summaryrefslogtreecommitdiffstats
path: root/compiler/image_writer.cc
Commit message (Collapse)AuthorAgeFilesLines
* Revert "Revert "Avoid compiling monster methods in boot image""Brian Carlstrom2014-03-051-5/+7
| | | | | | | | This reverts commit f679fab9be8905505d26a62bea257641f0bb0168. (cherry picked from commit 9d0bdfe63bec5a43d7668d9094a52caf77b6ed53) Change-Id: I72af1d44483edbd2a5f2e6b678e43f740d4613fd
* Support compiler filters for boot classpathBrian Carlstrom2014-03-041-7/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | image_writer.cc Remove assumption that all methods in the boot classpath are compiled oat_writer.cc Don't skip writing ArtMethod::quick_code_offset_ for methods that need resolution, leave that to ImageWriter dex2oat.cc Allow dex2dex compilation of image dex files by making the in memory pages writable in all cases, not just app case. oatdump.cc dump new OatHeader fields use ImageSpace.GetImageFilename, not command line image filename, since location may be in dalvik-cache remove inaccurate check about non-null GC map quick_trampoline_entrypoints.cc add and improve some DCHECKS that were useful while debugging class_linker.cc image_space.cc fix double facepalm parsed_options.cc fix zygote logging to not skip values to two part options like -classpath <foo> runtime.cc wireup parsed compiler options to runtime Change-Id: Iad314df0b80623c0663d61713d5098297ab9ac87
* Remove oat file location in the image.Nicolas Geoffray2014-03-041-2/+0
| | | | | | | | | The oat file is now always in the same directory, and has the same name as the image file. Only difference is the extension. This also removes the need for host-prefix. Change-Id: I16d1f7aeb1d58372d41921694664e9c321afc1ad
* Initial changes towards Generic JNI optionAndreas Gampe2014-03-031-3/+16
| | | | | | | | Some initial changes that lead to an UNIMPLEMENTED. Works by not compiling for JNI right now and tracking native methods which have neither quick nor portable code. Uses new trampoline. Change-Id: I5448654044eb2717752fd7359f4ef8bd5c17be6e
* (Experimental) Add Brooks pointers.Hiroshi Yamauchi2014-02-251-0/+6
| | | | | | | | | This feature is disabled by default. Verified that the Brooks pointers are installed correctly by using the CMS/SS collectors. Change-Id: Ia9be9814ab6e29169ac85edc4792ce8c81d552a9
* Avoid std::string allocations for finding an array class.Ian Rogers2014-02-241-1/+2
| | | | | | | | | | | | Introduce ClassLinker::FindArrayClass which performs an array class lookup given the element/component class. This has a 16 element cache of recently looked up arrays. Pass the current thread to ClassLinker Find .. Class routines to avoid calls to Thread::Current(). Avoid some uses of FindClass in the debugger where WellKnownClasses is a faster and more compacting GC friendly alternative. Change-Id: I60e231820b349543a7edb3ceb9cf1ce92db3c843
* Fix and optimize verify object.Mathieu Chartier2014-02-211-29/+33
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | VerifyObject no longer resides in heap. You can now enable VerifyObject for non-debug builds. VerifyStack is still slow, so it is now guarded by its own flag. Fixed the image writer to not use verification at places where verification fails due to invalid reads. Fixed RosAlloc to use SizeOf which doesn't call verify object. Added a flag paremeter to some of the mirror getters / setters to be able to selectively disable VerifyObject on certain calls. Optimized the GC to not verify each object multiple times during object scanning if verify object is enabled. Added 3 verification options: verify reads, verify this, and verify writes so that you can select how much verification you want for mirror getters and setters. Removed some useless DCHECKs which would slow debug builds without providing any benefits. TODO: RosAlloc verification doesn't currently work with verify objects. Bug: 12934910 Bug: 12879358 Change-Id: Ic61033104dfc334543f89b0fc0ad8cd4f4015d69
