)]}' { "log": [ { "commit": "18b17e97757a59718a23f8128dfc91c016df08b1", "tree": "90922f85d3f49e92b91290a5e97321963ae4e1cc", "parents": [ "19718f0cd605f55df26823c67b62ceab6fe99a00" ], "author": { "name": "Elliott Hughes", "email": "enh@google.com", "time": "Fri Jun 06 21:43:33 2014 -0700" }, "committer": { "name": "Elliott Hughes", "email": "enh@google.com", "time": "Fri Jun 06 21:55:27 2014 -0700" }, "message": "Add a regression test for two libm bugs we didn\u0027t have.\n\nReported on the OpenBSD list, but we already had the fix for one from FreeBSD,\nand I think the other only affected ld80 anyway. Worth having tests thuogh.\n\nChange-Id: Ic4bbeb2384fd578a3ef13e4907be83deda50815f\n" }, { "commit": "de9ac710534626e6796ea885f8480fc4cfd64158", "tree": "8c2a2b5a20da5fd47de17b941fe5dc71672628e6", "parents": [ "94a9ba2b4a4d8be7e3520e34bfc7bd47ff7af0ce" ], "author": { "name": "Elliott Hughes", "email": "enh@google.com", "time": "Mon May 19 16:58:52 2014 -0700" }, "committer": { "name": "Elliott Hughes", "email": "enh@google.com", "time": "Mon May 19 16:58:52 2014 -0700" }, "message": "Use -fvisibility\u003dhidden to build libm.\n\nBug: 11156955\nChange-Id: I77b7772a44313a1eb3970f5d2c8e2af63690efa4\n" }, { "commit": "9a5a3e8e74e2c6d7850ab29711189acf0f7378a2", "tree": "54310a7a9d342c88ce8f8073b6b90e675641af19", "parents": [ "185dd72a2c487502717903065e46d14039f17826" ], "author": { "name": "Elliott Hughes", "email": "enh@google.com", "time": "Mon May 05 20:28:28 2014 -0700" }, "committer": { "name": "Elliott Hughes", "email": "enh@google.com", "time": "Mon May 05 21:19:47 2014 -0700" }, "message": "Fix \u003cmath.h\u003e to quieten most of our warnings.\n\nI\u0027ve reported the wcsftime bug upstream, but we really just want to use -D\nto ensure the buggy code isn\u0027t built. (I\u0027ve also brought our strftime a bit\ncloser to upstream now we have the right define.)\n\nI don\u0027t think upstream is likely to fix all their sign-compare and\nuninitialized warnings, so let\u0027s just silence them.\n\nAs for libm, again upstream isn\u0027t likely to fix all their warnings, and\nsilencing those made the ones that were our fault stand out. I\u0027ve fixed\nour \u003cmath.h\u003e to fix the warnings caused by our lack of definitions for\nthe non-imprecise long-double functions. I checked the C99 standard, and\nall these functions are there.\n\nChange-Id: Iee8e1182c1db375058fb2c451eceb212bab47a37\n" }, { "commit": "1abc9ff6a5b5f8a9925f1b8d9d333bc5bc7d407f", "tree": "3ea8e2f0f7288d7019004b397847228678d29843", "parents": [ "4ff967f3301087ea6ccaa17cfb1f04a034448a68" ], "author": { "name": "Calin Juravle", "email": "calin@google.com", "time": "Thu Apr 17 18:17:32 2014 +0100" }, "committer": { "name": "Calin Juravle", "email": "calin@google.com", "time": "Wed Apr 23 19:03:06 2014 +0100" }, "message": "Clean-up _fpmath and fake_long_doubles\n\n- promoted IEEEld2bits to fpmath since most of the where the same for\ndiffrent archs\n- removed _fpmath\n- reinstated weak_references\n- moved isfinite and isnormal to libc\n- clean up fake_long_doubles\n- clean up some useless ifdefs\n- added missing nexttoward* tests\n\nBug: 14134235\nChange-Id: I95639c4885653fe47fd7dc0570ee5bb3389bbc6b\n" }, { "commit": "02c78a386739a8a2b3007efeb00a9ca04132100a", "tree": "6679bef40da8b7fbe6b3c5a479eb5c2e4e035708", "parents": [ "055890686636faddbb6d5d407c67f5dcc53ac865" ], "author": { "name": "Elliott Hughes", "email": "enh@google.com", "time": "Fri Apr 11 17:02:20 2014 -0700" }, "committer": { "name": "Elliott Hughes", "email": "enh@google.com", "time": "Mon Apr 14 14:35:47 2014 -0700" }, "message": "Reimplement isinf/isnan/fpclassify.\n\nAlso move isinf and isnan into libc like everyone else.\n\nAlso move fpclassify to libc like the BSDs (but unlike glibc). We need\nthis to be able to upgrade our float/double/long double parsing to gdtoa.