| Age | Commit message (Collapse) | Author |
|---|
|
Introduce a `CompiledCodeStorage` interface for callbacks to
`dex2oat` for storing code produced by `libart-compiler`.
Using this new interface, move `CompiledMethodStorage`
(this is a misnomer as it's also storing non-method thunk
code), `CompiledMethod` and `SwapSpace` to dex2oat/.
Test: m test-art-host-gtest
Test: testrunner.py --host --optimizing --jit
Test: run-gtests.sh
Test: testrunner.py --target --optimizing
Change-Id: Icb7ffc134d00e53c25bd665d044dd1c98bd49a64
|
|
And clean up some related test helpers in preparation for
moving `CompiledMethod` to dex2oat/.
Test: m test-art-host-gtest
Test: run-gtests.sh
Change-Id: I13b42bfb4f6ee3a7f7f22bbc8d22203c5d56e00b
|
|
Reduce dependencies on `compiled_method.h` in preparation
for moving it to dex2oat/.
Test: m test-art-host-gtest
Test: testrunner.py --host --optimizing
Test: run-gtests.sh
Test: testrunner.py --target --optimizing
Change-Id: I55ff807e0c7d729142b1a64a3a0b6c6267027dc3
|
|
Update the `GetInstructionSetAlignment()` function name to
`GetInstructionSetCodeAlignment()` in line with renames from
https://android-review.googlesource.com/2229498 .
Move the function to `instruction_set.h` and change the
handling of unknown instruction sets to the approach used
in other functions in the header file.
Test: buildbot-build.sh --target
Change-Id: I8239b94b56bf24f4e99a2d72ba10ec1f5a055d36
|
|
Also remove MIPS assembler/disassembler support.
Test: aosp_taimen-userdebug boots.
Test: m test-art-host-gtest
Test: testrunner.py --host --optimizing
Bug: 147346243
Change-Id: Id736074b97cd04987a7902741828b119508df1c0
|
|
Bug: http://b/111067277
Test: builds
Change-Id: I8b69ea3815e14bb6eb27f40c0dd01a85b340a355
|
|
Test: m test-art-host-gtest
Change-Id: Ibee78d5c54d3ff8162258963fde25065b579a000
|
|
There is no need to treat it specially any more,
because of the de-duplication at BitTable level.
This saves 0.6% of oat file size.
Test: test-art-host-gtest
Change-Id: Ife7927d736243879a41d6f325d49ebf6930a63f6
|
|
The information has been moved to CodeInfo,
where it is stored in much more compact way.
The old CL which added the data to CodeInfo cost 0.7%.
This CL saves 2.5% of .oat file size so a win overall.
Test: test-art-host-gtest
Change-Id: I07fcf6f2776c96218f995ba3b57a1e6ccbf5e317
|
|
Dedupe code items if they have the same contents, this is safe
because of the fixed quickening logic.
Removed duplicate code items in ManyMethods that caused the startup
methods section to be empty in dex2oat_test.
To avoid errors related to quickening and compilation of shared code
items, quickening is currently disabled for shared code items.
Bug: 63756964
Test: test-art-host
Change-Id: I7c62eb746785d787c5269effd396f7be4859d3a6
|
|
Fix varius bugs in the quickening logic where related to shared code
items for different methods.
Fixed the case where two methods quicken differently on the same code
item by checking that the quicken info is the same for all methods
that quicken the same code item. This is accomplished by requickening
and reverifying the contents of the quicken info.
Fixed the case where the dex to dex compiler would abort from a
DCHECK that there was no already quickened instructions.
Feature is tested by enabling deduping (aog/594315).
Test: test-art-host
Bug: 63756964
Change-Id: I52c2b89518f4e808594b450a5fcc373ab5a5863b
|
|
When compiling an intrinsic method, generate a graph that
invokes the same method and try to compile it. If the call
is actually intrinsified (or simplified to other HIR) and
yields a leaf method, use the result of this compilation
attempt, otherwise compile the actual code or JNI stub.
Note that CodeGenerator::CreateThrowingSlowPathLocations()
actually marks the locations as kNoCall if the throw is not
in a catch block, thus considering some throwing methods
(for example, String.charAt()) as leaf methods.
We would ideally want to use the intrinsic codegen for all
intrinsics that do not generate a slow-path call to the
default implementation. Relying on the leaf method is
suboptimal as we're missing out on methods that do other
types of calls, for example runtime calls. This shall be
fixed in a subsequent CL.
