Abseil Common Libraries (C++) (grcp 依赖) https://abseil.io/
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

191 lines
7.1 KiB

Export of internal Abseil changes -- f012012ef78234a6a4585321b67d7b7c92ebc266 by Laramie Leavitt <lar@google.com>: Slight restructuring of absl/random/internal randen implementation. Convert round-keys.inc into randen_round_keys.cc file. Consistently use a 128-bit pointer type for internal method parameters. This allows simpler pointer arithmetic in C++ & permits removal of some constants and casts. Remove some redundancy in comments & constexpr variables. Specifically, all references to Randen algorithm parameters use RandenTraits; duplication in RandenSlow removed. PiperOrigin-RevId: 312190313 -- dc8b42e054046741e9ed65335bfdface997c6063 by Abseil Team <absl-team@google.com>: Internal change. PiperOrigin-RevId: 312167304 -- f13d248fafaf206492c1362c3574031aea3abaf7 by Matthew Brown <matthewbr@google.com>: Cleanup StrFormat extensions a little. PiperOrigin-RevId: 312166336 -- 9d9117589667afe2332bb7ad42bc967ca7c54502 by Derek Mauro <dmauro@google.com>: Internal change PiperOrigin-RevId: 312105213 -- 9a12b9b3aa0e59b8ee6cf9408ed0029045543a9b by Abseil Team <absl-team@google.com>: Complete IGNORE_TYPE macro renaming. PiperOrigin-RevId: 311999699 -- 64756f20d61021d999bd0d4c15e9ad3857382f57 by Gennadiy Rozental <rogeeff@google.com>: Switch to fixed bytes specific default value. This fixes the Abseil Flags for big endian platforms. PiperOrigin-RevId: 311844448 -- bdbe6b5b29791dbc3816ada1828458b3010ff1e9 by Laramie Leavitt <lar@google.com>: Change many distribution tests to use pcg_engine as a deterministic source of entropy. It's reasonable to test that the BitGen itself has good entropy, however when testing the cross product of all random distributions x all the architecture variations x all submitted changes results in a large number of tests. In order to account for these failures while still using good entropy requires that our allowed sigma need to account for all of these independent tests. Our current sigma values are too restrictive, and we see a lot of failures, so we have to either relax the sigma values or convert some of the statistical tests to use deterministic values. This changelist does the latter. PiperOrigin-RevId: 311840096 GitOrigin-RevId: f012012ef78234a6a4585321b67d7b7c92ebc266 Change-Id: Ic84886f38ff30d7d72c126e9b63c9a61eb729a1a
5 years ago
//
// 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
//
// 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: casts.h
// -----------------------------------------------------------------------------
//
// This header file defines casting templates to fit use cases not covered by
// the standard casts provided in the C++ standard. As with all cast operations,
// use these with caution and only if alternatives do not exist.
#ifndef ABSL_BASE_CASTS_H_
#define ABSL_BASE_CASTS_H_
#include <cstring>
#include <memory>
#include <type_traits>
#include <utility>
#include "absl/base/internal/identity.h"
#include "absl/base/macros.h"
#include "absl/meta/type_traits.h"
namespace absl {
ABSL_NAMESPACE_BEGIN
namespace internal_casts {
template <class Dest, class Source>
struct is_bitcastable
: std::integral_constant<
bool,
sizeof(Dest) == sizeof(Source) &&
type_traits_internal::is_trivially_copyable<Source>::value &&
type_traits_internal::is_trivially_copyable<Dest>::value &&
std::is_default_constructible<Dest>::value> {};
} // namespace internal_casts
// implicit_cast()
//
// Performs an implicit conversion between types following the language
// rules for implicit conversion; if an implicit conversion is otherwise
// allowed by the language in the given context, this function performs such an
// implicit conversion.
//
// Example:
//
// // If the context allows implicit conversion:
// From from;
// To to = from;
//
// // Such code can be replaced by:
// implicit_cast<To>(from);
//
// An `implicit_cast()` may also be used to annotate numeric type conversions
// that, although safe, may produce compiler warnings (such as `long` to `int`).
// Additionally, an `implicit_cast()` is also useful within return statements to
// indicate a specific implicit conversion is being undertaken.
