)]}' { "log": [ { "commit": "69445f095c22aac2388f939bedebf224a6efcdaf", "tree": "8f09fc70cd4ef32613368c9ae3759cccc04b490b", "parents": [ "59231e5aff254f28b7ea7bcfd32cd759511d5844" ], "author": { "name": "Rainer Orth", "email": "ro@gcc.gnu.org", "time": "Tue Oct 22 08:44:25 2019" }, "committer": { "name": "Rainer Orth", "email": "ro@gcc.gnu.org", "time": "Tue Oct 22 08:44:25 2019" }, "message": "[builtins][test] Avoid unportable mmap call in clear_cache_test.c\n\nWithin the last two weeks, the Builtins-*-sunos :: clear_cache_test.c started to FAIL\non Solaris. Running it under truss shows\n\n mmap(0x00000000, 128, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE|MAP_ANON, 0, 0) Err#22 EINVAL\n _exit(1)\n\nWhile there are several possible reasons mmap can return EINVAL on Solaris, it turns\nout it\u0027s this one (from mmap(2)):\n\n MAP_ANON was specified, but the file descriptor was not\n -1.\n\nAnd indeed even the Linux mmap(2) documents this as unportable:\n\n MAP_ANONYMOUS\n The mapping is not backed by any file; its contents are initial‐\n ized to zero. The fd argument is ignored; however, some imple‐\n mentations require fd to be -1 if MAP_ANONYMOUS (or MAP_ANON) is\n specified, and portable applications should ensure this. The\n\nThis patch follows this advise. Tested on x86_64-pc-linux-gnu, amd64-pc-solaris2.11\nand sparcv9-sun-solaris2.11.\n\nDifferential Revision: https://reviews.llvm.org/D68455\n\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375490 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "59231e5aff254f28b7ea7bcfd32cd759511d5844", "tree": "d4fff985201a01bc6dd7f94b813c42cafa18fe86", "parents": [ "a3697b7f0eb500a70f161e7f8633d3936ab4ad81" ], "author": { "name": "Evgeniy Stepanov", "email": "eugeni.stepanov@gmail.com", "time": "Mon Oct 21 23:59:45 2019" }, "committer": { "name": "Evgeniy Stepanov", "email": "eugeni.stepanov@gmail.com", "time": "Mon Oct 21 23:59:45 2019" }, "message": "Fix lld detection in standalone compiler-rt.\n\nSummary:\nRight now all hwasan tests on Android are silently disabled because they\nrequire \"has_lld\" and standalone compiler-rt can not (and AFAIK was\nnever able to) set it.\n\nReviewers: pcc, dyung\n\nSubscribers: dberris, mgorny, #sanitizers, llvm-commits\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D69196\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375472 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "a3697b7f0eb500a70f161e7f8633d3936ab4ad81", "tree": "837f3f6d090c51ebc3e33153379400563088c3e6", "parents": [ "89a8e62c03aa2cfe044c9023ec3bbaefb835a7df" ], "author": { "name": "Evgeniy Stepanov", "email": "eugeni.stepanov@gmail.com", "time": "Mon Oct 21 23:57:12 2019" }, "committer": { "name": "Evgeniy Stepanov", "email": "eugeni.stepanov@gmail.com", "time": "Mon Oct 21 23:57:12 2019" }, "message": "[hwasan] Workaround unwinder issues in try-catch test.\n\nAndroid links the unwinder library to every DSO. The problem is,\nunwinder has global state, and hwasan implementation of personality\nfunction wrapper happens to rub it the wrong way.\n\nSwitch the test to static libc++ as a temporary workaround.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375471 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "89a8e62c03aa2cfe044c9023ec3bbaefb835a7df", "tree": "bf3fc3a3859406468e71587b726b252d00ceba66", "parents": [ "d59bde52167f18b585d2828a8fd83506ee14fd0e" ], "author": { "name": "Vedant Kumar", "email": "vsk@apple.com", "time": "Sat Oct 19 00:51:27 2019" }, "committer": { "name": "Vedant Kumar", "email": "vsk@apple.com", "time": "Sat Oct 19 00:51:27 2019" }, "message": "[profile] Use -fPIC -shared in a test instead of -dynamiclib\n\nThis is more portable than -dynamiclib. Also, fix the path to an input\nfile that broke when the test was moved in r375315.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375317 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "d59bde52167f18b585d2828a8fd83506ee14fd0e", "tree": "531aa058f79514312fc72afe4a91a82029abaf8d", "parents": [ "b19457ce72921f7eef9645b327337c933657e878" ], "author": { "name": "Vedant Kumar", "email": "vsk@apple.com", "time": "Sat Oct 19 00:46:53 2019" }, "committer": { "name": "Vedant Kumar", "email": "vsk@apple.com", "time": "Sat Oct 19 00:46:53 2019" }, "message": "[profile] Disable instrprof-get-filename-merge-mode.c on Windows\n\nThe Windows bots are failing with:\n\nclang: warning: argument unused during compilation: \u0027-dynamiclib\u0027 [-Wunused-command-line-argument]\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375315 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "b19457ce72921f7eef9645b327337c933657e878", "tree": "e7d1a6055e41c487783044206bf82a05d7df2504", "parents": [ "4aae16c11694f3d5aa65a5b7c05d2f6598f147ef" ], "author": { "name": "Vedant Kumar", "email": "vsk@apple.com", "time": "Fri Oct 18 23:33:40 2019" }, "committer": { "name": "Vedant Kumar", "email": "vsk@apple.com", "time": "Fri Oct 18 23:33:40 2019" }, "message": "[profile] Do not cache __llvm_profile_get_filename result\n\nWhen the %m filename pattern is used, the filename is unique to each\nimage, so the cached value is wrong.\n\nIt struck me that the full filename isn\u0027t something that\u0027s recomputed\noften, so perhaps it doesn\u0027t need to be cached at all. David Li pointed\nout we can go further and just hide lprofCurFilename. This may regress\nworkflows that depend on using the set-filename API to change filenames\nacross all loaded DSOs, but this is expected to be very rare.\n\nrdar://55137071\n\nDifferential Revision: https://reviews.llvm.org/D69137\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375301 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "4aae16c11694f3d5aa65a5b7c05d2f6598f147ef", "tree": "5f51c53ea308c470841fdc888f12bbb89b1e46a2", "parents": [ "2d6eaa654777895937c52efb727a0a4cb753fe5b" ], "author": { "name": "Peter Collingbourne", "email": "peter@pcc.me.uk", "time": "Fri Oct 18 22:51:38 2019" }, "committer": { "name": "Peter Collingbourne", "email": "peter@pcc.me.uk", "time": "Fri Oct 18 22:51:38 2019" }, "message": "hwasan: Add missing SANITIZER_INTERFACE_ATTRIBUTE on __hwasan_personality_wrapper.\n\nDifferential Revision: https://reviews.llvm.org/D69201\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375298 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "2d6eaa654777895937c52efb727a0a4cb753fe5b", "tree": "3fa06ec171ec7037e3e1ef925b4b1e3b8aa5a69c", "parents": [ "67ee8fcf65517112db0fd03ff2a9055a799fcb3d" ], "author": { "name": "Evgeniy Stepanov", "email": "eugeni.stepanov@gmail.com", "time": "Fri Oct 18 22:36:25 2019" }, "committer": { "name": "Evgeniy Stepanov", "email": "eugeni.stepanov@gmail.com", "time": "Fri Oct 18 22:36:25 2019" }, "message": "[hwasan] Remove system allocator fallback.\n\nSummary:\nThis has been an experiment with late malloc interposition, made\npossible by a non-standard feature of the Android dynamic loader.\n\nReviewers: pcc, mmalcomson\n\nSubscribers: srhines, #sanitizers, llvm-commits\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D69199\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375296 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "67ee8fcf65517112db0fd03ff2a9055a799fcb3d", "tree": "b65ea6de90484422f49551c05f340782c473e90a", "parents": [ "c15bffeb473d0b4a9f752b9fabc786832f407b64" ], "author": { "name": "Peter Collingbourne", "email": "peter@pcc.me.uk", "time": "Fri Oct 18 20:35:29 2019" }, "committer": { "name": "Peter Collingbourne", "email": "peter@pcc.me.uk", "time": "Fri Oct 18 20:35:29 2019" }, "message": "Update global_symbols.txt.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375284 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "c15bffeb473d0b4a9f752b9fabc786832f407b64", "tree": "98bb12907e7b7b30efb84a42f1801b426a20bab2", "parents": [ "749af53928a31afa3111f27cc41fd15849d86667" ], "author": { "name": "Peter Collingbourne", "email": "peter@pcc.me.uk", "time": "Fri Oct 18 20:00:32 2019" }, "committer": { "name": "Peter Collingbourne", "email": "peter@pcc.me.uk", "time": "Fri Oct 18 20:00:32 2019" }, "message": "scudo: Update TLS_SLOT_SANITIZER value.\n\nAndroid now allocates only 8 fixed TLS slots. Somehow we were getting away\nwith using a non-existent slot until now, but in some cases the TLS slots\nwere being placed at the end of a page, which led to a segfault at startup.\n\nDifferential Revision: https://reviews.llvm.org/D69191\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375276 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "749af53928a31afa3111f27cc41fd15849d86667", "tree": "1ad0345fba7485fd4eed1b554473cec918b74085", "parents": [ "6e86c1660cd0c8f0af4575b41f83150a560d3941" ], "author": { "name": "Sjoerd Meijer", "email": "sjoerd.meijer@arm.com", "time": "Fri Oct 18 11:01:45 2019" }, "committer": { "name": "Sjoerd Meijer", "email": "sjoerd.meijer@arm.com", "time": "Fri Oct 18 11:01:45 2019" }, "message": "[Arm][libsanitizer] Fix arm libsanitizer failure with bleeding edge glibc\n\nGlibc has recently introduced changed to the mode field in ipc_perm in commit\n2f959dfe849e0646e27403f2e4091536496ac0f0. For Arm this means that the mode\nfield no longer has the same size.\n\nThis causes an assert failure against libsanitizer\u0027s internal copy of ipc_perm.\nSince this change can\u0027t be easily detected I am adding arm to the list of\ntargets that are excluded from this check.\n\nPatch by: Tamar Christina\n\nDifferential Revision: https://reviews.llvm.org/D69104\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375220 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "6e86c1660cd0c8f0af4575b41f83150a560d3941", "tree": "c8a57e634bb86935da969757daf7eaaa3f0713cb", "parents": [ "45c83da934bc2f2f7cb68b2602753753cb0929f4" ], "author": { "name": "Evgeniy Stepanov", "email": "eugeni.stepanov@gmail.com", "time": "Thu Oct 17 20:32:54 2019" }, "committer": { "name": "Evgeniy Stepanov", "email": "eugeni.stepanov@gmail.com", "time": "Thu Oct 17 20:32:54 2019" }, "message": "libhwasan initialisation include kernel syscall ABI relaxation\n\nSummary:\nUntil now AArch64 development has been on patched kernels that have an always\non relaxed syscall ABI where tagged pointers are accepted.\nThe patches that have gone into the mainline kernel rely on each process opting\nin to this relaxed ABI.\n\nThis commit adds code to choose that ABI into __hwasan_init.\n\nThe idea has already been agreed with one of the hwasan developers\n(http://lists.llvm.org/pipermail/llvm-dev/2019-September/135328.html).\n\nThe patch ignores failures of `EINVAL` for Android, since there are older versions of the Android kernel that don\u0027t require this `prctl` or even have the relevant values. Avoiding EINVAL will let the library run on them.\n\nI\u0027ve tested this on an AArch64 VM running a kernel that requires this\nprctl, having compiled both with clang and gcc.\n\nPatch by Matthew Malcomson.