Commit Graph
353 Commits
Author SHA1 Message Date
Peter Dimov bde5937912 A header named _pop should presumably pop instead of push boost-1.69.0 boost-1.69.0-beta1 2018-09-15 03:40:33 +03:00
Peter Dimov e5753d13e2 Remove unused header 2018-09-15 03:38:52 +03:00
Peter Dimov 2837f3b9c5 g++ 4.5 and earlier do not support unrestricted unions 2018-09-11 20:18:13 +03:00
Peter Dimov 7edf9a030a Add libstdc++-4.9-dev to clang-3.6 2018-09-11 20:17:17 +03:00
Peter Dimov 553792d40c Update .travis.yml 2018-09-11 19:15:02 +03:00
Peter Dimov 07ec535600 boost/predef/detail/endian_compat.h is deprecated; switch to boost/predef/other/endian.h 2018-09-11 18:54:54 +03:00
Peter Dimov f6998d2406 Add .gitattributes boost-1.67.0 boost-1.68.0 2018-01-15 19:37:16 +02:00
Peter Dimov e93f6a2270 Fix unaligned loads and stores 2017-12-23 18:40:09 +02:00
Peter Dimov 62802fee96 Add unaligned test 2017-12-23 17:22:35 +02:00
Peter Dimov 10877d3667 Remove one more integer overflow 2017-12-23 04:47:54 +02:00
Peter Dimov d4738d8598 Avoid undefined behavior due to signed integer overflow 2017-12-23 00:04:34 +02:00
Peter Dimov 0d5adab378 Refactor boundary values tests 2017-12-22 16:35:21 +02:00
Peter Dimov 4648d237f0 Merge branch 'add-test-for-boundary-values' of https://github.com/arvidn/endian into develop 2017-12-22 16:28:00 +02:00
Peter Dimov 134b225378 Shifting a negative number is an -fsanitize=undefined violation, use the unsigned type 2017-12-22 16:24:21 +02:00
arvidn 674adbd592 add test to expose undefined behaviors endian_buffer when used with signed integers 2017-12-22 00:00:16 +01:00
Peter Dimov 69f5b950a2 Add -fno-sanitize-recover=undefined to Travis 2017-12-21 19:28:20 +02:00
Peter Dimov 34542134be Merge pull request #23 from arvidn/remove-trailing-whitespace
[cleanup] remove trailing whitespace from source code files
2017-12-21 17:42:03 +02:00
Peter Dimov 9c0e938c3c Add a macOs UBsan configuration to Travis 2017-12-21 17:19:59 +02:00
Peter Dimov 4bfe758320 Add -d0 to b2 headers on Appveyor 2017-12-21 17:19:09 +02:00
Peter Dimov b5a79243a2 Add -fsanitize=undefined to Travis 2017-12-21 03:49:45 +02:00
Peter Dimov b647051422 Add a construction/assignment/value test for buffer types 2017-12-21 03:32:38 +02:00
Peter Dimov c6aac76029 Add .travis.yml, appveyor.yml 2017-12-20 19:38:59 +02:00
Beman af744e9485 As presented in ABQ 2017-11-10 15:27:21 -07:00
arvidn 4b4123a0fa [cleanup] remove trailing whitespace from source code files 2017-11-04 13:07:54 +01:00
Beman a2b044f181 P0803r0 2017-10-16 07:14:41 -04:00
Beman ac7a8e1696 P0803R0 2017-10-16 07:12:29 -04:00
Beman d89e6ddc62 Initial commit 2017-10-15 10:03:37 -04:00
Beman 739bc40bfc Update tools version 2017-10-15 10:02:13 -04:00
Beman 754a4144ec Clear signed/unsigned warnings 2016-08-09 16:49:20 -04:00
Beman 2db73ab893 Document permanent removal of floating point types. boost-1.65.0 boost-1.66.0 boost-1.65.1 boost-1.64.0-beta2 boost-1.64.0-beta1 boost-1.64.0 boost-1.63.0 boost-1.62.0 boost-1.61.0 2016-04-05 13:57:27 -04:00
Beman Dawes dc263dd9ac Merge pull request #16 from BenWiederhake/develop
Fix a few typos
2016-04-04 17:22:05 -04:00
Beman Dawes 51ecbb06e2 Merge pull request #13 from niXman/patch-1
typo fixed
2016-04-04 17:11:42 -04:00
Ben Wiederhake b51593c766 Fix typos in documentation/comments
Credits go to the authors of codespell.
2016-04-03 19:13:04 +02:00
Ben Wiederhake d1865063ad Fix typos in documentation
Found and fixed while reading documentation
2016-04-03 19:12:49 +02:00
Beman 1b0a41e6cb Add Clang/c2 solution with conversion_test_clang project. Clang/c2 uses the Microsoft rather than GCC intrinsics, so check for defined(_MSC_VER) before defined(__clang__) 2015-12-11 08:15:25 -05:00
Beman 91985423e4 Also mention runtime ordering selection in the FAQ. boost-1.60.0 2015-10-16 10:57:35 -04:00
Beman 484c635316 Tweak the "Why are only big and little native endianness supported?" FAQ entry wording. 2015-10-16 10:43:45 -04:00
Beman f14f080359 Change the specification for native ordering to allow implementations to support orderings in addition to big and little. Thanks to Howard Hinnant for suggesting this change and providing draft wording. Also update headers to reflect actual inclusion hierarchy. 2015-10-16 10:30:34 -04:00
Beman 5c26514ded Add comment explaining omission of certain overloads. 2015-08-03 16:09:06 -04:00
Beman 117ee622c2 All Linux names FAQ entry. 2015-07-21 14:44:36 -04:00
niXman 55714f0d0a typo fixed 2015-04-16 14:18:59 +03:00
Beman bf0dec92d0 Add Dave Handley to acks at Tom Puverle's request. 2015-04-10 08:17:37 -04:00
Beman 636a7bfd6e Update README to reflect 1.58.0 inclusion and boostorg.github.io/endian availability. boost-1.58.0 boost-1.59.0 2015-04-03 08:33:33 -04:00
Beman cd57e01f9c The boost::uint64_t fix for old msvc compilers is also needed for buffer_test.cpp 2015-03-30 07:15:15 -04:00
Beman f0f2505f9b Remove fp_exhaustive_test from Jamfile 2015-03-30 07:01:19 -04:00
Beman d612bee5b9 Initial commit of work-in-progress 2015-03-27 08:06:33 -04:00
Beman fdc8965842 Some older Microsoft libraries did not have uint64_t, so use boost/cstdint.hpp for it. 2015-03-26 11:03:39 -04:00
Beman 3dffbdbcb4 Add notice explaining absence of floating point support. 2015-03-25 18:12:45 -04:00
Beman 01f7e359bb Update docs to reflect removal of floating point support from 1.58.0 2015-03-25 08:05:49 -04:00
Beman 5caf1d16c8 Remove floating point support for boost 1.58.0. As currently implemented, return by value operations are not reliable because normalization and other bit pattern changes can occur. Docs not updated as yet. Floating point support will be corrected and reinstated for boost 1.59.0, based on in-place customization and inplace and/or copy signatures to replace the unsafe (i.e. native_to_big/little ) by value returns. The big/little_to_native return by value functions are safe, but require reimplementation. 2015-03-24 17:24:56 -04:00