)]}' { "log": [ { "commit": "61ba8d2421a98e9b16510be4f9af7ca7bc4c9055", "tree": "9a79331dd5ebd46ed46f853b24ab072f43debf28", "parents": [ "eb369ce3669be74dd89b21f8b3ab31ace4a47086" ], "author": { "name": "Nicolas Geoffray", "email": "ngeoffray@google.com", "time": "Tue Aug 07 09:55:57 2018 +0100" }, "committer": { "name": "Nicolas Geoffray", "email": "ngeoffray@google.com", "time": "Mon Aug 20 17:18:31 2018 +0100" }, "message": "Fix longstanding bug around implicit NPEs and GC, version 2.\n\nThe TODO has been there since M (so forever :)):\nhttps://android-review.googlesource.com/c/platform/art/+/122794/13//COMMIT_MSG#13\n\nWe hardly see the issue in our tests as we need to have:\n1) A GC happening while creating the NPE object.\n2) ParallelMoves between the NullCheck and implicit null check operation\n that moves references.\n\nThe CL piggy backs on the \"IsEmittedAtUseSite\" flag, to set implicit\nnull checks with it. The liveness analysis then special cases implicit\nnull checks to record environment uses at the location of the actual\ninstruction that will do the implicit null check.\n\nTest: test.py --gcstress\nTest: run-libcore-tests --gcstress\nbug: 111545159\nChange-Id: I3ecea4fe0d7e483e93db83281ca10db47da228c5\n" }, { "commit": "a21eca6a7ddc7f986651261359f821771494c61e", "tree": "5afeffa6ea8384ab50e3e4f87c6b76d195b13134", "parents": [ "d925742e6c52237a262b4430f8b6b679ff1cc80e", "966b46fcba43764267069b6e19bcb2a092260418" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Aug 03 16:41:58 2018 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Fri Aug 03 16:41:58 2018 +0000" }, "message": "Merge \"Revert^2 \"ARM/ARM64: Use introspection marking for JITted code.\"\"" }, { "commit": "9d479254d0dc4043a15ab26205f40439eca15493", "tree": "af8a9c9c6f2c28e723a971c9d39c9d1cebd1f814", "parents": [ "ca20fb6cc4dda392e63bdc8ec9de54d89793373e" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Jul 24 11:35:20 2018 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Aug 03 11:43:30 2018 +0000" }, "message": "Rename type resolution entrypoints.\n\nRename the InitializeType and InitializeTypeAndVerifyAccess\nentrypoints to Resolve* to better match their semantics.\nKeep the InitializeStaticStorage name for now as the most\nappropriate name InitializeType would clash with the old\nname of the ResolveType entrypoint.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing\nChange-Id: Ide55b58c490d085ab37d8536f90699f7ed571d59\n" }, { "commit": "966b46fcba43764267069b6e19bcb2a092260418", "tree": "fe89667cbb09a981e67ebd1196d324038a6413ff", "parents": [ "98afa11c3cd8517bd28d1cad7aacaf0179c905f0" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Aug 03 10:20:19 2018 +0000" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Aug 03 12:35:34 2018 +0100" }, "message": "Revert^2 \"ARM/ARM64: Use introspection marking for JITted code.\"\n\nThis reverts commit 756e722c323c69a7c9891892602730e9c94b78f9.\n\nFix the introspection code to avoid avoid macro instructions\nfor unpoisoning references inside ExactAssemblyScope.\n\nChange-Id: I6effadb84de74aba0236ab84b52ca85770daf5be\nTest: m test-art-host-gtest\nTest: Pixel 2 XL boots.\nTest: m test-art-target-gtest\nTest: testrunner.py --target --optimizing --jit\nTest: ART_HEAP_POISONING\u003dtrue m test-art-target-gtest\nTest: ART_HEAP_POISONING\u003dtrue testrunner.py --target --optimizing --jit\nBug: 36141117\n" }, { "commit": "3232dbb6df866985089b13a36c56e2b39dd473ab", "tree": "055f3e8888bfb3cfd072a981e4733cfaad7b202c", "parents": [ "b27d874ebc0c067d96994a6ebe3c10eaeb2e4a75" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Jul 25 15:42:46 2018 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Aug 03 08:59:07 2018 +0000" }, "message": "Do not save/restore regs in ClinitCheck slow path.\n\nThe entrypoint is kSaveEverything, so the only register that\nneeds to be saved is the argument/return value register.\n\nThe size of the aosp_taimen-userdebug prebuilts:\n - before:\n arm/boot*.oat: 16811692\n arm64/boot*.oat: 19801032\n oat/arm64/services.odex: 20232208\n - after:\n arm/boot*.oat: 16798804 (-12.6KiB, -0.08%)\n arm64/boot*.oat: 19804392 (+3.3KiB, +0.02%)\n oat/arm64/services.odex: 20227784 (-4.3KiB, -0.02%)\nNote that though there is less code, the metadata for the\narm64/boot*.oat outweighs the code size reduction because of\nthe register map encoding as value+shift introduced in\n https://android-review.googlesource.com/695682\nwhich it\u0027s ill-suited for kSaveEverything entrypoints. We\nshould reconsider that encoding.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing\nTest: Pixel 2 XL boots.\nTest: testrunner.py --target --optimizing\nChange-Id: I5cd1deb90332a3b88a0a59d87925c557d9bff1ab\n" }, { "commit": "b27d874ebc0c067d96994a6ebe3c10eaeb2e4a75", "tree": "3084a0b326e98115f5d0621362d5b7c368775669", "parents": [ "8f5992d8c81d4d9a0805c649cdcf859328d5c1b6", "a9f303c089aa2b2fc82d97201352945678ef54ae" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Aug 03 08:57:16 2018 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Fri Aug 03 08:57:16 2018 +0000" }, "message": "Merge \"Rewrite Class init entrypoint to take a Class arg.\"" }, { "commit": "756e722c323c69a7c9891892602730e9c94b78f9", "tree": "a2bf360d95f0aef84f3bce43f43871910a0b7ed9", "parents": [ "450f1d0fa0c40198e63c3e016f02e40ac854b0cb" ], "author": { "name": "Nicolas Geoffray", "email": "ngeoffray@google.com", "time": "Thu Aug 02 17:53:46 2018 +0000" }, "committer": { "name": "Nicolas Geoffray", "email": "ngeoffray@google.com", "time": "Thu Aug 02 17:53:46 2018 +0000" }, "message": "Revert \"ARM/ARM64: Use introspection marking for JITted code.\"\n\nThis reverts commit 450f1d0fa0c40198e63c3e016f02e40ac854b0cb.\n\nReason for revert: breaks poisoning configuration\n\nBug: 36141117\nChange-Id: I198c20ca1db6d7d7602aa5318616e2b149de8772\n" }, { "commit": "a9f303c089aa2b2fc82d97201352945678ef54ae", "tree": "0df0eb5294a3ee72aea8ca670762c02ca9ffa8dd", "parents": [ "1bfd891d06e276d602b4a6ccf1a9f70967195218" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Jul 20 16:43:56 2018 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Aug 02 17:41:21 2018 +0100" }, "message": "Rewrite Class init entrypoint to take a Class arg.\n\nFixes invalid type index being passed to the entrypoint for\nclass init check across dex files when the target type does\nnot have a TypeId in the compilation unit\u0027s DexFile.\n\nThe size of the aosp_taimen-userdebug prebuilts:\n - before:\n arm/boot*.oat: 16782748\n arm64/boot*.oat: 19764400\n oat/arm64/services.odex: 20162432\n - after:\n arm/boot*.oat: 16811692 (+28.3KiB, +0.17%)\n arm64/boot*.oat: 19801032 (+35.8KiB, +0.19%)\n oat/arm64/services.odex: 20232208 (+68.1KiB, +0.35%)\nThis increase comes from doing two runtime calls instead of\none for HLoadClass/kBssEntry that MustGenerateClinitCheck().\n\nTest: Additional test in 476-clinit-inline-static-invoke\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing --jit\nTest: Pixel 2 XL boots.\nTest: testrunner.py --target --optimizing --jit\nTest: testrunner.py --jvm\nBug: 111433619\nChange-Id: I2fccd6944480ab4dac514f60d38e72c1014ae7b2\n" }, { "commit": "450f1d0fa0c40198e63c3e016f02e40ac854b0cb", "tree": "0606a5c722be0d706242c015cb1218021c5c1309", "parents": [ "da6220a29fae95f17edd5374dc6bc2d4870a84da" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Apr 25 17:27:45 2018 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Jul 25 18:34:19 2018 +0100" }, "message": "ARM/ARM64: Use introspection marking for JITted code.\n\nImpact on Golem benchmarks is within noise.\n\nTest: m test-art-host-gtest\nTest: Pixel 2 XL boots.\nTest: testrunner.py --target --optimizing --jit\nBug: 36141117\nChange-Id: Idf5177ee5cd34e2034d298a7907240b3e3e12d82\n" }, { "commit": "bd39d145e4986217bcb8dce1d4a9631d926a2781", "tree": "52dfd3307ab5279e960f9a1bf6e474e47440a3d8", "parents": [ "6f4cf6e8fa15de2f9bf7c6a649ea7a2fabef886a" ], "author": { "name": "Andreas Gampe", "email": "agampe@google.com", "time": "Thu Jul 19 11:14:42 2018 -0700" }, "committer": { "name": "Andreas Gampe", "email": "agampe@google.com", "time": "Thu Jul 19 11:37:05 2018 -0700" }, "message": "ART: Clean up unused using declarations\n\nMake tidy happy, and enable checking.\n\nTest: mmma art\nChange-Id: I9e18e80b3f37dd2aeb8ecd1c25abe4d5cf2f1c45\n" }, { "commit": "59d0872519e8abd1ec17003c17b5c3780ed831df", "tree": "718c0ff45c67d962a822527ae003eb75e720c4a4", "parents": [ "79deffcd99bb3c87ea1c2c36509ef156d4010b2e", "8e524ad3c690c183b1a71f6114796974a107c5dd" ], "author": { "name": "Treehugger Robot", "email": "treehugger-gerrit@google.com", "time": "Thu Jul 19 15:18:47 2018 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Thu Jul 19 15:18:47 2018 +0000" }, "message": "Merge \"Always produce PIC code for AOT compilation.\"" }, { "commit": "8e524ad3c690c183b1a71f6114796974a107c5dd", "tree": "6ee124814e8f33fd4706eb15fcad2cae0019546b", "parents": [ "5991b184a40e4ce181d67d683ced46caa6143b53" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Jul 13 10:27:43 2018 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Mon Jul 16 10:12:51 2018 +0100" }, "message": "Always produce PIC code for AOT compilation.\n\nChange sharpening to use PIC load kinds for AOT compilation\nand add \"Jit\" to the direct addressing load kind names. Use\nPIC code also for the Integer.valueOf() intrinsic codegen.\nRemove all support for non-PIC linker patches.\n\nThe dex2oat --compile-pic option is retained for now but\nignored by the compiler.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing\nTest: Pixel 2 XL boots.\nTest: testrunner.py --target --optimizing\nBug: 77856493\nChange-Id: I54d666f6522f160a1b6ece4045a15d19363acbb6\n" }, { "commit": "f58dc65c52f5e3f15eaaa1e25d7259e64649ade3", "tree": "4485299d9959a658909879b5a234fd807d7627ff", "parents": [ "0b4a439f808f4602c7b97364e49c5546f5100d51" ], "author": { "name": "Evgeny Astigeevich", "email": "evgeny.astigeevich@linaro.org", "time": "Mon Jun 25 17:54:07 2018 +0100" }, "committer": { "name": "Evgeny Astigeevich", "email": "evgeny.astigeevich@linaro.org", "time": "Fri Jul 13 16:30:32 2018 +0100" }, "message": "ART: Delete code optimizing a%1 and a%-1 from InstructionCodeGeneratorARM64\n\nIn InstructionWithAbsorbingInputSimplifier there is code optimizing a%1\nand a%-1. So the code in InstructionCodeGeneratorARM64 optimizing such\ncases can be deleted.