\n\nReviewers: eugenis, kcc, pcc\n\nReviewed By: eugenis\n\nSubscribers: srhines, kristof.beyls, #sanitizers, llvm-commits\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D68794\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375166 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "45c83da934bc2f2f7cb68b2602753753cb0929f4", "tree": "0d5668401a5375d14afeeadb61e4a3649830f823", "parents": [ "650d5e6d1a5b8f36cdeea4d94a7d4dbab3dcad6c" ], "author": { "name": "Dan Liew", "email": "dan@su-root.co.uk", "time": "Thu Oct 17 20:14:04 2019" }, "committer": { "name": "Dan Liew", "email": "dan@su-root.co.uk", "time": "Thu Oct 17 20:14:04 2019" }, "message": "[Builtins] Downgrade duplicate source file warning from a fatal error to a warning.\n\nThis is a follow up to r375150 to unbreak the `clang-ppc64be-linux` bot.\nThe commit caused running the tests to fail due to\n\n```\nllvm-lit:\n/home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/llvm/projects/compiler-rt/test/builtins/Unit/lit.cfg.py:116:\nfatal: builtins_source_features contains duplicates:\n[\u0027librt_has_divtc3\u0027]\n```\n\nThis commit should be reverted once the build system bug for powerpc is\nfixed.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375162 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "650d5e6d1a5b8f36cdeea4d94a7d4dbab3dcad6c", "tree": "5b611bd76c74c9cfd60c2c3a616923c66b1ce127", "parents": [ "3bd385c6f182169886bec74afd1115df10b95b80" ], "author": { "name": "Dan Liew", "email": "dan@su-root.co.uk", "time": "Thu Oct 17 18:12:49 2019" }, "committer": { "name": "Dan Liew", "email": "dan@su-root.co.uk", "time": "Thu Oct 17 18:12:49 2019" }, "message": "[Builtins] Provide a mechanism to selectively disable tests based on whether an implementation is provided by a builtin library.\n\nSummary:\nIf a platform removes some builtin implementations (e.g. via the\nDarwin-excludes mechanism) then this can lead to test failures because\nthe test expects an implementation to be available.\n\nTo solve this lit features are added for each configuration based\non which sources are included in the builtin library. The features\nare of the form `librt_has_\u003cname\u003e` where `\u003cname\u003e` is the name of the\nsource file with the file extension removed. This handles C and\nassembly sources.\n\nWith the lit features in place it is possible to make certain tests\nrequire them.\n\nExample:\n\n```\nREQUIRES: librt_has_comparedf2\n```\n\nAll top-level tests in `test/builtins/Unit` (i.e. not under\n`arm`, `ppc`, and `riscv`) have been annotated with the appropriate\n`REQUIRES: librt_has_*` statement.\n\nrdar://problem/55520987\n\nReviewers: beanz, steven_wu, arphaman, dexonsmith, phosek, thakis\n\nSubscribers: mgorny, #sanitizers, llvm-commits\n\nTags: #llvm, #sanitizers\n\nDifferential Revision: https://reviews.llvm.org/D68064\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375150 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "3bd385c6f182169886bec74afd1115df10b95b80", "tree": "ddea8c5f7222c3f4671ac502285c262edcb452a0", "parents": [ "5d842447aa515d35e44333f1cd6d3f40e7eba3d8" ], "author": { "name": "Reid Kleckner", "email": "rnk@google.com", "time": "Thu Oct 17 17:59:11 2019" }, "committer": { "name": "Reid Kleckner", "email": "rnk@google.com", "time": "Thu Oct 17 17:59:11 2019" }, "message": "[asan] Update Windows test expectations for LLVM\u0027s MS demangler\n\nAfter r375041 llvm-symbolizer uses it for demangling instead of\nUnDecorateSymbolName. LLVM puts spaces after commas while Microsoft does\nnot.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375147 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "5d842447aa515d35e44333f1cd6d3f40e7eba3d8", "tree": "f39f952730e2bd50dedef2d3971f58860bc4a63f", "parents": [ "a1b9a6b625e4e9c6e562d8fb02f76966dda2c577" ], "author": { "name": "Sam Elliott", "email": "selliott@lowrisc.org", "time": "Thu Oct 17 17:24:28 2019" }, "committer": { "name": "Sam Elliott", "email": "selliott@lowrisc.org", "time": "Thu Oct 17 17:24:28 2019" }, "message": "Revert [Sanitizers] Add support for RISC-V 64-bit\n\nThis reverts r375132 (git commit 00bbe990c5d4472d5413479a539b3d6edbb3ca7a)\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375136 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "a1b9a6b625e4e9c6e562d8fb02f76966dda2c577", "tree": "cd397fd982de2ae4d61774f6c9859e7fa7e4c0b9", "parents": [ "9729ecdb6e3c00c6a432ba9d05916d14c6fde0de" ], "author": { "name": "Sam Elliott", "email": "selliott@lowrisc.org", "time": "Thu Oct 17 16:36:27 2019" }, "committer": { "name": "Sam Elliott", "email": "selliott@lowrisc.org", "time": "Thu Oct 17 16:36:27 2019" }, "message": "[Sanitizers] Add support for RISC-V 64-bit\n\nSummary:\nThis has been tested with gcc trunk on openSUSE Tumbleweed on the HiFive Unleashed.\n\nPatch by Andreas Schwab (schwab)\n\nReviewers: luismarques\n\nReviewed By: luismarques\n\nSubscribers: mhorne, emaste, luismarques, asb, mgorny, fedor.sergeev, simoncook, kito-cheng, shiva0217, rogfer01, rkruppe, lenary, s.egerton, #sanitizers, llvm-commits\n\nTags: #llvm, #sanitizers\n\nDifferential Revision: https://reviews.llvm.org/D66870\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375132 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "9729ecdb6e3c00c6a432ba9d05916d14c6fde0de", "tree": "f39f952730e2bd50dedef2d3971f58860bc4a63f", "parents": [ "359fb9dafc705ce2f6c50acd9bdc34fdbfa66e11" ], "author": { "name": "Zoran Jovanovic", "email": "zoran.jovanovic@mips.com", "time": "Thu Oct 17 12:21:14 2019" }, "committer": { "name": "Zoran Jovanovic", "email": "zoran.jovanovic@mips.com", "time": "Thu Oct 17 12:21:14 2019" }, "message": "[mips] [builtins] Remove clear_mips_cache\nDifferential Revision: https://reviews.llvm.org/D69021\n\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@375110 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "359fb9dafc705ce2f6c50acd9bdc34fdbfa66e11", "tree": "c851172f3cd6a51da6c06e8d8a695e5c56cf5eb8", "parents": [ "19db58192301df651ff2baaad5305fe2835ae289" ], "author": { "name": "Dan Liew", "email": "dan@su-root.co.uk", "time": "Wed Oct 16 05:48:39 2019" }, "committer": { "name": "Dan Liew", "email": "dan@su-root.co.uk", "time": "Wed Oct 16 05:48:39 2019" }, "message": "[CMake] Disable building all Darwin libraries (except builtins) for macOS i386 when the SDK is \u003e\u003d 10.15.\n\nSummary:\nIn the macOS 10.15 SDK the ability to link i386 binaries was removed and\nin the corresponding OS it is not possible to run macOS i386 binaries.\n\nThe consequence of these changes meant that targets like `check-asan`\nwould fail because:\n\n* Unit tests could not be linked for i386\n* Lit tests for i386 would fail due to not being able to execute\n compiled binaries.\n\nThe simplest fix to this is to simply disable building for i386 for\nmacOS when using the 10.15 SDK (or newer). This disables building the\ni386 slice for most compiler-rt libraries and consequently disables the\nunit and lit tests for macOS i386.\n\nNote that because the `DARWIN_osx_ARCHS` CMake variable is a cache\nvariable this patch will have no affect on existing builds unless\nthe existing cache variable is deleted. The simplest way to deal with\nthis is delete existing builds and just do a fresh configure.\n\nNote this should not affect the builtins which are managed with\nthe `DARWIN_osx_BUILTIN_ARCHS` CMake cache variable.\n\nFor those who wish to force using a particular set of architectures when\nusing newer SDKs passing `-DDARWIN_osx_ARCHS\u003di386;x86_64;x86_64h` to\nCMake should provide a usable (but completely unsupported) workaround.\n\nrdar://problem/55668535\nrdar://problem/47939978\n\nReviewers: kubamracek, yln, azhar, kcc, dvyukov, vitalybuka, cryptoad, eugenis, thakis, phosek\n\nSubscribers: mgorny, #sanitizers, llvm-commits\n\nTags: #llvm, #sanitizers\n\nDifferential Revision: https://reviews.llvm.org/D68292\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374977 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "19db58192301df651ff2baaad5305fe2835ae289", "tree": "bb73fe48d2795551e20695cbfb6597557aa4ff28", "parents": [ "7bd52f9487abb8c59d578b14892ba4e10be56321" ], "author": { "name": "Julian Lettner", "email": "jlettner@apple.com", "time": "Wed Oct 16 02:13:16 2019" }, "committer": { "name": "Julian Lettner", "email": "jlettner@apple.com", "time": "Wed Oct 16 02:13:16 2019" }, "message": "Revert \"[ASan] Refine diagnoses messages\"\n\nThis reverts commit 4d1ecadda59ce82e5fa6e28dd15bf794eee88363.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374965 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "7bd52f9487abb8c59d578b14892ba4e10be56321", "tree": "cdc2538a8e715180cfdb55d7133ddf0c47109024", "parents": [ "011274a28238f524397923213db6e6514358d05f" ], "author": { "name": "Julian Lettner", "email": "jlettner@apple.com", "time": "Wed Oct 16 00:39:09 2019" }, "committer": { "name": "Julian Lettner", "email": "jlettner@apple.com", "time": "Wed Oct 16 00:39:09 2019" }, "message": "[ASan] Refine diagnoses messages\n\nThe provided PC is not reliable in every case, so don\u0027t suggest\nsomething that does not make sense.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374959 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "011274a28238f524397923213db6e6514358d05f", "tree": "e8d66cf13b4ef1c1216cf352af71a686839f3632", "parents": [ "810ae8ddac890a6613d814c0b5415c7fcb7f5cca" ], "author": { "name": "Dmitry Mikulin", "email": "dmitry.mikulin@sony.com", "time": "Tue Oct 15 16:32:50 2019" }, "committer": { "name": "Dmitry Mikulin", "email": "dmitry.mikulin@sony.com", "time": "Tue Oct 15 16:32:50 2019" }, "message": "CFI: wrong type passed to llvm.type.test with multiple inheritance devirtualization.\n\nDifferential Revision: https://reviews.llvm.org/D67985\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374909 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "810ae8ddac890a6613d814c0b5415c7fcb7f5cca", "tree": "3ac635e63afb7ac9125e40d1f18043b2a5f2a64c", "parents": [ "2f65e506ec18c7d9c683b6042e0d2d93f5afa78d" ], "author": { "name": "Dmitry Vyukov", "email": "dvyukov@google.com", "time": "Tue Oct 15 08:38:58 2019" }, "committer": { "name": "Dmitry Vyukov", "email": "dvyukov@google.com", "time": "Tue Oct 15 08:38:58 2019" }, "message": "tsan: fix Go ppc64le build\n\nThis #define is in the non-Go ppc64le build but not in the Go build.\n\nReviewed-in: https://reviews.llvm.org/D68046\nAuthor: randall77 (Keith Randall)\n\n\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374868 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "2f65e506ec18c7d9c683b6042e0d2d93f5afa78d", "tree": "9a85fcaaed11c20247c885c64ef1690f695e9523", "parents": [ "af03abc1602f31751fa66bb44e4da4d4beb7ca07" ], "author": { "name": "Jake Ehrlich", "email": "jakehehrlich@google.com", "time": "Fri Oct 11 23:35:13 2019" }, "committer": { "name": "Jake Ehrlich", "email": "jakehehrlich@google.com", "time": "Fri Oct 11 23:35:13 2019" }, "message": "[libFuzzer] Don\u0027t prefix absolute paths in fuchsia.