* Remove blacklistSebastien Hertz2014-02-171-18/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Removes the class initialization blacklist and use transaction to detect and revert class initialization attempting to invoke native method. This only concerns class initialization happening at compilation time when generating an image (like boot.art for the system). In transactional mode, we log every object's field assignment and array update. Therefore we're able to abort a transaction to restore values of fields and array as they were before the transaction starts. We also log changes to the intern string table so we can restore its state prior to transaction start. Since transactional mode only happens at compilation time, we don't need to log all these changes at runtime. In order to reduce the overhead of testing if transactional mode is on/off, we templatize interfaces of mirror::Object and mirror::Array, respectively responsible for setting a field and setting an array element. For various reasons, we skip some specific fields from transaction: - Object's class and array's length must remain unchanged so garbage collector can compute object's size. - Immutable fields only set during class loading: list of fields, method, dex caches, vtables, ... as all classes have been loaded and verified before a transaction occurs. - Object's monitor for performance reason. Before generating the image, we browse the heap to collect objects that need to be written into it. Since the heap may still holds references to unreachable objects due to aborted transactions, we trigger one collection at the end of the class preinitialization phase. Since the transaction is held by the runtime and all compilation threads share the same runtime, we need to ensure only one compilation thread has exclusive access to the runtime. To workaround this issue, we force class initialization phase to run with only one thread. Note this is only done when generating image so application compilation is not impacted. This issue will be addressed in a separate CL. Bug: 9676614 Change-Id: I221910a9183a5ba6c2b99a277f5a5a68bc69b5f9
* Support Direct Method/Type access for X86Mark Mendell2014-02-101-1/+13
| | | | | | | | | | | Thumb generates code to optimize calls to methods within core.oat. Implement this for X86 as well, but take advantage of mov with 32 bit immediate and call relative with 32 bit immediate. Fix some incorrect return locations for long inlines. Change-Id: I1907bdfc7574f3d0aa76c7fad13dc537acdf1ed3 Signed-off-by: Mark Mendell <mark.p.mendell@intel.com>
* Object model changes to support 64bit.Ian Rogers2014-02-061-68/+64
| | | | | | | | | | | | | | | | | | | | | | | 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
* Use direct class pointers at allocation sites in the compiled code.Hiroshi Yamauchi2014-01-231-13/+53
| | | | | | | | | | | | | | | | | | | | | | | - Rather than looking up a class from its type ID (and checking if it's resolved/initialized, resolving/initializing if not), use direct class pointers, if possible (boot-code-to-boot-class pointers and app-code-to-boot-class pointers.) - This results in a 1-2% speedup in Ritz MemAllocTest on Nexus 4. - Embedding the object size (along with class pointers) caused a 1-2% slowdown in MemAllocTest and isn't implemented in this change. - TODO: do the same for array allocations. - TODO: when/if an application gets its own image, implement app-code-to-app-class pointers. - Fix a -XX:gc bug. cf. https://android-review.googlesource.com/79460/ - Add /tmp/android-data/dalvik-cache to the list of locations to remove oat files in clean-oat-host. cf. https://android-review.googlesource.com/79550 - Add back a dropped UNLIKELY in FindMethodFromCode(). cf. https://android-review.googlesource.com/74205 Bug: 9986565 Change-Id: I590b96bd21f7a7472f88e36752e675547559a5b1
* Remove intialized static storage from dex cache.Ian Rogers2014-01-081-1/+0