\n\nAlso add some missing aliases. We now have all of:\n\n isnan, __isnan, isnanf, __isnanf, isnanl, __isnanl,\n isinf, __isinf, isinff, __isinff, isinfl, __isinfl,\n __fpclassify, __fpclassifyd, __fpclassifyf, __fpclassifyl.\n\nBug: 13469877\nChange-Id: I407ffbac06c765a6c5fffda8106c37d7db04f27d\n" }, { "commit": "d8d6feed3add3965f481dcdd6e1bb259a46f21c2", "tree": "14361f2be3d0195e28ce8cc50ad08f71e95f772e", "parents": [ "b7afa9df2552c6a6889a43c6a1722e1ce46f1969" ], "author": { "name": "Calin Juravle", "email": "calin@google.com", "time": "Tue Apr 01 16:45:53 2014 +0100" }, "committer": { "name": "Calin Juravle", "email": "calin@google.com", "time": "Wed Apr 02 21:54:39 2014 +0100" }, "message": "Fix modf* tests\n\nUse a value that can be exactly represented.\n\nChange-Id: Idf3b1db5e6042fb49c1f591f25f9eee656659ece\n" }, { "commit": "b7afa9df2552c6a6889a43c6a1722e1ce46f1969", "tree": "4e3a980860bbe66374939bfa5a8b644c2a2d3d8c", "parents": [ "b732b71dc10158a4c1d8ee2235d357f8694030ba" ], "author": { "name": "Calin Juravle", "email": "calin@google.com", "time": "Tue Apr 01 16:41:12 2014 +0100" }, "committer": { "name": "Calin Juravle", "email": "calin@google.com", "time": "Tue Apr 01 16:49:14 2014 +0100" }, "message": "Changed long double literal suffix from \u0027l\u0027 to \u0027L\u0027.\n\nChange-Id: Id569f2d335b0930704d12ad7f388cb98b65f3ab7\n" }, { "commit": "c8564f2df2a85a753e25bf48b4eefbf8379fe31a", "tree": "b0b1e6a7057b6bd1ddeb166d25a06438fa80b775", "parents": [ "a58b3f78e14500e7b10fc845ef0a6a8309664663" ], "author": { "name": "Calin Juravle", "email": "calin@google.com", "time": "Thu Mar 27 13:41:06 2014 +0000" }, "committer": { "name": "Calin Juravle", "email": "calin@google.com", "time": "Fri Mar 28 11:10:01 2014 +0000" }, "message": "Fix math tests.\n\nBug: 13657654\nChange-Id: I39b2f13b5b3d3e6897618ac3aed49a0a08458dd0\n" }, { "commit": "133bd09334cb28b45cca1ca0e5b4e22a9e5b68e5", "tree": "3bd86df9a54803d76a5a180eb93a209fbe9d271d", "parents": [ "989a80b75014eff71600a6a4e1445e710c2f94b4" ], "author": { "name": "Brian Carlstrom", "email": "bdc@google.com", "time": "Mon Mar 24 16:45:18 2014 -0700" }, "committer": { "name": "Brian Carlstrom", "email": "bdc@google.com", "time": "Mon Mar 24 16:53:22 2014 -0700" }, "message": "Add include of float.h for LDBL_MIN_EXP and LDBL_MANT_DIG\n\n(cherry picked from commit 1711c6ea3a0d891da665325c75738a68bd7d501f)\n\nChange-Id: Ia64676b2b904f042fec678003e3f2cd5bd82b510\n" }, { "commit": "4d77c1151c40010d137e4a2fa8629bff4bea72b0", "tree": "ba8f9dcf0b45afd26dff4e92a037295a6f78e19d", "parents": [ "849e162d3bedc5ca61bdf15486268b253b9f4e93" ], "author": { "name": "Calin Juravle", "email": "calin@google.com", "time": "Fri Mar 14 17:56:46 2014 +0000" }, "committer": { "name": "Calin Juravle", "email": "calin@google.com", "time": "Mon Mar 24 11:48:42 2014 +0000" }, "message": "Fix LP64 libm for 128-bit long doubles\n\n* reworked amd64/_fpmath.h and arm64/_fpmath.h to support 128-bit long\ndoubles.\n* improved tests to cover long double cases\n* made modfl an alias for LP32\n\nTests pass on x86, x86_64, arm, arm64 and mips.\n\nBug: 12921273\n\nChange-Id: Ibe39acde57972447a8950fa45b1e702acc68ebeb\n" }, { "commit": "5f6f4a956cd630f873fe41058e96262f897f4bc9", "tree": "d0c1a8d044d1597364dad94574e30a0bd8197bd7", "parents": [ "44ef37c1b445e864068cdcd885ab3d7b17079ad3" ], "author": { "name": "Narayan Kamath", "email": "narayan@google.com", "time": "Wed Feb 19 15:42:58 2014 +0000" }, "committer": { "name": "Narayan Kamath", "email": "narayan@google.com", "time": "Wed Feb 19 18:50:30 2014 +0000" }, "message": "Fix signbit / infinite / isinf / isnan unit tests.\n\nTurns out stlport isn\u0027t broken. \u003ccmath\u003e (included\ntransitively via gtest in our case) is not required\nto make C99 math macros (like signbit) available, nor is\nit required to preserve them if they\u0027re already defined.\nIt is only required to make the equivalent functions in\nnamespace std available.\n\nI couldn\u0027t find any documentation of required behaviour for\nprograms that include both \u003cmath.h\u003e and \u003ccmath\u003e.