\n\nThe ExecuteCommand function in fuchsia used to prefix the\ngetOutputFile for each command run with the artifact_prefix flag if\nit was available, because fuchsia components don\u0027t have a writable working\ndirectory. However, if a file with a global path is provided, fuchsia\nshould honor that.\n\nAn example of this is using the global /tmp directory to store stuff.\nIn fuchsia it ended up being translated to data///tmp, whereas we want\nto make sure it is using /tmp (which is available to components using the\nisolated-temp feature).\n\nTo test this I made the change, compiled fuchsia with this toolchain and\nran a fuzzer with the -fork\u003d1 flag (that mode makes use of the /tmp\ndirectory). I also tested that normal fuzzing workflow was not affected\nby this.\n\nAuthor: charco (Marco Vanotti)\nDifferential Revision: https://reviews.llvm.org/D68774\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374612 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "af03abc1602f31751fa66bb44e4da4d4beb7ca07", "tree": "7df9cb3161474c265d06fdefa5c924ba156b5ca0", "parents": [ "d12d7d30820ac62aaebbbf0d7eda66df1ea899c5" ], "author": { "name": "Reid Kleckner", "email": "rnk@google.com", "time": "Thu Oct 10 23:30:54 2019" }, "committer": { "name": "Reid Kleckner", "email": "rnk@google.com", "time": "Thu Oct 10 23:30:54 2019" }, "message": "Fix check-interception link error in compiler-rt debug mode\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374472 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "d12d7d30820ac62aaebbbf0d7eda66df1ea899c5", "tree": "8cce54330d54751a9da8f142ed8a816dd806b055", "parents": [ "6fa614831f7986043e08b25951fe7407b07e60bf" ], "author": { "name": "Evgeniy Stepanov", "email": "eugeni.stepanov@gmail.com", "time": "Thu Oct 10 20:47:22 2019" }, "committer": { "name": "Evgeniy Stepanov", "email": "eugeni.stepanov@gmail.com", "time": "Thu Oct 10 20:47:22 2019" }, "message": "Add a missing include in test.\n\nA fix for r373993.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374448 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "6fa614831f7986043e08b25951fe7407b07e60bf", "tree": "a0cc4cffd4f9f7a256c77285a9c73746ad9f6d10", "parents": [ "637654f8efbe34a83e792b1f38393fc723f14074" ], "author": { "name": "Julian Lettner", "email": "jlettner@apple.com", "time": "Thu Oct 10 17:19:58 2019" }, "committer": { "name": "Julian Lettner", "email": "jlettner@apple.com", "time": "Thu Oct 10 17:19:58 2019" }, "message": "Reland \"[ASan] Do not misrepresent high value address dereferences as null dereferences\"\n\nUpdated: Removed offending TODO comment.\n\nDereferences with addresses above the 48-bit hardware addressable range\nproduce \"invalid instruction\" (instead of \"invalid access\") hardware\nexceptions (there is no hardware address decoding logic for those bits),\nand the address provided by this exception is the address of the\ninstruction (not the faulting address). The kernel maps the \"invalid\ninstruction\" to SEGV, but fails to provide the real fault address.\n\nBecause of this ASan lies and says that those cases are null\ndereferences. This downgrades the severity of a found bug in terms of\nsecurity. In the ASan signal handler, we can not provide the real\nfaulting address, but at least we can try not to lie.\n\nrdar://50366151\n\nReviewed By: vitalybuka\n\nDifferential Revision: https://reviews.llvm.org/D68676\n\nllvm-svn: 374265\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374384 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "637654f8efbe34a83e792b1f38393fc723f14074", "tree": "c5abd596d13c86e244751e8ed81e50cab84daad8", "parents": [ "b8d5f6b4bb9527e27c74574d35683f3103868fd3" ], "author": { "name": "David Carlier", "email": "devnexen@gmail.com", "time": "Thu Oct 10 12:48:18 2019" }, "committer": { "name": "David Carlier", "email": "devnexen@gmail.com", "time": "Thu Oct 10 12:48:18 2019" }, "message": "[Sanitizers] Fix getrandom test\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374333 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "b8d5f6b4bb9527e27c74574d35683f3103868fd3", "tree": "fe0977fb8fead1e719d0dc3c6966577f79f1cd6f", "parents": [ "75ee7c0fc3e367e064e28510299aec35281c32e9" ], "author": { "name": "Russell Gallop", "email": "russell.gallop@gmail.com", "time": "Thu Oct 10 12:02:31 2019" }, "committer": { "name": "Russell Gallop", "email": "russell.gallop@gmail.com", "time": "Thu Oct 10 12:02:31 2019" }, "message": "Fix sanitizer lint check after r374315\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374321 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "75ee7c0fc3e367e064e28510299aec35281c32e9", "tree": "5eefec87716f1f26f040bf19c1d3215d8a12a40d", "parents": [ "f6e9da08badf010d1be8fc3b24d57c7ad6c8a0cb" ], "author": { "name": "Roman Lebedev", "email": "lebedev.ri@gmail.com", "time": "Thu Oct 10 11:32:06 2019" }, "committer": { "name": "Roman Lebedev", "email": "lebedev.ri@gmail.com", "time": "Thu Oct 10 11:32:06 2019" }, "message": "[UBSan] Appease linter\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374316 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "f6e9da08badf010d1be8fc3b24d57c7ad6c8a0cb", "tree": "a5cbd7a210617a84ea92b305bc83bedb6bacd11a", "parents": [ "eac233c8a8711e52d3ecd7f63815b384dce71e04" ], "author": { "name": "David Carlier", "email": "devnexen@gmail.com", "time": "Thu Oct 10 11:31:37 2019" }, "committer": { "name": "David Carlier", "email": "devnexen@gmail.com", "time": "Thu Oct 10 11:31:37 2019" }, "message": "[Sanitizers] Porting getrandom/getentropy interceptors to FreeBSD\n\n- Available from 12.x branch, by the time it lands next year in FreeBSD tree, the 11.x\u0027s might be EOL.\n- Intentionally changed the getrandom test to C code as with 12.0 (might be fixed in CURRENT since), there is a linkage issue in C++ context.\n\nReviewers: emaste, dim, vitalybuka\n\nReviewed-By: vitalybuka\n\nDifferential Revision: https://reviews.llvm.org/D68451\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374315 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "eac233c8a8711e52d3ecd7f63815b384dce71e04", "tree": "7b8db949d85496fead084b6287860b18e8d7d249", "parents": [ "fe7f65b081d5d0b32270cb1a82e2ee9a0391a6d1" ], "author": { "name": "Roman Lebedev", "email": "lebedev.ri@gmail.com", "time": "Thu Oct 10 11:03:41 2019" }, "committer": { "name": "Roman Lebedev", "email": "lebedev.ri@gmail.com", "time": "Thu Oct 10 11:03:41 2019" }, "message": "[UBSan] Split nullptr-and-nonzero-offset-variable.c in another direction\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374309 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "fe7f65b081d5d0b32270cb1a82e2ee9a0391a6d1", "tree": "539951269a99d4d5fd5a3194400a4fe61d451fec", "parents": [ "640d7e0a87981f53799ad7394c4cb93c928338c4" ], "author": { "name": "Russell Gallop", "email": "russell.gallop@gmail.com", "time": "Thu Oct 10 10:56:52 2019" }, "committer": { "name": "Russell Gallop", "email": "russell.gallop@gmail.com", "time": "Thu Oct 10 10:56:52 2019" }, "message": "Revert \"[ASan] Do not misrepresent high value address dereferences as null dereferences\"\n\nAs it was breaking bots running sanitizer lint check\n\nThis reverts r374265 (git b577efe4567f1f6a711ad36e1d17280dd1c4f009)\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374308 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "640d7e0a87981f53799ad7394c4cb93c928338c4", "tree": "6a8097a79298bb25567982fd2a1efc321ad08316", "parents": [ "335962ef8a976a8e2eae4ee0fe0b44682dc44680" ], "author": { "name": "Roman Lebedev", "email": "lebedev.ri@gmail.com", "time": "Thu Oct 10 10:41:42 2019" }, "committer": { "name": "Roman Lebedev", "email": "lebedev.ri@gmail.com", "time": "Thu Oct 10 10:41:42 2019" }, "message": "[UBSan] Split nullptr-and-nonzero-offset-variable.cpp into C and C++ variants\n\nI do not understand the BB failire, it fully passes locally.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374306 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "335962ef8a976a8e2eae4ee0fe0b44682dc44680", "tree": "2148f59733ffd39f39ddffa2752de8493fb71c71", "parents": [ "a8681c5ca1c1304e4b61c8cede444abc103b05bf" ], "author": { "name": "Roman Lebedev", "email": "lebedev.ri@gmail.com", "time": "Thu Oct 10 09:51:13 2019" }, "committer": { "name": "Roman Lebedev", "email": "lebedev.ri@gmail.com", "time": "Thu Oct 10 09:51:13 2019" }, "message": "[UBSan] Revisit nullptr-and-nonzero-offset-variable.cpp test to hopefully make it pass on sanitizer-windows BB\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374298 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "a8681c5ca1c1304e4b61c8cede444abc103b05bf", "tree": "42a2119d5ff3d3a5475e2998a7f01003bf0ed134", "parents": [ "51a6e79ff7df8a7a1ca79fdda4fe0aec571da26c" ], "author": { "name": "Roman Lebedev", "email": "lebedev.ri@gmail.com", "time": "Thu Oct 10 09:25:02 2019" }, "committer": { "name": "Roman Lebedev", "email": "lebedev.ri@gmail.com", "time": "Thu Oct 10 09:25:02 2019" }, "message": "[UBSan][clang][compiler-rt] Applying non-zero offset to nullptr is undefined behaviour\n\nSummary:\nQuote from http://eel.is/c++draft/expr.add#4:\n```\n4 When an expression J that has integral type is added to or subtracted\n from an expression P of pointer type, the result has the type of P.\n(4.1) If P evaluates to a null pointer value and J evaluates to 0,\n the result is a null pointer value.\n(4.2) Otherwise, if P points to an array element i of an array object x with n\n elements ([dcl.array]), the expressions P + J and J + P\n (where J has the value j) point to the (possibly-hypothetical) array\n element i+j of x if 0≤i+j≤n and the expression P - J points to the\n (possibly-hypothetical) array element i−j of x if 0≤i−j≤n.\n(4.3) Otherwise, the behavior is undefined.\n```\n\nTherefore, as per the standard, applying non-zero offset to `nullptr`\n(or making non-`nullptr` a `nullptr`, by subtracting pointer\u0027s integral value\nfrom the pointer itself) is undefined behavior. (*if* `nullptr` is not defined,\ni.e. e.g. `-fno-delete-null-pointer-checks` was *not* specified.)\n\nTo make things more fun, in C (6.5.6p8), applying *any* offset to null pointer\nis undefined, although Clang front-end pessimizes the code by not lowering\nthat info, so this UB is \"harmless\".\n\nSince rL369789 (D66608 `[InstCombine] icmp eq/ne (gep inbounds P, Idx..), null -\u003e icmp eq/ne P, null`)\nLLVM middle-end uses those guarantees for transformations.\nIf the source contains such UB\u0027s, said code may now be miscompiled.\nSuch miscompilations were already observed:\n* https://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20190826/687838.html\n* https://github.com/google/filament/pull/1566\n\nSurprisingly, UBSan does not catch those issues\n... until now. This diff teaches UBSan about these UB\u0027s.