Test: m test-art-host-gtest
Test: testrunner.py --host --optimizing
Bug: 67717501
Change-Id: I640fda7c22d4ff494b5ff77ebec3b7f5f75af652
|
|
Adding InstructionSet::kLast shall make it easier to encode
the InstructionSet in fewer bits using BitField<>. However,
introducing `kLast` into the `art` namespace is not a good
idea, so we change the InstructionSet to an enum class.
This also uncovered a case of InstructionSet::kNone being
erroneously used instead of vixl32::Condition::None(), so
it's good to remove `kNone` from the `art` namespace.
Test: m test-art-host-gtest
Test: testrunner.py --host --optimizing
Change-Id: I6fa6168dfba4ed6da86d021a69c80224f09997a6
|
|
Move LinkerPatch to compiler/linker/linker_patch.h .
Move SrcMapElem to compiler/debug/src_map_elem.h .
Introduce compiled_method-inl.h to reduce the number
of `#include`s in compiled_method.h .
Test: m test-art-host-gtest
Test: testrunner.py --host
Change-Id: Id211cdf94a63ad265bf4709f1a5e06dffbe30f64
|
|
The method info data is stored separately from the code info to
reduce oat size by improving deduplication of stack maps.
To reduce code size, this moves the invoke info and inline info
method indices to this table.
Oat size for a large app (arm64): 77746816 -> 74023552 (-4.8%)
Average oat size reduction for golem (arm64): 2%
Repurposed unused SrcMapElem deduping to be for MethodInfo.
TODO: Delete SrcMapElem in a follow up CL.
Bug: 36124906
Test: clean-oat-host && test-art-host-run-test
Change-Id: I2241362e728389030b959f42161ce817cf6e2009
|
|
This reduces the size of the pre-header by 8 bytes, reducing
oat file size and mmapped .text section size. The memory
needed to store a CompiledMethod by dex2oat is also reduced,
for 32-bit dex2oat by 8B and for 64-bit dex2oat by 16B. The
aosp_flounder-userdebug 32-bit and 64-bit boot.oat are each
about 1.1MiB smaller.
Disable the broken StubTest.IMT, b/27991555 .
Change-Id: I05fe45c28c8ffb7a0fa8b1117b969786748b1039
|
|
Use LengthPrefixedArray<>s instead of SwapVector<>s to store
CompiledMethod data and get rid of the unnecessary members
of CompiledMethod to reduce dex2oat memory usage. Refactor
the deduplication from CompilerDriver to a new class.
Use HashSet<> instead of std::set<> for the DedupeSet<> to
further decrease the memory usage and improve performance.
This reduces the dex2oat memory usage when compiling boot
image on Nexus 5 (with Optimizing, -j1) by ~6.75MiB (5%).
This also reduces the compile time by ~2.2% (~1.6% dex2oat
time; with Optimizing, without -j).
Change-Id: I974f1f5e58350de2bf487a2bca3907fa05fb80ea
|
|
quickening.""""
This reverts commit ed6195a514e3253576af27ea9ba13038509d29ac.
Change-Id: Icb58854301e8982147cdebe3edf2e0d9e0a63a56
|
|
quickening."""
Breaks again in some configurations I don't fully understand yet.
This reverts commit f075879649686e59b7a9065c5a061dbfdcdfbecc.
Change-Id: I0ac5533825e40b06462ee69b2740e4a96fb5c582
|
|
This reverts commit 327c5ed30a1f016ef3e1bb26ea7b4abd34eb63b9.
Change-Id: I0dc5d92e5d1ef98830fbd3c40ec59a93f9e0422d
|
|
Build failures on arm/arm64.
This reverts commit 6920703c8eae0d90528ea09945e742582b6f8198.
Change-Id: I0dd5426610150937dac6e4d9dd9aa759bdf7fca4
|
|
This is the starting piece of pushing more verification and
dex2dex data into an .oat file, to lower the dependency on verification
results at runtime.
Change-Id: I8e1b49a0207714bc5b84d1f606806718c5c7fb69
|
|
Also fixed some lines that were too long, and a few other minor
details.
Change-Id: I6efba5fb6e03eb5d0a300fddb2a75bf8e2f175cb
|
|
CFI is necessary for stack unwinding in gdb, lldb, and libunwind.
