614
Change-Id: Idcd0575684ee3cc0cec3f81b4fdd0d5988c11e8c
David Brazdil [Fri, 26 Jun 2015 15:58:14 +0000 (16:58 +0100)]
ART: Bail out immediately on try/catch
Optimizing builds the graph even if it knows it will delegate to
Quick. This patch moves the decision at the beginning of the
TryCompile function to save a little bit of compile time.
Bug:
22115561
Change-Id: I18b5a686e10592080ea3c49d53a0dd81aa3b3010
Roland Levillain [Fri, 26 Jun 2015 15:12:18 +0000 (16:12 +0100)]
MIPS: Initial version of optimizing compiler for MIPS64R6.
(cherry picked from commit
4dda3376b71209fae07f5c3c8ac3eb4b54207aa8)
(amended for mnc-dev)
Bug:
21555893
Change-Id: I874dc356eee6ab061a32f8f3df5f8ac3a4ab7dcf
Signed-off-by: Alexey Frunze <Alexey.Frunze@imgtec.com>
Signed-off-by: Douglas Leung <douglas.leung@imgtec.com>
Andreas Gampe [Fri, 26 Jun 2015 03:16:10 +0000 (03:16 +0000)]
Merge "ART: Fix kEverything compiler filter" into mnc-dev
Andreas Gampe [Tue, 19 May 2015 06:22:54 +0000 (23:22 -0700)]
ART: Only print stripped dex2oat command line
To curb logcat noise, strip many dex2oat parameters from the
logcat printout.
Bug:
20501758
(cherry picked from commit
046c706707ea0f16c804136e237ac7cbfdc897a1)
Change-Id: Ifc367f91f593916e0773af1ca950c798f129889f
buzbee [Thu, 25 Jun 2015 22:53:45 +0000 (15:53 -0700)]
ART: Fix kEverything compiler filter
Previously the kEverything filter failed to compile class
initializers. Now it will.
Bug:
19576257
(cherry picked from commit
c83329952b4a313e747c8835a73699e2cae5a6e2)
Change-Id: I189d2b5b379aee112c4e95f8d3e6c32abab6ed41
Andreas Gampe [Thu, 25 Jun 2015 17:26:40 +0000 (10:26 -0700)]
ART: Reset runtime_throw_failure flag
The flag is instruction-specific. It transports the info from Fail
to the main loop. It must be cleared after each iteration.
Introduce a second flag to store whether we saw such a failure at
all.
Update test expectations.
Bug:
22080519
(cherry picked from commit
d12e782bcee03ecb6dec41aa9673ef53b638dcea)
Change-Id: I32be914819946233babaa4cb7343844d97b61ba5
Andreas Gampe [Wed, 24 Jun 2015 23:25:33 +0000 (16:25 -0700)]
ART: Set methods to preverified in verify-none
Actually mark all methods to be verified, so that the interpreter
doesn't do access checks. Speedup is about 2x on a Nexus 6 when booting
in verify-none mode.
Bug:
22040916
(cherry picked from commit
1c284f2abdfe1b0ed87d5216021e69146f10b116)
Change-Id: I2d6c8df2a68eae6e358da2639d980cb8bfcc1560
Richard Uhler [Thu, 25 Jun 2015 15:21:12 +0000 (15:21 +0000)]
Merge "Do not try to compile resource-only dex files." into mnc-dev
Vladimir Marko [Tue, 23 Jun 2015 16:45:21 +0000 (17:45 +0100)]
Hard-fail get-/put-object to a non-reference field.
Bug:
21886894
(cherry picked from commit
414000ec4d728b5c85f8c6dee4f867fecde59b01)
Change-Id: Iafc32f0e45d26f3aaa2d521b98353e7cede16c6f
Nicolas Geoffray [Thu, 25 Jun 2015 09:01:47 +0000 (10:01 +0100)]
Only do some checks when compiling against the core image.
This will avoid false negatives when running dex2oatd on apks.
bug:
21865473
(cherry picked from commit
335005e2b3a179f26b7a8ae64ca60a1406b669bd)
Change-Id: Iac6dbe30c9d576077a0384b88696f79937d89471
Lazar Trsic [Wed, 24 Jun 2015 14:30:21 +0000 (16:30 +0200)]
[MIPS64] JNI Compiler: Sign-extend int function arguments
MIPS n64 ABI differs from arm64. Arguments smaller than the 8B stack
slot need to be sign-extended.
