tvl-depot/absl/base/config.h

666 lines
25 KiB
C
Raw Normal View History

2017-09-19 22:54:40 +02:00
//
// Copyright 2017 The Abseil Authors.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// https://www.apache.org/licenses/LICENSE-2.0
2017-09-19 22:54:40 +02:00
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
//
// -----------------------------------------------------------------------------
// File: config.h
// -----------------------------------------------------------------------------
//
// This header file defines a set of macros for checking the presence of
// important compiler and platform features. Such macros can be used to
// produce portable code by parameterizing compilation based on the presence or
// lack of a given feature.
//
// We define a "feature" as some interface we wish to program to: for example,
// a library function or system call. A value of `1` indicates support for
// that feature; any other value indicates the feature support is undefined.
//
// Example:
//
// Suppose a programmer wants to write a program that uses the 'mmap()' system
// call. The Abseil macro for that feature (`ABSL_HAVE_MMAP`) allows you to
// selectively include the `mmap.h` header and bracket code using that feature
// in the macro:
//
// #include "absl/base/config.h"
//
// #ifdef ABSL_HAVE_MMAP
// #include "sys/mman.h"
// #endif //ABSL_HAVE_MMAP
//
// ...
// #ifdef ABSL_HAVE_MMAP
// void *ptr = mmap(...);
// ...
// #endif // ABSL_HAVE_MMAP
#ifndef ABSL_BASE_CONFIG_H_
#define ABSL_BASE_CONFIG_H_
// Included for the __GLIBC__ macro (or similar macros on other systems).
#include <limits.h>
#ifdef __cplusplus
// Included for __GLIBCXX__, _LIBCPP_VERSION
#include <cstddef>
#endif // __cplusplus
#if defined(__APPLE__)
// Included for TARGET_OS_IPHONE, __IPHONE_OS_VERSION_MIN_REQUIRED,
// __IPHONE_8_0.
#include <Availability.h>
#include <TargetConditionals.h>
#endif
Export of internal Abseil changes -- 2f49cb9009386bc67bf54a2908c8720b749c8b7f by Greg Falcon <gfalcon@google.com>: docs: fix typo Import of https://github.com/abseil/abseil-cpp/pull/397 PiperOrigin-RevId: 277504420 -- f2bed362c1c12d3fa9c22d11f2b918668e8c37b7 by Abseil Team <absl-team@google.com>: Avoid our is_[copy/move]_assignable workarounds in MSVC 19.20 and on, since that release introduces a regression that breaks our workaround. We should ideally use the std forms in more cases, but branching when our workarounds fail is simpler to maintain. PiperOrigin-RevId: 277502334 -- e33de894ffd49848b9e088f59acc9743d1661948 by Derek Mauro <dmauro@google.com>: Update rules_cc version. The mirror.bazel.build URL does not exist (cache expiration?) PiperOrigin-RevId: 277498394 -- b23757b0747c64634d2d701433782c969effef19 by Abseil Team <absl-team@google.com>: Fix https://github.com/abseil/abseil-cpp/issues/394. PiperOrigin-RevId: 277491405 -- 54c75b8b29813531c52d67cf0ba7063baae4a4f3 by Abseil Team <absl-team@google.com>: Fix comment typos: waker => waiter. PiperOrigin-RevId: 277376952 -- 874eeaa3b3af808fc88b6355245f643674f5e36e by Abseil Team <absl-team@google.com>: Don't use atomic ops on waiter and wakeup counts in CONDVAR waiter mode. Just guard the waiter and wakeup counts with the mutex. This eliminates the race. Also fix a typo in the error message for pthread_cond_timedwait. PiperOrigin-RevId: 277366017 -- ce8c9a63109214519b5a7eaecef2c663c4d566df by Greg Falcon <gfalcon@google.com>: Implement the config options for our four main C++ forward compatibility types. These options control whether the names `any`, `optional`, `string_view`, and `variant` in namespace `absl` are aliases to the corresponding C++17 types in namespace `std`. By default, we continue to auto-detect the configuration of the compiler being run. These options are not intended to be modified on the command line (as -D flags, say). Instead, the options.h file can be modified by distributors of Abseil (e.g., binary packagers, maintainers of local copies of Abseil, etc.) Changing options will change Abseil in an ODR sense. Any program must only link in a single version of Abseil. Linking libraries that use Abseil configured with different options is an error: there is no ABI compatibility guarantee when linking different configurations, even if the Abseil versions used are otherwise exactly identical. PiperOrigin-RevId: 277364298 -- 5ed3ad42ae43a05862070f92f9ffd07f5c1f2b81 by Chris Kennelly <ckennelly@google.com>: Suppress -Wimplicit-int-float-conversion. On recent builds of Clang, this is an error/warning. PiperOrigin-RevId: 277346168 -- 9b9b0055243c705189bb27d912e6d45a7789cdee by Eric Fiselier <ericwf@google.com>: Allow building Abseil as a shared library with CMake. By default CMake's `add_library` creates the target as a static library. However, users can override the default using the builtin CMake option -DBUILD_SHARED_LIBS=ON. This changes Abseil's CMake to respect this configuration option by removing the explicit `STATIC` in our usages of `add_library`. PiperOrigin-RevId: 277337753 -- 63a8b7b8ede3a9d851916929d6b8537f4f2508ca by Abseil Team <absl-team@google.com>: Improve AlphaNum Hex performance by using absl::numbers_internal::FastHexToBufferZeroPad16. PiperOrigin-RevId: 277318108 -- dd047f7e92032682d94b27732df0e4d0670e24a4 by CJ Johnson <johnsoncj@google.com>: Internal change PiperOrigin-RevId: 277316913 -- d19ee7074929fed08973cc5b40a844573ce1e0a6 by Abseil Team <absl-team@google.com>: Handle invoking [[nodiscard]] functions correctly in our tests. PiperOrigin-RevId: 277301500 -- 5373f3737894ba9b8481e95e5f58c7957c00d26a by Chris Kennelly <ckennelly@google.com>: For internal reasons, loosen visibility restrictions of `//absl/base:malloc_internal`. As an internal-namespace interface, this module remains unsupported. We reserve the right to change, delete, or re-restrict visibility to this target at any time. PiperOrigin-RevId: 277118689 -- 44e4f6655e05393a375d09b3625c192b1fde5909 by Abseil Team <absl-team@google.com>: Fix error in example civil day comment. PiperOrigin-RevId: 277103315 -- 7657392b4ce48469106f11cdb952a0bc76a28df3 by Abseil Team <absl-team@google.com>: Internal change PiperOrigin-RevId: 277056076 -- c75bda76f40b01fa249b75b5a70c1f5907e56c76 by Abseil Team <absl-team@google.com>: Suppress lifetime constant-initialization tests when building with MSVC versions > 19.0. These are broken due to non-compliant initialization order in these versions: https://developercommunity.visualstudio.com/content/problem/336946/class-with-constexpr-constructor-not-using-static.html We don't know when Microsoft will fix this bug. PiperOrigin-RevId: 277049770 -- 16c3b9bf2a1796efa57f97b00bcd6227fbccca1f by Matt Calabrese <calabrese@google.com>: Avoid our is_[copy/move]_assignable workarounds in MSVC 19.20 and on, since that release introduces a regression that breaks our workaround. We should ideally use the std forms in more cases, but branching when our workarounds fail is simpler to maintain. PiperOrigin-RevId: 277048670 -- e91003fa3ee6026d9b80624a23fc144fa5d74810 by Chris Kennelly <ckennelly@google.com>: Fix -Wimplicit-int-float-conversion warning in latest clang PiperOrigin-RevId: 276771618 -- 53087ca6603e86ad815f3dd9ab795cc0f79792c1 by Andy Soffer <asoffer@google.com>: Add documentation on absl::SNPrintF. PiperOrigin-RevId: 276694032 -- a9426af8cbd4c3a8f9053e7446c821852e41ff61 by Jorg Brown <jorg@google.com>: Stop including kern/OSByteOrder.h in order to support __APPLE__ Apple compiles with clang now anyway, and clang has support for the built-in compiler swap functions that are much faster than any function call to the OS. PiperOrigin-RevId: 276625231 -- df974be5aa5b4dc1b09c356cb8816edfc7867e63 by Jorg Brown <jorg@google.com>: Fix the build for Android x86-64 builds, where __SSE4_2__ is defined but _bswap64 is not. PiperOrigin-RevId: 276542642 -- d99dc092b3a5ad17643005e55f3b3cb6b9187ccc by Jorg Brown <jorg@google.com>: Remove a byteswap from the non-SSE path of FastHexToBufferZeroPad16 Remove the need for including absl/base/internal/endian.h from the SSE case (since if we have the Intel SSE intrinsics, then clearly we also have the Intel Byte-Swap intrinsics.) PiperOrigin-RevId: 276532608 -- d67b106dc930d8558810ae3983613bb2ab1e0d36 by Abseil Team <absl-team@google.com>: Use explicit static_cast<double> for int64_t to double conversion This uses an explicit static_cast<double>() in the int64_t to double comparisons in duration.cc's SafeAddRepHi. This satisfies clang's -Wimplicit-int-to-float-conversion warning (with https://reviews.llvm.org/D64666). This may also make it easier for humans to realize that the comparison is happening between two floating point double precision values. It should have no impact on the behavior or generated code. Tested: No behavior change PiperOrigin-RevId: 276529211 GitOrigin-RevId: 2f49cb9009386bc67bf54a2908c8720b749c8b7f Change-Id: I71e0781893ce219960b8290d54b20532779cb0ff
2019-10-30 16:00:44 +01:00
#include "absl/base/options.h"
2017-09-19 22:54:40 +02:00
#include "absl/base/policy_checks.h"
Export of internal Abseil changes -- c42a234e2c186bf697ce8d77e85628601fa514a6 by Abseil Team <absl-team@google.com>: Enable the assertion in the iterator's operator++ PiperOrigin-RevId: 290134813 -- f8c53ba8e9c5bb16bbcc1e412a5c2519c912c83e by Abseil Team <absl-team@google.com>: Define operator== and operator!= for absl::{weak,strong}_equality and absl::{partial,weak,strong}_ordering types themselves. PiperOrigin-RevId: 290111564 -- 36bc574090cefad74a451719ce2761982647a51d by Tom Manshreck <shreck@google.com>: Specify Time library flag formats PiperOrigin-RevId: 289928010 -- 26dd40281add260baab2b60fec05dfb9c5304aaa by Mark Barolak <mbar@google.com>: Delete an extraneous forward declaration of absl::Cord. PiperOrigin-RevId: 289708481 -- e60aea7f33554ff66d7699bb70e7af1d26323f1d by Abseil Team <absl-team@google.com>: Release b-tree benchmarks. PiperOrigin-RevId: 289654429 -- 660aa83fa000d4bae072b2d1c790f81d0939bc7e by Greg Falcon <gfalcon@google.com>: Use https links. Import of https://github.com/abseil/abseil-cpp/pull/586 PiperOrigin-RevId: 289479559 -- 0611ea4482dcf23d6b0a0389fe041eeb9052449a by Derek Mauro <dmauro@google.com>: Removes the static initializer for LookupTables<absl::uint128>::kVmaxOverBase Uses template specialization to hard code the resulting array. Static initializers are problematic for a number of reasons. Not only are they responsible for the static initialization order fiasco, but they are in the critical path during program startup. For these reasons, the Google C++ style guide strongly discourages them (and forbids them when they are not trivially destructible), and Chromium even has a test forbidding them. https://google.github.io/styleguide/cppguide.html#Static_and_Global_Variables https://chromium.googlesource.com/chromium/src.git/+/master/docs/static_initializers.md http://neugierig.org/software/chromium/notes/2011/08/static-initializers.html PiperOrigin-RevId: 289458677 -- c869362f6bb7a872314f74750d38d81bdaa73f95 by Greg Falcon <gfalcon@google.com>: Step 2 of 2 to fix our CCTZ fork to respect inline namespaces. Re-import of CCTZ from GitHub, applying new changes to honor Abseil's optional inline namespace in MSVC. PiperOrigin-RevId: 289454407 -- fdb3474d76c2ee0371ccdf7593a78137c03a3f58 by Greg Falcon <gfalcon@google.com>: Step 1 of 2 to fix our CCTZ fork to respect inline namespaces. CCTZ uses a linker flag to simulate weak symbol support in MSVC. This takes the form of a #pragma that includes the mangled names of two types: the symbol to treat as weak, and the symbol to use as its default value if no override is provided. When Abseil is configured to use inline namespaces, the mangled names of these symbols change, and the pragma should change to reflect that. Fortunately for us, MSVC name mangling is simple enough that we can generate the needed string literals in the preprocessor. This CL introduces the new macros; the uses will be introduced in a follow-up CL. PiperOrigin-RevId: 289435599 -- 5f152cc36f008acb9ab78f30b5efa40ebaf2754b by Matt Kulukundis <kfm@google.com>: Improve documentation for lazy_emplace PiperOrigin-RevId: 289333112 GitOrigin-RevId: c42a234e2c186bf697ce8d77e85628601fa514a6 Change-Id: I139ce6c7044a70d083af53e428bcb987f0fd88c6
2020-01-16 22:38:39 +01:00
// Helper macro to convert a CPP variable to a string literal.