//
// Example:
//
// return implicit_cast<double>(size_in_bytes) / capacity_;
//
// Annotating code with `implicit_cast()` allows you to explicitly select
// particular overloads and template instantiations, while providing a safer
// cast than `reinterpret_cast()` or `static_cast()`.
//
// Additionally, an `implicit_cast()` can be used to allow upcasting within a
// type hierarchy where incorrect use of `static_cast()` could accidentally
// allow downcasting.
//
// Finally, an `implicit_cast()` can be used to perform implicit conversions
// from unrelated types that otherwise couldn't be implicitly cast directly;
// C++ will normally only implicitly cast "one step" in such conversions.
//
// That is, if C is a type which can be implicitly converted to B, with B being
// a type that can be implicitly converted to A, an `implicit_cast()` can be
// used to convert C to B (which the compiler can then implicitly convert to A
// using language rules).
//
// Example:
//
// // Assume an object C is convertible to B, which is implicitly convertible
// // to A
// A a = implicit_cast<B>(C);
//
// Such implicit cast chaining may be useful within template logic.
template <typename To>
constexpr To implicit_cast(typename absl::internal::identity_t<To> to) {
return to;
}
// bit_cast()
//
Export of internal Abseil changes -- 0db7f4046f9b59c0f8c3df2f0eb7fd88fc328439 by Abseil Team <absl-team@google.com>: Revise documentation of bit_cast: * Removes inappropriate examples (round-tripping pointers, serialization), for which reinterpret_cast is more appropriate. * Removes mention of "bit representation", which is not an explicit notion in C++. The best we get is "byte representation". * Removes a circular defition of "bitcast" as itself, and instead explains what it does. * Removes the mathism "for some values of", which is probably not totally accessible to a general audience, and in any case needless verbiage. * Fixes comments in the example. * Replaces some colloquialisms with simpler, more direct language. PiperOrigin-RevId: 421791786 -- e04e64df55d93c1b9a09c0483b97cc4d8763260d by Derek Mauro <dmauro@google.com>: Update Docker image to use GCC 11.2, Clang 14 (prerelease), CMake 3.22.1, and Bazel 4.2.2 PiperOrigin-RevId: 421658559 -- d002bb3dc5cd1fc5b4cbd79a450efc894caa567c by Chris Kennelly <ckennelly@google.com>: Add a small microbenchmark for absl::bit_width. PiperOrigin-RevId: 421604852 -- 131b057d1b76ecd7170421b48d661bb958ff676b by Evan Brown <ezb@google.com>: Adds a disabled test for EBO in nested `CompressedTuple`s. PiperOrigin-RevId: 421413134 -- e34c7876d3a1212d90c73c030ccae6169b682d43 by Jorg Brown <jorg@google.com>: Show users a better error message if they pass a pointer to absl::Uniform. PiperOrigin-RevId: 421090472 GitOrigin-RevId: 0db7f4046f9b59c0f8c3df2f0eb7fd88fc328439 Change-Id: I5a004e8d17e974fa4897a09d1466ae8fc65dfdbb
3 years ago
// Creates a value of the new type `Dest` whose representation is the same as
// that of the argument, which is of (deduced) type `Source` (a "bitwise cast";
// every bit in the value representation of the result is equal to the
// corresponding bit in the object representation of the source). Source and
// destination types must be of the same size, and both types must be trivially
// copyable.