\n\nThis patch deletes the code from InstructionCodeGeneratorARM64 and adds\nadditional tests.\n\nTest: 012-math, 014-math3, 411-optimizing-arith, 411-checker-hdiv-hrem-pow2\nTest: 701-easy-div-rem, 442-checker-constant-folding\nTest: test-art-host, test-art-target\nChange-Id: Ib80c0aa4c3e28b07fa79bb43783274c9d7fc456a\n" }, { "commit": "d109e30eab8ba25f8d89be2a83d9036e2d541af2", "tree": "24df91603efe9ce8c4a2efd09ac402aceb10df4e", "parents": [ "c916736ca1e375c276df251446baf2ac8ff3eb13" ], "author": { "name": "Alex Light", "email": "allight@google.com", "time": "Wed Jun 27 10:25:41 2018 -0700" }, "committer": { "name": "Alex Light", "email": "allight@google.com", "time": "Tue Jul 10 08:44:51 2018 -0700" }, "message": "Don\u0027t use StringFactory.newEmptyString in compiled code\n\nWhen compiling debuggable code we would compile a new-instance String\ninstruction into a StringFactory.newEmptyString invoke. This\nadditional invoke could be observed using tracing and is inconsistent\nwith the interpreter, where the string is simply allocated directly.\nIn order to bring these two modes into alignment we added a new\nAllocStringObject quick entrypoint that will be used instead of the\nnormal AllocObject\u003c...\u003e entrypoints when allocating a string. This\nentrypoint directly allocates a new string in the same manner the\ninterpreter does.\n\nNeeds next CL for test to work.\n\nBug: 110884646\nTest: ./test/testrunner/testrunner.py --host --runtime-option\u003d-Xjitthreshold:0 --jit\nTest: Manual inspection of compiled code.\nChange-Id: I7b4b084bcf7dd9a23485c0e3cd2cd04a04b43d3d\n" }, { "commit": "6fd1606a3f3fc2dd53ab4f8b371e420b3e33c74f", "tree": "9f944d267ce3616eb969c027665e4a451c2b3879", "parents": [ "bb089b6bf850c87e0e42917a383cc7298dcb09c5" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Jun 26 11:02:04 2018 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Jun 29 14:39:00 2018 +0100" }, "message": "Implement Integer.valueOf() intrinsic for boot image.\n\nAnd generate only one \"boot image live objects\" array rather\nthan one per boot*.art file.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing\nTest: Pixel 2 XL boots.\nTest: testrunner.py --target --optimizing\nBug: 71526895\nChange-Id: I23af7f47fea5150805f801cd2512f2d152ee5b73\n" }, { "commit": "86c84f7edbe58a7a18ac40abba0ef268ed367271", "tree": "84c8c19fd2f50bd9bc385df3ab2a339d908c0ef3", "parents": [ "de4efa65db3c38df49c680159bedf72afcc1fc71", "f9e90541479502840c19274cf4d5b7ff22e51193" ], "author": { "name": "Treehugger Robot", "email": "treehugger-gerrit@google.com", "time": "Wed Jun 27 13:18:12 2018 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Wed Jun 27 13:18:12 2018 +0000" }, "message": "Merge \"ART: Refactor Int64ConstantFrom to use Int64FromConstant; rename it to Int64FromLocation\"" }, { "commit": "f9e90541479502840c19274cf4d5b7ff22e51193", "tree": "a5ba846c0d493aae7db126ee91d635765250a6e9", "parents": [ "a3234e96206a841c83f9f5bf0d4e14fb07b72a5e" ], "author": { "name": "Evgeny Astigeevich", "email": "evgeny.astigeevich@linaro.org", "time": "Mon Jun 25 13:43:53 2018 +0100" }, "committer": { "name": "Evgeny Astigeevich", "email": "evgeny.astigeevich@linaro.org", "time": "Wed Jun 27 13:24:35 2018 +0100" }, "message": "ART: Refactor Int64ConstantFrom to use Int64FromConstant; rename it to Int64FromLocation\n\nInt64ConstantFrom function duplicates code of the Int64FromConstant. Its\ncode can be replaced with a call: Int64FromConstant(location.getConstant()).\n\nThe patch removes the duplicating code. It also changes the function name to\nInt64FromLocation to be consistent with its usage.\n\nTest: test-art-host, test-art-target\nChange-Id: I5624259aa72523f97ca8fc132a6152f338425c8e\n" }, { "commit": "f07d5617770c37d87447c8bddf105eb0469ab093", "tree": "e6b15b904e591a3b805b292ce9afd9b0cecbf40d", "parents": [ "2dc252e37d4df0c4160cd20b6fc852f5f28b7b87", "a043111e3a2c09b549708a6227a1f54d91da76aa" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Jun 27 12:23:54 2018 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Wed Jun 27 12:23:54 2018 +0000" }, "message": "Merge \"Move instruction_set_ to CompilerOptions.\"" }, { "commit": "a3234e96206a841c83f9f5bf0d4e14fb07b72a5e", "tree": "b71bab3d76984daa0def172ac9811a1f72c758e7", "parents": [ "a84b53a0a5404f4049f24300dbc81bc859b12105" ], "author": { "name": "Evgeny Astigeevich", "email": "evgeny.astigeevich@linaro.org", "time": "Tue Jun 19 23:26:15 2018 +0100" }, "committer": { "name": "Evgeny Astigeevich", "email": "evgeny.astigeevich@linaro.org", "time": "Wed Jun 27 10:03:49 2018 +0100" }, "message": "ARM64: Optimization of HRem and HDiv when a denominator is power of 2\n\nOn ARM64 when a denominator is a power of 2 fewer instructions can be\nused to represent HDiv and HRem. For example, a/2 can be lowered to\nadd+asr; a%2 to cmp+and+csneg. Currently four instructions\nare always used for the division by a power of 2 and five instructions for the\nremainder.\n\nThis patch optimizes the division by 2 (lowering to two instructions),\nthe remainder from the division by 2 (lowering to three instructions)\nand the remainder from the division by a power of 2 (lowering to four\ninstructions).\n\nOn Pixel 2, performance improvements, geomean of diff for a benchmark group (%),\nmax - the maximum seen diff of a single case in a benchmark group, higher better:\nBig core:\nalgorithm 0.664 (max: 1.6)\nintrinsics 5.813 (max: 19.0)\nmicro 4.734 (max: 22.0)\n\nLittle core:\nalgorithm 2.097 (max: 5.4)\nintrinsics 14.610 (max: 27.3)\nmicro 12.687 (max: 35.6)\n\nTest: 012-math, 014-math3, 411-optimizing-arith, 411-checker-hdiv-hrem-pow2\nTest: test-art-host, test-art-target\nChange-Id: Iaaec6dc8fc0ec5df2b2d0e8692d5dea573b8d284\n" }, { "commit": "a043111e3a2c09b549708a6227a1f54d91da76aa", "tree": "393fe11cfceccebf474e4bdf36ff79b70b97f589", "parents": [ "213ee2da6a1c58d0fc12c937bbd9c9974ca00aca" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Mon Jun 25 09:32:54 2018 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Mon Jun 25 18:43:19 2018 +0100" }, "message": "Move instruction_set_ to CompilerOptions.\n\nRemoves CompilerDriver dependency from ImageWriter and\nseveral other classes.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing\nTest: Pixel 2 XL boots.\nTest: m test-art-target-gtest\nTest: testrunner.py --target --optimizing\nChange-Id: I3c5b8ff73732128b9c4fad9405231a216ea72465\n" }, { "commit": "ccfc88af4ab94ff91f9b241d5113dfe7cb1f2b34", "tree": "68198426ee1a8ff1f4ee06f131be1e75fd961968", "parents": [ "ef3f32487a8a9a8b4272ea5ae372642c721ee41a", "eebb821b1adaf2db7662fc1c3ff4e9fcfe59a694" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Jun 22 09:03:30 2018 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Fri Jun 22 09:03:30 2018 +0000" }, "message": "Merge \"Implement Integer.valueOf() intrinsic for PIC.\"" }, { "commit": "bf711e388998c9233b9fc930bcf02511b6943706", "tree": "eeb66747dff7329516d85e29da22795ae51947b7", "parents": [ "5774f57afc997ffe765c32199bd0d5e55d23005a", "878f17d7737a91235013ed16ebe057a12367941b" ], "author": { "name": "Roland Levillain", "email": "rpl@google.com", "time": "Thu Jun 21 16:28:30 2018 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Thu Jun 21 16:28:30 2018 +0000" }, "message": "Merge \"ARM64: Splitting GenerateDivRem* functions into GenerateIntDiv and GenerateIntRem functions\"" }, { "commit": "eebb821b1adaf2db7662fc1c3ff4e9fcfe59a694", "tree": "a3d3cf5f8c20d03fccdc0808537904da63e74938", "parents": [ "7e56bd41cde4e489a11050d9e340bf8b5692d9e8" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Jun 05 14:57:24 2018 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Jun 21 16:12:28 2018 +0100" }, "message": "Implement Integer.valueOf() intrinsic for PIC.\n\nAnd fix the intrinsic for JIT even in case when someone\nmesses up the IntegerCache using reflection. Two cases are\nexposed with a regression test (one that previously failed\nrandomly and one that failed 100%) but other crashes were\npossible; for example, we would need a read barrier for\narray reads when elements are not guaranteed to be in the\nboot image.\n\nThe new approach loads references only from the boot image\nlive objects array which cannot be touched by reflection.\nThe referenced objects and IntegerCache.cache are exposed\nand can lead to weird behavior but not crashes.\n\nOn x86, the pc_relative_fixups_86 actually checks the cache\nan additional time but discrepancies between this check and\nthe location building at the beginning of codegen should be\nOK as the HIsX86ComputeBaseMethodAddress should be added\nfor PIC regardless of whether pc_relative_fixups_86 thinks\nthe method is intrinsified or not.\n\nTest: 717-integer-value-of\nTest: Pixel 2 XL boots.\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing --pictest --npictest\nTest: testrunner.py --host --jit\nTest: testrunner.py --target --optimizing --pictest --npictest\nTest: testrunner.py --target --jit\nBug: 71526895\nChange-Id: I89b3245a62aba22980c86a99e2af480bfa250af1\n" }, { "commit": "2227fe49558c5c5fc4820acb2cf357479e74b518", "tree": "bbfb6546c5da802132405569d2f06b459f12a0c3", "parents": [ "111b895dfaa271d8e9c32d1186615a0b73c106b5" ], "author": { "name": "Petre-Ionut Tudor", "email": "petre-ionut.tudor@linaro.org", "time": "Fri Apr 20 17:12:05 2018 +0100" }, "committer": { "name": "Nicolas Geoffray", "email": "ngeoffray@google.com", "time": "Thu Jun 14 07:42:40 2018 +0000" }, "message": "Small refactor of MIN/MAX compiler code.\n\nIntegrate instruction code generation and location creation with\nHandleBinaryOp. Code generation has been improved for constant\ninputs 0, 1 and -1.\n\nTest: 679-checker-minmax\nTest: test-art-host, test-art-target.\n\nChange-Id: Ib34eb8a4b29d22a2491d21656e1f64011ef9f986\n" }, { "commit": "878f17d7737a91235013ed16ebe057a12367941b", "tree": "f18731e59850baf5d0535a1253643884237e889b", "parents": [ "32e83b36601e080b01712aeb6e9ebaa512eb0f33" ], "author": { "name": "Evgeny Astigeevich", "email": "evgeny.astigeevich@linaro.org", "time": "Fri Jun 01 16:53:58 2018 +0100" }, "committer": { "name": "Evgeny Astigeevich", "email": "evgeny.astigeevich@linaro.org", "time": "Wed Jun 13 21:47:36 2018 +0100" }, "message": "ARM64: Splitting GenerateDivRem* functions into GenerateIntDiv and GenerateIntRem functions\n\nVisitDiv and VisitRem call GenerateDivRemIntegral.