#define ABSL_INTERNAL_DO_TOKEN_STR(x) #x
#define ABSL_INTERNAL_TOKEN_STR(x) ABSL_INTERNAL_DO_TOKEN_STR(x)
// -----------------------------------------------------------------------------
// Abseil namespace annotations
// -----------------------------------------------------------------------------
// ABSL_NAMESPACE_BEGIN/ABSL_NAMESPACE_END
//
// An annotation placed at the beginning/end of each `namespace absl` scope.
// This is used to inject an inline namespace.
//
// The proper way to write Abseil code in the `absl` namespace is:
//
// namespace absl {
// ABSL_NAMESPACE_BEGIN
//
// void Foo(); // absl::Foo().
//
// ABSL_NAMESPACE_END
// } // namespace absl
//
// Users of Abseil should not use these macros, because users of Abseil should
// not write `namespace absl {` in their own code for any reason. (Abseil does
// not support forward declarations of its own types, nor does it support
// user-provided specialization of Abseil templates. Code that violates these
// rules may be broken without warning.)
Export of internal Abseil changes -- 9beb68204986a015c9cb065b9fae4f9a8879a788 by Abseil Team <absl-team@google.com>: Move Base64EscapeInternal and CalculateBase64EscapedLenInternal to an internal header. PiperOrigin-RevId: 288917378 -- 90acfbe03b3f9f6de3ffa49c39343dfaa2c5d38c by Greg Falcon <gfalcon@google.com>: Update macos CI script to support the ALTERNATE_OPTIONS environment variable. PiperOrigin-RevId: 288913564 -- f1572e870678cdcda6b48cb39780d1ad984e4c1b by Derek Mauro <dmauro@google.com>: Makes absl::NullSafeStringView constexpr Fixes https://github.com/abseil/abseil-cpp/issues/583 PiperOrigin-RevId: 288906940 -- d28a8471e32c10caa64bfffe6d6d4d0a8d144013 by Abseil Team <absl-team@google.com>: absl::GetFlag is lock free for small trivially copyable types. PiperOrigin-RevId: 288768172 -- 2643b8ed1a1dc836b38ab9e46538a1af129ffd67 by Gennadiy Rozental <rogeeff@google.com>: Eliminate call to callback from flag initialization. We do not need to have this invocation inside FlagImpl::Init since SetCallback performs invocation anyways. Calling InitCallback from inside of Init complicates separation of value initialization from data guard initialization, which is about to happen. PiperOrigin-RevId: 288732526 -- 22caa880b7a4cb6da34e16a2e064a473c99e880b by Abseil Team <absl-team@google.com>: Fix the documentation on how to create a null string_view. PiperOrigin-RevId: 288727968 -- 10727c5cadc561837141176f4c9b9717cec9233a by Greg Falcon <gfalcon@google.com>: Change CI scripts for gcc to use the ALTERNATE_OPTIONS file as well. PiperOrigin-RevId: 288718855 -- 5d1e2dd6c7fb12af8aa4337a0f61872f5f0c5992 by Greg Falcon <gfalcon@google.com>: Add an option for using inline namespaces in Abseil. PiperOrigin-RevId: 288614491 GitOrigin-RevId: 9beb68204986a015c9cb065b9fae4f9a8879a788 Change-Id: If9acd46301e3df8cb231b4c16f7ed651bf4fb3c3
2020-01-09 18:58:48 +01:00
#if !defined(ABSL_OPTION_USE_INLINE_NAMESPACE) || \
!defined(ABSL_OPTION_INLINE_NAMESPACE_NAME)
#error options.h is misconfigured.
#endif
// Check that ABSL_OPTION_INLINE_NAMESPACE_NAME is neither "head" nor ""
#if defined(__cplusplus) && ABSL_OPTION_USE_INLINE_NAMESPACE == 1
#define ABSL_INTERNAL_INLINE_NAMESPACE_STR \
ABSL_INTERNAL_TOKEN_STR(ABSL_OPTION_INLINE_NAMESPACE_NAME)
static_assert(ABSL_INTERNAL_INLINE_NAMESPACE_STR[0] != '\0',
"options.h misconfigured: ABSL_OPTION_INLINE_NAMESPACE_NAME must "
"not be empty.");
static_assert(ABSL_INTERNAL_INLINE_NAMESPACE_STR[0] != 'h' ||
ABSL_INTERNAL_INLINE_NAMESPACE_STR[1] != 'e' ||
ABSL_INTERNAL_INLINE_NAMESPACE_STR[2] != 'a' ||
ABSL_INTERNAL_INLINE_NAMESPACE_STR[3] != 'd' ||
ABSL_INTERNAL_INLINE_NAMESPACE_STR[4] != '\0',
"options.h misconfigured: ABSL_OPTION_INLINE_NAMESPACE_NAME must "
"be changed to a new, unique identifier name.");
#endif
#if ABSL_OPTION_USE_INLINE_NAMESPACE == 0
#define ABSL_NAMESPACE_BEGIN
#define ABSL_NAMESPACE_END
Export of internal Abseil changes -- 9beb68204986a015c9cb065b9fae4f9a8879a788 by Abseil Team <absl-team@google.com>: Move Base64EscapeInternal and CalculateBase64EscapedLenInternal to an internal header. PiperOrigin-RevId: 288917378 -- 90acfbe03b3f9f6de3ffa49c39343dfaa2c5d38c by Greg Falcon <gfalcon@google.com>: Update macos CI script to support the ALTERNATE_OPTIONS environment variable. PiperOrigin-RevId: 288913564 -- f1572e870678cdcda6b48cb39780d1ad984e4c1b by Derek Mauro <dmauro@google.com>: Makes absl::NullSafeStringView constexpr Fixes https://github.com/abseil/abseil-cpp/issues/583 PiperOrigin-RevId: 288906940 -- d28a8471e32c10caa64bfffe6d6d4d0a8d144013 by Abseil Team <absl-team@google.com>: absl::GetFlag is lock free for small trivially copyable types. PiperOrigin-RevId: 288768172 -- 2643b8ed1a1dc836b38ab9e46538a1af129ffd67 by Gennadiy Rozental <rogeeff@google.com>: Eliminate call to callback from flag initialization. We do not need to have this invocation inside FlagImpl::Init since SetCallback performs invocation anyways. Calling InitCallback from inside of Init complicates separation of value initialization from data guard initialization, which is about to happen. PiperOrigin-RevId: 288732526 -- 22caa880b7a4cb6da34e16a2e064a473c99e880b by Abseil Team <absl-team@google.com>: Fix the documentation on how to create a null string_view. PiperOrigin-RevId: 288727968 -- 10727c5cadc561837141176f4c9b9717cec9233a by Greg Falcon <gfalcon@google.com>: Change CI scripts for gcc to use the ALTERNATE_OPTIONS file as well. PiperOrigin-RevId: 288718855 -- 5d1e2dd6c7fb12af8aa4337a0f61872f5f0c5992 by Greg Falcon <gfalcon@google.com>: Add an option for using inline namespaces in Abseil. PiperOrigin-RevId: 288614491 GitOrigin-RevId: 9beb68204986a015c9cb065b9fae4f9a8879a788 Change-Id: If9acd46301e3df8cb231b4c16f7ed651bf4fb3c3
2020-01-09 18:58:48 +01:00
#elif ABSL_OPTION_USE_INLINE_NAMESPACE == 1
#define ABSL_NAMESPACE_BEGIN \
inline namespace ABSL_OPTION_INLINE_NAMESPACE_NAME {
#define ABSL_NAMESPACE_END }
#else
#error options.h is misconfigured.
#endif
2017-09-19 22:54:40 +02:00
// -----------------------------------------------------------------------------
// Compiler Feature Checks
// -----------------------------------------------------------------------------
// ABSL_HAVE_BUILTIN()
//
// Checks whether the compiler supports a Clang Feature Checking Macro, and if
// so, checks whether it supports the provided builtin function "x" where x
// is one of the functions noted in
// https://clang.llvm.org/docs/LanguageExtensions.html
//
// Note: Use this macro to avoid an extra level of #ifdef __has_builtin check.
// http://releases.llvm.org/3.3/tools/clang/docs/LanguageExtensions.html
#ifdef __has_builtin
#define ABSL_HAVE_BUILTIN(x) __has_builtin(x)
#else
#define ABSL_HAVE_BUILTIN(x) 0
#endif
#if defined(__is_identifier)
#define ABSL_INTERNAL_HAS_KEYWORD(x) !(__is_identifier(x))
#else
#define ABSL_INTERNAL_HAS_KEYWORD(x) 0
#endif
2017-09-19 22:54:40 +02:00
// ABSL_HAVE_TLS is defined to 1 when __thread should be supported.
// We assume __thread is supported on Linux when compiled with Clang or compiled
// against libstdc++ with _GLIBCXX_HAVE_TLS defined.
#ifdef ABSL_HAVE_TLS
#error ABSL_HAVE_TLS cannot be directly set
#elif defined(__linux__) && (defined(__clang__) || defined(_GLIBCXX_HAVE_TLS))
#define ABSL_HAVE_TLS 1
#endif
// ABSL_HAVE_STD_IS_TRIVIALLY_DESTRUCTIBLE
//
// Checks whether `std::is_trivially_destructible<T>` is supported.
//
// Notes: All supported compilers using libc++ support this feature, as does
// gcc >= 4.8.1 using libstdc++, and Visual Studio.
#ifdef ABSL_HAVE_STD_IS_TRIVIALLY_DESTRUCTIBLE
#error ABSL_HAVE_STD_IS_TRIVIALLY_DESTRUCTIBLE cannot be directly set
#elif defined(_LIBCPP_VERSION) || \
(!defined(__clang__) && defined(__GNUC__) && defined(__GLIBCXX__) && \
(__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 8))) || \
defined(_MSC_VER)
#define ABSL_HAVE_STD_IS_TRIVIALLY_DESTRUCTIBLE 1
#endif
// ABSL_HAVE_STD_IS_TRIVIALLY_CONSTRUCTIBLE
//
// Checks whether `std::is_trivially_default_constructible<T>` and
// `std::is_trivially_copy_constructible<T>` are supported.
// ABSL_HAVE_STD_IS_TRIVIALLY_ASSIGNABLE
//
// Checks whether `std::is_trivially_copy_assignable<T>` is supported.
// Notes: Clang with libc++ supports these features, as does gcc >= 5.1 with
// either libc++ or libstdc++, and Visual Studio (but not NVCC).