//
// As with most casts, use with caution. A `bit_cast()` might be needed when you
// need to treat a value as the value of some other type, for example, to access
// the individual bits of an object which are not normally accessible through
// the object's type, such as for working with the binary representation of a
// floating point value:
Export of internal Abseil changes -- f012012ef78234a6a4585321b67d7b7c92ebc266 by Laramie Leavitt <lar@google.com>: Slight restructuring of absl/random/internal randen implementation. Convert round-keys.inc into randen_round_keys.cc file. Consistently use a 128-bit pointer type for internal method parameters. This allows simpler pointer arithmetic in C++ & permits removal of some constants and casts. Remove some redundancy in comments & constexpr variables. Specifically, all references to Randen algorithm parameters use RandenTraits; duplication in RandenSlow removed. PiperOrigin-RevId: 312190313 -- dc8b42e054046741e9ed65335bfdface997c6063 by Abseil Team <absl-team@google.com>: Internal change. PiperOrigin-RevId: 312167304 -- f13d248fafaf206492c1362c3574031aea3abaf7 by Matthew Brown <matthewbr@google.com>: Cleanup StrFormat extensions a little. PiperOrigin-RevId: 312166336 -- 9d9117589667afe2332bb7ad42bc967ca7c54502 by Derek Mauro <dmauro@google.com>: Internal change PiperOrigin-RevId: 312105213 -- 9a12b9b3aa0e59b8ee6cf9408ed0029045543a9b by Abseil Team <absl-team@google.com>: Complete IGNORE_TYPE macro renaming. PiperOrigin-RevId: 311999699 -- 64756f20d61021d999bd0d4c15e9ad3857382f57 by Gennadiy Rozental <rogeeff@google.com>: Switch to fixed bytes specific default value. This fixes the Abseil Flags for big endian platforms. PiperOrigin-RevId: 311844448 -- bdbe6b5b29791dbc3816ada1828458b3010ff1e9 by Laramie Leavitt <lar@google.com>: Change many distribution tests to use pcg_engine as a deterministic source of entropy. It's reasonable to test that the BitGen itself has good entropy, however when testing the cross product of all random distributions x all the architecture variations x all submitted changes results in a large number of tests. In order to account for these failures while still using good entropy requires that our allowed sigma need to account for all of these independent tests. Our current sigma values are too restrictive, and we see a lot of failures, so we have to either relax the sigma values or convert some of the statistical tests to use deterministic values. This changelist does the latter. PiperOrigin-RevId: 311840096 GitOrigin-RevId: f012012ef78234a6a4585321b67d7b7c92ebc266 Change-Id: Ic84886f38ff30d7d72c126e9b63c9a61eb729a1a
5 years ago
//
// float f = 3.14159265358979;
// int i = bit_cast<int32_t>(f);
// // i = 0x40490fdb
//
Export of internal Abseil changes -- 0db7f4046f9b59c0f8c3df2f0eb7fd88fc328439 by Abseil Team <absl-team@google.com>: Revise documentation of bit_cast: * Removes inappropriate examples (round-tripping pointers, serialization), for which reinterpret_cast is more appropriate. * Removes mention of "bit representation", which is not an explicit notion in C++. The best we get is "byte representation". * Removes a circular defition of "bitcast" as itself, and instead explains what it does. * Removes the mathism "for some values of", which is probably not totally accessible to a general audience, and in any case needless verbiage. * Fixes comments in the example. * Replaces some colloquialisms with simpler, more direct language. PiperOrigin-RevId: 421791786 -- e04e64df55d93c1b9a09c0483b97cc4d8763260d by Derek Mauro <dmauro@google.com>: Update Docker image to use GCC 11.2, Clang 14 (prerelease), CMake 3.22.1, and Bazel 4.2.2 PiperOrigin-RevId: 421658559 -- d002bb3dc5cd1fc5b4cbd79a450efc894caa567c by Chris Kennelly <ckennelly@google.com>: Add a small microbenchmark for absl::bit_width. PiperOrigin-RevId: 421604852 -- 131b057d1b76ecd7170421b48d661bb958ff676b by Evan Brown <ezb@google.com>: Adds a disabled test for EBO in nested `CompressedTuple`s. PiperOrigin-RevId: 421413134 -- e34c7876d3a1212d90c73c030ccae6169b682d43 by Jorg Brown <jorg@google.com>: Show users a better error message if they pass a pointer to absl::Uniform. PiperOrigin-RevId: 421090472 GitOrigin-RevId: 0db7f4046f9b59c0f8c3df2f0eb7fd88fc328439 Change-Id: I5a004e8d17e974fa4897a09d1466ae8fc65dfdbb
3 years ago
// Reinterpreting and accessing a value directly as a different type (as shown
// below) usually results in undefined behavior.