Change-Id: I1a3480e3a4a99f48bf7e6e63c4e83a80cfee40a2
|
|
Change-Id: I0eabd713d29475db9eb6e186f331dbfb00e0cf6b
|
|
Also simplify dex to java mapping and handle mapping
in prologues and epilogues.
Change-Id: I410f06024580f2a8788f2c93fe9bca132805029a
|
|
Added missing EntryPointToCodePointer.
This reverts commit a5ca888d715cd0c6c421313211caa1928be3e399.
Change-Id: Ia74df0ef3a7babbdcb0466fd24da28e304e3f5af
|
|
Sorry, run-test crashes on target:
0-05 12:15:51.633 I/DEBUG (27995): Abort message: 'art/runtime/mirror/art_method.cc:349] Check failed: PcIsWithinQuickCode(reinterpret_cast<uintptr_t>(code), pc) java.lang.Throwable java.lang.Throwable.fillInStackTrace() pc=71e3366b code=0x71e3362d size=ad000000'
10-05 12:15:51.633 I/DEBUG (27995): r0 00000000 r1 0000542b r2 00000006 r3 00000000
10-05 12:15:51.633 I/DEBUG (27995): r4 00000006 r5 b6f9addc r6 00000002 r7 0000010c
10-05 12:15:51.633 I/DEBUG (27995): r8 b63fe1e8 r9 be8e1418 sl b6427400 fp b63fcce0
10-05 12:15:51.633 I/DEBUG (27995): ip 0000542b sp be8e1358 lr b6e9a27b pc b6e9c280 cpsr 40070010
10-05 12:15:51.633 I/DEBUG (27995):
Bug: 17950037
This reverts commit 2535abe7d1fcdd0e6aca782b1f1932a703ed50a4.
Change-Id: I6f88849bc6f2befed0c0aaa0b7b2a08c967a83c3
|
|
Currently disabled by default unless -Xjit is passed in.
The proposed JIT is a method JIT which works by utilizing interpreter
instrumentation to request compilation of hot methods async during
runtime.
JIT options:
-Xjit / -Xnojit
-Xjitcodecachesize:N
-Xjitthreshold:integervalue
The JIT has a shared copy of a compiler driver which is accessed
by worker threads to compile individual methods.
Added JIT code cache and data cache, currently sized at 2 MB
capacity by default. Most apps will only fill a small fraction of
this cache however.
Added support to the compiler for compiling interpreter quickened
byte codes.
Added test target ART_TEST_JIT=TRUE and --jit for run-test.
TODO:
Clean up code cache.
Delete compiled methods after they are added to code cache.
Add more optimizations related to runtime checks e.g. direct pointers
for invokes.
Add method recompilation.
Move instrumentation to DexFile to improve performance and reduce
memory usage.
Bug: 17950037
Change-Id: Ifa5b2684a2d5059ec5a5210733900aafa3c51bca
|
|
Include enough infrastructure to allow cross-compiling for mips64,
interpret-only. This includes the instruction-set-features, frame
size info and utils assembler.
Also add a disassembler for oatdump, and support in patchoat.
Note: the runtime cannot run mips64, yet.
Change-Id: Id106581fa76b478984741c62a8a03be0f370d992
|
|
Introduce a swap-space and corresponding allocator to transparently
switch native allocations to memory backed by a file.
Bug: 18596910
(cherry picked from commit 62746d8d9c4400e4764f162b22bfb1a32be287a9)
Change-Id: I131448f3907115054a592af73db86d2b9257ea33
|
|
Change-Id: I3bf3250fa866fd2265f1b115d52fa5dedc48a7fc
|
|
With the recent move to gcc 4.9, we are hitting alignment
SIGBUS on ARM. The reason is that gcc will optimize two consecutive
32bits loads into one 64bits load, and the instruction (ldrd)
will fail if the data is not aligned.
Also removed the emission of mapping table when a method is optimized.
The information can be found in the StackMap itself.
Change-Id: Icf79406c18a3f4db3c05d52fc2c0dd2e35bf0f8f
|
|
Store the linker patches with each CompiledMethod instead of
keeping them in CompilerDriver. Reorganize oat file creation
to apply the patches as we're writing the method code. Add
framework for platform-specific relative call patches in the
OatWriter. Implement relative call patches for ARM.