2017-09-19 22:54:40 +02:00
#if defined(ABSL_HAVE_STD_IS_TRIVIALLY_CONSTRUCTIBLE)
#error ABSL_HAVE_STD_IS_TRIVIALLY_CONSTRUCTIBLE cannot be directly set
#elif defined(ABSL_HAVE_STD_IS_TRIVIALLY_ASSIGNABLE)
#error ABSL_HAVE_STD_IS_TRIVIALLY_ASSIGNABLE cannot directly set
#elif (defined(__clang__) && defined(_LIBCPP_VERSION)) || \
(!defined(__clang__) && defined(__GNUC__) && \
Export of internal Abseil changes -- f28b989d5161230c6561e923b458c797a96bcf90 by Greg Falcon <gfalcon@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 263586488 -- 8259484025b7de45358719fc6182a48cac8044c6 by Andy Soffer <asoffer@google.com>: Internal changes and combine namespaces into a single namespace. PiperOrigin-RevId: 263560576 -- 8d19f41661984a600d1f8bbfeb8a30fcb4dee7d6 by Mark Barolak <mbar@google.com>: Inside of absl::string_view::copy, use absl::string_view::traits_type::copy instead of std:copy to do the actual work. This both follows the C++ standard more closely and avoids avoid MSVC unchecked iterator warnings. PiperOrigin-RevId: 263430502 -- c06bf74236e12c7c1c97bfcbbc9d29bd65d6b36c by Andy Soffer <asoffer@google.com>: Remove force-inlining attributes. Benchmarking results indicate that they are creating meaningful performance differences. PiperOrigin-RevId: 263364896 -- ec4fa6eac958a9521456201b138784f55d3b17bc by Abseil Team <absl-team@google.com>: Make BM_Fill benchmarks more representative. PiperOrigin-RevId: 263349482 -- 4ae280b4eb31d9cb58e847eb670473340f7778c1 by Derek Mauro <dmauro@google.com>: Fix new -Wdeprecated-copy warning in gcc9 PiperOrigin-RevId: 263348118 -- d238a92f452a5c35686f9c71596fdd1fe62090a2 by Matt Calabrese <calabrese@google.com>: The std::is_trivially_xxx fail on versions of GCC up until 7.4 due to faulty underlying intrinsics, but our emulation succeeds. Update our traits to not compare against the standard library implementation in these versions. PiperOrigin-RevId: 263209457 GitOrigin-RevId: f28b989d5161230c6561e923b458c797a96bcf90 Change-Id: I4c41db5928ba71e243aeace4420e06d1a2df0b5b
2019-08-15 19:06:39 +02:00
(__GNUC__ > 7 || (__GNUC__ == 7 && __GNUC_MINOR__ >= 4)) && \
2017-09-19 22:54:40 +02:00
(defined(_LIBCPP_VERSION) || defined(__GLIBCXX__))) || \
(defined(_MSC_VER) && !defined(__NVCC__))
2017-09-19 22:54:40 +02:00
#define ABSL_HAVE_STD_IS_TRIVIALLY_CONSTRUCTIBLE 1
#define ABSL_HAVE_STD_IS_TRIVIALLY_ASSIGNABLE 1
#endif
// ABSL_HAVE_SOURCE_LOCATION_CURRENT
//
// Indicates whether `absl::SourceLocation::current()` will return useful
// information in some contexts.
#ifndef ABSL_HAVE_SOURCE_LOCATION_CURRENT
#if ABSL_INTERNAL_HAS_KEYWORD(__builtin_LINE) && \
ABSL_INTERNAL_HAS_KEYWORD(__builtin_FILE)
#define ABSL_HAVE_SOURCE_LOCATION_CURRENT 1
#endif
#endif
2017-09-19 22:54:40 +02:00
// ABSL_HAVE_THREAD_LOCAL
//
// Checks whether C++11's `thread_local` storage duration specifier is
// supported.
#ifdef ABSL_HAVE_THREAD_LOCAL
#error ABSL_HAVE_THREAD_LOCAL cannot be directly set
#elif defined(__APPLE__)
Export of internal Abseil changes. -- 07575526242a8e1275ac4223a3d2822795f46569 by CJ Johnson <johnsoncj@google.com>: Comment cleanup on InlinedVector PiperOrigin-RevId: 221322176 -- 49a5e643f85e34d53c41f5e6cc33357c55c9115d by Matt Kulukundis <kfm@google.com>: Internal cleanup PiperOrigin-RevId: 221309185 -- bb35be87ec9c74244b7d902e7e7d2d33ab139d76 by Abseil Team <absl-team@google.com>: Fix typo in comment. PiperOrigin-RevId: 221145354 -- afd4d7c106919708004e06aeea068a57c28aec44 by Derek Mauro <dmauro@google.com>: Update the debugging log message in CallOnceImpl() PiperOrigin-RevId: 221103254 -- 0b9dace8b88113777bf26a6d38f9bc0bcaf053a1 by Abseil Team <absl-team@google.com>: Workaround an MSVC 2015 bug in compile-time initialization. PiperOrigin-RevId: 220871483 -- ea0a3854511ed26beab827e5a5113766b334db86 by Marek Gilbert <mcg@google.com>: Fix ABSL_HAVE_THREAD_LOCAL when compiling for iOS 8 with Xcode 10. Xcode 10 has moved the check for thread_local to a link time, so clang reports __has_feature(cxx_thread_local) but then linking fails with messages like this: ld: targeted OS version does not support use of thread local variables PiperOrigin-RevId: 220815885 -- 485b6876c158c3dcf37eb32d7e512242d5d4ecc6 by Greg Falcon <gfalcon@google.com>: Make the absl::c_set_xxxx() algorithms refuse to compile when passed an unordered collection from std:: or absl::. These algorithms operate on sorted sequences; passing an unordered container to them is nearly certainly a bug. This change is technically an API break, but it only breaks incorrect code. We could try to be more clever and detect unordered collections from other libraries, but false positives will break legal code, and this would constitute an API break Abseil cannot afford. PiperOrigin-RevId: 220794190 -- c47cff7f9cc70a4c1604eee0131af552f40e46d6 by Jon Cohen <cohenjon@google.com>: MSVC 2017's STL throws a Structured Exception (not a C++ exception, essentially equivalent to SIGSEGV) when variant::emplace calls a throwing constructor when using the debug multithreaded MSVC runtime DLL. This manifests in dbg mode in Bazel builds. Disable tests which trigger this bug. It's impossible to specifically pull out MSVC 2017 -dbg modes because there's no way for Bazel to know when version of MSVC is being used -- you tell Bazel the directory where the MSVC tools live, not which version of MSVC tools to use. Thus the best we can do is switch on _DEBUG, which is set whenever the debug runtime is selected with the /MDd build flag, as in Bazel -dbg modes. See https://msdn.microsoft.com/en-us/library/b0084kay.aspx ctrl-f "_DEBUG" PiperOrigin-RevId: 220706161 -- 43993d4af309d92f4ebff38391dcc245f154ecc7 by Shaindel Schwartz <shaindel@google.com>: Internal change PiperOrigin-RevId: 220688429 -- 2448802972dcc261af153af464f2b022ef54a2a9 by Abseil Team <absl-team@google.com>: Speed up operator* for uint128 in WIN64. PiperOrigin-RevId: 220678790 -- 7b376403dd05ba10152fb52e40b29d8af79b58bb by Abseil Team <absl-team@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220654834 -- ae08af58111c3f838b8d4de25f501c3559c86002 by Abseil Team <absl-team@google.com>: CMake: Add absl_cc_test function PiperOrigin-RevId: 220603940 GitOrigin-RevId: 07575526242a8e1275ac4223a3d2822795f46569 Change-Id: Iba7f53eb394c8a9de564582a976793f9bb0596d9
2018-11-13 22:22:00 +01:00
// Notes:
// * Xcode's clang did not support `thread_local` until version 8, and
// even then not for all iOS < 9.0.
// * Xcode 9.3 started disallowing `thread_local` for 32-bit iOS simulator
// targeting iOS 9.x.
// * Xcode 10 moves the deployment target check for iOS < 9.0 to link time
// making __has_feature unreliable there.
//
// Otherwise, `__has_feature` is only supported by Clang so it has be inside
// `defined(__APPLE__)` check.
Export of internal Abseil changes. -- 07575526242a8e1275ac4223a3d2822795f46569 by CJ Johnson <johnsoncj@google.com>: Comment cleanup on InlinedVector PiperOrigin-RevId: 221322176 -- 49a5e643f85e34d53c41f5e6cc33357c55c9115d by Matt Kulukundis <kfm@google.com>: Internal cleanup PiperOrigin-RevId: 221309185 -- bb35be87ec9c74244b7d902e7e7d2d33ab139d76 by Abseil Team <absl-team@google.com>: Fix typo in comment. PiperOrigin-RevId: 221145354 -- afd4d7c106919708004e06aeea068a57c28aec44 by Derek Mauro <dmauro@google.com>: Update the debugging log message in CallOnceImpl() PiperOrigin-RevId: 221103254 -- 0b9dace8b88113777bf26a6d38f9bc0bcaf053a1 by Abseil Team <absl-team@google.com>: Workaround an MSVC 2015 bug in compile-time initialization. PiperOrigin-RevId: 220871483 -- ea0a3854511ed26beab827e5a5113766b334db86 by Marek Gilbert <mcg@google.com>: Fix ABSL_HAVE_THREAD_LOCAL when compiling for iOS 8 with Xcode 10. Xcode 10 has moved the check for thread_local to a link time, so clang reports __has_feature(cxx_thread_local) but then linking fails with messages like this: ld: targeted OS version does not support use of thread local variables PiperOrigin-RevId: 220815885 -- 485b6876c158c3dcf37eb32d7e512242d5d4ecc6 by Greg Falcon <gfalcon@google.com>: Make the absl::c_set_xxxx() algorithms refuse to compile when passed an unordered collection from std:: or absl::. These algorithms operate on sorted sequences; passing an unordered container to them is nearly certainly a bug. This change is technically an API break, but it only breaks incorrect code. We could try to be more clever and detect unordered collections from other libraries, but false positives will break legal code, and this would constitute an API break Abseil cannot afford. PiperOrigin-RevId: 220794190 -- c47cff7f9cc70a4c1604eee0131af552f40e46d6 by Jon Cohen <cohenjon@google.com>: MSVC 2017's STL throws a Structured Exception (not a C++ exception, essentially equivalent to SIGSEGV) when variant::emplace calls a throwing constructor when using the debug multithreaded MSVC runtime DLL. This manifests in dbg mode in Bazel builds. Disable tests which trigger this bug. It's impossible to specifically pull out MSVC 2017 -dbg modes because there's no way for Bazel to know when version of MSVC is being used -- you tell Bazel the directory where the MSVC tools live, not which version of MSVC tools to use. Thus the best we can do is switch on _DEBUG, which is set whenever the debug runtime is selected with the /MDd build flag, as in Bazel -dbg modes. See https://msdn.microsoft.com/en-us/library/b0084kay.aspx ctrl-f "_DEBUG" PiperOrigin-RevId: 220706161 -- 43993d4af309d92f4ebff38391dcc245f154ecc7 by Shaindel Schwartz <shaindel@google.com>: Internal change PiperOrigin-RevId: 220688429 -- 2448802972dcc261af153af464f2b022ef54a2a9 by Abseil Team <absl-team@google.com>: Speed up operator* for uint128 in WIN64. PiperOrigin-RevId: 220678790 -- 7b376403dd05ba10152fb52e40b29d8af79b58bb by Abseil Team <absl-team@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220654834 -- ae08af58111c3f838b8d4de25f501c3559c86002 by Abseil Team <absl-team@google.com>: CMake: Add absl_cc_test function PiperOrigin-RevId: 220603940 GitOrigin-RevId: 07575526242a8e1275ac4223a3d2822795f46569 Change-Id: Iba7f53eb394c8a9de564582a976793f9bb0596d9
2018-11-13 22:22:00 +01:00
#if __has_feature(cxx_thread_local) && \
!(TARGET_OS_IPHONE && __IPHONE_OS_VERSION_MIN_REQUIRED < __IPHONE_9_0)
#define ABSL_HAVE_THREAD_LOCAL 1
#endif
#else // !defined(__APPLE__)
2017-09-19 22:54:40 +02:00
#define ABSL_HAVE_THREAD_LOCAL 1
#endif
// There are platforms for which TLS should not be used even though the compiler
// makes it seem like it's supported (Android NDK < r12b for example).
// This is primarily because of linker problems and toolchain misconfiguration:
// Abseil does not intend to support this indefinitely. Currently, the newest
// toolchain that we intend to support that requires this behavior is the
// r11 NDK - allowing for a 5 year support window on that means this option
// is likely to be removed around June of 2021.
// TLS isn't supported until NDK r12b per
// https://developer.android.com/ndk/downloads/revision_history.html
// Since NDK r16, `__NDK_MAJOR__` and `__NDK_MINOR__` are defined in
// <android/ndk-version.h>. For NDK < r16, users should define these macros,
// e.g. `-D__NDK_MAJOR__=11 -D__NKD_MINOR__=0` for NDK r11.
#if defined(__ANDROID__) && defined(__clang__)
#if __has_include(<android/ndk-version.h>)
#include <android/ndk-version.h>
#endif // __has_include(<android/ndk-version.h>)
#if defined(__ANDROID__) && defined(__clang__) && defined(__NDK_MAJOR__) && \
defined(__NDK_MINOR__) && \
((__NDK_MAJOR__ < 12) || ((__NDK_MAJOR__ == 12) && (__NDK_MINOR__ < 1)))
#undef ABSL_HAVE_TLS
#undef ABSL_HAVE_THREAD_LOCAL
#endif
#endif // defined(__ANDROID__) && defined(__clang__)
2019-06-13 18:14:42 +02:00
// Emscripten doesn't yet support `thread_local` or `__thread`.
// https://github.com/emscripten-core/emscripten/issues/3502
#if defined(__EMSCRIPTEN__)
#undef ABSL_HAVE_TLS
#undef ABSL_HAVE_THREAD_LOCAL
#endif // defined(__EMSCRIPTEN__)
2017-09-19 22:54:40 +02:00
// ABSL_HAVE_INTRINSIC_INT128
//
// Checks whether the __int128 compiler extension for a 128-bit integral type is
// supported.