\n\n`getelementpointer inbounds` is a pretty frequent instruction,\nso this does have a measurable impact on performance;\nI\u0027ve addressed most of the obvious missing folds (and thus decreased the performance impact by ~5%),\nand then re-performed some performance measurements using my [[ https://github.com/darktable-org/rawspeed | RawSpeed ]] benchmark:\n(all measurements done with LLVM ToT, the sanitizer never fired.)\n* no sanitization vs. existing check: average `+21.62%` slowdown\n* existing check vs. check after this patch: average `22.04%` slowdown\n* no sanitization vs. this patch: average `48.42%` slowdown\n\nReviewers: vsk, filcab, rsmith, aaron.ballman, vitalybuka, rjmccall, #sanitizers\n\nReviewed By: rsmith\n\nSubscribers: kristof.beyls, nickdesaulniers, nikic, ychen, dtzWill, xbolva00, dberris, arphaman, rupprecht, reames, regehr, llvm-commits, cfe-commits\n\nTags: #clang, #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D67122\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374293 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "51a6e79ff7df8a7a1ca79fdda4fe0aec571da26c", "tree": "712268fabfbf0bbb7fbc6d9143e7295d7553268d", "parents": [ "beb0403319a2b449c0db2cacdfef77061cdc6029" ], "author": { "name": "Julian Lettner", "email": "jlettner@apple.com", "time": "Thu Oct 10 00:33:04 2019" }, "committer": { "name": "Julian Lettner", "email": "jlettner@apple.com", "time": "Thu Oct 10 00:33:04 2019" }, "message": "[ASan] Do not misrepresent high value address dereferences as null dereferences\n\nDereferences with addresses above the 48-bit hardware addressable range\nproduce \"invalid instruction\" (instead of \"invalid access\") hardware\nexceptions (there is no hardware address decoding logic for those bits),\nand the address provided by this exception is the address of the\ninstruction (not the faulting address). The kernel maps the \"invalid\ninstruction\" to SEGV, but fails to provide the real fault address.\n\nBecause of this ASan lies and says that those cases are null\ndereferences. This downgrades the severity of a found bug in terms of\nsecurity. In the ASan signal handler, we can not provide the real\nfaulting address, but at least we can try not to lie.\n\nrdar://50366151\n\nReviewed By: vitalybuka\n\nDifferential Revision: https://reviews.llvm.org/D68676\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374265 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "beb0403319a2b449c0db2cacdfef77061cdc6029", "tree": "386dd9226525cecc924368c32a962dee2248990c", "parents": [ "92e35995a6175bd255489b469d7179b504b6c7bf" ], "author": { "name": "Matt Morehouse", "email": "mascasa@google.com", "time": "Wed Oct 09 22:57:07 2019" }, "committer": { "name": "Matt Morehouse", "email": "mascasa@google.com", "time": "Wed Oct 09 22:57:07 2019" }, "message": "[sanitizer_common] Remove OnPrint from Go build.\n\nSummary: Go now uses __sanitizer_on_print instead.\n\nReviewers: vitalybuka, dvyukov\n\nReviewed By: vitalybuka\n\nSubscribers: llvm-commits\n\nTags: #llvm\n\nDifferential Revision: https://reviews.llvm.org/D68621\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374258 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "92e35995a6175bd255489b469d7179b504b6c7bf", "tree": "555bfcdca5a10dea26ab49e05fd2f62265ce13a8", "parents": [ "929efdcea500576c870fc84502b23cb64c1f9124" ], "author": { "name": "Jake Ehrlich", "email": "jakehehrlich@google.com", "time": "Wed Oct 09 21:01:50 2019" }, "committer": { "name": "Jake Ehrlich", "email": "jakehehrlich@google.com", "time": "Wed Oct 09 21:01:50 2019" }, "message": "[libFuzzer] Fix Alarm callback in fuchsia.\n\nThis patch adds an #if macro to skip the InFuzzingThread() comparison\nfor fuchsia, similar to what it is done for Windows and NetBSD.\n\nIn fuchsia, the alarm callback runs in a separate thread[0], making it fail\nthe comparison InFuzzingThread(), breaking the -timeout flag.\n\n[0]:\nhttps://github.com/llvm/llvm-project/blob/master/compiler-rt/lib/fuzzer/FuzzerUtilFuchsia.cpp#L323\n\nAuthor: charco (aka Marco Vanotti)\nDifferential Revision: https://reviews.llvm.org/D68166\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374228 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "929efdcea500576c870fc84502b23cb64c1f9124", "tree": "81f3cb411bc4c728a86c27c6d1ba00e292fc974b", "parents": [ "9db3b945ea8e4c9f4e78b9b66515336f6492b713" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Wed Oct 09 20:52:39 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Wed Oct 09 20:52:39 2019" }, "message": "[sanitizer, NFC] Fix grammar in comment\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374223 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "9db3b945ea8e4c9f4e78b9b66515336f6492b713", "tree": "c31ef6ed0d07dbd28cde2c6232da41ad562cddf3", "parents": [ "6e2db02a1839370609d08ea0135de7eead7f1135" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Wed Oct 09 20:48:50 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Wed Oct 09 20:48:50 2019" }, "message": "[sanitizer] Disable signal_trap_handler on s390\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374220 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "6e2db02a1839370609d08ea0135de7eead7f1135", "tree": "dc39cf8f4ce9faff49f5ea82f0bf5530474f553e", "parents": [ "2e6a8f1ebd7896db3546ac66bb6693d8be46ade9" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Wed Oct 09 20:22:14 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Wed Oct 09 20:22:14 2019" }, "message": "[sanitizer] Make signal_name a C test\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374213 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "2e6a8f1ebd7896db3546ac66bb6693d8be46ade9", "tree": "cfdc9d68b9ca3ec4f298726e3f7eb0ba64b15c42", "parents": [ "e5d05814ac964a37eb336fb5a743227e55e85a41" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Wed Oct 09 20:18:27 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Wed Oct 09 20:18:27 2019" }, "message": "[sanitizer] Use raise() in test and cover more signals\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374211 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "e5d05814ac964a37eb336fb5a743227e55e85a41", "tree": "5f523fd7bc248b230732ec33ca41cd2679d36b2c", "parents": [ "b185f4f2ef7c96095d31e09b041ffe81b4b24571" ], "author": { "name": "Kostya Kortchinsky", "email": "kostyak@google.com", "time": "Wed Oct 09 15:09:28 2019" }, "committer": { "name": "Kostya Kortchinsky", "email": "kostyak@google.com", "time": "Wed Oct 09 15:09:28 2019" }, "message": "[scudo][standalone] Get statistics in a char buffer\n\nSummary:\nFollowing up on D68471, this CL introduces some `getStats` APIs to\ngather statistics in char buffers (`ScopedString` really) instead of\nprinting them out right away. Ultimately `printStats` will just\noutput the buffer, but that allows us to potentially do some work\non the intermediate buffer, and can be used for a `mallocz` type\nof functionality. This allows us to pretty much get rid of all the\n`Printf` calls around, but I am keeping the function in for\ndebugging purposes.\n\nThis changes the existing tests to use the new APIs when required.\n\nI will add new tests as suggested in D68471 in another CL.\n\nReviewers: morehouse, hctim, vitalybuka, eugenis, cferris\n\nReviewed By: morehouse\n\nSubscribers: delcypher, #sanitizers, llvm-commits\n\nTags: #llvm, #sanitizers\n\nDifferential Revision: https://reviews.llvm.org/D68653\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374173 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "b185f4f2ef7c96095d31e09b041ffe81b4b24571", "tree": "e853edb69fd2b2065face645827d84385e240866", "parents": [ "d8c3df27dea12cca769a430033dd5a411f296b1b" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Tue Oct 08 22:09:51 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Tue Oct 08 22:09:51 2019" }, "message": "[sanitizer] Fix crypt.cpp on Android again\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374125 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "d8c3df27dea12cca769a430033dd5a411f296b1b", "tree": "97ba4ea8c994a6366fc9ae30a44fc18b7d97e512", "parents": [ "fc749b5cf3d06d3940f04fe69b7d3c66ca8a9368" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Tue Oct 08 20:50:46 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Tue Oct 08 20:50:46 2019" }, "message": "[sanitizer] Fix crypt.cpp test on Darwin\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374115 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "fc749b5cf3d06d3940f04fe69b7d3c66ca8a9368", "tree": "7b9668c366f8e0956c1e900d558a742ddd84b1a9", "parents": [ "6b1ce8376c9177e9e67668d81a46580efab5130e" ], "author": { "name": "Dan Liew", "email": "dan@su-root.co.uk", "time": "Tue Oct 08 20:06:01 2019" }, "committer": { "name": "Dan Liew", "email": "dan@su-root.co.uk", "time": "Tue Oct 08 20:06:01 2019" }, "message": "Fix `compiler_rt_logbf_test.c` test failure for Builtins-i386-darwin test suite.\n\nSummary:\nIt seems that compiler-rt\u0027s implementation and Darwin\nlibm\u0027s implementation of `logbf()` differ when given a NaN\nwith raised sign bit. Strangely this behaviour only happens with\ni386 Darwin libm. For x86_64 and x86_64h the existing compiler-rt\nimplementation matched Darwin libm.\n\nTo workaround this the `compiler_rt_logbf_test.c` has been modified\nto do a comparison on the `fp_t` type and if that fails check if both\nvalues are NaN. If both values are NaN they are equivalent and no\nerror needs to be raised.\n\nrdar://problem/55565503\n\nReviewers: rupprecht, scanon, compnerd, echristo\nSubscribers: #sanitizers, llvm-commits\nTags: #llvm, #sanitizers\nDifferential Revision: https://reviews.llvm.org/D67999\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374109 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "6b1ce8376c9177e9e67668d81a46580efab5130e", "tree": "a9b952f25872dcdc6f629ed534599dcca661a750", "parents": [ "757fcb8a374bbe7c19c3f75ec33053f6c4013918" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Tue Oct 08 17:06:27 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Tue Oct 08 17:06:27 2019" }, "message": "[sanitizer] Disable crypt*.cpp tests on Android\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374088 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "757fcb8a374bbe7c19c3f75ec33053f6c4013918", "tree": "3f7d69d6617eeb03d50f207b536275df97c8ce23", "parents": [ "52ad1a728a9295f34debe1541d82ae391df87b8b" ], "author": { "name": "David Carlier", "email": "devnexen@gmail.com", "time": "Tue Oct 08 15:45:35 2019" }, "committer": { "name": "David Carlier", "email": "devnexen@gmail.com", "time": "Tue Oct 08 15:45:35 2019" }, "message": "[builtins] Unbreak build on FreeBSD armv7 after D60351\n\nheaders include reordering.