Change-Id: Ie2effb3d92b61ac8f356140eba09dc37d62290f8
|
|
Now most of the methods supported by the compiler can be optimized,
instead of using the baseline.
Change-Id: I80ab36a34913fa4e7dd576c7bf55af63594dc1fa
|
|
OAT files have source line information enough for ART runtime needs like
jump to/from interpreter and thread suspension. But this information
is not enough for finer grained source level debugging and low-level
profiling (VTune or perf).
This patch adds to OAT files two additional sections:
.debug_line - DWARF formatted Elf32 section with detailed source line
information (mapping from native PC to Java source lines).
In addition to the debugging symbols added using the dex2oat option
--include-debug-symbols, the source line information is added to
the section .debug_line.
The source line info can be read by many Elf reading tools like objdump,
readelf, dwarfdump, gdb, perf, VTune, ...
gdb can use this debug line information in x86. In 64-bit mode
the information can be used if the oat file is mapped in the lower
address space (address has higher 32 bits zeroed). Relocation works.
Testing:
1. art/test/run-test --host --gdb [--64] 001-HelloWorld
2. in gdb: break Main.java:19
3. in gdb: break Runtime.java:111
4. in gdb: run - stops at void java.lang.Runtime.<init>()
5. in gdb: backtrace - shows call stack down to main()
6. in gdb: continue - stops at void Main.main() (only in 32-bit mode)
7. in gdb: backtrace - shows call stack down to main()
8. objdump -W <oat-file> - addresses are from VMA range of .text
section reported by objdump -h <file>
9. dwarfdump -ka <oat-file> - no errors expected
Size of aosp-x86-eng boot.oat increased by 11% from 80.5Mb to 89.2Mb
with two sections added .debug_line (7.2Mb) and .rel.debug (1.5Mb).
Change-Id: Ib8828832686e49782a63d5529008ff4814ed9cda
Signed-off-by: Yevgeny Rouban <yevgeny.y.rouban@intel.com>
|
|
Now for host-side x86 & x86_64 ART, we are able to get complete stacktrace with even mixed C/C++ & Java stack frames.
Testing:
1. art/test/run-test --host --gdb [--64] --no-relocate 005
2. In gdb, run 'b art::Class_classForName' which is implementation of a Java native method, then 'r'
3. In gdb, run 'bt'. You should see stack frames down to main()
Change-Id: I2d17e9aa0f6d42d374b5362a15ea35a2fce96302
|
|
Currently the entrypoint trampolines are compiled using the ARM32 assembler.
This changes that to use the thumb2 assembler for consistency with the
other compiled code.
Bug: 15455279
Change-Id: I6bacdf359e2ae4c0967fb5cd9dc694af11f802ec
|
|
The compiler driver is a single object delegating work to the compiler, rather
than passing it through to every Compiler call make it a member of Compiler so
that it maybe queried. This simplifies the Compiler API and makes the
relationship to CompilerDriver more explicit.
Remove reference arguments that contravene code style.
Change-Id: Iba47f2e3cbda679a7ec7588f26188d77643aa2c6
|
|
in instruction_set.h/cc
This allows to clean up some places that currently make explicit
comparisons.
Change-Id: I0dcc924c52fa53306f706aceea93a2d4a655c5df
|
|
This reverts commit f9487c039efb4112616d438593a2ab02792e0304.
Change-Id: Id48a4aae4ecce73db468587967968a3f7618b700
|
|
This reverts commit 081f73e888b3c246cf7635db37b7f1105cf1a2ff.
Change-Id: Ibd777f8ce73cf8ed6c4cb81d50bf6437ac28cb61
Conflicts:
compiler/dex/quick/mir_to_lir.h
|
|
This reverts commit 754ddad084ccb610d0cf486f6131bdc69bae5bc6.
Change-Id: Icd979adee1d8d781b40a5e75daf3719444cb72e8
|
|
This is an ARM specific optimization to the compiler
that uses trampoline islands to make calls to runtime
helper functions. The intention is to reduce the size
of the generated code (by 2 bytes per call) without
affecting performance.
By default this is on when generating an OAT file. It is
off when compiling to memory.