//
// Note: __SIZEOF_INT128__ is defined by Clang and GCC when __int128 is
// supported, but we avoid using it in certain cases:
// * On Clang:
// * Building using Clang for Windows, where the Clang runtime library has
// 128-bit support only on LP64 architectures, but Windows is LLP64.
// * On Nvidia's nvcc:
// * nvcc also defines __GNUC__ and __SIZEOF_INT128__, but not all versions
// actually support __int128.
2017-09-19 22:54:40 +02:00
#ifdef ABSL_HAVE_INTRINSIC_INT128
#error ABSL_HAVE_INTRINSIC_INT128 cannot be directly set
#elif defined(__SIZEOF_INT128__)
Export of internal Abseil changes. -- 22fceefcf070a0cf89bf1846bee16a9d36ad4161 by Derek Mauro <dmauro@google.com>: Use function static for once initialization of flag registry. This is a workaround for the MSVC debug constexpr initialization issue in absl::once_flag. GitHub #304 PiperOrigin-RevId: 248169007 -- 97bbe6a5233802b61e758c55f7ba8926539cc4ca by Chris Kennelly <ckennelly@google.com>: Internal change PiperOrigin-RevId: 248139347 -- e72640ee079b9fa44e2c7f925fa0a608bcfea515 by Derek Mauro <dmauro@google.com>: Re-write flags config. It doesn't have to be written in the convoluted way it currently is in the opensource-only code path. PiperOrigin-RevId: 248010502 -- 2a72552511b8086c78cb43012c1644e519b3807e by Abseil Team <absl-team@google.com>: Handle pthread_getschedparam() failure. Log an error message if pthread_getschedparam() fails. In Android's Media Framework, libminijail (which I believe is a sandbox) aborts the process if pthread_getschedparam() is called: media.swcodec: libminijail[7526]: blocked syscall: sched_getparam ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ /system/bin/tombstoned: received crash request for pid 7526 Although this CL cannot handle that extreme failure mode, it handles an error return from pthread_getschedparam() and won't use the uninitialized param.sched_priority value in that case. PiperOrigin-RevId: 247999953 -- bb154a92be37987d00d652c7c792594f2f515d83 by Abseil Team <absl-team@google.com>: Allow intrinsic int128 to be set for __aarch64__ targets. PiperOrigin-RevId: 247977594 GitOrigin-RevId: 22fceefcf070a0cf89bf1846bee16a9d36ad4161 Change-Id: I1f7ccfd82eb71446277a8e6f542fe835ac173d71
2019-05-14 19:53:39 +02:00
#if (defined(__clang__) && !defined(_WIN32)) || \
(defined(__CUDACC__) && __CUDACC_VER_MAJOR__ >= 9) || \
(defined(__GNUC__) && !defined(__clang__) && !defined(__CUDACC__))
2017-09-19 22:54:40 +02:00
#define ABSL_HAVE_INTRINSIC_INT128 1
#elif defined(__CUDACC__)
// __CUDACC_VER__ is a full version number before CUDA 9, and is defined to a
// string explaining that it has been removed starting with CUDA 9. We use
// nested #ifs because there is no short-circuiting in the preprocessor.
// NOTE: `__CUDACC__` could be undefined while `__CUDACC_VER__` is defined.
#if __CUDACC_VER__ >= 70000
#define ABSL_HAVE_INTRINSIC_INT128 1
#endif // __CUDACC_VER__ >= 70000
#endif // defined(__CUDACC__)
#endif // ABSL_HAVE_INTRINSIC_INT128
2017-09-19 22:54:40 +02:00
// ABSL_HAVE_EXCEPTIONS
//
// Checks whether the compiler both supports and enables exceptions. Many
// compilers support a "no exceptions" mode that disables exceptions.
//
// Generally, when ABSL_HAVE_EXCEPTIONS is not defined:
//
// * Code using `throw` and `try` may not compile.
// * The `noexcept` specifier will still compile and behave as normal.
// * The `noexcept` operator may still return `false`.
//
// For further details, consult the compiler's documentation.
#ifdef ABSL_HAVE_EXCEPTIONS
#error ABSL_HAVE_EXCEPTIONS cannot be directly set.
#elif defined(__clang__)
// TODO(calabrese)
// Switch to using __cpp_exceptions when we no longer support versions < 3.6.
// For details on this check, see:
// http://releases.llvm.org/3.6.0/tools/clang/docs/ReleaseNotes.html#the-exceptions-macro
#if defined(__EXCEPTIONS) && __has_feature(cxx_exceptions)
#define ABSL_HAVE_EXCEPTIONS 1
#endif // defined(__EXCEPTIONS) && __has_feature(cxx_exceptions)
// Handle remaining special cases and default to exceptions being supported.
#elif !(defined(__GNUC__) && (__GNUC__ < 5) && !defined(__EXCEPTIONS)) && \
!(defined(__GNUC__) && (__GNUC__ >= 5) && !defined(__cpp_exceptions)) && \
!(defined(_MSC_VER) && !defined(_CPPUNWIND))
#define ABSL_HAVE_EXCEPTIONS 1
#endif
// -----------------------------------------------------------------------------
// Platform Feature Checks
// -----------------------------------------------------------------------------
// Currently supported operating systems and associated preprocessor
// symbols:
//
// Linux and Linux-derived __linux__
// Android __ANDROID__ (implies __linux__)
// Linux (non-Android) __linux__ && !__ANDROID__
// Darwin (macOS and iOS) __APPLE__
2017-09-19 22:54:40 +02:00
// Akaros (http://akaros.org) __ros__
// Windows _WIN32
// NaCL __native_client__
// AsmJS __asmjs__
// WebAssembly __wasm__
// Fuchsia __Fuchsia__
2017-09-19 22:54:40 +02:00
//
// Note that since Android defines both __ANDROID__ and __linux__, one
// may probe for either Linux or Android by simply testing for __linux__.
// ABSL_HAVE_MMAP
//
// Checks whether the platform has an mmap(2) implementation as defined in
// POSIX.1-2001.
#ifdef ABSL_HAVE_MMAP
#error ABSL_HAVE_MMAP cannot be directly set
#elif defined(__linux__) || defined(__APPLE__) || defined(__FreeBSD__) || \
2017-12-13 21:02:50 +01:00
defined(__ros__) || defined(__native_client__) || defined(__asmjs__) || \
Export of internal Abseil changes. -- 5f1ab09522226336830d9ea6ef7276d37f536ac5 by Abseil Team <absl-team@google.com>: Clarify the documentation of ABSL_MUST_USE_RESULT. PiperOrigin-RevId: 221663609 -- af4c8359a20d56369fd1dce318220cf3be03ca66 by Greg Falcon <gfalcon@google.com>: Internal change PiperOrigin-RevId: 221538448 -- 487cd09bd1942bf607080deeae38fee6ce66f294 by Eric Fiselier <ericwf@google.com>: Work around emscripten bugs and missing features in absl/time:time_test. The emscripten toolchain has a couple of issues that cause time_test to fail. Specifically: 1) emscripten doesn't support signals. 2) The javascript implementation of strftime/strptime use different expansions of '%c' that mean it doesn't round-trip. PiperOrigin-RevId: 221523701 -- 5823652e6a200b97b07334bc47128dfac40e20fc by Xiaoyi Zhang <zhangxy@google.com>: Fix MSVC compiler warning by explicitly casting to char. Currently our MSVC build breaks with the following error: raw_hash_set.h(406): warning C4309: 'argument': truncation of constant value PiperOrigin-RevId: 221492585 -- c5806358320711a5efbe5c523df13e14ab53a17d by Greg Falcon <gfalcon@google.com>: Replace calls to getpagesize() with the more portable sysconf(_SC_PAGESIZE); the latter is in POSIX 1.0 and is called out in the Linux `getpagesize` man page as a more portable spelling. PiperOrigin-RevId: 221492471 -- 19ffe82851072229bb7ce73f754ffe4c18e8c575 by Abseil Team <absl-team@google.com>: Fix -Wundef error in absl/hash/internal/hash.h. PiperOrigin-RevId: 221444120 -- b30f3d0a848563b6e4ec33f3dc085831dfabb748 by Jon Cohen <cohenjon@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 221339736 GitOrigin-RevId: 5f1ab09522226336830d9ea6ef7276d37f536ac5 Change-Id: I96223d522d98bf6616dea88eb047c2d536eeddd0
2018-11-15 20:55:00 +01:00
defined(__wasm__) || defined(__Fuchsia__) || defined(__sun) || \
defined(__ASYLO__)
2017-09-19 22:54:40 +02:00
#define ABSL_HAVE_MMAP 1
#endif
// ABSL_HAVE_PTHREAD_GETSCHEDPARAM
//
// Checks whether the platform implements the pthread_(get|set)schedparam(3)
// functions as defined in POSIX.1-2001.
#ifdef ABSL_HAVE_PTHREAD_GETSCHEDPARAM
#error ABSL_HAVE_PTHREAD_GETSCHEDPARAM cannot be directly set
2017-12-13 21:02:50 +01:00
#elif defined(__linux__) || defined(__APPLE__) || defined(__FreeBSD__) || \
defined(__ros__)
2017-09-19 22:54:40 +02:00
#define ABSL_HAVE_PTHREAD_GETSCHEDPARAM 1
#endif
// ABSL_HAVE_SCHED_YIELD
//
// Checks whether the platform implements sched_yield(2) as defined in
// POSIX.1-2001.
#ifdef ABSL_HAVE_SCHED_YIELD
#error ABSL_HAVE_SCHED_YIELD cannot be directly set
#elif defined(__linux__) || defined(__ros__) || defined(__native_client__)
#define ABSL_HAVE_SCHED_YIELD 1
#endif
// ABSL_HAVE_SEMAPHORE_H
//
Export of internal Abseil changes -- e54b9c7bbb0c58475676c268e2e19c69f4bce48a by Jorg Brown <jorg@google.com>: Tweak ABSL_PREDICT_TRUE slightly, for better code on some platforms and/or optimization levels. "false || (x)" is more verbose than "!!(x)", but ultimately more efficient. For example, given this code: void InitIfNecessary() { if (ABSL_PREDICT_TRUE(NeedsInit())) { SlowInitIfNecessary(); } } Clang with default optimization level will produce: Before this CL After this CL InitIfNecessary: InitIfNecessary: push rbp push rbp mov rbp, rsp mov rbp, rsp call NeedsInit call NeedsInit xor al, -1 xor al, -1 test al, 1 test al, 1 jne .LBB2_1 jne .LBB3_1 jmp .LBB2_2 jmp .LBB3_2 .LBB2_1: .LBB3_1: call SlowInitIfNecessary call SlowInitIfNecessary .LBB2_2: .LBB3_2: pop rbp pop rbp ret ret PiperOrigin-RevId: 276401386 -- 0a3c4dfd8342bf2b1b11a87f1c662c883f73cab7 by Abseil Team <absl-team@google.com>: Fix comment nit: sem_open => sem_init. The code calls sem_init, not sem_open, to initialize an unnamed semaphore. (sem_open creates or opens a named semaphore.) PiperOrigin-RevId: 276344072 -- b36a664e9459057509a90e83d3482e1d3a4c44c7 by Abseil Team <absl-team@google.com>: Fix typo in flat_hash_map.h: exchaged -> exchanged PiperOrigin-RevId: 276295792 -- 7bbd8d18276eb110c8335743e35fceb662ddf3d6 by Samuel Benzaquen <sbenza@google.com>: Add assertions to verify use of iterators. PiperOrigin-RevId: 276283300 -- 677398a8ffcb1f59182cffe57a4fe7ff147a0404 by Laramie Leavitt <lar@google.com>: Migrate distribution_impl.h/cc to generate_real.h/cc. Combine the methods RandU64To<Float,Double> into a single method: GenerateRealFromBits(). Remove rejection sampling from absl::uniform_real_distribution. PiperOrigin-RevId: 276158675 -- c60c9d11d24b0c546329d998e78e15a84b3153f5 by Abseil Team <absl-team@google.com>: Internal change PiperOrigin-RevId: 276126962 -- 4c840cab6a8d86efa29b397cafaf7520eece68cc by Andy Soffer <asoffer@google.com>: Update CMakeLists.txt to address https://github.com/abseil/abseil-cpp/issues/365. This does not cover every platform, but it does at least address the first-order issue of assuming gcc implies x86. PiperOrigin-RevId: 276116253 -- 98da366e6b5d51afe5d7ac6722126aca23d85ee6 by Abseil Team <absl-team@google.com>: Internal change PiperOrigin-RevId: 276097452 GitOrigin-RevId: e54b9c7bbb0c58475676c268e2e19c69f4bce48a Change-Id: I02d84454bb71ab21ad3d39650acf6cc6e36f58d7
2019-10-24 04:35:39 +02:00
// Checks whether the platform supports the <semaphore.h> header and sem_init(3)
2017-09-19 22:54:40 +02:00
// family of functions as standardized in POSIX.1-2001.