\n\nReviewers: phosek, echristo\n\nReviewed-By: phosek\n\nDifferential Revsion: https://reviews.llvm.org/D68045\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374070 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "52ad1a728a9295f34debe1541d82ae391df87b8b", "tree": "cc57014e02e9a6fb10e7aa4d42766e5ab8fd0daf", "parents": [ "12a9be541d8e64026106c6f453702c779c7f1e97" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Tue Oct 08 02:00:53 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Tue Oct 08 02:00:53 2019" }, "message": "[sanitizer] Fix signal_trap_handler.cpp on android\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@374010 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "12a9be541d8e64026106c6f453702c779c7f1e97", "tree": "15f6abc26b55cb037f3c3036a65e76af16a180cb", "parents": [ "8ea8db5364fa341ce6dc7538be9311a725321848" ], "author": { "name": "Evgeniy Stepanov", "email": "eugeni.stepanov@gmail.com", "time": "Tue Oct 08 00:00:30 2019" }, "committer": { "name": "Evgeniy Stepanov", "email": "eugeni.stepanov@gmail.com", "time": "Tue Oct 08 00:00:30 2019" }, "message": "[msan] Add interceptors: crypt, crypt_r.\n\nReviewers: vitalybuka\n\nSubscribers: srhines, #sanitizers, llvm-commits\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D68431\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373993 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "8ea8db5364fa341ce6dc7538be9311a725321848", "tree": "595cb8f6ddc0146af4a5958558c3c2a73e7210cf", "parents": [ "634be25039e35525c078b512e6a8f98678198f80" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Mon Oct 07 23:11:07 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Mon Oct 07 23:11:07 2019" }, "message": "[tsan, go] fix Go windows build\n\nSummary:\nDon\u0027t use weak exports when building tsan into a shared library for Go. gcc can\u0027t handle the pragmas used to make the weak references.\n\nInclude files that have been added since the last update to build.bat. (We should really find a better way to list all the files needed.)\n\nAdd windows version defines (WINVER and _WIN32_WINNT) to get AcquireSRWLockExclusive and ReleaseSRWLockExclusive defined.\n\nDefine GetProcessMemoryInfo to use the kernel32 version. This is kind of a hack, the windows header files should do this translation for us. I think we\u0027re not in the right family partition (we\u0027re using Desktop, but that translation only happens for App and System partitions???), but hacking the family partition seems equally gross and I have no idea what the consequences of that might be.\n\nPatch by Keith Randall.\n\nReviewers: dvyukov, vitalybuka\n\nReviewed By: vitalybuka\n\nSubscribers: jfb, delcypher, #sanitizers, llvm-commits\n\nTags: #llvm, #sanitizers\n\nDifferential Revision: https://reviews.llvm.org/D68599\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373984 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "634be25039e35525c078b512e6a8f98678198f80", "tree": "bb45eba0245d6d63037fbb8e2c71a3653cb576f1", "parents": [ "e52028d098cedaa904cad6d48affafe27f6fe4b9" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Mon Oct 07 23:04:16 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Mon Oct 07 23:04:16 2019" }, "message": "[tsan, go] break commands into multiple lines\n\nSummary: Patch by Keith Randall.\n\nReviewers: dvyukov, vitalybuka\n\nSubscribers: delcypher, jfb, #sanitizers, llvm-commits\n\nTags: #llvm, #sanitizers\n\nDifferential Revision: https://reviews.llvm.org/D68596\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373983 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "e52028d098cedaa904cad6d48affafe27f6fe4b9", "tree": "f2fd81a6558a395270b44ad532296c87c6bfcd44", "parents": [ "50d8df97befe6f5804b38d1d974f9b3c65282744" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Mon Oct 07 22:43:19 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Mon Oct 07 22:43:19 2019" }, "message": "[sanitizer] Print SIGTRAP for corresponding signal\n\nReviewers: eugenis, jfb\n\nSubscribers: #sanitizers, llvm-commits\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D68603\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373979 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "50d8df97befe6f5804b38d1d974f9b3c65282744", "tree": "92ada8a4ba7841f45d642bf1b449cf8f5da5b186", "parents": [ "5b3c533a7b7066ffe881b7e9a15b026682e04427" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Mon Oct 07 22:43:17 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Mon Oct 07 22:43:17 2019" }, "message": "[tsan] Don\u0027t delay SIGTRAP handler\n\nReviewers: eugenis, jfb\n\nSubscribers: #sanitizers, llvm-commits\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D68604\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373978 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "5b3c533a7b7066ffe881b7e9a15b026682e04427", "tree": "d684fc80a38285242ee8b1e494c462f40a361886", "parents": [ "3154ce262c73bf5daecedbad78183d1d3d1fc648" ], "author": { "name": "Kostya Kortchinsky", "email": "kostyak@google.com", "time": "Mon Oct 07 17:37:39 2019" }, "committer": { "name": "Kostya Kortchinsky", "email": "kostyak@google.com", "time": "Mon Oct 07 17:37:39 2019" }, "message": "[scudo][standalone] Correct releaseToOS behavior\n\nSummary:\nThere was an issue in `releaseToOSMaybe`: one of the criteria to\ndecide if we should proceed with the release was wrong. Namely:\n\n```\nconst uptr N \u003d Sci-\u003eStats.PoppedBlocks - Sci-\u003eStats.PushedBlocks;\nif (N * BlockSize \u003c PageSize)\n return; // No chance to release anything.\n```\n\nI meant to check if the amount of bytes in the free list was lower\nthan a page, but this actually checks if the amount of **in use** bytes\nwas lower than a page.\n\nThe correct code is:\n\n```\nconst uptr BytesInFreeList \u003d\n Region-\u003eAllocatedUser -\n (Region-\u003eStats.PoppedBlocks - Region-\u003eStats.PushedBlocks) * BlockSize;\nif (BytesInFreeList \u003c PageSize)\n return 0; // No chance to release anything.\n```\n\nConsequences of the bug:\n- if a class size has less than a page worth of in-use bytes (allocated\n or in a cache), reclaiming would not occur, whatever the amount of\n blocks in the free list; in real world scenarios this is unlikely to\n happen and be impactful;\n- if a class size had less than a page worth of free bytes (and enough\n in-use bytes, etc), then reclaiming would be attempted, with likely\n no result. This means the reclaiming was overzealous at times.\n\nI didn\u0027t have a good way to test for this, so I changed the prototype\nof the function to return the number of bytes released, allowing to\nget the information needed. The test added fails with the initial\ncriteria.\n\nAnother issue is that `ReleaseToOsInterval` can actually be 0, meaning\nwe always try to release (side note: it\u0027s terrible for performances).\nso change a `\u003e 0` check to `\u003e\u003d 0`.\n\nAdditionally, decrease the `CanRelease` threshold to `PageSize / 32`.\nI still have to make that configurable but I will do it at another time.\n\nFinally, rename some variables in `printStats`: I feel like \"available\"\nwas too ambiguous, so change it to \"total\".\n\nReviewers: morehouse, hctim, eugenis, vitalybuka, cferris\n\nReviewed By: morehouse\n\nSubscribers: delcypher, #sanitizers, llvm-commits\n\nTags: #llvm, #sanitizers\n\nDifferential Revision: https://reviews.llvm.org/D68471\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373930 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "3154ce262c73bf5daecedbad78183d1d3d1fc648", "tree": "bd87eb3bccdb0d5b7652b63c7e72748de79755e6", "parents": [ "337963066ac61242a234ba0a231656531c1edfed" ], "author": { "name": "Kostya Kortchinsky", "email": "kostyak@google.com", "time": "Fri Oct 04 15:46:34 2019" }, "committer": { "name": "Kostya Kortchinsky", "email": "kostyak@google.com", "time": "Fri Oct 04 15:46:34 2019" }, "message": "[scudo][standalone] Make malloc_info return a minimal XML\n\nSummary:\nInitially, our malloc_info was returning ENOTSUP, but Android would\nrather have it return successfully and write a barebone XML to the\nstream, so we will oblige.\n\nAdd an associated test.\n\nReviewers: cferris, morehouse, hctim, eugenis, vitalybuka\n\nReviewed By: morehouse\n\nSubscribers: delcypher, #sanitizers, llvm-commits\n\nTags: #llvm, #sanitizers\n\nDifferential Revision: https://reviews.llvm.org/D68427\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373754 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "337963066ac61242a234ba0a231656531c1edfed", "tree": "34348e4c9e8100ff089b4e3293f29cdbd58caa72", "parents": [ "272c4f3f4656a6308f98a5c738e03995def59e0f" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Fri Oct 04 07:25:53 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Fri Oct 04 07:25:53 2019" }, "message": "[compiler-rt] Remove O1 tests from signal_line.cpp\n\nIt does not work on arm\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373702 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "272c4f3f4656a6308f98a5c738e03995def59e0f", "tree": "314a314fc584eef056f67e94c25861cdcb48b3a3", "parents": [ "0376cd8c6878bc8f8dacb570c0c2eeee67b866d6" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Fri Oct 04 00:43:05 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Fri Oct 04 00:43:05 2019" }, "message": "[compiler-rt] Fix signal_line.cpp test\n\nr373682 committed wrong experimental version\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373684 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "0376cd8c6878bc8f8dacb570c0c2eeee67b866d6", "tree": "f5b7c946ee70c4c8cdf52011dc5670916457ee40", "parents": [ "d503f912411ab41de7ee446d604034419e38c322" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Fri Oct 04 00:38:08 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Fri Oct 04 00:38:08 2019" }, "message": "[compiler-rt] Remove O2, O3 from signal_line test for fix android tests\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373682 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "d503f912411ab41de7ee446d604034419e38c322", "tree": "18a6da4be8808783110815b80c039104d4d1bb31", "parents": [ "abca7a8505b57c210b042bad8b0e89bcb454d00b" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Fri Oct 04 00:38:07 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Fri Oct 04 00:38:07 2019" }, "message": "compiler-rt: use __GLIBC_PREREQ for SANITIZER_INTERCEPT_GETRANDOM\n\nSummary: Fixes https://github.com/google/oss-fuzz/issues/2836\n\nReviewers: eugenis\n\nSubscribers: dberris, #sanitizers, llvm-commits\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D68178\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373681 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "abca7a8505b57c210b042bad8b0e89bcb454d00b", "tree": "9a14275c7e822bc0badb6a864af6b5534f7c2466", "parents": [ "5bbc57793fbcbd2d992cb675bb13453c635d2a30" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Thu Oct 03 18:18:35 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Thu Oct 03 18:18:35 2019" }, "message": "[compiler-rt] More optimization levels in signal_line.cpp test\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373642 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "5bbc57793fbcbd2d992cb675bb13453c635d2a30", "tree": "f22e61fc8c9a95e77e40edc3debfc2ae397b0f01", "parents": [ "cd9ff1a7763ea4a6996d4fa640640908fffd2db4" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Thu Oct 03 17:46:07 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Thu Oct 03 17:46:07 2019" }, "message": "compiler-rt: move all __GLIBC_PREREQ into own header file\n\nReviewers: eugenis\n\nSubscribers: dberris, #sanitizers, llvm-commits\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D68177\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373632 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "cd9ff1a7763ea4a6996d4fa640640908fffd2db4", "tree": "7823d66d450a104efdfea300bb3c2792088b67aa", "parents": [ "f1c1315d806c358795f78be74dcedebed8ebc937" ], "author": { "name": "Matt Morehouse", "email": "mascasa@google.com", "time": "Wed Oct 02 22:57:37 2019" }, "committer": { "name": "Matt Morehouse", "email": "mascasa@google.com", "time": "Wed Oct 02 22:57:37 2019" }, "message": "Move SanitizerInitializeUnwinder outside anonymous namespace.