Export of internal Abseil changes -- f012012ef78234a6a4585321b67d7b7c92ebc266 by Laramie Leavitt <lar@google.com>: Slight restructuring of absl/random/internal randen implementation. Convert round-keys.inc into randen_round_keys.cc file. Consistently use a 128-bit pointer type for internal method parameters. This allows simpler pointer arithmetic in C++ & permits removal of some constants and casts. Remove some redundancy in comments & constexpr variables. Specifically, all references to Randen algorithm parameters use RandenTraits; duplication in RandenSlow removed. PiperOrigin-RevId: 312190313 -- dc8b42e054046741e9ed65335bfdface997c6063 by Abseil Team <absl-team@google.com>: Internal change. PiperOrigin-RevId: 312167304 -- f13d248fafaf206492c1362c3574031aea3abaf7 by Matthew Brown <matthewbr@google.com>: Cleanup StrFormat extensions a little. PiperOrigin-RevId: 312166336 -- 9d9117589667afe2332bb7ad42bc967ca7c54502 by Derek Mauro <dmauro@google.com>: Internal change PiperOrigin-RevId: 312105213 -- 9a12b9b3aa0e59b8ee6cf9408ed0029045543a9b by Abseil Team <absl-team@google.com>: Complete IGNORE_TYPE macro renaming. PiperOrigin-RevId: 311999699 -- 64756f20d61021d999bd0d4c15e9ad3857382f57 by Gennadiy Rozental <rogeeff@google.com>: Switch to fixed bytes specific default value. This fixes the Abseil Flags for big endian platforms. PiperOrigin-RevId: 311844448 -- bdbe6b5b29791dbc3816ada1828458b3010ff1e9 by Laramie Leavitt <lar@google.com>: Change many distribution tests to use pcg_engine as a deterministic source of entropy. It's reasonable to test that the BitGen itself has good entropy, however when testing the cross product of all random distributions x all the architecture variations x all submitted changes results in a large number of tests. In order to account for these failures while still using good entropy requires that our allowed sigma need to account for all of these independent tests. Our current sigma values are too restrictive, and we see a lot of failures, so we have to either relax the sigma values or convert some of the statistical tests to use deterministic values. This changelist does the latter. PiperOrigin-RevId: 311840096 GitOrigin-RevId: f012012ef78234a6a4585321b67d7b7c92ebc266 Change-Id: Ic84886f38ff30d7d72c126e9b63c9a61eb729a1a
5 years ago
//
// Example:
//
// // WRONG
Export of internal Abseil changes -- 0db7f4046f9b59c0f8c3df2f0eb7fd88fc328439 by Abseil Team <absl-team@google.com>: Revise documentation of bit_cast: * Removes inappropriate examples (round-tripping pointers, serialization), for which reinterpret_cast is more appropriate. * Removes mention of "bit representation", which is not an explicit notion in C++. The best we get is "byte representation". * Removes a circular defition of "bitcast" as itself, and instead explains what it does. * Removes the mathism "for some values of", which is probably not totally accessible to a general audience, and in any case needless verbiage. * Fixes comments in the example. * Replaces some colloquialisms with simpler, more direct language. PiperOrigin-RevId: 421791786 -- e04e64df55d93c1b9a09c0483b97cc4d8763260d by Derek Mauro <dmauro@google.com>: Update Docker image to use GCC 11.2, Clang 14 (prerelease), CMake 3.22.1, and Bazel 4.2.2 PiperOrigin-RevId: 421658559 -- d002bb3dc5cd1fc5b4cbd79a450efc894caa567c by Chris Kennelly <ckennelly@google.com>: Add a small microbenchmark for absl::bit_width. PiperOrigin-RevId: 421604852 -- 131b057d1b76ecd7170421b48d661bb958ff676b by Evan Brown <ezb@google.com>: Adds a disabled test for EBO in nested `CompressedTuple`s. PiperOrigin-RevId: 421413134 -- e34c7876d3a1212d90c73c030ccae6169b682d43 by Jorg Brown <jorg@google.com>: Show users a better error message if they pass a pointer to absl::Uniform. PiperOrigin-RevId: 421090472 GitOrigin-RevId: 0db7f4046f9b59c0f8c3df2f0eb7fd88fc328439 Change-Id: I5a004e8d17e974fa4897a09d1466ae8fc65dfdbb
3 years ago
// float f = 3.14159265358979;
// int i = reinterpret_cast<int&>(f); // Wrong
// int j = *reinterpret_cast<int*>(&f); // Equally wrong
// int k = *bit_cast<int*>(&f); // Equally wrong
//
// Reinterpret-casting results in undefined behavior according to the ISO C++
// specification, section [basic.lval]. Roughly, this section says: if an object
// in memory has one type, and a program accesses it with a different type, the
// result is undefined behavior for most "different type".