\nGenerateDivRemIntegral does not know whether it is invoked for HDiv or\nHRem and has to check this. As a result all GenerateDivRem* functions\nhave such checks. Code for optimizing HRem and HDiv, e.g. a denominator is\npower of 2, can be specific for HRem or HDiv. So having it in\nGenerateDivRem would create issues with code maintenance.\n\nThis patch split GenerateDivRem* functions into GenerateIntDiv* and\nGenerateIntRem. BTW \u0027Integral\u0027 meaning is not \u0027Integer\u0027. So changed it\nas well. It also removes the case \u0027division by 1 or -1\u0027 because the case\nis handled in InstructionSimplifierVisitor. As there is a commonly used\nfunction Int64ConstantFrom(Location) it is used instead of\nInt64FromConstant(HConstant). This removes some code as well.\n\nTest: 012-math, 014-math3, 411-optimizing-arith\nTest: test-art-host, test-art-target\nChange-Id: I972129b24a206c8230d304be551cd2c18dbc7c9c\n" }, { "commit": "4c8e12e66968929b36fac6a2237ca4b04160161e", "tree": "d8bbfd72a978c69ef2eef98c37e7869673c52295", "parents": [ "20c64f8d802cc575cc9a1a1f6c493a611b23e2ee" ], "author": { "name": "Orion Hodson", "email": "oth@google.com", "time": "Fri May 18 08:33:20 2018 +0100" }, "committer": { "name": "Orion Hodson", "email": "oth@google.com", "time": "Mon Jun 11 10:57:17 2018 +0100" }, "message": "ART: Adds an entrypoint for invoke-custom\n\nAdd support for the compiler to call into the runtime for\ninvoke-custom bytecodes.\n\nBug: 35337872\nTest: art/test.py --host -r -t 952\nTest: art/test.py --target --64 -r -t 952\nTest: art/test.py --target --32 -r -t 952\nChange-Id: I821432e7e5248c91b8e1d36c3112974c34171803\n" }, { "commit": "dbaa5c7ba8935cf87ceb40a4054f9842929e9a51", "tree": "5037625c80cb97a0e13026dc450db28e59ff72ca", "parents": [ "51dda39549033b3c50a7fce5522ffc81325db54b" ], "author": { "name": "Orion Hodson", "email": "oth@google.com", "time": "Thu May 10 08:22:46 2018 +0100" }, "committer": { "name": "Orion Hodson", "email": "oth@google.com", "time": "Fri May 11 11:55:30 2018 +0100" }, "message": "ART: Compiler support for const-method-handle\n\nImplemented as a runtime call.\n\nBug: 66890674\nTest: art/test.py --target -r -t 979\nTest: art/test.py --target --64 -r -t 979\nTest: art/test.py --host -r -t 979\nChange-Id: I67f461c819a7d528d7455afda8b4a59e9aed381c\n" }, { "commit": "18259d7fb7164a5e029df4f883b3a79ccc2403e8", "tree": "ba378bfdef4127bb0607215186e3b150fd38bcdf", "parents": [ "922501b4bbf724e4259477a27764291684eedffb" ], "author": { "name": "Orion Hodson", "email": "oth@google.com", "time": "Thu Apr 12 11:18:23 2018 +0100" }, "committer": { "name": "Orion Hodson", "email": "oth@google.com", "time": "Thu May 10 15:04:09 2018 +0100" }, "message": "ART: Compiler support for const-method-type\n\nImplemented as a runtime call.\n\nBug: 66890674\nTest: art/test.py --target -r -t 979\nTest: art/test.py --target --64 -r -t 979\nTest: art/test.py --host -r -t 979\nChange-Id: I4b3d3969d455d0198cfe122eea8abd54e0ea20ee\n" }, { "commit": "7a69505c2f65aef8f016891456c35475f8264d46", "tree": "3926019c2d1407da1ecba7282ede28bb2c430265", "parents": [ "6c3533991522d036cbb5a656c44f63bf633a2925" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Apr 12 10:26:50 2018 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Apr 12 13:09:38 2018 +0100" }, "message": "Rewrite null checks in read barrier introspection thunks.\n\nRely on the implicit null check in the fast path.\n\nTest: Manual; run-test --gdb 160, break in the introspection\n entrypoint, find the mf.testField0000 read barrier\n code in the caller (this one has a stack map for null\n check, most other reads do not need one), break there,\n step into the thunk, overwrite the base register with\n 0 and observe the NPE being thrown. Repeat with --64.\nTest: Pixel 2 XL boots.\nTest: testrunner.py --target --optimizing\nBug: 36141117\nChange-Id: I61f879f22f5697a4108f1021eb0e3add742c8755\n" }, { "commit": "ca1e038eb94694f0f1f94ed3781572411c85d365", "tree": "90ad2c2494821c2f3d904eb42e61fbecd7acaf9d", "parents": [ "3f967b25650e44cd61f5a1112727a8218f2b0804" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Apr 11 09:58:41 2018 +0000" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Apr 11 11:10:10 2018 +0100" }, "message": "Revert^2 \"Compile link-time thunks in codegen.\"\n\nThe linker crash (the reason for revert) is flaky and maybe\nwe shall not see it with this CL now that unrelated source\ncode has changed.\n\nTest: Rely on TreeHugger\nBug: 36141117\nBug: 77581732\n\nThis reverts commit 5806a9ec99b5494b511e84c74f494f0b3a8ebec5.\n\nChange-Id: I3a4a058847dff601681ba391abf45833424fa06d\n" }, { "commit": "5806a9ec99b5494b511e84c74f494f0b3a8ebec5", "tree": "bb50d00ff0890c2e10f351f462b47b56b01e78ea", "parents": [ "c9dd2207dfdab42586b1d6a5e7f11cf2fcea3a7a" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Apr 04 17:23:28 2018 +0000" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Apr 04 17:23:28 2018 +0000" }, "message": "Revert \"Compile link-time thunks in codegen.\"\n\nReason for revert: This caused clang linker crash\nin several branches.\n\nBug: 77581732\n\nThis reverts commit c9dd2207dfdab42586b1d6a5e7f11cf2fcea3a7a.\n\nChange-Id: I1923809083cf41c4f19e3e60df03ae80517aaedb\n" }, { "commit": "c9dd2207dfdab42586b1d6a5e7f11cf2fcea3a7a", "tree": "879df31fd10658093b8931117ee617064ce82519", "parents": [ "30a2d9c61da75359dee4ce90236d19fc6341b07a" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Nov 23 16:05:19 2017 +0000" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Apr 04 10:34:36 2018 +0100" }, "message": "Compile link-time thunks in codegen.\n\nPrepare for experimenting with Baker read barrier marking\nintrospection entrypoints for JIT.\n\nTest: m test-art-host-gtest\nTest: Compare compiled boot*.oat before and after (no diff).\nTest: Pixel 2 XL boots.\nBug: 36141117\nChange-Id: Idb413a31b158db4bf89a8707ea46dd167a06f110\n" }, { "commit": "175e7862dbdb44089ef327fc43ba00c791fd3838", "tree": "7b7ff4327b51b57e47e4b22af8d771edb9d462c1", "parents": [ "77c6fc7341143dd27c74cddd786398688d7b4c91" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Mar 27 09:03:13 2018 +0000" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Mar 27 10:04:25 2018 +0100" }, "message": "Revert^4 \"Compiler changes for bitstring based type checks.\"\n\nDisabled the build time flag. (No image version bump needed.)\n\nBug: 26687569\nBug: 64692057\nBug: 76420366\n\nThis reverts commit 3fbd3ad99fad077e5c760e7238bcd55b07d4c06e.\n\nChange-Id: I5d83c4ce8a7331c435d5155ac6e0ce1c77d60004\n" }, { "commit": "3fbd3ad99fad077e5c760e7238bcd55b07d4c06e", "tree": "e8bc33fa60c38f7e1c85f8d4acf6a738df9b426a", "parents": [ "3f41323cc9da335e9aa4f3fbad90a86caa82ee4d" ], "author": { "name": "Andreas Gampe", "email": "agampe@google.com", "time": "Mon Mar 26 21:14:46 2018 +0000" }, "committer": { "name": "Andreas Gampe", "email": "agampe@google.com", "time": "Mon Mar 26 21:14:46 2018 +0000" }, "message": "Revert^3 \"Compiler changes for bitstring based type checks.\"\n\nThis reverts commit 3f41323cc9da335e9aa4f3fbad90a86caa82ee4d.\n\nReason for revert: Fails sporadically.\n\nBug: 26687569\nBug: 64692057\nBug: 76420366\nChange-Id: I84d1e9e46c58aeecf17591ff71fbac6a1e583909\n" }, { "commit": "3f41323cc9da335e9aa4f3fbad90a86caa82ee4d", "tree": "ce41c620d2cd411da3c20aa95fb9a69328e77c42", "parents": [ "9ec1e24ebc683b15bb9c6db5554ac2ff9458adae" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Mon Feb 12 18:39:15 2018 +0000" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Mar 22 12:12:39 2018 +0000" }, "message": "Revert^2 \"Compiler changes for bitstring based type checks.\"\n\nAdd extra output for debugging failures and re-enable\nthe bitstring type checks.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing --jit\nTest: testrunner.py --host -t 670-bitstring-type-check\nTest: Pixel 2 XL boots.\nTest: testrunner.py --target --optimizing --jit\nTest: testrunner.py --target -t 670-bitstring-type-check\nBug: 64692057\nBug: 26687569\n\nThis reverts commit bff7a52e2c6c9e988c3ed1f12a2da0fa5fd37cfb.\n\nChange-Id: I090e241983f3ac6ed8394d842e17716087d169ac\n" }, { "commit": "9992e095643f6746361df03c4c98e742d9ad5899", "tree": "8abf49af54ee57fc0acebf2a3d9cafd87d6ec48e", "parents": [ "a5867bfeb34529dad71220046e7327cef23af207", "e47f60c482648172334aaca59e6c1ab7a3d42610" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Mar 09 10:17:11 2018 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Fri Mar 09 10:17:11 2018 +0000" }, "message": "Merge \"Retrieve String/Class references from .data.bimg.rel.ro.\"" }, { "commit": "8ba5641ddc43fc13cdb0158bd9f3237c4a90a356", "tree": "4dad508f24b675e87dd31ff26e597289a329c5cc", "parents": [ "66f40dbc3e56c7102820842ec49a55b70cf0e151", "b066d43b1d9184899aff32b1f243d092611ad9c6" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Mar 08 17:45:59 2018 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Thu Mar 08 17:45:59 2018 +0000" }, "message": "Merge \"Load ArtMethod* from .data.bimg.rel.ro entries.\"" }, { "commit": "e47f60c482648172334aaca59e6c1ab7a3d42610", "tree": "ae0672b12a6ad200e1c38962c77bccfc3e5cb531", "parents": [ "b066d43b1d9184899aff32b1f243d092611ad9c6" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Feb 21 13:43:28 2018 +0000" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Mar 08 10:40:12 2018 +0000" }, "message": "Retrieve String/Class references from .data.bimg.rel.ro.\n\nFor PIC AOT-compiled app, use the .data.bimg.rel.ro to load\nthe boot image String/Class references instead of using the\nmmapped boot image ClassTable and InternTable.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing --pictest --npictest\nTest: Pixel 2 XL boots.\nTest: testrunner.py --target --optimizing --pictest --npictest\nBug: 71526895\nChange-Id: Id5703229777aecb589a933a41f92e44d3ec02a3d\n" }, { "commit": "b066d43b1d9184899aff32b1f243d092611ad9c6", "tree": "5409177f52b1f1c648297913cb0e0b2808b9048d", "parents": [ "fe491c7b9cdd64ff4ccc10f6b212cb92a59fc765" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Jan 03 13:14:37 2018 +0000" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Mar 08 10:40:10 2018 +0000" }, "message": "Load ArtMethod* from .data.bimg.rel.ro entries.