| | | | | | | | | The initialized static storage array is used by compiled code to determine if for a sget/sput class initialization is necessary. The compiled code typically doesn't require this test as the class is pre-initialized or the class being accessed is the same as the current method. Change-Id: Icbc45e692b3d0ac61e559e69edb6c9b29439e571
* Don't rely on gcc extensionsBernhard Rosenkränzer2013-12-121-2/+2
| | | | | | | | | | | | | | Make the code more compatible with different compilers. clang doesn't allow extra static qualifiers on template specializations, const qualifiers on function types, or inline attributes on lambda functions, and is more picky about casting away constness with reinterpret_cast. These modifications are compatible with both gcc and clang. Change-Id: I739b10df2780bec537827a13679fd2bcc2cc7188 Signed-off-by: Bernhard Rosenkränzer <Bernhard.Rosenkranzer@linaro.org>
* Compact zygote.Mathieu Chartier2013-11-221-1/+1
| | | | | | | | | | | | | | | | | | | | | We now start out using the bump pointer allocator and switch to the free list collector (ROSAlloc) after the zygote forks. Before compaction: Zygote size: 9060352 After compaction Zygote size: 8425864 The main reason the size doesn't reduce more is that most of the zygote space is non-movable objects allocated by VMRuntime.newNonMovableObject. The objects which are non-movable but could be movable include around 10000 classes and some number of fields and methods. Bug: 8981901 Change-Id: Iea21b70fb7af27cb7e92d72070d278a5cd4026ac
* Don't read characters from 0 length strings.Ian Rogers2013-11-141-1/+6
| | | | | | Caught by valgrind. Bug: 11670287 Change-Id: I3acf4855c8662b804cf0c24680fc21c50c435bdb
* Fix image writer bitmap size calculation.Mathieu Chartier2013-11-131-2/+4
| | | | | | | | Wasn't properly dividing by 8 inside of the image header creation. This resulted in a bitmap size which was larger than it should be inside of the image. Change-Id: I344d1f3c1794a7cff3c9e22afc7fdabedf74413c
* Compacting collector.Mathieu Chartier2013-11-111-153/+193
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The compacting collector is currently similar to semispace. It works by copying objects back and forth between two bump pointer spaces. There are types of objects which are "non-movable" due to current runtime limitations. These are Classes, Methods, and Fields. Bump pointer spaces are a new type of continuous alloc space which have no lock in the allocation code path. When you allocate from these it uses atomic operations to increase an index. Traversing the objects in the bump pointer space relies on Object::SizeOf matching the allocated size exactly. Runtime changes: JNI::GetArrayElements returns copies objects if you attempt to get the backing data of a movable array. For GetArrayElementsCritical, we return direct backing storage for any types of arrays, but temporarily disable the GC until the critical region is completed. Added a new runtime call called VisitObjects, this is used in place of the old pattern which was flushing the allocation stack and walking the bitmaps. Changed image writer to be compaction safe and use object monitor word for forwarding addresses. Added a bunch of added SIRTs to ClassLinker, MethodLinker, etc.. TODO: Enable switching allocators, compacting on background, etc.. Bug: 8981901 Change-Id: I3c886fd322a6eef2b99388d19a765042ec26ab99
* Lazily compute object identity hash codes.Mathieu Chartier2013-10-311-2/+7
| | | | | | | | | | | | | | | Before, we computed identity hashcodes whenever we inflated a monitor. This caused issues since it meant that we would have all of these hash codes in the image, causing locks to excessively inflate during application run time. This change makes it so that we lazily compute hash codes. When a thin lock gets inflated, we assign a hash code of 0 assigned to it. This value signifies no hash code. When we try to get the identity hash code of an object with an inflated monitor, it gets computed if it is 0. Change-Id: Iae6acd1960515a36e74644e5b1323ff336731806
* Merge "Don't use UTF16 length as length for MUTF8." into dalvik-devIan Rogers2013-10-301-2/+2
|\