Use combination (lw,sd), instead of (lw,sw) for 4B values.
Change fixes software keyboard crash on mips64.
Bug:
21555893
(cherry picked from commit
f652d605753f1387e7797461b47116c5dcdf928d)
Change-Id: I7574d37f6039e9e8c9e0047254be71d28d4c829a
Andreas Gampe [Thu, 25 Jun 2015 04:12:06 +0000 (04:12 +0000)]
Merge "ART: Fix arm32 instrumentation exit stub" into mnc-dev
Mathieu Chartier [Thu, 25 Jun 2015 00:04:17 +0000 (17:04 -0700)]
Fix force copy
We now correctly pass the returned pointer back onto the release functions.
Bug:
22056708
Change-Id: I1a7300d3a4522a3c81b432ec742ae1c0bd00b51e
(cherry picked from commit
b735bd9c04aa291d0a1bdc2c0a094a1a75ad0596)
Andreas Gampe [Wed, 24 Jun 2015 06:28:59 +0000 (23:28 -0700)]
ART: Fix arm32 instrumentation exit stub
Properly store, pass and restore floating point result.
Bug:
22047833
Change-Id: If5d9efbe63334c9b6353f76ce6e4fe519602482a
Mathieu Chartier [Wed, 24 Jun 2015 20:25:25 +0000 (20:25 +0000)]
Merge "Fix imt conflict trampoline for mips64" into mnc-dev
Mathieu Chartier [Wed, 24 Jun 2015 20:11:06 +0000 (13:11 -0700)]
Fix imt conflict trampoline for mips64
Changed lwu to ld
Bug:
21815962
Change-Id: I137e871d9c00f83cd5efc9d3a87fa05e84dc5d23
Christopher Ferris [Wed, 24 Jun 2015 16:37:39 +0000 (09:37 -0700)]
Increase alt signal stack to 32K on host.
On host, when there is a crash, the code tries to do an unwind
from the signal handler. Under some conditions this can require
a stack larger than 16K, so increase this value to 32K.
The symptom of this is that you get an empty backtrace running
art tests when there is a crash.
Bug:
21949571
(cherry picked from commit
e34c71ce21dbc9fac3427df887e90e1311b6fa8e)
Change-Id: Ib1105a3470064d98843bd46f32d4fadb1d20e416
Jeff Hao [Wed, 24 Jun 2015 17:05:42 +0000 (17:05 +0000)]
Merge "Adding optimizing compiler test case for String.<init>." into mnc-dev
Nicolas Geoffray [Wed, 24 Jun 2015 14:53:03 +0000 (15:53 +0100)]
Use a flag from the verifier to know if we should compile.
Only used for the lack of bottom type in the aget-object case
for now. Could be used for more.
bug:
21865466
(cherry picked from commit
4824c27988c8eeb302791624bb3ce1d557b0db6c)
Change-Id: I2bb7fe1d4737bd92c1076b5193607d74d8761ee7
Nicolas Geoffray [Wed, 24 Jun 2015 13:57:44 +0000 (14:57 +0100)]
BCE: don't assume a bounds check always gets a HArrayLength.
Deoptimizations may change it to a HPhi.
bug:
22056703
(cherry picked from commit
8df886b9214802ad689316a1dedb00a6d102555c)
Change-Id: I8afcf88e3a12dbe4d87101e6a7cefb8b81e2bb96
Richard Uhler [Wed, 24 Jun 2015 15:44:19 +0000 (08:44 -0700)]
Do not try to compile resource-only dex files.
This changes behavior in the case where we are asked to load a dex
file that does not exist or has no classes.dex entry.
Previously we would run dex2oat, which would log an error message and
fail. Now we skip running dex2oat, we report the DexOptStatus as
kNoDexOptNeeded, and we do not try to fall back to the missing
original dex files.
Bug:
21722039
Change-Id: I90b1f4165138daac57bb9d7a354319005652c593
Nicolas Geoffray [Wed, 24 Jun 2015 13:25:09 +0000 (14:25 +0100)]
Fix codegen_test after DCE fix.