Export of internal Abseil changes -- f012012ef78234a6a4585321b67d7b7c92ebc266 by Laramie Leavitt <lar@google.com>: Slight restructuring of absl/random/internal randen implementation. Convert round-keys.inc into randen_round_keys.cc file. Consistently use a 128-bit pointer type for internal method parameters. This allows simpler pointer arithmetic in C++ & permits removal of some constants and casts. Remove some redundancy in comments & constexpr variables. Specifically, all references to Randen algorithm parameters use RandenTraits; duplication in RandenSlow removed. PiperOrigin-RevId: 312190313 -- dc8b42e054046741e9ed65335bfdface997c6063 by Abseil Team <absl-team@google.com>: Internal change. PiperOrigin-RevId: 312167304 -- f13d248fafaf206492c1362c3574031aea3abaf7 by Matthew Brown <matthewbr@google.com>: Cleanup StrFormat extensions a little. PiperOrigin-RevId: 312166336 -- 9d9117589667afe2332bb7ad42bc967ca7c54502 by Derek Mauro <dmauro@google.com>: Internal change PiperOrigin-RevId: 312105213 -- 9a12b9b3aa0e59b8ee6cf9408ed0029045543a9b by Abseil Team <absl-team@google.com>: Complete IGNORE_TYPE macro renaming. PiperOrigin-RevId: 311999699 -- 64756f20d61021d999bd0d4c15e9ad3857382f57 by Gennadiy Rozental <rogeeff@google.com>: Switch to fixed bytes specific default value. This fixes the Abseil Flags for big endian platforms. PiperOrigin-RevId: 311844448 -- bdbe6b5b29791dbc3816ada1828458b3010ff1e9 by Laramie Leavitt <lar@google.com>: Change many distribution tests to use pcg_engine as a deterministic source of entropy. It's reasonable to test that the BitGen itself has good entropy, however when testing the cross product of all random distributions x all the architecture variations x all submitted changes results in a large number of tests. In order to account for these failures while still using good entropy requires that our allowed sigma need to account for all of these independent tests. Our current sigma values are too restrictive, and we see a lot of failures, so we have to either relax the sigma values or convert some of the statistical tests to use deterministic values. This changelist does the latter. PiperOrigin-RevId: 311840096 GitOrigin-RevId: f012012ef78234a6a4585321b67d7b7c92ebc266 Change-Id: Ic84886f38ff30d7d72c126e9b63c9a61eb729a1a
5 years ago
//
Export of internal Abseil changes -- 0db7f4046f9b59c0f8c3df2f0eb7fd88fc328439 by Abseil Team <absl-team@google.com>: Revise documentation of bit_cast: * Removes inappropriate examples (round-tripping pointers, serialization), for which reinterpret_cast is more appropriate. * Removes mention of "bit representation", which is not an explicit notion in C++. The best we get is "byte representation". * Removes a circular defition of "bitcast" as itself, and instead explains what it does. * Removes the mathism "for some values of", which is probably not totally accessible to a general audience, and in any case needless verbiage. * Fixes comments in the example. * Replaces some colloquialisms with simpler, more direct language. PiperOrigin-RevId: 421791786 -- e04e64df55d93c1b9a09c0483b97cc4d8763260d by Derek Mauro <dmauro@google.com>: Update Docker image to use GCC 11.2, Clang 14 (prerelease), CMake 3.22.1, and Bazel 4.2.2 PiperOrigin-RevId: 421658559 -- d002bb3dc5cd1fc5b4cbd79a450efc894caa567c by Chris Kennelly <ckennelly@google.com>: Add a small microbenchmark for absl::bit_width. PiperOrigin-RevId: 421604852 -- 131b057d1b76ecd7170421b48d661bb958ff676b by Evan Brown <ezb@google.com>: Adds a disabled test for EBO in nested `CompressedTuple`s. PiperOrigin-RevId: 421413134 -- e34c7876d3a1212d90c73c030ccae6169b682d43 by Jorg Brown <jorg@google.com>: Show users a better error message if they pass a pointer to absl::Uniform. PiperOrigin-RevId: 421090472 GitOrigin-RevId: 0db7f4046f9b59c0f8c3df2f0eb7fd88fc328439 Change-Id: I5a004e8d17e974fa4897a09d1466ae8fc65dfdbb
3 years ago
// Using bit_cast on a pointer and then dereferencing it is no better than using
// reinterpret_cast. You should only use bit_cast on the value itself.