| * Don't use UTF16 length as length for MUTF8.Ian Rogers2013-10-301-2/+2
| | | | | | | | | | | | Bug 11367555. Change-Id: Ia0b07072a1a49d435c3b71ed9a668b316b7ff5d8
* | Merge "Implement Interface Method Tables (IMT)." into dalvik-devJeff Hao2013-10-291-0/+12
|\ \
| * | Implement Interface Method Tables (IMT).Jeff Hao2013-10-291-0/+12
| |/ | | | | | | Change-Id: Idf7fe85e1293453a8ad862ff2380dcd5db4e3a39
* | Fix object identity hash.Mathieu Chartier2013-10-291-1/+25
|/ | | | | | | | | | | | | The object identity hash is now stored in the monitor word after being computed. Hashes are computed by a pseudo random number generator. When we write the image, we eagerly compute object hashes to prevent pages getting dirtied. Bug: 8981901 Change-Id: Ic8edacbacb0afc7055fd740a52444929f88ed564
* Throw IOException at source of failing to open a dex file.Ian Rogers2013-10-211-7/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Before is: java.lang.ClassNotFoundException: Didn't find class "GCBench" on path: DexPathList[[zip file "/disk2/dalvik-dev/out/host/linux-x86/framework/GCBench.jar"],nativeLibraryDirectories=[/disk2/dalvik-dev/out/host/linux-x86/lib]] at dalvik.system.BaseDexClassLoader.findClass(BaseDexClassLoader.java:56) at java.lang.ClassLoader.loadClass(ClassLoader.java:511) at java.lang.ClassLoader.loadClass(ClassLoader.java:469) Suppressed: java.lang.ClassNotFoundException: GCBench at java.lang.Class.classForName(Native Method) at java.lang.BootClassLoader.findClass(ClassLoader.java:781) at java.lang.BootClassLoader.loadClass(ClassLoader.java:841) at java.lang.ClassLoader.loadClass(ClassLoader.java:504) ... 1 more Caused by: java.lang.NoClassDefFoundError: Class "LGCBench;" not found ... 5 more And after is: java.lang.ClassNotFoundException: Didn't find class "GCBench" on path: DexPathList[[zip file "/disk2/dalvik-dev/out/host/linux-x86/framework/GCBench.jar"],nativeLibraryDirectories=[/disk2/dalvik-dev/out/host/linux-x86/lib]] at dalvik.system.BaseDexClassLoader.findClass(BaseDexClassLoader.java:56) at java.lang.ClassLoader.loadClass(ClassLoader.java:511) at java.lang.ClassLoader.loadClass(ClassLoader.java:469) Suppressed: java.io.IOException: Zip archive '/disk2/dalvik-dev/out/host/linux-x86/framework/GCBench.jar' doesn't contain classes.dex at dalvik.system.DexFile.openDexFile(Native Method) at dalvik.system.DexFile.<init>(DexFile.java:80) at dalvik.system.DexFile.<init>(DexFile.java:59) at dalvik.system.DexPathList.loadDexFile(DexPathList.java:268) at dalvik.system.DexPathList.makeDexElements(DexPathList.java:235) at dalvik.system.DexPathList.<init>(DexPathList.java:113) at dalvik.system.BaseDexClassLoader.<init>(BaseDexClassLoader.java:48) at dalvik.system.PathClassLoader.<init>(PathClassLoader.java:38) at java.lang.ClassLoader.createSystemClassLoader(ClassLoader.java:128) at java.lang.ClassLoader.access$000(ClassLoader.java:65) at java.lang.ClassLoader$SystemClassLoader.<clinit>(ClassLoader.java:81) at java.lang.ClassLoader.getSystemClassLoader(ClassLoader.java:137) Suppressed: java.lang.ClassNotFoundException: GCBench at java.lang.Class.classForName(Native Method) at java.lang.BootClassLoader.findClass(ClassLoader.java:781) at java.lang.BootClassLoader.loadClass(ClassLoader.java:841) at java.lang.ClassLoader.loadClass(ClassLoader.java:504) ... 1 more Caused by: java.lang.NoClassDefFoundError: Class "LGCBench;" not found ... 5 more Also, move dex file verifier messages out of logs. In the process the ClassLinker::dex_lock_ needed tidying to cover a smaller scope. Bug 11301553. Change-Id: I80058652e11e7ea63457cc01a0cb48afe1c15543
* StringPiece clean up.Ian Rogers2013-09-241-2/+2
| | | | | | | | | | | Profile guided clean up. Try to avoid creating StringPieces with the contents of a dex file where the length is known. Try to avoid RegTypeCache::FromDescriptor when there's a class available. Make ConstantType::ConstantValue inlinable. Saving of about 50ms from a 2 threaded ThinkFree compile on host. Change-Id: I47a12c3c76f46e2c9805be1c3a3e3870fe1f5d85