//
// Note: While Apple provides <semaphore.h> for both iOS and macOS, it is
2018-04-19 20:11:44 +02:00
// explicitly deprecated and will cause build failures if enabled for those
2017-09-19 22:54:40 +02:00
// platforms. We side-step the issue by not defining it here for Apple
// platforms.
#ifdef ABSL_HAVE_SEMAPHORE_H
#error ABSL_HAVE_SEMAPHORE_H cannot be directly set
#elif defined(__linux__) || defined(__ros__)
#define ABSL_HAVE_SEMAPHORE_H 1
#endif
// ABSL_HAVE_ALARM
//
// Checks whether the platform supports the <signal.h> header and alarm(2)
// function as standardized in POSIX.1-2001.
#ifdef ABSL_HAVE_ALARM
#error ABSL_HAVE_ALARM cannot be directly set
#elif defined(__GOOGLE_GRTE_VERSION__)
// feature tests for Google's GRTE
#define ABSL_HAVE_ALARM 1
#elif defined(__GLIBC__)
// feature test for glibc
#define ABSL_HAVE_ALARM 1
#elif defined(_MSC_VER)
// feature tests for Microsoft's library
#elif defined(__MINGW32__)
// mingw32 doesn't provide alarm(2):
// https://osdn.net/projects/mingw/scm/git/mingw-org-wsl/blobs/5.2-trunk/mingwrt/include/unistd.h
// mingw-w64 provides a no-op implementation:
// https://sourceforge.net/p/mingw-w64/mingw-w64/ci/master/tree/mingw-w64-crt/misc/alarm.c
Export of internal Abseil changes. -- 5f1ab09522226336830d9ea6ef7276d37f536ac5 by Abseil Team <absl-team@google.com>: Clarify the documentation of ABSL_MUST_USE_RESULT. PiperOrigin-RevId: 221663609 -- af4c8359a20d56369fd1dce318220cf3be03ca66 by Greg Falcon <gfalcon@google.com>: Internal change PiperOrigin-RevId: 221538448 -- 487cd09bd1942bf607080deeae38fee6ce66f294 by Eric Fiselier <ericwf@google.com>: Work around emscripten bugs and missing features in absl/time:time_test. The emscripten toolchain has a couple of issues that cause time_test to fail. Specifically: 1) emscripten doesn't support signals. 2) The javascript implementation of strftime/strptime use different expansions of '%c' that mean it doesn't round-trip. PiperOrigin-RevId: 221523701 -- 5823652e6a200b97b07334bc47128dfac40e20fc by Xiaoyi Zhang <zhangxy@google.com>: Fix MSVC compiler warning by explicitly casting to char. Currently our MSVC build breaks with the following error: raw_hash_set.h(406): warning C4309: 'argument': truncation of constant value PiperOrigin-RevId: 221492585 -- c5806358320711a5efbe5c523df13e14ab53a17d by Greg Falcon <gfalcon@google.com>: Replace calls to getpagesize() with the more portable sysconf(_SC_PAGESIZE); the latter is in POSIX 1.0 and is called out in the Linux `getpagesize` man page as a more portable spelling. PiperOrigin-RevId: 221492471 -- 19ffe82851072229bb7ce73f754ffe4c18e8c575 by Abseil Team <absl-team@google.com>: Fix -Wundef error in absl/hash/internal/hash.h. PiperOrigin-RevId: 221444120 -- b30f3d0a848563b6e4ec33f3dc085831dfabb748 by Jon Cohen <cohenjon@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 221339736 GitOrigin-RevId: 5f1ab09522226336830d9ea6ef7276d37f536ac5 Change-Id: I96223d522d98bf6616dea88eb047c2d536eeddd0
2018-11-15 20:55:00 +01:00
#elif defined(__EMSCRIPTEN__)
// emscripten doesn't support signals
#elif defined(__Fuchsia__)
// Signals don't exist on fuchsia.
2017-09-19 22:54:40 +02:00
#elif defined(__native_client__)
#else
// other standard libraries
#define ABSL_HAVE_ALARM 1
#endif
// ABSL_IS_LITTLE_ENDIAN
// ABSL_IS_BIG_ENDIAN
//
// Checks the endianness of the platform.
//
// Notes: uses the built in endian macros provided by GCC (since 4.6) and
// Clang (since 3.2); see
// https://gcc.gnu.org/onlinedocs/cpp/Common-Predefined-Macros.html.
// Otherwise, if _WIN32, assume little endian. Otherwise, bail with an error.
#if defined(ABSL_IS_BIG_ENDIAN)
#error "ABSL_IS_BIG_ENDIAN cannot be directly set."
#endif
#if defined(ABSL_IS_LITTLE_ENDIAN)
#error "ABSL_IS_LITTLE_ENDIAN cannot be directly set."
#endif
#if (defined(__BYTE_ORDER__) && defined(__ORDER_LITTLE_ENDIAN__) && \
__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
#define ABSL_IS_LITTLE_ENDIAN 1
#elif defined(__BYTE_ORDER__) && defined(__ORDER_BIG_ENDIAN__) && \
__BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
#define ABSL_IS_BIG_ENDIAN 1
#elif defined(_WIN32)
#define ABSL_IS_LITTLE_ENDIAN 1
#else
#error "absl endian detection needs to be set up for your compiler"
#endif
// macOS 10.13 and iOS 10.11 don't let you use <any>, <optional>, or <variant>
// even though the headers exist and are publicly noted to work. See
Export of internal Abseil changes. -- da04b8cd21f6225d71397471474d34a77df0efd6 by Jon Cohen <cohenjon@google.com>: Don't use std::any, std::optional, std::variant, and friends on MacOS versions older than 10.14. Although Xcode 10 includes those headers and makes the types available to use, according to https://developer.apple.com/documentation/xcode_release_notes/xcode_10_release_notes, on MacOS 10.13 and earlier use of any functions (std::get, for example) results in an error message to upgrade to MacOS 10.14. This fixes https://github.com/abseil/abseil-cpp/issues/207. See that issue for more information on the error generated. PiperOrigin-RevId: 221844618 -- 1d99f77b4c60c5b0d7984f46e8ed63a3f969c635 by Jon Cohen <cohenjon@google.com>: raw_hash_set_test is still flaky under gcc 4.8. Since we now have the probe_test, we don't need the PerfectRatio tests. Just remove them. PiperOrigin-RevId: 221843042 -- 135cbb2a5d90963256518b3b59fe6710815e5dfa by Abseil Team <absl-team@google.com>: Update absl/algorithm/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 221828348 -- 1a5abde4f17f998ae89d87155d59f982a70202d8 by Jon Cohen <cohenjon@google.com>: Internal change PiperOrigin-RevId: 221708245 -- e03e031d4de39275989f695c768b0940cce1ff16 by Matt Armstrong <marmstrong@google.com>: Log to FATAL in throw_delegate.h ABSL_RAW_LOG(FATAL, ...) is guaranteed to abort. Previously, the code was logging to ERROR and calling abort() explicitly, which defeated any integration with absl::raw_logging_internal::AbortHook(). These changes are limited to Abseil internal APIs. PiperOrigin-RevId: 221696513 -- d13691523a3f9a5367fd1194cf9604bf4a969029 by Shahriar Rouf <nafi@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 221694877 -- f4044c56d44ba0ac2a9f218ed55f1b1f9e985eae by Abseil Team <absl-team@google.com>: Update absl/base/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 221676669 GitOrigin-RevId: da04b8cd21f6225d71397471474d34a77df0efd6 Change-Id: If6621e10d096a39b6a056a072c2727a0df0b0620
2018-11-16 22:36:37 +01:00
// https://github.com/abseil/abseil-cpp/issues/207 and
// https://developer.apple.com/documentation/xcode_release_notes/xcode_10_release_notes
// libc++ spells out the availability requirements in the file
// llvm-project/libcxx/include/__config via the #define
// _LIBCPP_AVAILABILITY_BAD_OPTIONAL_ACCESS.
Export of internal Abseil changes. -- da04b8cd21f6225d71397471474d34a77df0efd6 by Jon Cohen <cohenjon@google.com>: Don't use std::any, std::optional, std::variant, and friends on MacOS versions older than 10.14. Although Xcode 10 includes those headers and makes the types available to use, according to https://developer.apple.com/documentation/xcode_release_notes/xcode_10_release_notes, on MacOS 10.13 and earlier use of any functions (std::get, for example) results in an error message to upgrade to MacOS 10.14. This fixes https://github.com/abseil/abseil-cpp/issues/207. See that issue for more information on the error generated. PiperOrigin-RevId: 221844618 -- 1d99f77b4c60c5b0d7984f46e8ed63a3f969c635 by Jon Cohen <cohenjon@google.com>: raw_hash_set_test is still flaky under gcc 4.8. Since we now have the probe_test, we don't need the PerfectRatio tests. Just remove them. PiperOrigin-RevId: 221843042 -- 135cbb2a5d90963256518b3b59fe6710815e5dfa by Abseil Team <absl-team@google.com>: Update absl/algorithm/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 221828348 -- 1a5abde4f17f998ae89d87155d59f982a70202d8 by Jon Cohen <cohenjon@google.com>: Internal change PiperOrigin-RevId: 221708245 -- e03e031d4de39275989f695c768b0940cce1ff16 by Matt Armstrong <marmstrong@google.com>: Log to FATAL in throw_delegate.h ABSL_RAW_LOG(FATAL, ...) is guaranteed to abort. Previously, the code was logging to ERROR and calling abort() explicitly, which defeated any integration with absl::raw_logging_internal::AbortHook(). These changes are limited to Abseil internal APIs. PiperOrigin-RevId: 221696513 -- d13691523a3f9a5367fd1194cf9604bf4a969029 by Shahriar Rouf <nafi@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 221694877 -- f4044c56d44ba0ac2a9f218ed55f1b1f9e985eae by Abseil Team <absl-team@google.com>: Update absl/base/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 221676669 GitOrigin-RevId: da04b8cd21f6225d71397471474d34a77df0efd6 Change-Id: If6621e10d096a39b6a056a072c2727a0df0b0620
2018-11-16 22:36:37 +01:00
#if defined(__APPLE__) && defined(_LIBCPP_VERSION) && \
((defined(__ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__) && \
__ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__ < 101400) || \
(defined(__ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__) && \
__ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__ < 120000) || \
(defined(__ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__) && \
__ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__ < 120000) || \
(defined(__ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__) && \
__ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__ < 50000))
#define ABSL_INTERNAL_APPLE_CXX17_TYPES_UNAVAILABLE 1
Export of internal Abseil changes. -- da04b8cd21f6225d71397471474d34a77df0efd6 by Jon Cohen <cohenjon@google.com>: Don't use std::any, std::optional, std::variant, and friends on MacOS versions older than 10.14. Although Xcode 10 includes those headers and makes the types available to use, according to https://developer.apple.com/documentation/xcode_release_notes/xcode_10_release_notes, on MacOS 10.13 and earlier use of any functions (std::get, for example) results in an error message to upgrade to MacOS 10.14. This fixes https://github.com/abseil/abseil-cpp/issues/207. See that issue for more information on the error generated. PiperOrigin-RevId: 221844618 -- 1d99f77b4c60c5b0d7984f46e8ed63a3f969c635 by Jon Cohen <cohenjon@google.com>: raw_hash_set_test is still flaky under gcc 4.8. Since we now have the probe_test, we don't need the PerfectRatio tests. Just remove them. PiperOrigin-RevId: 221843042 -- 135cbb2a5d90963256518b3b59fe6710815e5dfa by Abseil Team <absl-team@google.com>: Update absl/algorithm/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 221828348 -- 1a5abde4f17f998ae89d87155d59f982a70202d8 by Jon Cohen <cohenjon@google.com>: Internal change PiperOrigin-RevId: 221708245 -- e03e031d4de39275989f695c768b0940cce1ff16 by Matt Armstrong <marmstrong@google.com>: Log to FATAL in throw_delegate.h ABSL_RAW_LOG(FATAL, ...) is guaranteed to abort. Previously, the code was logging to ERROR and calling abort() explicitly, which defeated any integration with absl::raw_logging_internal::AbortHook(). These changes are limited to Abseil internal APIs. PiperOrigin-RevId: 221696513 -- d13691523a3f9a5367fd1194cf9604bf4a969029 by Shahriar Rouf <nafi@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 221694877 -- f4044c56d44ba0ac2a9f218ed55f1b1f9e985eae by Abseil Team <absl-team@google.com>: Update absl/base/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 221676669 GitOrigin-RevId: da04b8cd21f6225d71397471474d34a77df0efd6 Change-Id: If6621e10d096a39b6a056a072c2727a0df0b0620
2018-11-16 22:36:37 +01:00
#else
#define ABSL_INTERNAL_APPLE_CXX17_TYPES_UNAVAILABLE 0
Export of internal Abseil changes. -- da04b8cd21f6225d71397471474d34a77df0efd6 by Jon Cohen <cohenjon@google.com>: Don't use std::any, std::optional, std::variant, and friends on MacOS versions older than 10.14. Although Xcode 10 includes those headers and makes the types available to use, according to https://developer.apple.com/documentation/xcode_release_notes/xcode_10_release_notes, on MacOS 10.13 and earlier use of any functions (std::get, for example) results in an error message to upgrade to MacOS 10.14. This fixes https://github.com/abseil/abseil-cpp/issues/207. See that issue for more information on the error generated. PiperOrigin-RevId: 221844618 -- 1d99f77b4c60c5b0d7984f46e8ed63a3f969c635 by Jon Cohen <cohenjon@google.com>: raw_hash_set_test is still flaky under gcc 4.8. Since we now have the probe_test, we don't need the PerfectRatio tests. Just remove them. PiperOrigin-RevId: 221843042 -- 135cbb2a5d90963256518b3b59fe6710815e5dfa by Abseil Team <absl-team@google.com>: Update absl/algorithm/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 221828348 -- 1a5abde4f17f998ae89d87155d59f982a70202d8 by Jon Cohen <cohenjon@google.com>: Internal change PiperOrigin-RevId: 221708245 -- e03e031d4de39275989f695c768b0940cce1ff16 by Matt Armstrong <marmstrong@google.com>: Log to FATAL in throw_delegate.h ABSL_RAW_LOG(FATAL, ...) is guaranteed to abort. Previously, the code was logging to ERROR and calling abort() explicitly, which defeated any integration with absl::raw_logging_internal::AbortHook(). These changes are limited to Abseil internal APIs. PiperOrigin-RevId: 221696513 -- d13691523a3f9a5367fd1194cf9604bf4a969029 by Shahriar Rouf <nafi@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 221694877 -- f4044c56d44ba0ac2a9f218ed55f1b1f9e985eae by Abseil Team <absl-team@google.com>: Update absl/base/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 221676669 GitOrigin-RevId: da04b8cd21f6225d71397471474d34a77df0efd6 Change-Id: If6621e10d096a39b6a056a072c2727a0df0b0620
2018-11-16 22:36:37 +01:00
#endif
2017-09-19 22:54:40 +02:00
// ABSL_HAVE_STD_ANY
//
// Checks whether C++17 std::any is available by checking whether <any> exists.