Export of internal Abseil changes -- f012012ef78234a6a4585321b67d7b7c92ebc266 by Laramie Leavitt <lar@google.com>: Slight restructuring of absl/random/internal randen implementation. Convert round-keys.inc into randen_round_keys.cc file. Consistently use a 128-bit pointer type for internal method parameters. This allows simpler pointer arithmetic in C++ & permits removal of some constants and casts. Remove some redundancy in comments & constexpr variables. Specifically, all references to Randen algorithm parameters use RandenTraits; duplication in RandenSlow removed. PiperOrigin-RevId: 312190313 -- dc8b42e054046741e9ed65335bfdface997c6063 by Abseil Team <absl-team@google.com>: Internal change. PiperOrigin-RevId: 312167304 -- f13d248fafaf206492c1362c3574031aea3abaf7 by Matthew Brown <matthewbr@google.com>: Cleanup StrFormat extensions a little. PiperOrigin-RevId: 312166336 -- 9d9117589667afe2332bb7ad42bc967ca7c54502 by Derek Mauro <dmauro@google.com>: Internal change PiperOrigin-RevId: 312105213 -- 9a12b9b3aa0e59b8ee6cf9408ed0029045543a9b by Abseil Team <absl-team@google.com>: Complete IGNORE_TYPE macro renaming. PiperOrigin-RevId: 311999699 -- 64756f20d61021d999bd0d4c15e9ad3857382f57 by Gennadiy Rozental <rogeeff@google.com>: Switch to fixed bytes specific default value. This fixes the Abseil Flags for big endian platforms. PiperOrigin-RevId: 311844448 -- bdbe6b5b29791dbc3816ada1828458b3010ff1e9 by Laramie Leavitt <lar@google.com>: Change many distribution tests to use pcg_engine as a deterministic source of entropy. It's reasonable to test that the BitGen itself has good entropy, however when testing the cross product of all random distributions x all the architecture variations x all submitted changes results in a large number of tests. In order to account for these failures while still using good entropy requires that our allowed sigma need to account for all of these independent tests. Our current sigma values are too restrictive, and we see a lot of failures, so we have to either relax the sigma values or convert some of the statistical tests to use deterministic values. This changelist does the latter. PiperOrigin-RevId: 311840096 GitOrigin-RevId: f012012ef78234a6a4585321b67d7b7c92ebc266 Change-Id: Ic84886f38ff30d7d72c126e9b63c9a61eb729a1a
5 years ago
//
// Such casting results in type punning: holding an object in memory of one type
// and reading its bits back using a different type. A `bit_cast()` avoids this
Export of internal Abseil changes -- 0db7f4046f9b59c0f8c3df2f0eb7fd88fc328439 by Abseil Team <absl-team@google.com>: Revise documentation of bit_cast: * Removes inappropriate examples (round-tripping pointers, serialization), for which reinterpret_cast is more appropriate. * Removes mention of "bit representation", which is not an explicit notion in C++. The best we get is "byte representation". * Removes a circular defition of "bitcast" as itself, and instead explains what it does. * Removes the mathism "for some values of", which is probably not totally accessible to a general audience, and in any case needless verbiage. * Fixes comments in the example. * Replaces some colloquialisms with simpler, more direct language. PiperOrigin-RevId: 421791786 -- e04e64df55d93c1b9a09c0483b97cc4d8763260d by Derek Mauro <dmauro@google.com>: Update Docker image to use GCC 11.2, Clang 14 (prerelease), CMake 3.22.1, and Bazel 4.2.2 PiperOrigin-RevId: 421658559 -- d002bb3dc5cd1fc5b4cbd79a450efc894caa567c by Chris Kennelly <ckennelly@google.com>: Add a small microbenchmark for absl::bit_width. PiperOrigin-RevId: 421604852 -- 131b057d1b76ecd7170421b48d661bb958ff676b by Evan Brown <ezb@google.com>: Adds a disabled test for EBO in nested `CompressedTuple`s. PiperOrigin-RevId: 421413134 -- e34c7876d3a1212d90c73c030ccae6169b682d43 by Jorg Brown <jorg@google.com>: Show users a better error message if they pass a pointer to absl::Uniform. PiperOrigin-RevId: 421090472 GitOrigin-RevId: 0db7f4046f9b59c0f8c3df2f0eb7fd88fc328439 Change-Id: I5a004e8d17e974fa4897a09d1466ae8fc65dfdbb
3 years ago
// issue by copying the object representation to a new value, which avoids
// introducing this undefined behavior (since the original value is never
// accessed in the wrong way).