\n\nIntroduce a new .data.bimg.rel.ro section in oat files where\nwe store offsets of boot image objects from the beginning of\nthe boot image. At runtime we relocate these entries using\nthe actual boot image address to turn offsets to pointers.\n\nUse the .data.bimg.rel.ro to prepare the boot image methods\nused by HInvokeStaticOrDirect for PIC AOT app compilation.\nLoading the ArtMethod* from .data.bimg.rel.ro instead of the\n.bss avoids the initial call to the resolution trampoline.\n\nTest: Additional test in 522-checker-sharpening\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing --pictest --npictest\nTest: Pixel 2 XL boots.\nTest: testrunner.py --target --optimizing --pictest --npictest\nBug: 71526895\nChange-Id: Ie5f5b1f622704877b36730377146e59092e46c0c\n" }, { "commit": "351df3e70521ebbe00ed6c7ac4ea25a0c26f4034", "tree": "8a22bf2eb06d2a5e57c6d6272ac070d368467118", "parents": [ "72efc159bf4e5cde85cd28e78316681effdceb5b" ], "author": { "name": "Aart Bik", "email": "ajcbik@google.com", "time": "Wed Mar 07 11:54:57 2018 -0800" }, "committer": { "name": "Aart Bik", "email": "ajcbik@google.com", "time": "Wed Mar 07 12:00:42 2018 -0800" }, "message": "Minor cleanup of MIN/MAX code.\n\nRationale:\nShare the type dispatching code better.\n\nBug: b/65164101\n\nTest: test-art-host,target\nChange-Id: Ib06c915d570fd0a53f7734cdb316d2d16310db74\n" }, { "commit": "1f8d51bc03cbc607ae32fadf3a90f385adeffb95", "tree": "70e18902051ce47e0d524525b83709efbe6f250f", "parents": [ "7a02c66fd9ed174fc2e49ccc9f582dd661b7de9e" ], "author": { "name": "Aart Bik", "email": "ajcbik@google.com", "time": "Thu Feb 15 10:42:37 2018 -0800" }, "committer": { "name": "Aart Bik", "email": "ajcbik@google.com", "time": "Wed Mar 07 09:22:09 2018 -0800" }, "message": "Introduce MIN/MAX/ABS as HIR nodes.\n\nRationale:\nHaving explicit MIN/MAX/ABS operations (in contrast\nwith intrinsics) simplifies recognition and optimization\nof these common operations (e.g. constant folding, hoisting,\ndetection of saturation arithmetic). Furthermore, mapping\nconditionals, selectors, intrinsics, etc. (some still TBD)\nonto these operations generalizes the way they are optimized\ndownstream substantially.\n\nBug: b/65164101\n\nTest: test-art-host,target\n\nChange-Id: I69240683339356e5a012802f179298f0b04c6726\n" }, { "commit": "3dad341ed027b760d9b4ee402cb2c93ac484a07a", "tree": "a69817c3b2d455273b23d083267ef08c8b9adaee", "parents": [ "d961043ff1dd6fddb68aa90c1f939cfafec24219" ], "author": { "name": "Aart Bik", "email": "ajcbik@google.com", "time": "Wed Feb 28 12:01:46 2018 -0800" }, "committer": { "name": "Aart Bik", "email": "ajcbik@google.com", "time": "Thu Mar 01 13:07:04 2018 -0800" }, "message": "Introduce ABS as HIR nodes.\n\nNOTE: step 1 of 2 for\n\"Introduce MIN/MAX/ABS as HIR nodes.\"\n\nRationale:\nHaving explicit MIN/MAX/ABS operations (in contrast\nwith intrinsics) simplifies recognition and optimization\nof these common operations (e.g. constant folding, hoisting,\ndetection of saturation arithmetic). Furthermore, mapping\nconditionals, selectors, intrinsics, etc. (some still TBD)\nonto these operations generalizes the way they are optimized\ndownstream substantially.\n\nBug: b/65164101\n\nTest: test-art-host,target\nChange-Id: I9c93987197216158ba02c8aca2385086adedabc4\n" }, { "commit": "59eb30f96d87e3e72a060099a292ae14dd5fe1c8", "tree": "49f9334a44a28eef8d0c35c6061c61743d4db4a8", "parents": [ "5919f737facdebbe8b738272e681ae33e085de98" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Feb 20 11:52:34 2018 +0000" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Feb 20 14:27:43 2018 +0000" }, "message": "ART: Clean up patching data in codegens.\n\nReuse PatchInfo\u003c\u003e for additional architectures and make the\nnaming more consistent across architectures. Change the\nDexFile reference to pointer in preparation for patching\nreferences to the upcoming .data.bimg.rel.ro section.\n\nUpdate obsolete comments; instead of referencing dex cache\narrays which were used in the past, reference the .bss and\nthe .data.bimg.rel.ro which shall be used in upcoming CLs.\n\nTest: Rely on TreeHugger.\nChange-Id: I03be4c4118918189e55c62105bb594500c6a42c1\n" }, { "commit": "66c158ef6b2a16257f1590b3ace78848a7c2407b", "tree": "f17f7eee70aa43711c7eb764c1789f4ec17aef37", "parents": [ "92d0c8b68c24a2fa21f95d63a1ff2fb00fdb9aaf" ], "author": { "name": "Aart Bik", "email": "ajcbik@google.com", "time": "Wed Jan 31 12:55:04 2018 -0800" }, "committer": { "name": "Aart Bik", "email": "ajcbik@google.com", "time": "Thu Feb 01 13:26:03 2018 -0800" }, "message": "Clean up signed/unsigned in vectorizer.\n\nRationale:\nCurrently we have some remaining ugliness around signed and unsigned\nSIMD operations due to lack of kUint32 and kUint64 in the HIR. By\n\"softly\" introducing these types, ABS/MIN/MAX/HALVING_ADD/SAD_ACCUMULATE\noperations can solely rely on the packed data types to distinguish\nbetween signed and unsigned operations. Cleaner, and also allows for\nsome code removal in the current loop optimizer.\n\nBug: 72709770\n\nTest: test-art-host test-art-target\nChange-Id: I68e4cdfba325f622a7256adbe649735569cab2a3\n" }, { "commit": "8d728324571b720a952b297787eed70c7a1d1acb", "tree": "f24a774aa56c236f2f756b1b439e4c61830bec56", "parents": [ "0530796b73d0a33a5df27e3a7765c0835994769a" ], "author": { "name": "Nicolas Geoffray", "email": "ngeoffray@google.com", "time": "Thu Jan 18 22:44:32 2018 +0000" }, "committer": { "name": "Nicolas Geoffray", "email": "ngeoffray@google.com", "time": "Mon Jan 29 09:34:05 2018 +0000" }, "message": "Add compiler option for counting hotness in compiled code.\n\nFor eventually easier profiling of boot classpath and\nsystem server.\n\nbug: 30934496\nTest: 674-hotness-compiled\n\nChange-Id: I0f63c644527b74f6ef2649f481c2a1c731bb9f21\n" }, { "commit": "bff7a52e2c6c9e988c3ed1f12a2da0fa5fd37cfb", "tree": "e281a8dde61e396ed5f20c31d41086b1b1b18389", "parents": [ "83af48e9f4cdfcf3f0069c63561bab4c176bd2f1" ], "author": { "name": "Nicolas Geoffray", "email": "ngeoffray@google.com", "time": "Thu Jan 25 13:33:07 2018 +0000" }, "committer": { "name": "Nicolas Geoffray", "email": "ngeoffray@google.com", "time": "Thu Jan 25 15:05:16 2018 +0000" }, "message": "Revert \"Compiler changes for bitstring based type checks.\"\n\nBug: 64692057\nBug: 71853552\nBug: 26687569\n\nThis reverts commit eb0ebed72432b3c6b8c7b38f8937d7ba736f4567.\n\nChange-Id: I7daeaa077960ba41b2ed42bc47f17501621be4be\n" }, { "commit": "eb0ebed72432b3c6b8c7b38f8937d7ba736f4567", "tree": "74d95eb4bfbf01ef6fd3a68695f5d7cec69338d7", "parents": [ "e57043081e6b091a9fd23a84043373148ae72f1f" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Jan 10 18:26:38 2018 +0000" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Jan 23 13:02:59 2018 +0000" }, "message": "Compiler changes for bitstring based type checks.\n\nWe guard the use of this feature with a compile-time flag,\nset to true in this CL.\n\nBoot image size for aosp_taimen-userdebug in AOSP master:\n - before:\n arm boot*.oat: 63604740\n arm64 boot*.oat: 74237864\n - after:\n arm boot*.oat: 63531172 (-72KiB, -0.1%)\n arm64 boot*.oat: 74135008 (-100KiB, -0.1%)\n\nThe new TypeCheckBenchmark yields the following changes\nusing the little cores of taimen fixed at 1.4016GHz:\n 32-bit 64-bit\n timeCheckCastLevel1ToLevel1 11.48-\u003e15.80 11.47-\u003e15.78\n timeCheckCastLevel2ToLevel1 15.08-\u003e15.79 15.08-\u003e15.79\n timeCheckCastLevel3ToLevel1 19.01-\u003e15.82 17.94-\u003e15.81\n timeCheckCastLevel9ToLevel1 42.55-\u003e15.79 42.63-\u003e15.81\n timeCheckCastLevel9ToLevel2 39.70-\u003e14.36 39.70-\u003e14.35\n timeInstanceOfLevel1ToLevel1 13.74-\u003e17.93 13.76-\u003e17.95\n timeInstanceOfLevel2ToLevel1 17.02-\u003e17.95 16.99-\u003e17.93\n timeInstanceOfLevel3ToLevel1 24.03-\u003e17.95 24.45-\u003e17.95\n timeInstanceOfLevel9ToLevel1 47.13-\u003e17.95 47.14-\u003e18.00\n timeInstanceOfLevel9ToLevel2 44.19-\u003e16.52 44.27-\u003e16.51\nThis suggests that the bitstring typecheck should not be\nused for exact type checks which would be equivalent to the\n\"Level1ToLevel1\" benchmark. Whether the implementation is\na beneficial replacement for the kClassHierarchyCheck and\nkAbstractClassCheck on average depends on how many levels\nfrom the target class (or Object for a negative result) is\na typical object\u0027s class.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing --jit\nTest: testrunner.py --host -t 670-bitstring-type-check\nTest: Pixel 2 XL boots.\nTest: testrunner.py --target --optimizing --jit\nTest: testrunner.py --target -t 670-bitstring-type-check\nBug: 64692057\nBug: 71853552\nBug: 26687569\nChange-Id: I538d7e036b5a8ae2cc3fe77662a5903d74854562\n" }, { "commit": "a8b8e9b12a9740d71cff2fa65d47825b74f72c37", "tree": "301275759cf145711175992a503fcc7d710c2d2f", "parents": [ "6d4c343ee5db18f039aeb3e07ff8d3c1fd37c3a0" ], "author": { "name": "Aart Bik", "email": "ajcbik@google.com", "time": "Tue Jan 09 11:01:02 2018 -0800" }, "committer": { "name": "Aart Bik", "email": "ajcbik@google.com", "time": "Tue Jan 16 09:44:28 2018 -0800" }, "message": "Improve code sinking near \"always throwing\" method calls\n\nRationale:\nWith simple dex bytecode analysis, the inliner marks methods\nthat always throw to help subsequent code sinking. This reduces\noverhead of non-nullable enforcing calls found in e.g the Kotlin\nruntime library (1%-2% improvement on tree microbenchmark, about\n5% on Denis\u0027 benchmark).\n\nTest: test-art-host test-art-target\n\nChange-Id: I45348f049721476828eb5443738021720d2857c0\n" }, { "commit": "dc682aa9d0eae1a851af059434adb6f6cf8f06f8", "tree": "f93f00493ee5887b05b42a6a5dd99eb6794daad4", "parents": [ "d6b7e8c63f8eca25460f56f66dcae15eaa897ff0" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Jan 04 18:42:57 2018 +0000" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Jan 10 14:30:26 2018 +0000" }, "message": "Use 28 bits for type check bit string.\n\nAnd reverse the order of fields in the Class::status_. This\navoids generated code size increase:\n - ClassStatus in high bits allows class initialization\n check using \"status_high_byte \u003c (kInitialized \u003c\u003c 4)\"\n which is unaffected by the low 4 bits of LHS instead of\n needing to extract the status bits,\n - the type check bit string in the bottom bits instead of\n somewehere in the middle allows the comparison on ARM\n to be done using the same code size as with the old\n layout in most cases (except when the compared value is\n 9-16 bits and not a modified immediate: 2 bytes less for\n 9-12 bits and sometimes 2 bytes more for 13-16 bits; the\n latter could be worked around using LDRH if the second\n character\u0027s boundary is at 16 bits).