2017-09-19 22:54:40 +02:00
#ifdef ABSL_HAVE_STD_ANY
#error "ABSL_HAVE_STD_ANY cannot be directly set."
#endif
#ifdef __has_include
Export of internal Abseil changes. -- da04b8cd21f6225d71397471474d34a77df0efd6 by Jon Cohen <cohenjon@google.com>: Don't use std::any, std::optional, std::variant, and friends on MacOS versions older than 10.14. Although Xcode 10 includes those headers and makes the types available to use, according to https://developer.apple.com/documentation/xcode_release_notes/xcode_10_release_notes, on MacOS 10.13 and earlier use of any functions (std::get, for example) results in an error message to upgrade to MacOS 10.14. This fixes https://github.com/abseil/abseil-cpp/issues/207. See that issue for more information on the error generated. PiperOrigin-RevId: 221844618 -- 1d99f77b4c60c5b0d7984f46e8ed63a3f969c635 by Jon Cohen <cohenjon@google.com>: raw_hash_set_test is still flaky under gcc 4.8. Since we now have the probe_test, we don't need the PerfectRatio tests. Just remove them. PiperOrigin-RevId: 221843042 -- 135cbb2a5d90963256518b3b59fe6710815e5dfa by Abseil Team <absl-team@google.com>: Update absl/algorithm/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 221828348 -- 1a5abde4f17f998ae89d87155d59f982a70202d8 by Jon Cohen <cohenjon@google.com>: Internal change PiperOrigin-RevId: 221708245 -- e03e031d4de39275989f695c768b0940cce1ff16 by Matt Armstrong <marmstrong@google.com>: Log to FATAL in throw_delegate.h ABSL_RAW_LOG(FATAL, ...) is guaranteed to abort. Previously, the code was logging to ERROR and calling abort() explicitly, which defeated any integration with absl::raw_logging_internal::AbortHook(). These changes are limited to Abseil internal APIs. PiperOrigin-RevId: 221696513 -- d13691523a3f9a5367fd1194cf9604bf4a969029 by Shahriar Rouf <nafi@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 221694877 -- f4044c56d44ba0ac2a9f218ed55f1b1f9e985eae by Abseil Team <absl-team@google.com>: Update absl/base/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 221676669 GitOrigin-RevId: da04b8cd21f6225d71397471474d34a77df0efd6 Change-Id: If6621e10d096a39b6a056a072c2727a0df0b0620
2018-11-16 22:36:37 +01:00
#if __has_include(<any>) && __cplusplus >= 201703L && \
!ABSL_INTERNAL_APPLE_CXX17_TYPES_UNAVAILABLE
2017-09-19 22:54:40 +02:00
#define ABSL_HAVE_STD_ANY 1
#endif
#endif
// ABSL_HAVE_STD_OPTIONAL
//
// Checks whether C++17 std::optional is available.
#ifdef ABSL_HAVE_STD_OPTIONAL
#error "ABSL_HAVE_STD_OPTIONAL cannot be directly set."
#endif
#ifdef __has_include
Export of internal Abseil changes. -- da04b8cd21f6225d71397471474d34a77df0efd6 by Jon Cohen <cohenjon@google.com>: Don't use std::any, std::optional, std::variant, and friends on MacOS versions older than 10.14. Although Xcode 10 includes those headers and makes the types available to use, according to https://developer.apple.com/documentation/xcode_release_notes/xcode_10_release_notes, on MacOS 10.13 and earlier use of any functions (std::get, for example) results in an error message to upgrade to MacOS 10.14. This fixes https://github.com/abseil/abseil-cpp/issues/207. See that issue for more information on the error generated. PiperOrigin-RevId: 221844618 -- 1d99f77b4c60c5b0d7984f46e8ed63a3f969c635 by Jon Cohen <cohenjon@google.com>: raw_hash_set_test is still flaky under gcc 4.8. Since we now have the probe_test, we don't need the PerfectRatio tests. Just remove them. PiperOrigin-RevId: 221843042 -- 135cbb2a5d90963256518b3b59fe6710815e5dfa by Abseil Team <absl-team@google.com>: Update absl/algorithm/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 221828348 -- 1a5abde4f17f998ae89d87155d59f982a70202d8 by Jon Cohen <cohenjon@google.com>: Internal change PiperOrigin-RevId: 221708245 -- e03e031d4de39275989f695c768b0940cce1ff16 by Matt Armstrong <marmstrong@google.com>: Log to FATAL in throw_delegate.h ABSL_RAW_LOG(FATAL, ...) is guaranteed to abort. Previously, the code was logging to ERROR and calling abort() explicitly, which defeated any integration with absl::raw_logging_internal::AbortHook(). These changes are limited to Abseil internal APIs. PiperOrigin-RevId: 221696513 -- d13691523a3f9a5367fd1194cf9604bf4a969029 by Shahriar Rouf <nafi@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 221694877 -- f4044c56d44ba0ac2a9f218ed55f1b1f9e985eae by Abseil Team <absl-team@google.com>: Update absl/base/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 221676669 GitOrigin-RevId: da04b8cd21f6225d71397471474d34a77df0efd6 Change-Id: If6621e10d096a39b6a056a072c2727a0df0b0620
2018-11-16 22:36:37 +01:00
#if __has_include(<optional>) && __cplusplus >= 201703L && \
!ABSL_INTERNAL_APPLE_CXX17_TYPES_UNAVAILABLE
2017-09-19 22:54:40 +02:00
#define ABSL_HAVE_STD_OPTIONAL 1
#endif
#endif
// ABSL_HAVE_STD_VARIANT
//
// Checks whether C++17 std::variant is available.
#ifdef ABSL_HAVE_STD_VARIANT
#error "ABSL_HAVE_STD_VARIANT cannot be directly set."
#endif
#ifdef __has_include
Export of internal Abseil changes. -- da04b8cd21f6225d71397471474d34a77df0efd6 by Jon Cohen <cohenjon@google.com>: Don't use std::any, std::optional, std::variant, and friends on MacOS versions older than 10.14. Although Xcode 10 includes those headers and makes the types available to use, according to https://developer.apple.com/documentation/xcode_release_notes/xcode_10_release_notes, on MacOS 10.13 and earlier use of any functions (std::get, for example) results in an error message to upgrade to MacOS 10.14. This fixes https://github.com/abseil/abseil-cpp/issues/207. See that issue for more information on the error generated. PiperOrigin-RevId: 221844618 -- 1d99f77b4c60c5b0d7984f46e8ed63a3f969c635 by Jon Cohen <cohenjon@google.com>: raw_hash_set_test is still flaky under gcc 4.8. Since we now have the probe_test, we don't need the PerfectRatio tests. Just remove them. PiperOrigin-RevId: 221843042 -- 135cbb2a5d90963256518b3b59fe6710815e5dfa by Abseil Team <absl-team@google.com>: Update absl/algorithm/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 221828348 -- 1a5abde4f17f998ae89d87155d59f982a70202d8 by Jon Cohen <cohenjon@google.com>: Internal change PiperOrigin-RevId: 221708245 -- e03e031d4de39275989f695c768b0940cce1ff16 by Matt Armstrong <marmstrong@google.com>: Log to FATAL in throw_delegate.h ABSL_RAW_LOG(FATAL, ...) is guaranteed to abort. Previously, the code was logging to ERROR and calling abort() explicitly, which defeated any integration with absl::raw_logging_internal::AbortHook(). These changes are limited to Abseil internal APIs. PiperOrigin-RevId: 221696513 -- d13691523a3f9a5367fd1194cf9604bf4a969029 by Shahriar Rouf <nafi@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 221694877 -- f4044c56d44ba0ac2a9f218ed55f1b1f9e985eae by Abseil Team <absl-team@google.com>: Update absl/base/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 221676669 GitOrigin-RevId: da04b8cd21f6225d71397471474d34a77df0efd6 Change-Id: If6621e10d096a39b6a056a072c2727a0df0b0620
2018-11-16 22:36:37 +01:00
#if __has_include(<variant>) && __cplusplus >= 201703L && \
!ABSL_INTERNAL_APPLE_CXX17_TYPES_UNAVAILABLE
#define ABSL_HAVE_STD_VARIANT 1
#endif
#endif
2017-09-19 22:54:40 +02:00
// ABSL_HAVE_STD_STRING_VIEW
//
// Checks whether C++17 std::string_view is available.
#ifdef ABSL_HAVE_STD_STRING_VIEW
#error "ABSL_HAVE_STD_STRING_VIEW cannot be directly set."
#endif
#ifdef __has_include
#if __has_include(<string_view>) && __cplusplus >= 201703L
#define ABSL_HAVE_STD_STRING_VIEW 1
#endif
#endif
// For MSVC, `__has_include` is supported in VS 2017 15.3, which is later than
// the support for <optional>, <any>, <string_view>, <variant>. So we use
// _MSC_VER to check whether we have VS 2017 RTM (when <optional>, <any>,
// <string_view>, <variant> is implemented) or higher. Also, `__cplusplus` is
// not correctly set by MSVC, so we use `_MSVC_LANG` to check the language
// version.
// TODO(zhangxy): fix tests before enabling aliasing for `std::any`.