\n\nFixes the Android build breakage introduced in r373528.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373539 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "f1c1315d806c358795f78be74dcedebed8ebc937", "tree": "fffe268bbf41a525b9676ebdb55533e92565733b", "parents": [ "8cfda3057d3cd96793aecfe4514906cb4040f4ff" ], "author": { "name": "Matt Morehouse", "email": "mascasa@google.com", "time": "Wed Oct 02 22:00:29 2019" }, "committer": { "name": "Matt Morehouse", "email": "mascasa@google.com", "time": "Wed Oct 02 22:00:29 2019" }, "message": "[sanitizer_common] Declare __sanitizer_on_print in Windows interception.\n\nSpeculative fix for Windows bot.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373532 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "8cfda3057d3cd96793aecfe4514906cb4040f4ff", "tree": "6bf1ee8b833974ea829eb5183237cb25769ff8e7", "parents": [ "2666d540c2c6b0d177df6bc20c8572f53399c27a" ], "author": { "name": "Matt Morehouse", "email": "mascasa@google.com", "time": "Wed Oct 02 21:38:22 2019" }, "committer": { "name": "Matt Morehouse", "email": "mascasa@google.com", "time": "Wed Oct 02 21:38:22 2019" }, "message": "[sanitizer_common] Disable onprint.cpp on Android.\n\nThe test fails to find the written file on Android.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373531 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "2666d540c2c6b0d177df6bc20c8572f53399c27a", "tree": "51e7ef0419da4e11ca1af9fa08f95861a7450579", "parents": [ "177d0c91d6f86adfca2f466141c7a4532209935f" ], "author": { "name": "Matt Morehouse", "email": "mascasa@google.com", "time": "Wed Oct 02 21:21:18 2019" }, "committer": { "name": "Matt Morehouse", "email": "mascasa@google.com", "time": "Wed Oct 02 21:21:18 2019" }, "message": "[sanitizer_common] Add __sanitizer_on_print to interface list.\n\nShould fix the current Windows buildbot failure.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373530 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "177d0c91d6f86adfca2f466141c7a4532209935f", "tree": "7fe1d35fb91f5707cdf020dca8284d1abb7671b0", "parents": [ "ade11291cd1f477fcd0c20a0448237e671932442" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Wed Oct 02 21:20:37 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Wed Oct 02 21:20:37 2019" }, "message": "[compiler-rt] Use GetNextInstructionPc in signal handlers\n\nSummary:\nAll other stack trace callers assume that PC contains return address.\nHWAsan already use GetNextInstructionPc in similar code.\n\nPR43339\n\nReviewers: eugenis, kcc, jfb\n\nSubscribers: dexonsmith, dberris, #sanitizers, llvm-commits\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D68313\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373529 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "ade11291cd1f477fcd0c20a0448237e671932442", "tree": "72dc8416ad8567b330c4ad917d5153a10fcfa2d9", "parents": [ "c88ce270afe17bf35376a89c8b929eee93764e07" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Wed Oct 02 21:20:32 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Wed Oct 02 21:20:32 2019" }, "message": "NFC: add namespace {}\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373528 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "c88ce270afe17bf35376a89c8b929eee93764e07", "tree": "b696e8ea2d8cd0d90b6617d50c5b77c357bcd6a6", "parents": [ "5f8a0110ec12ac0b7c86545c5058ab287c070b5e" ], "author": { "name": "Matt Morehouse", "email": "mascasa@google.com", "time": "Wed Oct 02 20:13:21 2019" }, "committer": { "name": "Matt Morehouse", "email": "mascasa@google.com", "time": "Wed Oct 02 20:13:21 2019" }, "message": "[sanitizer_common] Rename OnPrint to __sanitizer_on_print.\n\nSummary:\nhttps://reviews.llvm.org/D28596 exposed OnPrint in the global namespace,\nwhich can cause collisions with user-defined OnPrint() functions.\n\nReviewers: vitalybuka, dvyukov\n\nReviewed By: vitalybuka, dvyukov\n\nSubscribers: llvm-commits\n\nTags: #llvm\n\nDifferential Revision: https://reviews.llvm.org/D67987\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373518 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "5f8a0110ec12ac0b7c86545c5058ab287c070b5e", "tree": "29c61705b3a7ff00220fdbaf48f2f542f29a11d1", "parents": [ "156cec6232ae2971456bc206bfd0fce1133a1e6d" ], "author": { "name": "Dan Liew", "email": "dan@su-root.co.uk", "time": "Tue Oct 01 23:08:18 2019" }, "committer": { "name": "Dan Liew", "email": "dan@su-root.co.uk", "time": "Tue Oct 01 23:08:18 2019" }, "message": "[CMake] Fix the value of `config.target_cflags` for non-macOS Apple platforms. Attempt #3.\n\nThe main problem here is that `-*-version_min\u003d` was not being passed to\nthe compiler when building test cases. This can cause problems when\ntesting on devices running older OSs because Clang would previously\nassume the minimum deployment target is the the latest OS in the SDK\nwhich could be much newer than what the device is running.\n\nPreviously the generated value looked like this:\n\n`-arch arm64 -isysroot\n\u003cpath_to_xcode\u003e/Contents/Developer/Platforms/iPhoneOS.platform/Developer/SDKs/iPhoneOS12.1.sdk`\n\nWith this change it now looks like:\n\n`-arch arm64 -stdlib\u003dlibc++ -miphoneos-version-min\u003d8.0 -isysroot\n\u003cpath_to_xcode\u003e/Contents/Developer/Platforms/iPhoneOS.platform/Developer/SDKs/iPhoneOS12.1.sdk`\n\nThis mirrors the setting of config.target_cflags on macOS.\n\nThis change is made for ASan, LibFuzzer, TSan, and UBSan.\n\nTo implement this a new `get_test_cflags_for_apple_platform()` function\nhas been added that when given an Apple platform name and architecture\nreturns a string containing the C compiler flags to use when building\ntests. This also calls a new helper function `is_valid_apple_platform()`\nthat validates Apple platform names.\n\nThis is the third attempt at landing the patch.\n\nThe first attempt (r359305) had to be reverted (r359327) due to a buildbot\nfailure. The problem was that calling `get_test_cflags_for_apple_platform()`\ncan trigger a CMake error if the provided architecture is not supported by the\ncurrent CMake configuration. Previously, this could be triggered by passing\n`-DCOMPILER_RT_ENABLE_IOS\u003dOFF` to CMake. The root cause is that we were\ngenerating test configurations for a list of architectures without checking if\nthe relevant Sanitizer actually supported that architecture. We now intersect\nthe list of architectures for an Apple platform with\n`\u003cSANITIZER\u003e_SUPPORTED_ARCH` (where `\u003cSANITIZER\u003e` is a Sanitizer name) to\niterate through the correct list of architectures.\n\nThe second attempt (r363633) had to be reverted (r363779) due to a build\nfailure. The failed build was using a modified Apple toolchain where the iOS\nsimulator SDK was missing. This exposed a bug in the existing UBSan test\ngeneration code where it was assumed that `COMPILER_RT_ENABLE_IOS` implied that\nthe toolchain supported both iOS and the iOS simulator. This is not true. This\nhas been fixed by using the list `SANITIZER_COMMON_SUPPORTED_OS` for the list\nof supported Apple platforms for UBSan. For consistency with the other\nSanitizers we also now intersect the list of architectures with\nUBSAN_SUPPORTED_ARCH.\n\nrdar://problem/50124489\n\nDifferential Revision: https://reviews.llvm.org/D61242\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373405 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "156cec6232ae2971456bc206bfd0fce1133a1e6d", "tree": "4b3c71ab1d5336be65bb4ff9ab8a46bce921ed51", "parents": [ "d69ff12cfe5254a0862ba2950523d3a44a205ed4" ], "author": { "name": "Matt Morehouse", "email": "mascasa@google.com", "time": "Tue Oct 01 22:49:06 2019" }, "committer": { "name": "Matt Morehouse", "email": "mascasa@google.com", "time": "Tue Oct 01 22:49:06 2019" }, "message": "[libFuzzer] Remove lazy counters.\n\nSummary: Lazy counters haven\u0027t improved performance for large fuzz targets.\n\nReviewers: kcc\n\nReviewed By: kcc\n\nSubscribers: llvm-commits\n\nTags: #llvm\n\nDifferential Revision: https://reviews.llvm.org/D67476\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373403 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "d69ff12cfe5254a0862ba2950523d3a44a205ed4", "tree": "99a4f0e29f44be860619595e0e3c5fb434e2acfd", "parents": [ "8ea41dbb8a53edd40b3b5562b1585137efbeab4e" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Tue Oct 01 18:03:11 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Tue Oct 01 18:03:11 2019" }, "message": "Revert \"compiler-rt: use __GLIBC_PREREQ for SANITIZER_INTERCEPT_GETRANDOM\"\nRevert \"compiler-rt: move all __GLIBC_PREREQ into own header file\"\n\n\"move all __GLIBC_PREREQ\" breaks build on some bots\n\nThis reverts commit 2d75ee937397c209dbd95aefc88da6301fed07da.\nThis reverts commit 7a6461fcc2ed8e28c43993c561721af0bbe97f3a.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373367 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "8ea41dbb8a53edd40b3b5562b1585137efbeab4e", "tree": "db4f5df661886120fef7183dc4c2639f242caeff", "parents": [ "f4c36bc0240c173724c1b8126882b50dd667ff5a" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Tue Oct 01 00:58:29 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Tue Oct 01 00:58:29 2019" }, "message": "compiler-rt: use __GLIBC_PREREQ for SANITIZER_INTERCEPT_GETRANDOM\n\nSummary: Fixes https://github.com/google/oss-fuzz/issues/2836\n\nReviewers: eugenis\n\nSubscribers: dberris, #sanitizers, llvm-commits\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D68178\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373284 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "f4c36bc0240c173724c1b8126882b50dd667ff5a", "tree": "500494162f755aebbf778bb2411e41577d5754a5", "parents": [ "eb7816fac22e926dd5c24dd48ebfb0379a53f957" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Tue Oct 01 00:58:28 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Tue Oct 01 00:58:28 2019" }, "message": "compiler-rt: move all __GLIBC_PREREQ into own header file\n\nReviewers: eugenis\n\nSubscribers: dberris, #sanitizers, llvm-commits\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D68177\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373283 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "eb7816fac22e926dd5c24dd48ebfb0379a53f957", "tree": "99a4f0e29f44be860619595e0e3c5fb434e2acfd", "parents": [ "0f6ecea9555533c4414cf6addd2ffed02c1938ec" ], "author": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Tue Oct 01 00:58:26 2019" }, "committer": { "name": "Vitaly Buka", "email": "vitalybuka@google.com", "time": "Tue Oct 01 00:58:26 2019" }, "message": "Rename tsan_interceptors.cpp into tsan_interceptors_posix.cpp\n\nSummary:\nIt\u0027s needed to use __GLIBC_PREREQ from \u003cfeatures.h\u003e\n\ntsan didn\u0027t let us to include \u003cfeatures.h\u003e by using --sysroot\u003d. to disable system includes on\nanything that is not named as \"tsan*posix*\", \"tsan*mac*\", \"tsan*linux*\".