To switch this off in dex2oat, use the command line option:
--no-helper-trampolines
Enhances disassembler to print the trampoline entry on the
BL instruction like this:
0xb6a850c0: f7ffff9e bl -196 (0xb6a85000) ; pTestSuspend
Bug: 12607709
Change-Id: I9202bdb7cf21252ad807bd48701f1f6ce8e3d0fe
|
|
Passed all tests from jni_compiler_test and art/test on host with jni_copiler.
Incoming argument spill is enabled, entry_spills refactored. Now each entry spill
contains data type size (4 or 8) and offset which should be used for spill.
Assembler REX support implemented in opcodes used in JNI compiler.
Please note, JNI compiler is not enabled by default yet (see compiler_driver.cc:1875).
Change-Id: I5fd19cca72122b197aec07c3708b1e80c324be44
Signed-off-by: Dmitry Petrochenko <dmitry.petrochenko@intel.com>
|
|
Adds support for arm64 to ART. Assembler stubs are sufficient for
down calls into interpreter. JNI compiler and generics are not finished.
Basic Generic JNI functionality.
Change-Id: I4a07c79d1e037b9f5746673480e32cf456867b82
|
|
Also, hack x86 assembler for use as a x86-64 trampoline compiler's assembler.
Implement missing x86-64 quick resolution trampoline.
Add x86-64 to the quick elf writer.
Change-Id: I08216c67014a83492ada12898ab8000218ba7bb4
|
|
This is actually a lot of work. To do this, we need:
.debug_info
.debug_abbrev
.debug_frame
.debug_str
These are generated into the OAT file by OatWriter and ElfWriterQuick.
Since the Quick ART runtime doesn't use dlopen to load the OAT files,
GDB can't find this information. Use the alternate GDB JIT interface,
which can be invoked at runtime. To use this interface, an ELF image
needs to be built in memory. Read the information from the OAT file,
fixup the addresses to point to the real locations, add a symbol table
to hold the .text symbol, and then let GDB know about the information,
which will be read from the runtime address space.
This is quite primitive now, and could be cleaned up considerably. It
probably needs symbol table entries for the methods, and descriptions of
parameters and return types.
Currently only supported for X86.
This defaults to enabled for debug builds. Added dexoat --gen-gdb-info
and --no-gen-gdb-info flags to override.
Change-Id: I4d18b2370f6dfaa00c8cc1925f10717be3bd1a62
Signed-off-by: Mark Mendell <mark.p.mendell@intel.com>
|
|
Modify mirror objects so that references between them use an ObjectReference
value type rather than an Object* so that functionality to compress larger
references can be captured in the ObjectRefererence implementation.
ObjectReferences are 32bit and all other aspects of object layout remain as
they are currently.
Expand fields in objects holding pointers so they can hold 64bit pointers. Its
expected the size of these will come down by improving where we hold compiler
meta-data.
Stub out x86_64 architecture specific runtime implementation.
Modify OutputStream so that reads and writes are of unsigned quantities.
Make the use of portable or quick code more explicit.
Templatize AtomicInteger to support more than just int32_t as a type.
Add missing, and fix issues relating to, missing annotalysis information on the
mutator lock.
Refactor and share implementations for array copy between System and uses
elsewhere in the runtime.
Fix numerous 64bit build issues.
Change-Id: I1a5694c251a42c9eff71084dfdd4b51fff716822
|
|
Moved deduplication to be in the compiler driver instead of oat
writer. This enables deduplication to be performed on multiple
threads. Also added a hash function to avoid excessive comparison
of byte arrays.
Improvements:
Before (alloats host):
real 1m6.967s
user 4m22.940s
sys 1m22.610s
Thinkfree.apk (target mako):
0m23.74s real 0m50.95s user 0m9.50s system
0m24.62s real 0m50.61s user 0m10.07s system
0m24.22s real 0m51.44s user 0m10.09s system
0m23.70s real 0m51.05s user 0m9.97s system
0m23.50s real 0m50.74s user 0m10.63s system
After (alloats host):
real 1m5.705s
user 4m44.030s
sys 1m29.990s
Thinkfree.apk (target mako):
0m23.32s real 0m51.38s user 0m10.00s system
0m23.49s real 0m51.20s user 0m9.80s system
0m23.18s real 0m50.80s user 0m9.77s system
0m23.52s real 0m51.22s user 0m10.02s system
0m23.50s real 0m51.55s user 0m9.46s system
Bug: 10552630
Change-Id: Ia6d06a747b86b0bfc4473b3cd68f8ce1a1c7eb22
|