summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Merge "Print lock level in ExclusiveUnlock"Mathieu Chartier2015-01-281-1/+2
|\
| * Print lock level in ExclusiveUnlockMathieu Chartier2015-01-271-1/+2
|/ | | | | Bug: 18713034 Change-Id: I83e1bf1e2023b595d8ddf20008935c75b2b0aa0c
* Merge "ART: Even more Quick cleanup"Andreas Gampe2015-01-288-78/+34
|\
| * ART: Even more Quick cleanupAndreas Gampe2015-01-278-78/+34
| | | | | | | | | | | | Remove Backend. Change-Id: I247cc65ccda6a362ba1a8f5e73e7f12ecd980a87
* | Merge "Revert "Revert "ART: Implement X86 hard float (Quick/JNI/Baseline)"""Andreas Gampe2015-01-2718-167/+601
|\ \
| * | Revert "Revert "ART: Implement X86 hard float (Quick/JNI/Baseline)""Mark P Mendell2015-01-2718-167/+601
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 949c91fb91f40a4a80b2b492913cf8541008975e. This time, don't clobber EBX before saving it. Redo some of the macros to make register usage explicit. Change-Id: I8db8662877cd006816e16a28f42444ab7c36bfef
* | | Merge "ART: Clean up some build flags"Andreas Gampe2015-01-271-4/+1
|\ \ \ | |_|/ |/| |
| * | ART: Clean up some build flagsAndreas Gampe2015-01-271-4/+1
|/ / | | | | | | | | | | Remove some no-longer necessary build flags. Change-Id: Iea85a21689c20931b88d0e84b84410652b66b38d
* | Merge "ART: reduce UpdateModUnionTable time for partial GC"Hiroshi Yamauchi2015-01-271-6/+12
|\ \ | |/ |/|
| * ART: reduce UpdateModUnionTable time for partial GCLei Li2015-01-231-6/+12
| | | | | | | | | | | | | | | | | | | | UpdateModUnionTable is to scan objects with dirty cards in the immune region. The immune region of partial GC are image space and zygote space. As image space is always immuned for any kind of GC, there is no need to scan dirty cards of zygote space pointing to image space for partial GC. Change-Id: I2e1f61c2b361985c211076a441418797abf324c2
* | Merge "ART: clear dirty cards of alloc space for MS/CMS partial and full GCs"Hiroshi Yamauchi2015-01-277-17/+47
|\ \
| * | ART: clear dirty cards of alloc space for MS/CMS partial and full GCsLei Li2015-01-267-17/+47
| | | | | | | | | | | | | | | | | | | | | | | | | | | For MS/CMS partial and full GCs, we could clear the dirty cards of alloc space when we process cards as we care about the dirty cards after GC starts. Change-Id: I1f9b32b20d75979387bc5d26b0cf9a256dcf20b6 Signed-off-by: Lei Li <lei.l.li@intel.com>
* | | Merge "Checker: Allow don't-care output on a line"David Brazdil2015-01-272-47/+65
|\ \ \
| * | | Checker: Allow don't-care output on a lineDavid Brazdil2015-01-272-47/+65
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch changes the behaviour of whitespace characters in CHECK lines, allowing for additional content between verified parts of the matched output line. Tests therefore won't need to explicitly match attributes which are not tested. The way attributes are printed ensures that the right part of the line is matched against. Example: - output line: i32 Div [ i4 i8 ] ( loop_header:null ) - CHECK before: Div [ {{i\d+}} {{i\d+}} ] ( loop_header:null ) - CHECK now: Div ( loop_header:null ) Change-Id: Icf6bacfb285ae288bea21640e860a871a94cc386
* | | | Merge "ART: Further refactor use lists"David Brazdil2015-01-274-20/+15
|\ \ \ \ | |/ / / |/| | |
| * | | ART: Further refactor use listsDavid Brazdil2015-01-274-20/+15
|/ / / | | | | | | | | | Change-Id: I9e3219575a508ca5141d851bfcaf848302480c32
* | | Merge "run-test: enable checker in dev mode"Calin Juravle2015-01-272-4/+15
|\ \ \
| * | | run-test: enable checker in dev modeCalin Juravle2015-01-272-4/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Also, use explicit flush (instead of passing flush parameter to print) to enable better compatibility across python versions. Change-Id: Iaf294d88e932b778d5dce7f3c2b8eca775849973