\nSee compiler-rt/lib/tsan/CMakeLists.txt\n\nReviewers: eugenis, dvyukov, kcc\n\nReviewed By: kcc\n\nSubscribers: mgorny, #sanitizers, llvm-commits\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D68176\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373282 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "0f6ecea9555533c4414cf6addd2ffed02c1938ec", "tree": "b3cf0f719a54b3947fd72c859bafe02bc2b8607e", "parents": [ "2b7e04031eae52b103c2dd20d4aaf6c4a9f7251e" ], "author": { "name": "Evgeniy Stepanov", "email": "eugeni.stepanov@gmail.com", "time": "Mon Sep 30 17:49:48 2019" }, "committer": { "name": "Evgeniy Stepanov", "email": "eugeni.stepanov@gmail.com", "time": "Mon Sep 30 17:49:48 2019" }, "message": "[msan] Intercept __getrlimit.\n\nSummary:\nThis interceptor is useful on its own, but the main purpose of this\nchange is to intercept libpthread initialization on linux/glibc in\norder to run __msan_init before any .preinit_array constructors.\n\nWe used to trigger on pthread_initialize_minimal -\u003e getrlimit(), but\nthat call has changed to __getrlimit at some point.\n\nReviewers: vitalybuka, pcc\n\nSubscribers: jfb, #sanitizers, llvm-commits\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D68168\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373239 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "2b7e04031eae52b103c2dd20d4aaf6c4a9f7251e", "tree": "303fb17aec69fbe385c3dba2197bce3de3cc0534", "parents": [ "298de05963130a8c282d2a67d1a0971c38696043" ], "author": { "name": "Mitch Phillips", "email": "mitchphillips@outlook.com", "time": "Fri Sep 27 22:04:36 2019" }, "committer": { "name": "Mitch Phillips", "email": "mitchphillips@outlook.com", "time": "Fri Sep 27 22:04:36 2019" }, "message": "[libFuzzer] Dump trace and provide correct msg for overwritten input.\n\nSummary:\nNow crashes with a stacktrace and uses \u0027overwrites-const-input\u0027 as the error\nmessage instead of \u0027out-of-memory\u0027.\n\nReviewers: morehouse, Dor1s\n\nReviewed By: morehouse, Dor1s\n\nSubscribers: #sanitizers, llvm-commits, metzman, Dor1s\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D68067\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373130 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "298de05963130a8c282d2a67d1a0971c38696043", "tree": "28e40041c168331c704955494dc72cec7cc785cb", "parents": [ "c1ac214eef56861e2ba85cb5f55b82b73c0d48be" ], "author": { "name": "Vedant Kumar", "email": "vsk@apple.com", "time": "Fri Sep 27 20:24:48 2019" }, "committer": { "name": "Vedant Kumar", "email": "vsk@apple.com", "time": "Fri Sep 27 20:24:48 2019" }, "message": "Revert \"[profile] Add a test dependency on cxx-headers\"\n\nThis reverts commit b539350f7d006b7d6f42c5c4b5715da87a52e5d8.\n\nSee: http://lab.llvm.org:8011/builders/sanitizer-windows/builds/52140/steps/annotate/logs/stdio\n\nThe cxx-headers target doesn\u0027t exist everywhere.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373123 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "c1ac214eef56861e2ba85cb5f55b82b73c0d48be", "tree": "4e12c82583b408b02c62402b574ed594f8999581", "parents": [ "4f62b857fcd59fb3c7398928fb5e34c0ec36ae22" ], "author": { "name": "Vedant Kumar", "email": "vsk@apple.com", "time": "Fri Sep 27 20:12:38 2019" }, "committer": { "name": "Vedant Kumar", "email": "vsk@apple.com", "time": "Fri Sep 27 20:12:38 2019" }, "message": "[profile] Mark instrprof-gcov-fork.test UNSUPPORTED on Darwin as well\n\nThis test remains flaky everywhere, I think. We should consider deleting\nit and accompanying support code in GCOVProfiling: I\u0027ve stopped short of\ndoing that now as the gcov exec* tests appear to be stable.\n\nSee the thread re: r347779.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373121 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "4f62b857fcd59fb3c7398928fb5e34c0ec36ae22", "tree": "d253b63e387247cd451984aa812f4226a360e8c2", "parents": [ "bd083fac50c327010787396271c799b5cc576090" ], "author": { "name": "Vedant Kumar", "email": "vsk@apple.com", "time": "Fri Sep 27 20:12:35 2019" }, "committer": { "name": "Vedant Kumar", "email": "vsk@apple.com", "time": "Fri Sep 27 20:12:35 2019" }, "message": "[profile] Add a test dependency on cxx-headers\n\nThis enables running profile runtime tests which #include \u003cstring\u003e, etc.\nvia just `check-profile`.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373120 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "bd083fac50c327010787396271c799b5cc576090", "tree": "445bd1bc7cd19ce43560298f9cddb597c58f1a97", "parents": [ "8bc91a47c8c1c42eb99c5d6030a0d8c785548f0a" ], "author": { "name": "Peter Collingbourne", "email": "peter@pcc.me.uk", "time": "Fri Sep 27 01:02:10 2019" }, "committer": { "name": "Peter Collingbourne", "email": "peter@pcc.me.uk", "time": "Fri Sep 27 01:02:10 2019" }, "message": "hwasan: Compatibility fixes for short granules.\n\nWe can\u0027t use short granules with stack instrumentation when targeting older\nAPI levels because the rest of the system won\u0027t understand the short granule\ntags stored in shadow memory.\n\nMoreover, we need to be able to let old binaries (which won\u0027t understand\nshort granule tags) run on a new system that supports short granule\ntags. Such binaries will call the __hwasan_tag_mismatch function when their\noutlined checks fail. We can compensate for the binary\u0027s lack of support\nfor short granules by implementing the short granule part of the check in\nthe __hwasan_tag_mismatch function. Unfortunately we can\u0027t do anything about\ninline checks, but I don\u0027t believe that we can generate these by default on\naarch64, nor did we do so when the ABI was fixed.\n\nA new function, __hwasan_tag_mismatch_v2, is introduced that lets code\ntargeting the new runtime avoid redoing the short granule check. Because tag\nmismatches are rare this isn\u0027t important from a performance perspective; the\nmain benefit is that it introduces a symbol dependency that prevents binaries\ntargeting the new runtime from running on older (i.e. incompatible) runtimes.\n\nDifferential Revision: https://reviews.llvm.org/D68059\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@373035 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "8bc91a47c8c1c42eb99c5d6030a0d8c785548f0a", "tree": "27093c4f24affe244212172db029674ff686f45b", "parents": [ "e01a28f8cd0672b26a2cf80ad7ebafe953063293" ], "author": { "name": "Mitch Phillips", "email": "mitchphillips@outlook.com", "time": "Thu Sep 26 00:54:30 2019" }, "committer": { "name": "Mitch Phillips", "email": "mitchphillips@outlook.com", "time": "Thu Sep 26 00:54:30 2019" }, "message": "[libFuzzer] [NFC] Fix grammar error with \"it\u0027s\"\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372937 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "e01a28f8cd0672b26a2cf80ad7ebafe953063293", "tree": "26594a576804dc76493c662f06a492bb0e2044c1", "parents": [ "1db063ce1e1c7e244f429fafc0c55df631bd12f5" ], "author": { "name": "Nico Weber", "email": "nicolasweber@gmx.de", "time": "Wed Sep 25 11:57:51 2019" }, "committer": { "name": "Nico Weber", "email": "nicolasweber@gmx.de", "time": "Wed Sep 25 11:57:51 2019" }, "message": "builtins test: Move clear_cache_test.c from a mprotect()ed global to a mmap()ed variable\n\nld64 in the macOS 10.15 SDK gives __DATA a maxprot of 3, meaning it\ncan\u0027t be made executable at runtime by default.\n\nChange clear_cache_test.c to use mmap()ed data that\u0027s mapped as writable\nand executable from the beginning, instead of trying to mprotect()ing a\n__DATA variable as executable. This fixes the test on macOS with the\n10.15 SDK.\n\nPR43407.\n\nDifferential Revision: https://reviews.llvm.org/D67929\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372849 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "1db063ce1e1c7e244f429fafc0c55df631bd12f5", "tree": "1b4b1e931f73ddf1e2c31e87203def60cd03ea89", "parents": [ "f754c5f52bf33370d9444463f37ce22d9316f204" ], "author": { "name": "Puyan Lotfi", "email": "puyan@puyan.org", "time": "Tue Sep 24 23:48:42 2019" }, "committer": { "name": "Puyan Lotfi", "email": "puyan@puyan.org", "time": "Tue Sep 24 23:48:42 2019" }, "message": "[NFC] Invoke lipo from CMAKE_LIPO.\n\nThis shouldn\u0027t change anything, except that a cmake cache file that specifies\nCMAKE_LIPO can specify an alternate lipo to use.\n\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372790 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "f754c5f52bf33370d9444463f37ce22d9316f204", "tree": "5f5760bdbbc20d85f7d65dbc0fb3a7417857ff0a", "parents": [ "b5c751dcf99002b7cecdc1975ac107dc4fdef652" ], "author": { "name": "Joachim Protze", "email": "protze@itc.rwth-aachen.de", "time": "Tue Sep 24 11:19:02 2019" }, "committer": { "name": "Joachim Protze", "email": "protze@itc.rwth-aachen.de", "time": "Tue Sep 24 11:19:02 2019" }, "message": "[TSAN] Add read/write range interface functions with PC\n\nAdding annotation function variants __tsan_write_range_pc and\n__tsan_read_range_pc to annotate ranged access to memory while providing a\nprogram counter for the access.\n\nDifferential Revision: https://reviews.llvm.org/D66885\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372730 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "b5c751dcf99002b7cecdc1975ac107dc4fdef652", "tree": "ea39e115c6fb1c6282ac05842edaec9d0b46f77e", "parents": [ "fe62c81099a0b246783b8f1d1434219a5fdc08e6" ], "author": { "name": "Dmitry Vyukov", "email": "dvyukov@google.com", "time": "Tue Sep 24 08:27:51 2019" }, "committer": { "name": "Dmitry Vyukov", "email": "dvyukov@google.com", "time": "Tue Sep 24 08:27:51 2019" }, "message": "sanitizer_common: fix freebsd build error\n\nVariable flags is not used. Remove it.\n\nSuggested-by: randall77 (Keith Randall)\nReview: https://reviews.llvm.org/D67928\n\n\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372698 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "fe62c81099a0b246783b8f1d1434219a5fdc08e6", "tree": "426d3bf16adadb6a63fd1513a0bf9ab93bb8da20", "parents": [ "0690f4e61e8a2fdf221d6c061d7a05e7c2fa52dc" ], "author": { "name": "Rumeet Dhindsa", "email": "rdhindsa@google.com", "time": "Tue Sep 24 02:59:02 2019" }, "committer": { "name": "Rumeet Dhindsa", "email": "rdhindsa@google.com", "time": "Tue Sep 24 02:59:02 2019" }, "message": "\nFix int to bool errors exposed due to r372612.