\n\nAdd one of the extra bits to the 2nd character to push its\nboundary to 16 bits so that we can test an implementation\nusing 16-bit loads in a subsequent CL, arbitrarily add the\nother three bits to the 3rd character. This CL is only\nabout making those bits available and allowing testing, the\ndetermination of how to use the additonal bits for the best\nimpact (whether to have a 4th character or distribute them\ndifferently among the three characters) shall be done later.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing\nTest: Pixel 2 XL boots.\nTest: testrunner.py --target --optimizing\nBug: 64692057\nChange-Id: I38c59837e3df3accb813fb1e04dc42e9afcd2d73\n" }, { "commit": "2c64a837e62c2839521c89060b5bb0dcb237ddda", "tree": "65475ed2e313ff17354e741bac7e9c85739b8b95", "parents": [ "6cd0005698181e4cef2247b632d396e605d58fa3" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Jan 04 11:31:56 2018 +0000" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Jan 05 11:27:50 2018 +0000" }, "message": "Change ClassStatus to fit into 4 bits.\n\nIn preparation for extending the type check bit string from\n24 to 28 bits, rewrite ClassStatus to fit into 4 bits. Also\nperform a proper cleanup of the ClassStatus, i.e. change it\nto an enum class, remove the \"Status\" word from enumerator\nnames, replace \"Max\" with \"Last\" in line with other\nenumerations and remove aliases from mirror::Class.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing\nTest: Pixel 2 XL boots.\nTest: testrunner.py --target --optimizing\nBug: 64692057\nBug: 65318848\nChange-Id: Iec1610ba5dac2c527b36c12819f132e1a77f2d45\n" }, { "commit": "8758454d380a2b0de1f4a99e9623cfac5460ccdf", "tree": "0c23241949fc3ab94c826a18ffa2185218521bd4", "parents": [ "e619f6c51d0ff99a00c63d3d092e0132522e8fc8" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Dec 12 17:47:52 2017 +0000" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Dec 14 16:05:15 2017 +0000" }, "message": "Clean up InstanceOf/CheckCast.\n\nAvoid read barriers for boot image class InstanceOf. Boot\nimage classes are non-moveable, so comparing them against\nfrom-space and to-space reference yields the same result.\n\nChange the notion of a \"fatal\" type check slow path to mean\nthat the runtime call shall not return by normal path, i.e.\n\"fatal\" now includes certainly throwing in a try-block. This\navoids unnecessary code to restore registers and jump back.\nFor boot image classes the CheckCast comparisons do not need\nread barriers (for the same reason as for InstanceOf), so we\nshall not have any false negatives and can treat the check\u0027s\nslow paths as final in the same cases as in non-CC configs.\n\nBoot image size for aosp_taimen-userdebug in AOSP master:\n - before:\n arm boot*.oat: 37075460\n arm64 boot*.oat: 43431768\n - after:\n arm boot*.oat: 36894292 (-177KiB, -0.5%)\n arm64 boot*.oat: 43201256 (-225KiB, -0.5%)\n\nAlso remove some obsolete helpers from CodeGenerator.\n\nTest: Additional test in 603-checker-instanceof.\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing\nTest: Pixel 2 XL boots.\nTest: testrunner.py --target --optimizing\nBug: 12687968\nChange-Id: Ib1381084e46a10e70320dcc618f0502ad725f0b8\n" }, { "commit": "f3c52b42a035902245d00a619fed0275afb063d2", "tree": "c46dab07826be55e9ca92ab301eed586c2f307ca", "parents": [ "b360bff818ad0bf59668cd2bebaaeeaa8a3b5dfe" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Nov 17 17:32:12 2017 +0000" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Nov 21 17:09:44 2017 +0000" }, "message": "Fill Class and String .bss slots in runtime.\n\nShift the responsibility for filling Class and String .bss\nslots from compiled code to runtime. This reduces the size\nof the compiled code.\n\nMake oatdump list .bss slot mappings (ArtMethod, Class and\nString) for each dex file.\n\naosp_taimen-userdebug boot image size:\n - before:\n arm boot*.oat: 36534524\n arm64 boot*.oat: 42723256\n - after:\n arm boot*.oat: 36431448 (-101KiB, -0.3%)\n arm64 boot*.oat: 42645016 (-76KiB, -0.2%)\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing\nTest: Pixel 2 XL boots.\nTest: testrunner.py --target --optimizing\nTest: m dump-oat, manually inspect output.\nBug: 65737953\nChange-Id: I1330d070307410107e12c309d4c7f8121baba83c\n" }, { "commit": "cefd676fb79d225fcd7e8e8c0ef141d70a2f45b8", "tree": "be625fa0bc1255cbea5c5066dd878fb0b2a1c454", "parents": [ "dbc26ad5e8ded15688d20a39344c677077311279", "86083f7cd118f3d6c757191e83b4e4abaabdc5d7" ], "author": { "name": "Treehugger Robot", "email": "treehugger-gerrit@google.com", "time": "Wed Nov 08 03:26:30 2017 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Wed Nov 08 03:26:30 2017 +0000" }, "message": "Merge \"runtime: Bitstring implementation for subtype checking (4/4).\"" }, { "commit": "86083f7cd118f3d6c757191e83b4e4abaabdc5d7", "tree": "8e5b81ae0d09d41bfd90284a1b6b16b2332435e5", "parents": [ "495e783c9180c3fc033ce459ee0a783e633f7754" ], "author": { "name": "Igor Murashkin", "email": "iam@google.com", "time": "Fri Oct 27 10:59:04 2017 -0700" }, "committer": { "name": "Igor Murashkin", "email": "iam@google.com", "time": "Tue Nov 07 14:47:51 2017 -0800" }, "message": "runtime: Bitstring implementation for subtype checking (4/4).\n\nIntegrate the previous CLs into ART Runtime. Subsequent CLs to add\noptimizing compiler support.\n\nUse spare 24-bits from \"Class#status_\" field to\nimplement faster subtype checking in the runtime. Does not incur any extra memory overhead,\nand (when in compiled code) this is always as fast or faster than the original check.\n\nThe new subtype checking is O(1) of the form:\n\n src \u003c: target :\u003d\n (*src).status \u003e\u003e #imm_target_mask \u003d\u003d #imm_target_shifted\n\nBased on the original prototype CL by Zhengkai Wu:\nhttps://android-review.googlesource.com/#/c/platform/art/+/440996/\n\nTest: art/test.py -b -j32 --host\nBug: 64692057\nChange-Id: Iec3c54af529055a7f6147eebe5611d9ecd46942b\n" }, { "commit": "33bff25bcd7a02d35c54f63740eadb1a4833fc92", "tree": "553db4f60878acf2a0fa7036a739d406df9a29b7", "parents": [ "321b3ca9a36d769283c64d4bdee0798db80af524" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Nov 01 14:35:42 2017 +0000" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Nov 02 10:11:02 2017 +0000" }, "message": "ART: Make InstructionSet an enum class and add kLast.\n\nAdding InstructionSet::kLast shall make it easier to encode\nthe InstructionSet in fewer bits using BitField\u003c\u003e. However,\nintroducing `kLast` into the `art` namespace is not a good\nidea, so we change the InstructionSet to an enum class.\nThis also uncovered a case of InstructionSet::kNone being\nerroneously used instead of vixl32::Condition::None(), so\nit\u0027s good to remove `kNone` from the `art` namespace.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing\nChange-Id: I6fa6168dfba4ed6da86d021a69c80224f09997a6\n" }, { "commit": "96b7474ebf313abdaf99e657e4ba9758e2467fb1", "tree": "175c7a007ef033c3fccc1e74da2e4424e74da336", "parents": [ "6247604714ae7fb2b64451b225cc0ecd3d4b716f", "174b2e27ebf933b80f4e8b64b4b024ab4306aaac" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Oct 20 10:31:25 2017 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Fri Oct 20 10:31:25 2017 +0000" }, "message": "Merge \"Use ScopedArenaAllocator for code generation.\"" }, { "commit": "61b922847403ac0e74b6477114c81a28ac2e01a0", "tree": "02674602fb2592f758f51389b3c7b276ab4df3ee", "parents": [ "6783118d2ad9d759f0617b1219a9e29a10a569f7" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Oct 11 13:23:17 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Oct 18 15:52:51 2017 +0100" }, "message": "ART: Introduce Uint8 loads in compiled code.\n\nSome vectorization patterns are not recognized anymore.\nThis shall be fixed later.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing\nTest: testrunner.py --target --optimizing on Nexus 5X\nTest: Nexus 5X boots.\nBug: 23964345\nBug: 67935418\nChange-Id: I587a328d4799529949c86fa8045c6df21e3a8617\n" }, { "commit": "174b2e27ebf933b80f4e8b64b4b024ab4306aaac", "tree": "968cdd8d7fd68571115db77cc288807c3b257911", "parents": [ "6783118d2ad9d759f0617b1219a9e29a10a569f7" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Oct 12 13:34:49 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Oct 17 11:12:08 2017 +0100" }, "message": "Use ScopedArenaAllocator for code generation.\n\nReuse the memory previously allocated on the ArenaStack by\noptimization passes.\n\nThis CL handles only the architecture-independent codegen\nand slow paths, architecture-dependent codegen allocations\nshall be moved to the ScopedArenaAllocator in a follow-up.\n\nMemory needed to compile the two most expensive methods for\naosp_angler-userdebug boot image:\n BatteryStats.dumpCheckinLocked() : 19.6MiB -\u003e 18.5MiB (-1189KiB)\n BatteryStats.dumpLocked(): 39.3MiB -\u003e 37.0MiB (-2379KiB)\n\nAlso move definitions of functions that use bit_vector-inl.h\nfrom bit_vector.h also to bit_vector-inl.h .\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing\nBug: 64312607\nChange-Id: I84688c3a5a95bf90f56bd3a150bc31fedc95f29c\n" }, { "commit": "bea75ff0835324076fed6ff5d443b9e02c65d223", "tree": "61ae2e8fe552938fcae1e277f51823ba2a4f6e74", "parents": [ "567563a9c6ccc06c2c9889d1c3c4feaa3c2b2dab" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Oct 11 20:39:54 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Oct 12 10:58:02 2017 +0100" }, "message": "Fix using LiveIntervals beyond their lifetime.\n\nFixes a bug introduced by\n https://android-review.googlesource.com/504041\n\nTest: test-art-host-gtest\nTest: testrunner.py --host --optimizing\nBug: 64312607\nChange-Id: I7fd2d55c2a657f736eaed7c94c41d1237ae2ec0b\n" }, { "commit": "ca6fff898afcb62491458ae8bcd428bfb3043da1", "tree": "195a6b16d3a4b34acc2faf91ce56f448efb15e07", "parents": [ "aa7273e56fbafc2692c8d20a31b50d2f4bdd2aa1" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Oct 03 14:49:14 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Oct 06 17:53:50 2017 +0100" }, "message": "ART: Use ScopedArenaAllocator for pass-local data.