* | | | Merge "ART: More Quick cleanup"Andreas Gampe2015-01-276-302/+208
|\ \ \ \
| * | | | ART: More Quick cleanupAndreas Gampe2015-01-266-302/+208
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Remove frontend, move code to QuickCompiler. Fix some includes. Change-Id: Iedd8f339ec76635e48b166e285f7fc571fc948ac
* | | | | Merge "Revert "ART: Implement X86 hard float (Quick/JNI/Baseline)""Vladimir Marko2015-01-2717-593/+167
|\ \ \ \ \
| * | | | | Revert "ART: Implement X86 hard float (Quick/JNI/Baseline)"Vladimir Marko2015-01-2717-593/+167
|/ / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | And the 3 Mac build fixes. Fix conflicts in context_x86.* . This reverts commits 3d2c8e74c27efee58e24ec31441124f3f21384b9 , 34eda1dd66b92a361797c63d57fa19e83c08a1b4 , f601d1954348b71186fa160a0ae6a1f4f1c5aee6 , bc503348a1da573488503cc2819c9e30807bea31 . Bug: 19150481 Change-Id: I6650ee30a7d261159380fe2119e14379e4dc9970
* | | | | Merge "Refactor register access from StackVisitor"Sebastien Hertz2015-01-2715-269/+253
|\ \ \ \ \
| * | | | | Refactor register access from StackVisitorSebastien Hertz2015-01-2615-269/+253
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Moves register access checking up to StackVisitor by adding methods IsAccessibleGPR and IsAccessibleFPR in Context class. It allows to simplify GetGPR/FPR and SetGPR/FPR methods in the Context class (and its subclasses). Also simplifies code in StackVisitor by adding IsAccessibleRegister, GetRegister and SetRegister methods which then call either GPR or FPR specific methods in Context depending on the nature of the accessed register. Bug: 18547544 Bug: 19106446 Change-Id: I6e707608d935a71571d0e975a6e766053de3763a
* | | | | | Merge "JDWP: allocate single step control only when active"Sebastien Hertz2015-01-274-90/+115
|\ \ \ \ \ \ | |_|/ / / / |/| | | | |
| * | | | | JDWP: allocate single step control only when activeSebastien Hertz2015-01-264-90/+115
| |/ / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Only allocates SingleStepControl when a single step is requested in a thread. Bug: 19142632 Change-Id: I830b7eaea86b516d3fac80ce14d41c66b4bd5415
* | | | | Merge "ART: Fix inlining of Mir2Lir"Andreas Gampe2015-01-273-3/+3
|\ \ \ \ \
| * | | | | ART: Fix inlining of Mir2LirAndreas Gampe2015-01-263-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Missed -inl includes. Change-Id: I39e6d603c7f5d36693aca3816653594488bff63f
* | | | | | Merge "Fix 114-ParallelGC."Hiroshi Yamauchi2015-01-271-1/+5
|\ \ \ \ \ \ | |/ / / / / |/| | | | |
| * | | | | Fix 114-ParallelGC.Hiroshi Yamauchi2015-01-261-1/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In main(), the outer ArrayList may still be reachable after the try-catch under the interpreter or a compiler without a liveness analysis and the last ArrayList allocation may fail due to an OOME. This fixes 114-ParallelGC under the CC and the GSS collectors. Bug: 12687968 Change-Id: Ie1082d38b2a677ec70fdc23b0187ae8ce0612808
* | | | | | Merge "ART: Some Quick cleanup"Andreas Gampe2015-01-2679-277/+366
|\ \ \ \ \ \
| * | | | | | ART: Some Quick cleanupAndreas Gampe2015-01-2679-277/+366
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Make several fields const in CompilationUnit. May benefit some Mir2Lir code that repeats tests, and in general immutability is good. Remove compiler_internals.h and refactor some other headers to reduce overly broad imports (and thus forced recompiles on changes). Change-Id: I898405907c68923581373b5981d8a85d2e5d185a