#if defined(_MSC_VER) && _MSC_VER >= 1910 && \
((defined(_MSVC_LANG) && _MSVC_LANG > 201402) || __cplusplus > 201402)
// #define ABSL_HAVE_STD_ANY 1
#define ABSL_HAVE_STD_OPTIONAL 1
#define ABSL_HAVE_STD_VARIANT 1
#define ABSL_HAVE_STD_STRING_VIEW 1
#endif
Export of internal Abseil changes -- 2f49cb9009386bc67bf54a2908c8720b749c8b7f by Greg Falcon <gfalcon@google.com>: docs: fix typo Import of https://github.com/abseil/abseil-cpp/pull/397 PiperOrigin-RevId: 277504420 -- f2bed362c1c12d3fa9c22d11f2b918668e8c37b7 by Abseil Team <absl-team@google.com>: Avoid our is_[copy/move]_assignable workarounds in MSVC 19.20 and on, since that release introduces a regression that breaks our workaround. We should ideally use the std forms in more cases, but branching when our workarounds fail is simpler to maintain. PiperOrigin-RevId: 277502334 -- e33de894ffd49848b9e088f59acc9743d1661948 by Derek Mauro <dmauro@google.com>: Update rules_cc version. The mirror.bazel.build URL does not exist (cache expiration?) PiperOrigin-RevId: 277498394 -- b23757b0747c64634d2d701433782c969effef19 by Abseil Team <absl-team@google.com>: Fix https://github.com/abseil/abseil-cpp/issues/394. PiperOrigin-RevId: 277491405 -- 54c75b8b29813531c52d67cf0ba7063baae4a4f3 by Abseil Team <absl-team@google.com>: Fix comment typos: waker => waiter. PiperOrigin-RevId: 277376952 -- 874eeaa3b3af808fc88b6355245f643674f5e36e by Abseil Team <absl-team@google.com>: Don't use atomic ops on waiter and wakeup counts in CONDVAR waiter mode. Just guard the waiter and wakeup counts with the mutex. This eliminates the race. Also fix a typo in the error message for pthread_cond_timedwait. PiperOrigin-RevId: 277366017 -- ce8c9a63109214519b5a7eaecef2c663c4d566df by Greg Falcon <gfalcon@google.com>: Implement the config options for our four main C++ forward compatibility types. These options control whether the names `any`, `optional`, `string_view`, and `variant` in namespace `absl` are aliases to the corresponding C++17 types in namespace `std`. By default, we continue to auto-detect the configuration of the compiler being run. These options are not intended to be modified on the command line (as -D flags, say). Instead, the options.h file can be modified by distributors of Abseil (e.g., binary packagers, maintainers of local copies of Abseil, etc.) Changing options will change Abseil in an ODR sense. Any program must only link in a single version of Abseil. Linking libraries that use Abseil configured with different options is an error: there is no ABI compatibility guarantee when linking different configurations, even if the Abseil versions used are otherwise exactly identical. PiperOrigin-RevId: 277364298 -- 5ed3ad42ae43a05862070f92f9ffd07f5c1f2b81 by Chris Kennelly <ckennelly@google.com>: Suppress -Wimplicit-int-float-conversion. On recent builds of Clang, this is an error/warning. PiperOrigin-RevId: 277346168 -- 9b9b0055243c705189bb27d912e6d45a7789cdee by Eric Fiselier <ericwf@google.com>: Allow building Abseil as a shared library with CMake. By default CMake's `add_library` creates the target as a static library. However, users can override the default using the builtin CMake option -DBUILD_SHARED_LIBS=ON. This changes Abseil's CMake to respect this configuration option by removing the explicit `STATIC` in our usages of `add_library`. PiperOrigin-RevId: 277337753 -- 63a8b7b8ede3a9d851916929d6b8537f4f2508ca by Abseil Team <absl-team@google.com>: Improve AlphaNum Hex performance by using absl::numbers_internal::FastHexToBufferZeroPad16. PiperOrigin-RevId: 277318108 -- dd047f7e92032682d94b27732df0e4d0670e24a4 by CJ Johnson <johnsoncj@google.com>: Internal change PiperOrigin-RevId: 277316913 -- d19ee7074929fed08973cc5b40a844573ce1e0a6 by Abseil Team <absl-team@google.com>: Handle invoking [[nodiscard]] functions correctly in our tests. PiperOrigin-RevId: 277301500 -- 5373f3737894ba9b8481e95e5f58c7957c00d26a by Chris Kennelly <ckennelly@google.com>: For internal reasons, loosen visibility restrictions of `//absl/base:malloc_internal`. As an internal-namespace interface, this module remains unsupported. We reserve the right to change, delete, or re-restrict visibility to this target at any time. PiperOrigin-RevId: 277118689 -- 44e4f6655e05393a375d09b3625c192b1fde5909 by Abseil Team <absl-team@google.com>: Fix error in example civil day comment. PiperOrigin-RevId: 277103315 -- 7657392b4ce48469106f11cdb952a0bc76a28df3 by Abseil Team <absl-team@google.com>: Internal change PiperOrigin-RevId: 277056076 -- c75bda76f40b01fa249b75b5a70c1f5907e56c76 by Abseil Team <absl-team@google.com>: Suppress lifetime constant-initialization tests when building with MSVC versions > 19.0. These are broken due to non-compliant initialization order in these versions: https://developercommunity.visualstudio.com/content/problem/336946/class-with-constexpr-constructor-not-using-static.html We don't know when Microsoft will fix this bug. PiperOrigin-RevId: 277049770 -- 16c3b9bf2a1796efa57f97b00bcd6227fbccca1f by Matt Calabrese <calabrese@google.com>: Avoid our is_[copy/move]_assignable workarounds in MSVC 19.20 and on, since that release introduces a regression that breaks our workaround. We should ideally use the std forms in more cases, but branching when our workarounds fail is simpler to maintain. PiperOrigin-RevId: 277048670 -- e91003fa3ee6026d9b80624a23fc144fa5d74810 by Chris Kennelly <ckennelly@google.com>: Fix -Wimplicit-int-float-conversion warning in latest clang PiperOrigin-RevId: 276771618 -- 53087ca6603e86ad815f3dd9ab795cc0f79792c1 by Andy Soffer <asoffer@google.com>: Add documentation on absl::SNPrintF. PiperOrigin-RevId: 276694032 -- a9426af8cbd4c3a8f9053e7446c821852e41ff61 by Jorg Brown <jorg@google.com>: Stop including kern/OSByteOrder.h in order to support __APPLE__ Apple compiles with clang now anyway, and clang has support for the built-in compiler swap functions that are much faster than any function call to the OS. PiperOrigin-RevId: 276625231 -- df974be5aa5b4dc1b09c356cb8816edfc7867e63 by Jorg Brown <jorg@google.com>: Fix the build for Android x86-64 builds, where __SSE4_2__ is defined but _bswap64 is not. PiperOrigin-RevId: 276542642 -- d99dc092b3a5ad17643005e55f3b3cb6b9187ccc by Jorg Brown <jorg@google.com>: Remove a byteswap from the non-SSE path of FastHexToBufferZeroPad16 Remove the need for including absl/base/internal/endian.h from the SSE case (since if we have the Intel SSE intrinsics, then clearly we also have the Intel Byte-Swap intrinsics.) PiperOrigin-RevId: 276532608 -- d67b106dc930d8558810ae3983613bb2ab1e0d36 by Abseil Team <absl-team@google.com>: Use explicit static_cast<double> for int64_t to double conversion This uses an explicit static_cast<double>() in the int64_t to double comparisons in duration.cc's SafeAddRepHi. This satisfies clang's -Wimplicit-int-to-float-conversion warning (with https://reviews.llvm.org/D64666). This may also make it easier for humans to realize that the comparison is happening between two floating point double precision values. It should have no impact on the behavior or generated code. Tested: No behavior change PiperOrigin-RevId: 276529211 GitOrigin-RevId: 2f49cb9009386bc67bf54a2908c8720b749c8b7f Change-Id: I71e0781893ce219960b8290d54b20532779cb0ff
2019-10-30 16:00:44 +01:00
// ABSL_USES_STD_ANY
//
// Indicates whether absl::any is an alias for std::any.
#if !defined(ABSL_OPTION_USE_STD_ANY)
#error options.h is misconfigured.
#elif ABSL_OPTION_USE_STD_ANY == 0 || \
(ABSL_OPTION_USE_STD_ANY == 2 && !defined(ABSL_HAVE_STD_ANY))
#undef ABSL_USES_STD_ANY
#elif ABSL_OPTION_USE_STD_ANY == 1 || \
(ABSL_OPTION_USE_STD_ANY == 2 && defined(ABSL_HAVE_STD_ANY))
#define ABSL_USES_STD_ANY 1
#else
#error options.h is misconfigured.
#endif
// ABSL_USES_STD_OPTIONAL
//
// Indicates whether absl::optional is an alias for std::optional.
#if !defined(ABSL_OPTION_USE_STD_OPTIONAL)
#error options.h is misconfigured.
#elif ABSL_OPTION_USE_STD_OPTIONAL == 0 || \
(ABSL_OPTION_USE_STD_OPTIONAL == 2 && !defined(ABSL_HAVE_STD_OPTIONAL))
#undef ABSL_USES_STD_OPTIONAL
#elif ABSL_OPTION_USE_STD_OPTIONAL == 1 || \
(ABSL_OPTION_USE_STD_OPTIONAL == 2 && defined(ABSL_HAVE_STD_OPTIONAL))
#define ABSL_USES_STD_OPTIONAL 1
#else
#error options.h is misconfigured.
#endif
// ABSL_USES_STD_VARIANT
//
// Indicates whether absl::variant is an alias for std::variant.
#if !defined(ABSL_OPTION_USE_STD_VARIANT)
#error options.h is misconfigured.
#elif ABSL_OPTION_USE_STD_VARIANT == 0 || \
(ABSL_OPTION_USE_STD_VARIANT == 2 && !defined(ABSL_HAVE_STD_VARIANT))
#undef ABSL_USES_STD_VARIANT
#elif ABSL_OPTION_USE_STD_VARIANT == 1 || \
(ABSL_OPTION_USE_STD_VARIANT == 2 && defined(ABSL_HAVE_STD_VARIANT))
#define ABSL_USES_STD_VARIANT 1
#else
#error options.h is misconfigured.
#endif
// ABSL_USES_STD_STRING_VIEW
//
// Indicates whether absl::string_view is an alias for std::string_view.
#if !defined(ABSL_OPTION_USE_STD_STRING_VIEW)
#error options.h is misconfigured.
#elif ABSL_OPTION_USE_STD_STRING_VIEW == 0 || \
(ABSL_OPTION_USE_STD_STRING_VIEW == 2 && \
!defined(ABSL_HAVE_STD_STRING_VIEW))
#undef ABSL_USES_STD_STRING_VIEW
#elif ABSL_OPTION_USE_STD_STRING_VIEW == 1 || \
(ABSL_OPTION_USE_STD_STRING_VIEW == 2 && \
defined(ABSL_HAVE_STD_STRING_VIEW))
#define ABSL_USES_STD_STRING_VIEW 1
#else
#error options.h is misconfigured.