\n \nDifferential Revision: https://reviews.llvm.org/D67937\nM lib/builtins/fp_add_impl.inc\nM lib/builtins/fp_lib.h\nM lib/builtins/fp_trunc_impl.inc\n\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372684 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "0690f4e61e8a2fdf221d6c061d7a05e7c2fa52dc", "tree": "914325e636f1cc9e9bc3db0082c7007db8631ef4", "parents": [ "2f240f447422b9f657f5449eb18f54ec1ed3fec1" ], "author": { "name": "Artem Dergachev", "email": "artem.dergachev@gmail.com", "time": "Tue Sep 24 00:01:51 2019" }, "committer": { "name": "Artem Dergachev", "email": "artem.dergachev@gmail.com", "time": "Tue Sep 24 00:01:51 2019" }, "message": "[llvm-cov] NFC: Specify a specific C++ standard in the test.\n\nMakes life easier for downstream users with customized default standard.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372674 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "2f240f447422b9f657f5449eb18f54ec1ed3fec1", "tree": "16d9f436e33fc47dc86fcf191ba99aa8540c70a4", "parents": [ "16dfa3fc2f2ac2a3ec23e61a448c8595397ffbaf" ], "author": { "name": "Jinsong Ji", "email": "jji@us.ibm.com", "time": "Sun Sep 22 15:31:03 2019" }, "committer": { "name": "Jinsong Ji", "email": "jji@us.ibm.com", "time": "Sun Sep 22 15:31:03 2019" }, "message": "[compiler-rt] Fix lint check failure on comments\n\nThis fixes buildbot failures for https://reviews.llvm.org/rL372459.\n(at least on PowerPC/Z )\n\nThe fix is generated by running clang-format on the error lines only.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372511 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "16dfa3fc2f2ac2a3ec23e61a448c8595397ffbaf", "tree": "a2cb08c4a1565cbbbcb3b5d30ef209e6675cc237", "parents": [ "54f5dca4b6446825d51b1d0b4bd313079dbfda70" ], "author": { "name": "Kamil Rytarowski", "email": "n54@gmx.com", "time": "Sat Sep 21 07:45:02 2019" }, "committer": { "name": "Kamil Rytarowski", "email": "n54@gmx.com", "time": "Sat Sep 21 07:45:02 2019" }, "message": "Add __lsan::ScopedInterceptorDisabler for strerror(3)\n\nSummary:\nstrerror(3) on NetBSD uses internally TSD with a destructor that is never\nfired for exit(3). It\u0027s correctly called for pthread_exit(3) scenarios.\n\nThis is a case when a leak on exit(3) is expected, unavoidable and harmless.\n\nReviewers: joerg, vitalybuka, dvyukov, mgorny\n\nReviewed By: vitalybuka\n\nSubscribers: dmgreen, kristof.beyls, jfb, llvm-commits, #sanitizers\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D67337\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372461 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "54f5dca4b6446825d51b1d0b4bd313079dbfda70", "tree": "b13c1a65c42365a9fd5da0cba6e8caa09e6ad0e6", "parents": [ "96b852a95bebc7c2d9da09975a4764c365167621" ], "author": { "name": "Kamil Rytarowski", "email": "n54@gmx.com", "time": "Sat Sep 21 07:43:55 2019" }, "committer": { "name": "Kamil Rytarowski", "email": "n54@gmx.com", "time": "Sat Sep 21 07:43:55 2019" }, "message": "Avoid memory leak in ASan test\n\nSummary:\nAdd missing free(3) for the malloc(3) call.\n\nDetected on NetBSD with LSan.\n\nReviewers: joerg, mgorny, vitalybuka, dvyukov\n\nReviewed By: vitalybuka\n\nSubscribers: llvm-commits, #sanitizers\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D67330\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372460 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "96b852a95bebc7c2d9da09975a4764c365167621", "tree": "ef6cf465bc7fac9830d1ba6a74ab43472b7a0681", "parents": [ "dc2e8074dc573a070e71da4be3d97c5f6b765dba" ], "author": { "name": "Kamil Rytarowski", "email": "n54@gmx.com", "time": "Sat Sep 21 07:30:42 2019" }, "committer": { "name": "Kamil Rytarowski", "email": "n54@gmx.com", "time": "Sat Sep 21 07:30:42 2019" }, "message": "Stop tracking atexit/__cxa_atexit/pthread_atfork allocations in LSan/NetBSD\n\nSummary:\nThe atexit(3) and __cxa_atexit() calls allocate internally memory and free on exit,\nafter executing all callback. This causes false positives as DoLeakCheck() is called\nfrom the atexit handler. In the LSan/ASan tests there are strict checks triggering\nfalse positives here.\n\nIntercept all atexit(3) and __cxa_atexit() calls and disable LSan when calling the\nreal functions.\n\nStop tracing allocations in pthread_atfork(3) funtions, as there are performed\ninternal allocations that are not freed for the time of running StopTheWorld()\ncode. This avoids false-positives.\n\nThe same changes have to be replicated in the ASan and LSan runtime.\n\nNon-NetBSD OSs are not tested and this code is restricted to NetBSD only.\n\nReviewers: dvyukov, joerg, mgorny, vitalybuka, eugenis\n\nReviewed By: vitalybuka\n\nSubscribers: jfb, llvm-commits, #sanitizers\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D67331\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372459 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "dc2e8074dc573a070e71da4be3d97c5f6b765dba", "tree": "c8d2c70dd7a4ac0503148c7a052078c9c9645a79", "parents": [ "a5996120c6cf50aaaa9f137945ae07176f37b612" ], "author": { "name": "Evgeniy Stepanov", "email": "eugeni.stepanov@gmail.com", "time": "Thu Sep 19 19:52:57 2019" }, "committer": { "name": "Evgeniy Stepanov", "email": "eugeni.stepanov@gmail.com", "time": "Thu Sep 19 19:52:57 2019" }, "message": "[lsan] Fix deadlock in dl_iterate_phdr.\n\nSummary:\nDo not grab the allocator lock before calling dl_iterate_phdr. This may\ncause a lock order inversion with (valid) user code that uses malloc\ninside a dl_iterate_phdr callback.\n\nReviewers: vitalybuka, hctim\n\nSubscribers: jfb, #sanitizers, llvm-commits\n\nTags: #sanitizers, #llvm\n\nDifferential Revision: https://reviews.llvm.org/D67738\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372348 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "a5996120c6cf50aaaa9f137945ae07176f37b612", "tree": "adb5fa13756a6e34033f34f789ebe572d07ac143", "parents": [ "121faed8f831cc672d10460e0ae2e65296b3e0d4" ], "author": { "name": "Hans Wennborg", "email": "hans@hanshq.net", "time": "Thu Sep 19 11:28:32 2019" }, "committer": { "name": "Hans Wennborg", "email": "hans@hanshq.net", "time": "Thu Sep 19 11:28:32 2019" }, "message": "[cmake] Strip quotes in try_compile_only\n\nAfter r372209, the compile command can end up including an argument with\nquotes in it, e.g.\n\n -fprofile-instr-use\u003d\"/foo/bar.profdata\"\n\nwhen invoking the compiler with execute_process, the compiler ends up\ngetting that argument with quotes and all, and fails to open the file.\n\nThis all seems horribly broken, but one way of working around it is to\nsimply strip the quotes from the string here. If they were there to\nprotect a path that\u0027s got spaces in it, that wasn\u0027t going to work\nanyway because the string is later split by spaces.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372312 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "121faed8f831cc672d10460e0ae2e65296b3e0d4", "tree": "04babb5f58341a3c896fa25df0b4d0b905169d1c", "parents": [ "d5f68535b6a1fbb4f8afed3d999d965861bb8a4d" ], "author": { "name": "Dmitry Vyukov", "email": "dvyukov@google.com", "time": "Wed Sep 18 09:18:04 2019" }, "committer": { "name": "Dmitry Vyukov", "email": "dvyukov@google.com", "time": "Wed Sep 18 09:18:04 2019" }, "message": "tsan: allow the Go runtime to return multiple stack frames for a single PC\n\nThis fix allows tsan to report stack traces correctly even in the\npresence of mid-stack inlining by the Go compiler.\n\nSee https://go-review.googlesource.com/c/go/+/195781 for the Go runtime side of this change.\n\nAuthor: randall77 (Keith Randall)\nReviewed: https://reviews.llvm.org/D67671\n\n\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372205 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "d5f68535b6a1fbb4f8afed3d999d965861bb8a4d", "tree": "6e78a3ca810787f0893084ee9e787f76136f9c60", "parents": [ "0254289ab2423cec855044848ccca817ae0f45de" ], "author": { "name": "David Bolvansky", "email": "david.bolvansky@gmail.com", "time": "Tue Sep 17 17:17:30 2019" }, "committer": { "name": "David Bolvansky", "email": "david.bolvansky@gmail.com", "time": "Tue Sep 17 17:17:30 2019" }, "message": "Remove asan test for strncat(x, y, 0)\n\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372143 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "0254289ab2423cec855044848ccca817ae0f45de", "tree": "edb02b7f9a524c2103bd5184f01b3aef1707e10b", "parents": [ "c47e688f48a50aaf27ac05e073630057517d4674" ], "author": { "name": "David Bolvansky", "email": "david.bolvansky@gmail.com", "time": "Tue Sep 17 17:07:31 2019" }, "committer": { "name": "David Bolvansky", "email": "david.bolvansky@gmail.com", "time": "Tue Sep 17 17:07:31 2019" }, "message": "[ASAN] Adjust asan tests due to new optimizations\n\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372141 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "c47e688f48a50aaf27ac05e073630057517d4674", "tree": "5bb0e8aa673ef21d6bd1fff9fd52696af69871d8", "parents": [ "e6a6d6c9c3259a7e9b7cc1541be5dc5e2c1c8445" ], "author": { "name": "Jake Ehrlich", "email": "jakehehrlich@google.com", "time": "Tue Sep 17 00:34:41 2019" }, "committer": { "name": "Jake Ehrlich", "email": "jakehehrlich@google.com", "time": "Tue Sep 17 00:34:41 2019" }, "message": "[libFuzzer] Always print DSO map on Fuchsia libFuzzer launch\n\nFuchsia doesn\u0027t have /proc/id/maps, so it relies on the kernel logging system\nto provide the DSO map to be able to symbolize in the context of ASLR. The DSO\nmap is logged automatically on Fuchsia when encountering a crash or writing to\nthe sanitizer log for the first time in a process. There are several cases\nwhere libFuzzer doesn\u0027t encounter a crash, e.g. on timeouts, OOMs, and when\nconfigured to print new PCs as they become covered, to name a few. Therefore,\nthis change always writes to the sanitizer log on startup to ensure the DSO map\nis available in the log.\n\nAuthor: aarongreen\nDifferential Revision: https://reviews.llvm.org/D66233\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372056 91177308-0d34-0410-b5e6-96231b3b80d8\n" }, { "commit": "e6a6d6c9c3259a7e9b7cc1541be5dc5e2c1c8445", "tree": "c3aa9ac30f966b973bc3691797dfb1cb875bf665", "parents": [ "8626a35a71f4398bc9b654b7375a2cdb17bd735d" ], "author": { "name": "Vedant Kumar", "email": "vsk@apple.com", "time": "Mon Sep 16 22:32:18 2019" }, "committer": { "name": "Vedant Kumar", "email": "vsk@apple.com", "time": "Mon Sep 16 22:32:18 2019" }, "message": "[test] Clean up previous raw profile before merging into it\n\nThis fixes a test failure in instrprof-set-file-object-merging.c which\nseems to have been caused by reuse of stale data in old raw profiles.\n\ngit-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@372041 91177308-0d34-0410-b5e6-96231b3b80d8\n" } ], "next": "8626a35a71f4398bc9b654b7375a2cdb17bd735d" }