\n\nPasses using local ArenaAllocator were hiding their memory\nusage from the allocation counting, making it difficult to\ntrack down where memory was used. Using ScopedArenaAllocator\nreveals the memory usage.\n\nThis changes the HGraph constructor which requires a lot of\nchanges in tests. Refactor these tests to limit the amount\nof work needed the next time we change that constructor.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host\nTest: Build with kArenaAllocatorCountAllocations \u003d true.\nBug: 64312607\nChange-Id: I34939e4086b500d6e827ff3ef2211d1a421ac91a\n" }, { "commit": "d5d2f2ce627aa0f6920d7ae05197abd1a396e035", "tree": "e8e780780c832e3614a22438a23fb60ee4960ca3", "parents": [ "efac0df8c738764823c637deeca1f3be33912064" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Sep 26 12:37:26 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Oct 03 10:40:51 2017 +0100" }, "message": "ART: Introduce Uint8 compiler data type.\n\nThis CL adds all the necessary codegen for the Uint8 type\nbut does not add code transformations that use that code.\nVectorization codegens are modified to use Uint8 as the\npacked type when appropriate. The side effects are now\ndisconnected from the instruction\u0027s type after the graph has\nbeen built to allow changing HArrayGet/H*FieldGet/HVecLoad\nto use a type different from the underlying field or array.\n\nNote: HArrayGet for String.charAt() is modified to have\nno side effects whatsoever; Strings are immutable.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host --optimizing --jit\nTest: testrunner.py --target --optimizing on Nexus 6P\nTest: Nexus 6P boots.\nBug: 23964345\nChange-Id: If2dfffedcfb1f50db24570a1e9bd517b3f17bfd0\n" }, { "commit": "0ebe0d83138bba1996e9c8007969b5381d972b32", "tree": "a5ee66ebc5b587ade97e56ac8fc7d832fbbed4af", "parents": [ "e1e347dace0ded83774999bb26c37527dcdb1d5a" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Sep 21 22:50:39 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Mon Sep 25 15:45:01 2017 +0100" }, "message": "ART: Introduce compiler data type.\n\nReplace most uses of the runtime\u0027s Primitive in compiler\nwith a new class DataType. This prepares for introducing\nnew types, such as Uint8, that the runtime does not need\nto know about.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host\nBug: 23964345\nChange-Id: Iec2ad82454eec678fffcd8279a9746b90feb9b0c\n" }, { "commit": "d8dbc8da0e5cc6b5c2176ce2d3877e6194d72c0c", "tree": "af6e9fb02471d75ebdea46190a0aa3e9dbdb892d", "parents": [ "93780a60090356921b844dbefdc13442c9f18b52" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Sep 20 13:37:47 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Sep 20 15:55:10 2017 +0100" }, "message": "Refactor compiled_method.h .\n\nMove LinkerPatch to compiler/linker/linker_patch.h .\nMove SrcMapElem to compiler/debug/src_map_elem.h .\nIntroduce compiled_method-inl.h to reduce the number\nof `#include`s in compiled_method.h .\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host\nChange-Id: Id211cdf94a63ad265bf4709f1a5e06dffbe30f64\n" }, { "commit": "09659c22dc2f2c85a0ade965d1fc5160944b8692", "tree": "66fd5729395d27569c4d9d255a5ce9b44cb000bf", "parents": [ "4d159807a4854caa6396b708a38bbd6fa49d736f" ], "author": { "name": "Andreas Gampe", "email": "agampe@google.com", "time": "Mon Sep 18 18:23:32 2017 -0700" }, "committer": { "name": "Andreas Gampe", "email": "agampe@google.com", "time": "Tue Sep 19 10:26:51 2017 -0700" }, "message": "ART: Remove heap poisoning from globals.h\n\nRemove mostly-unused include and move it to its users.\n\nTest: m\nChange-Id: Ibb40f919db64a490290c6e18cf1123aaf44199fc\n" }, { "commit": "fc8b422c286501346b5b797420fb616aaa5e952a", "tree": "61c857a895cdad9ce387a899f92824701259df32", "parents": [ "7090dfe84f78b1928fcbdfd664d0dd9ea52633ff" ], "author": { "name": "Mathieu Chartier", "email": "mathieuc@google.com", "time": "Sun Sep 17 13:44:24 2017 -0700" }, "committer": { "name": "Mathieu Chartier", "email": "mathieuc@google.com", "time": "Mon Sep 18 10:57:06 2017 -0700" }, "message": "Clean up AtomicDexRefMap\n\nMake ClassReference, TypeReference, and MethodReference extend\nDexFileReference. This enables using all of these types as the key\nfor AtomicDexRefMap.\n\nTest: test-art-host\nBug: 63851220\nBug: 63756964\n\nChange-Id: Ida3c94cadb53272cb5057e5cebc5971c1ab4d366\n" }, { "commit": "94ec2db21332ee1dcdbbf254b99a9a999a304fe0", "tree": "6ced7e596731b61f95a3693f336527f55ea3cf3a", "parents": [ "6cfbdbc359ec5414d3e49f70d28f8c0e65b98d63" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Sep 06 17:21:03 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Mon Sep 11 15:12:51 2017 +0100" }, "message": "Use mmapped boot image class table for PIC app HLoadClass.\n\nImplement new HLoadClass load kind for boot image classes\nreferenced by PIC-compiled apps (i.e. prebuilts) that uses\nPC-relative load from a boot image ClassTable mmapped into\nthe apps .bss. This reduces the size of the PIC prebuilts\nthat reference boot image classes compared to the kBssEntry\nas we can completely avoid the slow path and stack map\nunless we need to do the class initialization check.\n\nPrebuilt services.odex for aosp_angler-userdebug (arm64):\n - before: 20312800\n - after: 19775352 (-525KiB)\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host\nTest: testrunner.py --host --pictest\nTest: testrunner.py --target on Nexus 6P.\nTest: testrunner.py --target --pictest on Nexus 6P.\nTest: Nexus 6P boots.\nBug: 31951624\nChange-Id: I13adb19a1fa7d095a72a41f09daa6101876e77a8\n" }, { "commit": "6cfbdbc359ec5414d3e49f70d28f8c0e65b98d63", "tree": "f92b309ddc43c2254b6067346a653170fbbf7316", "parents": [ "0f3c7003e08a42a4ed8c9f8dfffb1bee1118de59" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Jul 25 13:26:39 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Sep 07 17:52:35 2017 +0100" }, "message": "Use mmapped boot image intern table for PIC app HLoadString.\n\nImplement new HLoadString load kind for boot image strings\nreferenced by PIC-compiled apps (i.e. prebuilts) that uses\nPC-relative load from a boot image InternTable mmapped into\nthe apps .bss. This reduces the size of the PIC prebuilts\nthat reference boot image strings compared to the kBssEntry\nas we can completely avoid the slow path and stack map.\n\nWe separate the InternedStrings and ClassTable sections of\nthe boot image (.art) file from the rest, aligning the\nstart of the InternedStrings section to a page boundary.\nThis may actually increase the size of the boot image file\nby a page but it also allows mprotecting() these tables as\nread-only. The ClassTable section is included in\nanticipation of a similar load kind for HLoadClass.\n\nPrebuilt services.odex for aosp_angler-userdebug (arm64):\n - before: 20862776\n - after: 20308512 (-541KiB)\nNote that 92KiB savings could have been achieved by simply\navoiding the read barrier, similar to the HLoadClass flag\nIsInBootImage(). Such flag is now unnecessary.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host\nTest: testrunner.py --host --pictest\nTest: testrunner.py --target on Nexus 6P.\nTest: testrunner.py --target --pictest on Nexus 6P.\nTest: Nexus 6P boots.\nBug: 31951624\nChange-Id: I5f2bf1fc0bb36a8483244317cfdfa69e192ef6c5\n" }, { "commit": "2b03a1f24600c8c9558fb13d3f8bca1ef0f8ee40", "tree": "4d5fb728acd7f98f8949c84364375a804ff0ba63", "parents": [ "461ec567f16039374dff35e2f3b808986c100249" ], "author": { "name": "Roland Levillain", "email": "rpl@google.com", "time": "Tue Jun 06 16:09:59 2017 +0100" }, "committer": { "name": "Roland Levillain", "email": "rpl@google.com", "time": "Thu Aug 10 18:17:47 2017 +0100" }, "message": "Instrument ARM64 generated code to check the Marking Register.\n\nGenerate run-time code in the Optimizing compiler checking that\nthe Marking Register\u0027s value matches `self.tls32_.is.gc_marking`\nin debug mode (on target; and on host with JIT, or with AOT when\ncompiling the core image). If a check fails, abort.\n\nTest: m test-art-target\nTest: m test-art-target with tree built with ART_USE_READ_BARRIER\u003dfalse\nTest: ARM64 device boot test with libartd.\nBug: 37707231\nChange-Id: Ie9b322b22b3d26654a06821e1db71dbda3c43061\n" }, { "commit": "8cf9cb386cd9286d67e879f1ee501ec00d72a4e1", "tree": "88e86e214b425e444760fe4e0ffeee677e1558a2", "parents": [ "914b7b6a6c9f399b26b41e9160e9871ef749e0db" ], "author": { "name": "Andreas Gampe", "email": "agampe@google.com", "time": "Wed Jul 19 09:28:38 2017 -0700" }, "committer": { "name": "Andreas Gampe", "email": "agampe@google.com", "time": "Mon Jul 24 16:07:10 2017 -0700" }, "message": "ART: Include cleanup\n\nLet clang-format reorder the header includes.\n\nDerived with:\n\n* .clang-format:\n BasedOnStyle: Google\n IncludeIsMainRegex: \u0027(_test|-inl)?$\u0027\n\n* Steps:\n find . -name \u0027*.cc\u0027 -o -name \u0027*.h\u0027 | xargs sed -i.bak -e \u0027s/^#include/ #include/\u0027 ; git commit -a -m \u0027ART: Include cleanup\u0027\n git-clang-format -style\u003dfile HEAD^\n manual inspection\n git commit -a --amend\n\nTest: mmma art\nChange-Id: Ia963a8ce3ce5f96b5e78acd587e26908c7a70d02\n" }, { "commit": "97c46466aea25ab63a99b3d1afc558f0d9f55abb", "tree": "afd225f51d28a77329bc2590a025400e088f260c", "parents": [ "00cca3a275562d110a8b35094b9b12fac37f67ab" ], "author": { "name": "Roland Levillain", "email": "rpl@google.com", "time": "Thu May 11 14:04:03 2017 +0100" }, "committer": { "name": "Roland Levillain", "email": "rpl@google.com", "time": "Tue Jul 11 17:43:27 2017 +0100" }, "message": "Introduce a Marking Register in ARM64 code generation.\n\nWhen generating code for ARM64, maintain the status of\nThread::Current()-\u003eGetIsGcMarking() in register X20,\ndubbed MR (Marking Register), and check the value of that\nregister (instead of loading and checking a read barrier\nmarking entrypoint) in read barriers.\n\nTest: m test-art-target\nTest: m test-art-target with tree built with ART_USE_READ_BARRIER\u003dfalse\nTest: ARM64 device boot test\nBug: 37707231\nChange-Id: Ibe9bc5c99a2176b0a0476e9e9ad7fcc9f745017b\n" }, { "commit": "6fda42718a348cfb758d8714e223cab7e855765b", "tree": "d6f0f4d0c1ca2eec26a56e2aadaf00d3205b70a0", "parents": [ "a0e63dfbfe2f2513a709e94b8a1ac17418396fdf" ], "author": { "name": "Nicolas Geoffray", "email": "ngeoffray@google.com", "time": "Mon Jun 26 09:12:45 2017 +0100" }, "committer": { "name": "Nicolas Geoffray", "email": "ngeoffray@google.com", "time": "Mon Jun 26 10:59:17 2017 +0100" }, "message": "Fix braino when handling branches fallthrough in arm backend.