#endif
Export of internal Abseil changes. -- 07575526242a8e1275ac4223a3d2822795f46569 by CJ Johnson <johnsoncj@google.com>: Comment cleanup on InlinedVector PiperOrigin-RevId: 221322176 -- 49a5e643f85e34d53c41f5e6cc33357c55c9115d by Matt Kulukundis <kfm@google.com>: Internal cleanup PiperOrigin-RevId: 221309185 -- bb35be87ec9c74244b7d902e7e7d2d33ab139d76 by Abseil Team <absl-team@google.com>: Fix typo in comment. PiperOrigin-RevId: 221145354 -- afd4d7c106919708004e06aeea068a57c28aec44 by Derek Mauro <dmauro@google.com>: Update the debugging log message in CallOnceImpl() PiperOrigin-RevId: 221103254 -- 0b9dace8b88113777bf26a6d38f9bc0bcaf053a1 by Abseil Team <absl-team@google.com>: Workaround an MSVC 2015 bug in compile-time initialization. PiperOrigin-RevId: 220871483 -- ea0a3854511ed26beab827e5a5113766b334db86 by Marek Gilbert <mcg@google.com>: Fix ABSL_HAVE_THREAD_LOCAL when compiling for iOS 8 with Xcode 10. Xcode 10 has moved the check for thread_local to a link time, so clang reports __has_feature(cxx_thread_local) but then linking fails with messages like this: ld: targeted OS version does not support use of thread local variables PiperOrigin-RevId: 220815885 -- 485b6876c158c3dcf37eb32d7e512242d5d4ecc6 by Greg Falcon <gfalcon@google.com>: Make the absl::c_set_xxxx() algorithms refuse to compile when passed an unordered collection from std:: or absl::. These algorithms operate on sorted sequences; passing an unordered container to them is nearly certainly a bug. This change is technically an API break, but it only breaks incorrect code. We could try to be more clever and detect unordered collections from other libraries, but false positives will break legal code, and this would constitute an API break Abseil cannot afford. PiperOrigin-RevId: 220794190 -- c47cff7f9cc70a4c1604eee0131af552f40e46d6 by Jon Cohen <cohenjon@google.com>: MSVC 2017's STL throws a Structured Exception (not a C++ exception, essentially equivalent to SIGSEGV) when variant::emplace calls a throwing constructor when using the debug multithreaded MSVC runtime DLL. This manifests in dbg mode in Bazel builds. Disable tests which trigger this bug. It's impossible to specifically pull out MSVC 2017 -dbg modes because there's no way for Bazel to know when version of MSVC is being used -- you tell Bazel the directory where the MSVC tools live, not which version of MSVC tools to use. Thus the best we can do is switch on _DEBUG, which is set whenever the debug runtime is selected with the /MDd build flag, as in Bazel -dbg modes. See https://msdn.microsoft.com/en-us/library/b0084kay.aspx ctrl-f "_DEBUG" PiperOrigin-RevId: 220706161 -- 43993d4af309d92f4ebff38391dcc245f154ecc7 by Shaindel Schwartz <shaindel@google.com>: Internal change PiperOrigin-RevId: 220688429 -- 2448802972dcc261af153af464f2b022ef54a2a9 by Abseil Team <absl-team@google.com>: Speed up operator* for uint128 in WIN64. PiperOrigin-RevId: 220678790 -- 7b376403dd05ba10152fb52e40b29d8af79b58bb by Abseil Team <absl-team@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220654834 -- ae08af58111c3f838b8d4de25f501c3559c86002 by Abseil Team <absl-team@google.com>: CMake: Add absl_cc_test function PiperOrigin-RevId: 220603940 GitOrigin-RevId: 07575526242a8e1275ac4223a3d2822795f46569 Change-Id: Iba7f53eb394c8a9de564582a976793f9bb0596d9
2018-11-13 22:22:00 +01:00
// In debug mode, MSVC 2017's std::variant throws a EXCEPTION_ACCESS_VIOLATION
// SEH exception from emplace for variant<SomeStruct> when constructing the
// struct can throw. This defeats some of variant_test and
// variant_exception_safety_test.
#if defined(_MSC_VER) && _MSC_VER >= 1700 && defined(_DEBUG)
#define ABSL_INTERNAL_MSVC_2017_DBG_MODE
#endif
Export of internal Abseil changes -- c42a234e2c186bf697ce8d77e85628601fa514a6 by Abseil Team <absl-team@google.com>: Enable the assertion in the iterator's operator++ PiperOrigin-RevId: 290134813 -- f8c53ba8e9c5bb16bbcc1e412a5c2519c912c83e by Abseil Team <absl-team@google.com>: Define operator== and operator!= for absl::{weak,strong}_equality and absl::{partial,weak,strong}_ordering types themselves. PiperOrigin-RevId: 290111564 -- 36bc574090cefad74a451719ce2761982647a51d by Tom Manshreck <shreck@google.com>: Specify Time library flag formats PiperOrigin-RevId: 289928010 -- 26dd40281add260baab2b60fec05dfb9c5304aaa by Mark Barolak <mbar@google.com>: Delete an extraneous forward declaration of absl::Cord. PiperOrigin-RevId: 289708481 -- e60aea7f33554ff66d7699bb70e7af1d26323f1d by Abseil Team <absl-team@google.com>: Release b-tree benchmarks. PiperOrigin-RevId: 289654429 -- 660aa83fa000d4bae072b2d1c790f81d0939bc7e by Greg Falcon <gfalcon@google.com>: Use https links. Import of https://github.com/abseil/abseil-cpp/pull/586 PiperOrigin-RevId: 289479559 -- 0611ea4482dcf23d6b0a0389fe041eeb9052449a by Derek Mauro <dmauro@google.com>: Removes the static initializer for LookupTables<absl::uint128>::kVmaxOverBase Uses template specialization to hard code the resulting array. Static initializers are problematic for a number of reasons. Not only are they responsible for the static initialization order fiasco, but they are in the critical path during program startup. For these reasons, the Google C++ style guide strongly discourages them (and forbids them when they are not trivially destructible), and Chromium even has a test forbidding them. https://google.github.io/styleguide/cppguide.html#Static_and_Global_Variables https://chromium.googlesource.com/chromium/src.git/+/master/docs/static_initializers.md http://neugierig.org/software/chromium/notes/2011/08/static-initializers.html PiperOrigin-RevId: 289458677 -- c869362f6bb7a872314f74750d38d81bdaa73f95 by Greg Falcon <gfalcon@google.com>: Step 2 of 2 to fix our CCTZ fork to respect inline namespaces. Re-import of CCTZ from GitHub, applying new changes to honor Abseil's optional inline namespace in MSVC. PiperOrigin-RevId: 289454407 -- fdb3474d76c2ee0371ccdf7593a78137c03a3f58 by Greg Falcon <gfalcon@google.com>: Step 1 of 2 to fix our CCTZ fork to respect inline namespaces. CCTZ uses a linker flag to simulate weak symbol support in MSVC. This takes the form of a #pragma that includes the mangled names of two types: the symbol to treat as weak, and the symbol to use as its default value if no override is provided. When Abseil is configured to use inline namespaces, the mangled names of these symbols change, and the pragma should change to reflect that. Fortunately for us, MSVC name mangling is simple enough that we can generate the needed string literals in the preprocessor. This CL introduces the new macros; the uses will be introduced in a follow-up CL. PiperOrigin-RevId: 289435599 -- 5f152cc36f008acb9ab78f30b5efa40ebaf2754b by Matt Kulukundis <kfm@google.com>: Improve documentation for lazy_emplace PiperOrigin-RevId: 289333112 GitOrigin-RevId: c42a234e2c186bf697ce8d77e85628601fa514a6 Change-Id: I139ce6c7044a70d083af53e428bcb987f0fd88c6
2020-01-16 22:38:39 +01:00
// ABSL_INTERNAL_MANGLED_NS
// ABSL_INTERNAL_MANGLED_BACKREFERENCE
//
// Internal macros for building up mangled names in our internal fork of CCTZ.
// This implementation detail is only needed and provided for the MSVC build.
//
// These macros both expand to string literals. ABSL_INTERNAL_MANGLED_NS is
// the mangled spelling of the `absl` namespace, and
// ABSL_INTERNAL_MANGLED_BACKREFERENCE is a back-reference integer representing
// the proper count to skip past the CCTZ fork namespace names. (This number
// is one larger when there is an inline namespace name to skip.)
#if defined(_MSC_VER)
#if ABSL_OPTION_USE_INLINE_NAMESPACE == 0
#define ABSL_INTERNAL_MANGLED_NS "absl"
#define ABSL_INTERNAL_MANGLED_BACKREFERENCE "5"
#else
#define ABSL_INTERNAL_MANGLED_NS \
ABSL_INTERNAL_TOKEN_STR(ABSL_OPTION_INLINE_NAMESPACE_NAME) "@absl"
#define ABSL_INTERNAL_MANGLED_BACKREFERENCE "6"
#endif
#endif
#undef ABSL_INTERNAL_HAS_KEYWORD
Export of internal Abseil changes -- 8bdb2020150ed0fd4a4e520e454dc5f54e33f776 by Eric Fiselier <ericwf@google.com>: Workaround bug in GCC 9.2 and after. PiperOrigin-RevId: 291982551 -- 47ff4820e595f96c082a90d733725f6882d83e3b by Abseil Team <absl-team@google.com>: Improve ABSL_ATTRIBUTE_PACKED documentation Recommend to apply ABSL_ATTRIBUTE_PACKED to structure members instead of to an entire structure because applying this attribute to an entire structure may cause the compiler to generate suboptimal code. It reduces the alignment of the data structure from a value larger than one to one. When applied to a structure, ABSL_ATTRIBUTE_PACKED reduces the alignment of a structure (alignof()) to 1. As a result, the compiler can no longer assume that e.g. uint32 members are aligned on a four byte boundary and hence is forced to use single-byte load and store instructions on CPU architectures that do not support non-aligned loads or stores. PiperOrigin-RevId: 291977920 -- 902b7a86f860da699d3a2e5c738be5ef73ede3b4 by Mark Barolak <mbar@google.com>: Internal change PiperOrigin-RevId: 291963048 -- bb3bd3247e376d53a3080b105f13ec7566d3ae50 by Abseil Team <absl-team@google.com>: Support the C++17 insert_or_assign() API in btree_map. PiperOrigin-RevId: 291945474 -- ff3b3cfcbbc64f086f95501f48d49426bcde356f by Gennadiy Rozental <rogeeff@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 291861110 -- fd465cd9cbbacd3962f67a7346d6462edaddd809 by Derek Mauro <dmauro@google.com>: Add flaky=1 to beta_distribution_test. PiperOrigin-RevId: 291757364 -- 3603adfb59c4128c542b670952cce250d59e1f67 by Derek Mauro <dmauro@google.com>: Separate the initialization of NumCPUs() and NominalCPUFrequency() The OSS version of Abseil never needs to call NominalCPUFrequency(). In some configurations, initializing NominalCPUFrequency() requires spending at least 3ms measuring the CPU frequency. By separating the initialization from NumCPUs(), which is called in most configurations, we can save at least 3ms of program startup time. PiperOrigin-RevId: 291737273 -- bea9e4a6bff5a0351d340deab966641867e08c4d by Abseil Team <absl-team@google.com>: Change the cmake library names not to have a redundant `absl_` prefix. PiperOrigin-RevId: 291640501 -- 501b602ef260cd7c8c527342581ceffb3c5b6d4c by Gennadiy Rozental <rogeeff@google.com>: Introducing benchmark for absl::GetFlag. PiperOrigin-RevId: 291433394 -- 4eeaddc788da4b91c272a8adca77ca6dbbbc1d44 by Xiaoyi Zhang <zhangxy@google.com>: fix: Add support for more ARM processors detection Import of https://github.com/abseil/abseil-cpp/pull/608 PiperOrigin-RevId: 291420397 -- a3087a8e883c5d71de7d9bd4ec8f4db5142dfcf5 by Derek Mauro <dmauro@google.com>: Removes the flaky raw_hash_set prefetch test PiperOrigin-RevId: 291197079 -- aad6c2121c102ac36216e771c83227cf3e3bfd66 by Andy Soffer <asoffer@google.com>: Enable building Abseil as a DLL. This is currently experimental and unsupported. This CL does a few things: 1. Adds the ABSL_DLL macro to any class holding a static data member, or to global constants in headers. 2. Adds a whitelist of all files in the DLL and all the build targets that are conglomerated into the DLL. 3. When BUILD_SHARED_LIBS is specified, any build target that would be in the DLL still exists, but we swap out all of it's dependencies so it just depends on abseil_dll PiperOrigin-RevId: 291192055 -- 5e888cd6f2a7722805d41f872108a03a84e421c7 by Mark Barolak <mbar@google.com>: Move absl/strings/internal/escaping.{cc,h} into internal build targets. This puts absl/strings/internal/escaping.h behind a whitelist and it also resolves https://github.com/abseil/abseil-cpp/issues/604. PiperOrigin-RevId: 291173320 -- 166836d24970da87587c1728036f53f05a28f0af by Eric Fiselier <ericwf@google.com>: Internal Change. PiperOrigin-RevId: 291012718 -- 996ddb3dffda02440fa93f30ca5d71b14b688875 by Abseil Team <absl-team@google.com>: Fix shared libraries log spam for built-in types in absl::GetFlag PiperOrigin-RevId: 290772743 GitOrigin-RevId: 8bdb2020150ed0fd4a4e520e454dc5f54e33f776 Change-Id: I8bf2265dd14ebbace220a1b6b982bb5040ad2a26
2020-01-28 20:50:11 +01:00
// ABSL_DLL
//
// When building Abseil as a DLL, this macro expands to `__declspec(dllexport)`
// so we can annotate symbols appropriately as being exported. When used in
// headers consuming a DLL, this macro expands to `__declspec(dllimport)` so
// that consumers know the symbol is defined inside the DLL. In all other cases,
// the macro expands to nothing.
#if defined(_MSC_VER)
#if defined(ABSL_BUILD_DLL)
#define ABSL_DLL __declspec(dllexport)
#elif defined(ABSL_CONSUME_DLL)
#define ABSL_DLL __declspec(dllimport)
#else
#define ABSL_DLL
#endif
#else
#define ABSL_DLL
#endif // defined(_MSC_VER)
2017-09-19 22:54:40 +02:00
#endif // ABSL_BASE_CONFIG_H_