* | | | | | | Merge "Add missing EndTiming in CompileMethod"Mathieu Chartier2015-01-261-0/+4
|\ \ \ \ \ \ \
| * | | | | | | Add missing EndTiming in CompileMethodMathieu Chartier2015-01-261-0/+4
| | |/ / / / / | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | DCHECK failure when running tests with JIT. Bug: 17950037 Change-Id: I4f886758b2a9322c2d7a8e8449e84c9a1709c813
* | | | | | | Merge "ART: Take apart stream output"Andreas Gampe2015-01-261-2/+3
|\ \ \ \ \ \ \ | |/ / / / / / |/| | | | | |
| * | | | | | ART: Take apart stream outputAndreas Gampe2015-01-261-2/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | To investigate bug. Bug: 19028453 Change-Id: Ib109caf2747f9adeef15dfc0f4708498e395dacf
* | | | | | | Merge "Change AtomicStack to use StackReference"Mathieu Chartier2015-01-2613-107/+140
|\ \ \ \ \ \ \
| * | | | | | | Change AtomicStack to use StackReferenceMathieu Chartier2015-01-2513-107/+140
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Previously used Object*, using StackReference saves memory on 64 bit devices. Bug: 12935052 Bug: 17643507 Change-Id: I035878690054eeeb24d655a900b8f26c837703ff
* | | | | | | | Merge "ART: Fix assembler labels' usage in macros for Mac."Vladimir Marko2015-01-261-2/+2
|\ \ \ \ \ \ \ \ | |_|/ / / / / / |/| | | | | | |
| * | | | | | | ART: Fix assembler labels' usage in macros for Mac.Vladimir Marko2015-01-261-2/+2
|/ / / / / / / | | | | | | | | | | | | | | | | | | | | | Change-Id: I99552528329a7f2cd5656eef0210c831cfc5fe09
* | | | | | | Merge "Optimizing: Speed up HEnvironment use removal"David Brazdil2015-01-2613-151/+213
|\ \ \ \ \ \ \
| * | | | | | | Optimizing: Speed up HEnvironment use removalDavid Brazdil2015-01-2613-151/+213
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Removal of use records from HEnvironment vregs involved iterating over potentially large linked lists which made compilation of huge methods very slow. This patch turns use lists into doubly-linked lists, stores pointers to the relevant nodes inside HEnvironment and subsequently turns the removals into constant-time operations. Change-Id: I0e1d4d782fd624e7b8075af75d4adf0a0634a1ee
* | | | | | | | Merge "Move code around and address growable_array comment."Nicolas Geoffray2015-01-269-325/+381
|\ \ \ \ \ \ \ \
| * | | | | | | | Move code around and address growable_array comment.Nicolas Geoffray2015-01-269-325/+381
| |/ / / / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - Move SideEffectsAnalysis to its own file. - Move most of gvn.h to gvn.cc. - Don't call Resize in GrowableArray constructor, but just set num_used directly. Change-Id: I1f1291207945d678d3c99cc0ec1ec155bcae82f6
* | | | | | | | Merge "Quick: Fix range check for intrinsic String.charAt() on x86."Vladimir Marko2015-01-262-2/+38
|\ \ \ \ \ \ \ \
| * | | | | | | | Quick: Fix range check for intrinsic String.charAt() on x86.Vladimir Marko2015-01-262-2/+38
| | |_|_|/ / / / | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Bug: 19125146 Change-Id: I274190a7a60cd2e29a854738ed1ec99a9e611969
* | | | | | | | Merge "Fix Mac build."Vladimir Marko2015-01-261-1/+1
|\ \ \ \ \ \ \ \ | |_|/ / / / / / |/| | | | | | |
| * | | | | | | Fix Mac build.Vladimir Marko2015-01-261-1/+1
|/ / / / / / / | | | | | | | | | | | | | | | | | | | | | Change-Id: I88c0b3c7b799df8dd0aab86b6801c37093045129
* | | | | | | Merge "Fix build breakage after GVN change."Nicolas Geoffray2015-01-262-44/+62
|\ \ \ \ \ \ \
| * | | | | | | Fix build breakage after GVN change.Nicolas Geoffray2015-01-262-44/+62
|/ / / / / / / | | | | | | | | | | | | | | | | | | | | | Change-Id: I120717f8cae452cad9967559c466c1605a3e7b14