* Use class def index from java.lang.Class.Ian Rogers2013-09-191-0/+1
| | | | | | | | | | | | Bug: 10244719 Depends on: https://googleplex-android-review.git.corp.google.com/362363 This removes the computation of the dex file index, when necessary this is computed by searching the dex file. Its only necessary in dalvik.system.DexFile.defineClassNative and DexFile::FindInClassPath, the latter not showing up significantly in profiling with this change. Change-Id: I20c73a3b17d86286428ab0fd21bc13f51f36c85c
* Write out image bitmap inside of image file.Mathieu Chartier2013-08-291-4/+38
| | | | | | | | | | | | We now create the image bitmap when we generate the image. The image bitmap is written after the image inside of the image file. This speeds up dex2oat by making walking the image during heap creation unnecessary. This should also help memory pressure by enabling the image bitmap to be swappable. Bug: 10432288 Change-Id: Idebf459ed15edbb41a7d9b9b353934155bce2f19
* Fix up TODO: c++0x, update cpplint.Mathieu Chartier2013-08-161-35/+13
| | | | | | | | | | | | Needed to update cpplint to handle const auto. Fixed a few cpplint errors that were being missed before. Replaced most of the TODO c++0x with ranged based loops. Loops which do not have a descriptive container name have a concrete type instead of auto. Change-Id: Id7cc0f27030f56057c544e94277300b3f298c9c5
* Add flock(2)ing on dex-cache files to prevent racesBrian Carlstrom2013-08-151-2/+2
| | | | | Bug: 9071417 Change-Id: I1ee9ff281867f90fba7a8ed8bbf06b33ac29d511
* Refactor java.lang.reflect implementationBrian Carlstrom2013-08-131-23/+23
| | | | | | | | | | | Cherry-picked from commit ed41d5c44299ec5d44b8514f6e17f802f48094d1. Move to ArtMethod/Field instead of AbstractMethod/Field and have java.lang.reflect APIs delegate to ArtMethod/ArtField. Bug: 10014286. Change-Id: Iafc1d8c5b62562c9af8fb9fd8c5e1d61270536e7
* Uleb128 compression of vmap and mapping table.Ian Rogers2013-08-091-2/+2
| | | | | | Bug 9437697. Change-Id: I30bcb97d12cd8b46d3b2cdcbdd358f08fbb9947a
* Entry point clean up.Ian Rogers2013-08-071-45/+62
| | | | | | | | | | | | | | | | | | Create set of entry points needed for image methods to avoid fix-up at load time: - interpreter - bridge to interpreter, bridge to compiled code - jni - dlsym lookup - quick - resolution and bridge to interpreter - portable - resolution and bridge to interpreter Fix JNI work around to use JNI work around argument rewriting code that'd been accidentally disabled. Remove abstact method error stub, use interpreter bridge instead. Consolidate trampoline (previously stub) generation in generic helper. Simplify trampolines to jump directly into assembly code, keeps stack crawlable. Dex: replace use of int with ThreadOffset for values that are thread offsets. Tidy entry point routines between interpreter, jni, quick and portable. Change-Id: I52a7c2bbb1b7e0ff8a3c3100b774212309d0828e
* Fix cpplint whitespace/comments issuesBrian Carlstrom2013-07-261-3/+3
| | | | Change-Id: Iae286862c85fb8fd8901eae1204cd6d271d69496
* Remove OatWriter buffering to memory for ElfWriterQuickBrian Carlstrom2013-07-241-0/+4
| | | | | | This allows the oat contents to be directly written to the file. Change-Id: Ibc7ddf57477b152f07784b52f7334be73fd22833
* Fix cpplint build/namespaces issuesBrian Carlstrom2013-07-181-1/+8
| | | | Change-Id: I19c68703270c1482d6c6aad8cdf97d3d2924360a
* Fix cpplint whitespace/braces issuesBrian Carlstrom2013-07-171-1/+1
| | | | Change-Id: Ide80939faf8e8690d8842dde8133902ac725ed1a
* Create separate Android.mk for main build targetsBrian Carlstrom2013-07-121-0/+680
The runtime, compiler, dex2oat, and oatdump now are in seperate trees to prevent dependency creep. They can now be individually built without rebuilding the rest of the art projects. dalvikvm and jdwpspy were already this way. Builds in the art directory should behave as before, building everything including tests. Change-Id: Ic6b1151e5ed0f823c3dd301afd2b13eb2d8feb81
generated by cgit v1.1 at 2025-02-13 16:55:43 +0000