\n\nI\u0027ve verified experimentally that llvm\u0027s libc++ and gnu\nstl behave the same as stlport.\n\nbug: 12935307\n\nChange-Id: I9dc5cc0fd9f4f259abc8eefb280177cdd092a94b\n" }, { "commit": "f04935c85e0b466f0d30d2cd4c0fa2fff62e7d6d", "tree": "8f1c6af75517892648f49adad9d4d5d44c20f49c", "parents": [ "e163a3986aa7d1c67fe3274eb9204812be0a6245" ], "author": { "name": "Christopher Ferris", "email": "cferris@google.com", "time": "Fri Dec 20 18:43:21 2013 -0800" }, "committer": { "name": "Christopher Ferris", "email": "cferris@google.com", "time": "Thu Feb 06 20:12:21 2014 -0800" }, "message": "Make sure that the same tests are on all platforms.\n\nIn order to be able to generate a list of tests for cts, the same set of\ntests must exist across all platforms. This CL adds empty tests where a\ntest was conditionally compiled out.\n\nThis CL creates a single library libBionicTests that includes all of\nthe tests found in bionic-unit-tests-static.\n\nAlso fix a few missing include files in some test files.\n\nTested by running and compiling the tests for every platform and\nverifying the same number of tests are on each platform.\n\nChange-Id: I9989d4bfebb0f9c409a0ce7e87169299eac605a2\n" }, { "commit": "af64dad4796ab4f077e2f2ba1e90cc56b6575d63", "tree": "b219e1771210f5db0dc42dc20e889d77d9a39981", "parents": [ "f246c589d66e5dc0e3cddc3c37261fb0e3fc67e9" ], "author": { "name": "Narayan Kamath", "email": "narayan@google.com", "time": "Mon Nov 18 18:47:48 2013 +0000" }, "committer": { "name": "Narayan Kamath", "email": "narayan@google.com", "time": "Mon Nov 18 19:00:59 2013 +0000" }, "message": "Add test case for 1^NaN\n\nShould be 1.\n\nChange-Id: I6cf723a419f51d3bda58286a538774b71276c7b5\n" }, { "commit": "6a8f00d49c30107903f2449ae563cfee51b48468", "tree": "2d3ad149e0357491a6cc228ad0e639285aff7794", "parents": [ "44cbe94bba2e7e4b1369c86b11d66d6b60f0e333" ], "author": { "name": "Elliott Hughes", "email": "enh@google.com", "time": "Wed Nov 13 13:29:23 2013 -0800" }, "committer": { "name": "Elliott Hughes", "email": "enh@google.com", "time": "Wed Nov 13 13:29:23 2013 -0800" }, "message": "Add unit tests for pow\u0027s behavior with NaNs.\n\nBug: 11669804\nChange-Id: Idfa6527640c8d0c2cf3e5343bf7afe013e237545\n" }, { "commit": "5227663d2ffd70dc32f03a7a5b103ef0d3fc0584", "tree": "215ba2dfc40261cbe16e229b3276d3441a2d9f6d", "parents": [ "59aeff94178e03c908791695f6dd4c9bd8c7115b" ], "author": { "name": "Elliott Hughes", "email": "enh@google.com", "time": "Tue Feb 12 20:18:49 2013 -0800" }, "committer": { "name": "Elliott Hughes", "email": "enh@google.com", "time": "Tue Feb 12 20:18:49 2013 -0800" }, "message": "Put the right number of Ls after 64-bit constants.\n\nChange-Id: I9f96259f21e42a84b9ebe20655fe0edb31f41892\n" }, { "commit": "a0ee07829a9ba7e99ef68e8c12551301cc797f0f", "tree": "fefc432ee572779579a16d2868d4181a33399281", "parents": [ "a990cf5b3392c5aef767aee1e67b4d7ef651afc6" ], "author": { "name": "Elliott Hughes", "email": "enh@google.com", "time": "Wed Jan 30 19:06:37 2013 -0800" }, "committer": { "name": "Elliott Hughes", "email": "enh@google.com", "time": "Fri Feb 01 14:51:19 2013 -0800" }, "message": "Upgrade libm.\n\nThis brings us up to date with FreeBSD HEAD, fixes various bugs, unifies\nthe set of functions we support on ARM, MIPS, and x86, fixes \"long double\",\nadds ISO C99 support, and adds basic unit tests.\n\nIt turns out that our \"long double\" functions have always been broken\nfor non-normal numbers. This patch fixes that by not using the upstream\nimplementations and just forwarding to the regular \"double\" implementation\ninstead (since \"long double\" on Android is just \"double\" anyway, which is\nwhat BSD doesn\u0027t support).\n\nAll the tests pass on ARM, MIPS, and x86, plus glibc on x86-64.\n\nBug: 3169850\nBug: 8012787\nBug: https://code.google.com/p/android/issues/detail?id\u003d6697\nChange-Id: If0c343030959c24bfc50d4d21c9530052c581837\n" } ] }