Export of internal Abseil changes -- f012012ef78234a6a4585321b67d7b7c92ebc266 by Laramie Leavitt <lar@google.com>: Slight restructuring of absl/random/internal randen implementation. Convert round-keys.inc into randen_round_keys.cc file. Consistently use a 128-bit pointer type for internal method parameters. This allows simpler pointer arithmetic in C++ & permits removal of some constants and casts. Remove some redundancy in comments & constexpr variables. Specifically, all references to Randen algorithm parameters use RandenTraits; duplication in RandenSlow removed. PiperOrigin-RevId: 312190313 -- dc8b42e054046741e9ed65335bfdface997c6063 by Abseil Team <absl-team@google.com>: Internal change. PiperOrigin-RevId: 312167304 -- f13d248fafaf206492c1362c3574031aea3abaf7 by Matthew Brown <matthewbr@google.com>: Cleanup StrFormat extensions a little. PiperOrigin-RevId: 312166336 -- 9d9117589667afe2332bb7ad42bc967ca7c54502 by Derek Mauro <dmauro@google.com>: Internal change PiperOrigin-RevId: 312105213 -- 9a12b9b3aa0e59b8ee6cf9408ed0029045543a9b by Abseil Team <absl-team@google.com>: Complete IGNORE_TYPE macro renaming. PiperOrigin-RevId: 311999699 -- 64756f20d61021d999bd0d4c15e9ad3857382f57 by Gennadiy Rozental <rogeeff@google.com>: Switch to fixed bytes specific default value. This fixes the Abseil Flags for big endian platforms. PiperOrigin-RevId: 311844448 -- bdbe6b5b29791dbc3816ada1828458b3010ff1e9 by Laramie Leavitt <lar@google.com>: Change many distribution tests to use pcg_engine as a deterministic source of entropy. It's reasonable to test that the BitGen itself has good entropy, however when testing the cross product of all random distributions x all the architecture variations x all submitted changes results in a large number of tests. In order to account for these failures while still using good entropy requires that our allowed sigma need to account for all of these independent tests. Our current sigma values are too restrictive, and we see a lot of failures, so we have to either relax the sigma values or convert some of the statistical tests to use deterministic values. This changelist does the latter. PiperOrigin-RevId: 311840096 GitOrigin-RevId: f012012ef78234a6a4585321b67d7b7c92ebc266 Change-Id: Ic84886f38ff30d7d72c126e9b63c9a61eb729a1a
5 years ago
//
Export of internal Abseil changes -- 0db7f4046f9b59c0f8c3df2f0eb7fd88fc328439 by Abseil Team <absl-team@google.com>: Revise documentation of bit_cast: * Removes inappropriate examples (round-tripping pointers, serialization), for which reinterpret_cast is more appropriate. * Removes mention of "bit representation", which is not an explicit notion in C++. The best we get is "byte representation". * Removes a circular defition of "bitcast" as itself, and instead explains what it does. * Removes the mathism "for some values of", which is probably not totally accessible to a general audience, and in any case needless verbiage. * Fixes comments in the example. * Replaces some colloquialisms with simpler, more direct language. PiperOrigin-RevId: 421791786 -- e04e64df55d93c1b9a09c0483b97cc4d8763260d by Derek Mauro <dmauro@google.com>: Update Docker image to use GCC 11.2, Clang 14 (prerelease), CMake 3.22.1, and Bazel 4.2.2 PiperOrigin-RevId: 421658559 -- d002bb3dc5cd1fc5b4cbd79a450efc894caa567c by Chris Kennelly <ckennelly@google.com>: Add a small microbenchmark for absl::bit_width. PiperOrigin-RevId: 421604852 -- 131b057d1b76ecd7170421b48d661bb958ff676b by Evan Brown <ezb@google.com>: Adds a disabled test for EBO in nested `CompressedTuple`s. PiperOrigin-RevId: 421413134 -- e34c7876d3a1212d90c73c030ccae6169b682d43 by Jorg Brown <jorg@google.com>: Show users a better error message if they pass a pointer to absl::Uniform. PiperOrigin-RevId: 421090472 GitOrigin-RevId: 0db7f4046f9b59c0f8c3df2f0eb7fd88fc328439 Change-Id: I5a004e8d17e974fa4897a09d1466ae8fc65dfdbb
3 years ago
// NOTE: The requirements here are stricter than the bit_cast of standard
// proposal P0476 due to the need for workarounds and lack of intrinsics.