\n\nbug: 62210114\nTest: 657-branches\nChange-Id: I753a9a57e404c792cd4375ea66c91839684bdee2\n" }, { "commit": "5678db5b3a0275d04bc610236f89fac9f76b5b1e", "tree": "efc4ffe5d59a0c6c5f4c15a886459962d24de4aa", "parents": [ "83b140474aa1759739c8ee4464bf226c4fa0f6d7" ], "author": { "name": "Andreas Gampe", "email": "agampe@google.com", "time": "Thu Jun 08 14:11:18 2017 -0700" }, "committer": { "name": "Andreas Gampe", "email": "agampe@google.com", "time": "Thu Jun 08 14:11:18 2017 -0700" }, "message": "ART: Refactor bit_utils and stl_util\n\nMove iterator code from bit_utils.h into bit_utils_iterator.h. Move\nIdentity into stl_util_identity.h. Remove now unnecessary includes,\nand fix up transitive users.\n\nTest: m\nChange-Id: Id1ce9cda66827c5d00584f39ed310b6b37629906\n" }, { "commit": "0eb882bfc5d260e8014c26adfda11602065aa5d8", "tree": "e66dbebfb1e9a254c20954a2f2f98541aebfd5af", "parents": [ "b5f5d746ac3f2c3088292395603cb1470e7749d2" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Mon May 15 13:39:18 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Jun 07 15:39:24 2017 +0100" }, "message": "Use ArtMethod* .bss entries for HInvokeStaticOrDirect.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host\nTest: testrunner.py --target\nTest: Nexus 6P boots.\nTest: Build aosp_mips64-userdebug.\nBug: 30627598\nChange-Id: I0e54fdd2e91e983d475b7a04d40815ba89ae3d4f\n" }, { "commit": "3c8a91250b3e4e87548ec16bf1ab1ea46dbb84a4", "tree": "b5da100b358d1335eab403372e4f616c5c2d607c", "parents": [ "0a87f31513e5f9da27856af054d2241452898b22", "e7197bf7d58c705a048e13e241d7ca320502cd40" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Jun 07 10:38:10 2017 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Wed Jun 07 10:38:11 2017 +0000" }, "message": "Merge \"Replace invoke kind kDexCacheViaMethod with kRuntimeCall.\"" }, { "commit": "e7197bf7d58c705a048e13e241d7ca320502cd40", "tree": "496032b40145660b40002b9d5b7a78f1c2eeb44f", "parents": [ "4ee8e291a7d5b7b98f35f495eb97705836910871" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Jun 02 17:00:23 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Jun 06 15:26:42 2017 +0100" }, "message": "Replace invoke kind kDexCacheViaMethod with kRuntimeCall.\n\nIn preparation for replacing the dex cache method array\nwith a hash-based array, get rid of one unnecessary use.\nThis method load kind is currently used only on mips for\nirreducible loops and OSR, so this should have no impact\non x86/x86-64/arm/arm64.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host\nTest: Repeat the above tests with manually changing\n kDexCachePcRelative to kRuntimeCall in sharpening.cc.\n (Ignore failures in 552-checker-sharpening.)\nBug: 30627598\nChange-Id: Ifce42645f2dcc350bbb88c2f4642e88fc5f98152\n" }, { "commit": "ab6393400f0dd213d335092c6e83f6a8743f00c2", "tree": "43f35b71321e7b96af7ad5ddc557638e365d2f06", "parents": [ "2c97600c1107931825bf9f7f25517e89b7210ab4", "d254f5c0d7b43397e8b8885a56ec4d36e9b61602" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Mon Jun 05 12:11:07 2017 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Mon Jun 05 12:11:09 2017 +0000" }, "message": "Merge \"Revert \"ART: Reference.getReferent intrinsic for arm and arm64\"\"" }, { "commit": "d254f5c0d7b43397e8b8885a56ec4d36e9b61602", "tree": "ef645025a42f88a2c9eb0ab9483ff519886f0a4c", "parents": [ "cff1b21b3e19c5d3a2d726fdb60dacd7de2a6f0d" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Jun 02 15:18:36 2017 +0000" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Jun 02 16:24:33 2017 +0100" }, "message": "Revert \"ART: Reference.getReferent intrinsic for arm and arm64\"\n\nReverting because GenerateCalleeMethodStaticOrDirectCall()\nprevents replacing kDexCacheViaMethod with kRuntimeCall\nwhere we would not retrieve the target method at all and\nleave the runtime to retrieve and call it just like for\nunresolved methods.\n\nThe intrinsic should be re-implemented by loading the\nflags through HLoadClass.\n\nNote that the intrinsic was unimplemented for CC and a bit\nbroken for non-CC, using LDR instead of LDRB for loading\nthe flags.\n\nTest: Rely on TreeHugger.\nBug: 32535355\nBug: 30627598\n\nThis reverts commit d8c052ac0aa3382c4807add33afa32580ffeecbb.\n\nChange-Id: I81fd14dac60c94ac543e336f4f3c888259fc8bd7\n" }, { "commit": "847e6ce98b4b822fd94c631975763845978ebaa3", "tree": "760e26dea1597d8219d8c515317d978b0213cdc1", "parents": [ "cff1b21b3e19c5d3a2d726fdb60dacd7de2a6f0d" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Jun 02 13:55:07 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Jun 02 14:03:28 2017 +0100" }, "message": "Rename kDexCacheViaMethod to kRuntimeCall for HLoadClass/String.\n\nThe old name does not reflect the actual code anymore.\n\nTest: testrunner.py --host\nChange-Id: I2e13cf727bba9d901c4d3fc821bb526d38a775b8\n" }, { "commit": "d49012909625c3bf87bf51138fe79315ce1b1bdc", "tree": "349ef2cdcb7255d042244046601bd0fd5eb3a092", "parents": [ "726e1793d3f54470705e5b84e7860074e029b0ed" ], "author": { "name": "Andreas Gampe", "email": "agampe@google.com", "time": "Tue May 30 18:41:34 2017 -0700" }, "committer": { "name": "Andreas Gampe", "email": "agampe@google.com", "time": "Tue May 30 20:44:33 2017 -0700" }, "message": "ART: Clean up heap headers\n\nUse more forward declarations for accounting structures and spaces.\nFactor out structs to reduce header surface. Remove heap include where\nunnecessary. Fix up transitive users. Move some debug-only code out\nof line.\n\nTest: m test-art-host\nChange-Id: I16db4aaa803f39e155ce6e1b0778b7e393dcbb17\n" }, { "commit": "6597946d29be9108e2cc51223553d3db9290a3d9", "tree": "cea6647a45cc59fa1423751179e647124f269990", "parents": [ "a654e0378a8d0bb149362399917e4da2959e6991" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri May 19 17:25:12 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Mon May 22 18:35:05 2017 +0100" }, "message": "Use PC-relative pointer to boot image methods.\n\nIn preparation for adding ArtMethod entries to the .bss\nsection, add direct PC-relative pointers to methods so that\nthe number of needed .bss entries for boot image is small.\n\nTest: m test-art-host-gtest\nTest: testrunner.py --host\nTest: testrunner.py --target on Nexus 6P\nTest: Nexus 6P boots.\nTest: Build aosp_mips64-userdebug\nBug: 30627598\nChange-Id: Ia89f5f9975b741ddac2816e1570077ba4b4c020f\n" }, { "commit": "764d454d1d51448deb81f6e8d2d7d317c7f4d1b4", "tree": "029f9e65cfe7e953df55b7af45e099924ffdbbac", "parents": [ "7974468d32b817be1d49dfcf513605f646c481bc" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue May 16 10:31:41 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue May 16 15:08:55 2017 +0100" }, "message": "Remove LoadString/Class kind kBootImageLinkTimeAddress.\n\nWe no longer support non-PIC boot image compilation.\n\nAlso clean up some obsolete code for method patches\nand make JIT correctly report itself as non-PIC.\n\nTest: testrunner.py --host\nTest: testrunner.py --target\nBug: 33192586\nChange-Id: I593289c5c1b0e88b82b86a933038be97bbb15ad2\n" }, { "commit": "de6fc4a429b69f074dc6d28b52197b425aacd78c", "tree": "879e5eacda2af3bad1aa9dc81ea08922c6c26cc8", "parents": [ "99a77162e9dc351da14bb7819a2718395d7277ed", "e1811ed6b57a54dc8ebd327e4bd2c4422092a3a0" ], "author": { "name": "Treehugger Robot", "email": "treehugger-gerrit@google.com", "time": "Thu May 11 11:59:40 2017 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Thu May 11 11:59:42 2017 +0000" }, "message": "Merge \"ARM64: Share address computation across SIMD LDRs/STRs.\"" }, { "commit": "7d157fcaaae137cc98dbfb872aa1bdc0105a898f", "tree": "2b7d8affda23908e5bfbfaad446079db2ef1ee09", "parents": [ "58d7ddc678e5bcd2364c24c4bdc8a3cfbcfc5358" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed May 10 16:29:23 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu May 11 11:14:54 2017 +0100" }, "message": "Clean up some uses of \"auto\".\n\nMake actual types more explicit, either by replacing \"auto\"\nwith actual type or by assigning std::pair\u003c\u003e elements of\nan \"auto\" variable to typed variables. Avoid binding const\nreferences to temporaries. Avoid copying a container.\n\nTest: m test-art-host-gtest\nChange-Id: I1a59f9ba1ee15950cacfc5853bd010c1726de603\n" }, { "commit": "e1811ed6b57a54dc8ebd327e4bd2c4422092a3a0", "tree": "e3ce48e66190c11a8b5342f4ec0d1046ba28d788", "parents": [ "7113885fcd983b33ee1e350865d21517d6297843" ], "author": { "name": "Artem Serov", "email": "artem.serov@linaro.org", "time": "Thu Apr 27 16:50:47 2017 +0100" }, "committer": { "name": "Artem Serov", "email": "artem.serov@linaro.org", "time": "Thu May 11 10:06:04 2017 +0100" }, "message": "ARM64: Share address computation across SIMD LDRs/STRs.\n\nFor array accesses the element address has the following structure:\nAddress \u003d CONST_OFFSET + base_addr + index \u003c\u003c ELEM_SHIFT\n\nTaking into account ARM64 LDR/STR addressing modes address part\n(CONST_OFFSET + index \u003c\u003c ELEM_SHIFT) can be shared across array\naccess with the same data type and index.\n\nFor example, for the following loop 5 accesses can share address\ncomputation:\n\nvoid foo(int[] a, int[] b, int[] c) {\n for (i...) {\n a[i] \u003d a[i] + 5;\n b[i] \u003d b[i] + c[i];\n }\n}\n\nTest: test-art-host, test-art-target\n\nChange-Id: I46af3b4e4a55004336672cdba3296b7622d815ca\n" }, { "commit": "92f4672f811a4eccdc596f7c2235804abd196fde", "tree": "03449148ef022c9c3afc50bc960f2851e0098f97", "parents": [ "5528d1c7cf686d2edea51ff12881e495c2124f09", "952b23505e2512c9327e6d20c8304493bf8fcf7c" ], "author": { "name": "Roland Levillain", "email": "rpl@google.com", "time": "Wed May 10 17:24:56 2017 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Wed May 10 17:24:58 2017 +0000" }, "message": "Merge \"Improve the documentation of an ARM64 parallel move resolver corner case.\"" }, { "commit": "270970e660d3c99e62a88b18144d159dd8699c55", "tree": "910248487abab475a9109925a9089fedd08b45c1", "parents": [ "201c81947f23b3f43c2ff78d9c1078dd53157fe5", "88abba2b0cb0151d89e16da3e64025878dc2f142" ], "author": { "name": "Treehugger Robot", "email": "treehugger-gerrit@google.com", "time": "Wed May 10 14:04:18 2017 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Wed May 10 14:04:19 2017 +0000" }, "message": "Merge \"ARM/AOT: Allow 16-bit LDR for Baker read barrier loads.