(cherry picked from commit
ea80942c39cb5e9bc3e38a388583d1646c1e2a3f)
bug:
22031382
Change-Id: If410045779adb9e7e611af22c988232b5bf9a902
Nicolas Geoffray [Wed, 24 Jun 2015 11:20:24 +0000 (12:20 +0100)]
Recompute dominator tree after DCE.
bug:
22031382
(cherry picked from commit
1f82ecc6a0c9f88d03d6d1a6d95eeb8707bd06c1)
Change-Id: I9a74edb185cb806045903dfe9695d9cc1a02e86b
Nicolas Geoffray [Wed, 24 Jun 2015 09:38:27 +0000 (10:38 +0100)]
BCE: Narrow instead of unconditionnaly overwrite the range.
bug:
21862741
(cherry picked from commit
a09ff9c11f07863ac57e6120a824f0d20dfaa284)
Change-Id: Ia8e903e09a7f9c2b8ef7cf3522f73f154534b81f
Nicolas Geoffray [Wed, 24 Jun 2015 11:41:20 +0000 (12:41 +0100)]
Fix another case of un-verified dead code.
bug:
22042796
https://code.google.com/p/android/issues/detail?id=178008
(cherry picked from commit
1efcc22cd1895c48adccbe49270d8e8583c2b12d)
Change-Id: I5c0d783e842da39cd3dcbb2f18ccf784e797a64f
Jeff Hao [Wed, 24 Jun 2015 00:34:04 +0000 (17:34 -0700)]
Adding optimizing compiler test case for String.<init>.
Bug:
21991156
Change-Id: Ibd0bd7d3b20680ac06ac08cb857595288c4e9b69
Andreas Gampe [Tue, 23 Jun 2015 23:15:18 +0000 (23:15 +0000)]
Andreas Gampe [Tue, 23 Jun 2015 21:18:17 +0000 (14:18 -0700)]
ART: Follow-up to
507cc6f83bf6379728f2dd20391f2ed5fbfe6371
Do not trigger DCHECK aborts for final and abstract classes when
they are used from other classes. Instead issue a soft error that
leads to a runtime failure.
Bug:
21873151
(cherry picked from commit
f23f33de8b3e9abd16720e0f48d43d63d41dee16)
Change-Id: Ie5251a176722811b3c74585222874313a53f7e02
Mathieu Chartier [Tue, 23 Jun 2015 22:54:48 +0000 (22:54 +0000)]
Merge "Fix field ordering for String in hprof" into mnc-dev
Mathieu Chartier [Tue, 23 Jun 2015 22:45:15 +0000 (15:45 -0700)]
Fix field ordering for String in hprof
We were inserting value field after the Object field instead of
the String fields.
Bug:
22043800
Change-Id: Ibaddbd7299562949e32eceb625feac325539e3ce
Jeff Hao [Tue, 23 Jun 2015 03:20:30 +0000 (20:20 -0700)]
Make dex file verifier check that indexes are increasing.
Note that this still allows duplicates.
Bug:
21868011
Change-Id: Ib0e91c816eccfaf3b0dac340c05ffd5dc4df17cc
Nicolas Geoffray [Mon, 22 Jun 2015 22:12:45 +0000 (23:12 +0100)]
Do not overwrite an input register in shift operations.
'second_reg' is an input register that can survive the instruction.
Instead use the output register as a temporary result.
bug:
21667432
(cherry picked from commit
a4f3581da73b83484a30ab499c4f8ad43b378dab)
Change-Id: Ic1f399964911b8a9fc57352130c92b2a0a1b8e0d
Mathieu Chartier [Tue, 23 Jun 2015 00:14:43 +0000 (00:14 +0000)]
Merge "Fix another miranda method moving GC bug" into mnc-dev
Mathieu Chartier [Sun, 14 Jun 2015 21:14:37 +0000 (14:14 -0700)]
New experimental GC stress mode
Tries to do a GC for every unique call stack (up to 16 frames).
The goal is to catch moving GC bugs and lock violations without being
rediculously slow. Some tests fail on 64 bits, 32 bit host doesn't
work. N5 is booting.
Added runtime -Xgc options: gcstress and nogcstress.
Bug:
21664466
(cherry picked from commit
310008008c90fea246efd00cb99ee7ded97c5209)
Change-Id: Icb8e420f2048e8ee83bcca7937563166a2638f5c
Jeff Hao [Mon, 22 Jun 2015 22:26:54 +0000 (22:26 +0000)]
Merge "Change GC to allow strings to use large object space." into mnc-dev