\"" }, { "commit": "ddc38fe3e5618e3922ecc445193dacb2f39ef736", "tree": "090cf77abda6a5755371673966ceea83f858048c", "parents": [ "45f9865cc974d344c9a859508c8ec8ce101f4c52", "4e92c3ce7ef354620a785553bbada554fca83a67" ], "author": { "name": "Nicolas Geoffray", "email": "ngeoffray@google.com", "time": "Wed May 10 10:55:45 2017 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Wed May 10 10:55:47 2017 +0000" }, "message": "Merge \"Add runtime reasons for deopt.\"" }, { "commit": "4e92c3ce7ef354620a785553bbada554fca83a67", "tree": "42029deff4d3ba7f89b5fdbf79ff410da575f431", "parents": [ "549844e9ccf432d1396b19af890eedb602b8ba04" ], "author": { "name": "Nicolas Geoffray", "email": "ngeoffray@google.com", "time": "Mon May 08 09:34:26 2017 +0100" }, "committer": { "name": "Nicolas Geoffray", "email": "ngeoffray@google.com", "time": "Wed May 10 09:27:33 2017 +0100" }, "message": "Add runtime reasons for deopt.\n\nCurrently to help investigate. Also:\n1) Log when deoptimization happens (which method and what reason)\n2) Trace when deoptimization happens (to make it visible in systrace)\n\nbug:37655083\nTest: test-art-host test-art-target\nChange-Id: I0c2d87b40db09e8e475cf97a7c784a034c585e97\n" }, { "commit": "88abba2b0cb0151d89e16da3e64025878dc2f142", "tree": "231e5551a1b8d3c8bf162c9d0f30916b36ba2742", "parents": [ "b9c3a99096c746b09af611e55e11b86600374011" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed May 03 17:09:25 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue May 09 15:43:34 2017 +0100" }, "message": "ARM/AOT: Allow 16-bit LDR for Baker read barrier loads.\n\nTest: m test-art-target-gtest\nTest: testrunner.py --target on Nexus 6P.\nTest: testrunner.py --target on Nexus 6P with heap poisoning enabled.\nTest: Repeat the above tests with ART_USE_OLD_ARM_BACKEND\u003dtrue.\nBug: 29516974\nBug: 30126666\nBug: 36141117\nChange-Id: I458f2ec5fe9abead4db06c7595d992945096fb68\n" }, { "commit": "93029177d766269e2685cf60625f61e38820e78d", "tree": "744e11e81a301a59c2be15adf3bc7fe4190ccdd5", "parents": [ "91f956c925f015b8cd355e8ec2a697a4ba69f7ab", "ff48700df9da9aa0c6a8c1f65c9d862f936e1a89" ], "author": { "name": "Roland Levillain", "email": "rpl@google.com", "time": "Mon May 08 08:58:16 2017 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Mon May 08 08:58:18 2017 +0000" }, "message": "Merge \"Improve the implementation of UnsafeCASObject with Baker read barriers.\"" }, { "commit": "ff48700df9da9aa0c6a8c1f65c9d862f936e1a89", "tree": "932dd5749a82d9477d24200db6d747d6008a0a5d", "parents": [ "15cb9753075bcaa5b91a6497a2d35e8bd98af1f0" ], "author": { "name": "Roland Levillain", "email": "rpl@google.com", "time": "Tue Mar 07 16:50:01 2017 +0000" }, "committer": { "name": "Roland Levillain", "email": "rpl@google.com", "time": "Fri May 05 16:08:50 2017 +0100" }, "message": "Improve the implementation of UnsafeCASObject with Baker read barriers.\n\nOn ARM and ARM64, avoid loading the reference altogether when the\nGC is not marking.\n\nAlso, extract the code logic for updating a reference field from\nGenerateReferenceLoadWithBakerReadBarrier routines and move it to\nnew routines (UpdateReferenceFieldWithBakerReadBarrier), to make\nthe implementation more legible.\n\nTest: Run ART target tests in Baker read barrier configuration.\nBug: 29516974\nChange-Id: I11c53f0607e997cd02ec7911725e98ef3dc97d90\n" }, { "commit": "952b23505e2512c9327e6d20c8304493bf8fcf7c", "tree": "ff82866fef9b738a3d90df30fd31c0fdafb8c4de", "parents": [ "15cb9753075bcaa5b91a6497a2d35e8bd98af1f0" ], "author": { "name": "Roland Levillain", "email": "rpl@google.com", "time": "Wed May 03 19:49:14 2017 +0100" }, "committer": { "name": "Roland Levillain", "email": "rpl@google.com", "time": "Wed May 03 19:52:06 2017 +0100" }, "message": "Improve the documentation of an ARM64 parallel move resolver corner case.\n\nTest: m test-art-host-gtest-codegen_test\nBug: 34760542\nBug: 34834461\nChange-Id: I7e716c4b665ed51af9908042f88fb2e4bcefb849\n" }, { "commit": "d01745ef88bfd25df574a885d90a1a7785db5f5b", "tree": "058eb1593dbb0fe8a8e26b901909bec8aa01d474", "parents": [ "a57c334075b193de9690fff97acf6c1b1d1283fc" ], "author": { "name": "Igor Murashkin", "email": "iam@google.com", "time": "Wed Apr 05 16:40:31 2017 -0700" }, "committer": { "name": "Igor Murashkin", "email": "iam@google.com", "time": "Tue May 02 09:45:45 2017 -0700" }, "message": "optimizing: constructor fence redundancy elimination - remove dmb after LSE\n\nPart one of a few upcoming CLs to optimize constructor fences.\n\nThis improves load-store-elimination; all singleton objects that are not\nreturned will have their associated constructor fence removed.\n\nIf the allocation is removed, so is the fence. Even if allocation is not\nremoved, fences can sometimes be removed.\n\nThis change is enabled by tracking the \"this\" object associated with the\nconstructor fence as an input. Fence inputs are considered weak; they do not keep\nthe \"this\" object alive; if the instructions for \"this\" are all deleted,\nthe fence can also be deleted.\n\nBug: 36656456\nTest: art/test.py --host \u0026\u0026 art/test.py --target\nChange-Id: I05659ab07e20d6e2ecd4be051b722726776f4ab1\n" }, { "commit": "66d691de219e840b3f84385d8bd1b7001562b0e5", "tree": "4034e867246db26acaa1a36b89b823ce5d0f1a58", "parents": [ "38870a8a2717ccf1bcd3faddc53b1999985bb29f" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Fri Apr 07 17:53:39 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Apr 25 18:29:33 2017 +0100" }, "message": "ARM64: Link-time generated thunks for ArrayGet Baker CC read barrier.\n\nTest: Added a test to relative_patcher_arm64\nTest: m test-art-target-gtest on Nexus 6P.\nTest: Nexus 6P boots.\nTest: testrunner.py --target on Nexus 6P.\nTest: Nexus 6P boots with heap poisoning.\nTest: testrunner.py --target on Nexus 6P with heap poisoning.\nBug: 29516974\nBug: 30126666\nBug: 36141117\nChange-Id: Id0f23089c55cbb53b84305c11bb4b03718561ade\n" }, { "commit": "d1ef87339c5af782652678d5849d1dfab14d79ce", "tree": "8b8b936ce58043248b2b58497e7b210c49a704b1", "parents": [ "1cae72e6cfc0b4cd4ad20467dfea243139296998" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Tue Apr 18 13:55:13 2017 +0100" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Apr 19 10:31:50 2017 +0100" }, "message": "ARM64: Heap poisoning for link-time Baker CC read barrier thunks.\n\nAnd fix running out of scratch registers for HArraySet\nwith large constant index and a reference to poison.\n\nTest: Nexus 6P boots with heap poisoning enabled.\nTest: testrunner.py --target with heap poisoning enabled on Nexus 6P.\nBug: 29516974\nBug: 30126666\nBug: 36141117\nChange-Id: Ifb38f4a0e23a2963468772f34f294febfc340b8c\n" }, { "commit": "dda7360caf19d44cfef5307e8a6ec4d7cba6b474", "tree": "9f9a7f0c7197b72c9eca9665efae0d4fe17d9268", "parents": [ "27fb1dc467effbd8df43e6207743fdb7bcee4044" ], "author": { "name": "Mathieu Chartier", "email": "mathieuc@google.com", "time": "Mon Apr 17 12:50:45 2017 -0700" }, "committer": { "name": "Mathieu Chartier", "email": "mathieuc@google.com", "time": "Mon Apr 17 12:50:45 2017 -0700" }, "message": "Disable read barrier thunks for heap poisoning\n\nLogic to unpoison is not yet implemented, this causes SIGSEGVs in\nvarious places.\n\nBug: 29516974\nBug: 30126666\nBug: 36141117\n\nTest: ./test/testrunner/testrunner.py -j4 --optimizing --debuggable --ndebuggable --target --verbose\nChange-Id: I8317a142d07af36090d5f05ce46100dfa07c17e7\n" }, { "commit": "7b331261c6bdb6316a649ab591813f4dd1a5892f", "tree": "22b60227265c73903428ea55cde3123ea992ac4e", "parents": [ "b1a52116a1ca418dcccad2ca2acd6cb36f8ca0e7", "f4f2daafb38c9c07ea74044a0fb89a2a19288b7a" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Thu Apr 13 12:54:22 2017 +0000" }, "committer": { "name": "Gerrit Code Review", "email": "noreply-gerritcodereview@google.com", "time": "Thu Apr 13 12:54:23 2017 +0000" }, "message": "Merge \"ARM64: Use link-time generated thunks for Baker CC read barrier.\"" }, { "commit": "f4f2daafb38c9c07ea74044a0fb89a2a19288b7a", "tree": "13fd63a65c12e60074bc2bc1e693fbb3b788ed8e", "parents": [ "26c25d5da32fe1bdd94dd1404197c14994ecab60" ], "author": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Mon Mar 20 18:26:59 2017 +0000" }, "committer": { "name": "Vladimir Marko", "email": "vmarko@google.com", "time": "Wed Apr 12 17:02:06 2017 +0100" }, "message": "ARM64: Use link-time generated thunks for Baker CC read barrier.\n\nRemaining work for follow-up CLs:\n - array loads,\n - volatile field loads,\n - use implicit null check in field thunk.\n\nTest: Added tests to relative_patcher_arm64\nTest: New run-test 160-read-barrier-stress\nTest: m test-art-target-gtest on Nexus 6P.\nTest: testrunner.py --target on Nexus 6P.\nBug: 29516974\nBug: 30126666\nBug: 36141117\nChange-Id: Id68ff171c55a3f1bf1ac1b657f480531aa7b3710\n" }, { "commit": "d4bccf1ece319a3a99e03ecbcbbf40bb82b9e331", "tree": "2890740d9cab3eee2be223666f528c6707b89f90", "parents": [ "903b8169074c01590ab3f5ad9190d9c7e3fe795b" ], "author": { "name": "Artem Serov", "email": "artem.serov@linaro.org", "time": "Mon Apr 03 18:47:32 2017 +0100" }, "committer": { "name": "Artem Serov", "email": "artem.serov@linaro.org", "time": "Mon Apr 10 11:43:33 2017 +0100" }, "message": "ARM64: Support 128-bit registers for SIMD.\n\nTest: test-art-host, test-art-target\n\nChange-Id: Ifb931a99d34ea77602a0e0781040ed092de9faaa\n" }, { "commit": "7957d9500ca53b78acb1eba754416535d4418a89", "tree": "518792ef3c5556686d7b075dbc4f2e3dc8af8632", "parents": [ "8309d9b173b205bb9530dbd94e48731c20b8e4f4" ], "author": { "name": "Artem Serov", "email": "artem.serov@linaro.org", "time": "Tue Apr 04 15:44:09 2017 +0100" }, "committer": { "name": "Artem Serov", "email": "artem.serov@linaro.org", "time": "Thu Apr 06 16:38:32 2017 +0100" }, "message": "ARM64: Saves 128-bit regs state along SuspendCheckSlowPath.\n\nWe need to save 128 bits of data (default ABI of ART runtime\nonly saves 64 bits). Note that this is *only* done for Q-registers\nthat are live, so overhead is not too big.\n\nTest: test-art-target, test-art-host.\nChange-Id: I1f018a708c316f9d426db13b2e3b3071aa4c999b\n" } ], "next": "7b7f8b4e07b57962cacbd9a05d9f5fad4f0a91aa" }