Export of internal Abseil changes -- f012012ef78234a6a4585321b67d7b7c92ebc266 by Laramie Leavitt <lar@google.com>: Slight restructuring of absl/random/internal randen implementation. Convert round-keys.inc into randen_round_keys.cc file. Consistently use a 128-bit pointer type for internal method parameters. This allows simpler pointer arithmetic in C++ & permits removal of some constants and casts. Remove some redundancy in comments & constexpr variables. Specifically, all references to Randen algorithm parameters use RandenTraits; duplication in RandenSlow removed. PiperOrigin-RevId: 312190313 -- dc8b42e054046741e9ed65335bfdface997c6063 by Abseil Team <absl-team@google.com>: Internal change. PiperOrigin-RevId: 312167304 -- f13d248fafaf206492c1362c3574031aea3abaf7 by Matthew Brown <matthewbr@google.com>: Cleanup StrFormat extensions a little. PiperOrigin-RevId: 312166336 -- 9d9117589667afe2332bb7ad42bc967ca7c54502 by Derek Mauro <dmauro@google.com>: Internal change PiperOrigin-RevId: 312105213 -- 9a12b9b3aa0e59b8ee6cf9408ed0029045543a9b by Abseil Team <absl-team@google.com>: Complete IGNORE_TYPE macro renaming. PiperOrigin-RevId: 311999699 -- 64756f20d61021d999bd0d4c15e9ad3857382f57 by Gennadiy Rozental <rogeeff@google.com>: Switch to fixed bytes specific default value. This fixes the Abseil Flags for big endian platforms. PiperOrigin-RevId: 311844448 -- bdbe6b5b29791dbc3816ada1828458b3010ff1e9 by Laramie Leavitt <lar@google.com>: Change many distribution tests to use pcg_engine as a deterministic source of entropy. It's reasonable to test that the BitGen itself has good entropy, however when testing the cross product of all random distributions x all the architecture variations x all submitted changes results in a large number of tests. In order to account for these failures while still using good entropy requires that our allowed sigma need to account for all of these independent tests. Our current sigma values are too restrictive, and we see a lot of failures, so we have to either relax the sigma values or convert some of the statistical tests to use deterministic values. This changelist does the latter. PiperOrigin-RevId: 311840096 GitOrigin-RevId: f012012ef78234a6a4585321b67d7b7c92ebc266 Change-Id: Ic84886f38ff30d7d72c126e9b63c9a61eb729a1a
5 years ago
// Specifically, this implementation also requires `Dest` to be
// default-constructible.
template <
typename Dest, typename Source,
typename std::enable_if<internal_casts::is_bitcastable<Dest, Source>::value,
int>::type = 0>
inline Dest bit_cast(const Source& source) {
Dest dest;
memcpy(static_cast<void*>(std::addressof(dest)),
static_cast<const void*>(std::addressof(source)), sizeof(dest));
return dest;
}
// NOTE: This overload is only picked if the requirements of bit_cast are
// not met. It is therefore UB, but is provided temporarily as previous
// versions of this function template were unchecked. Do not use this in
// new code.
Export of internal Abseil changes -- f012012ef78234a6a4585321b67d7b7c92ebc266 by Laramie Leavitt <lar@google.com>: Slight restructuring of absl/random/internal randen implementation. Convert round-keys.inc into randen_round_keys.cc file. Consistently use a 128-bit pointer type for internal method parameters. This allows simpler pointer arithmetic in C++ & permits removal of some constants and casts. Remove some redundancy in comments & constexpr variables. Specifically, all references to Randen algorithm parameters use RandenTraits; duplication in RandenSlow removed. PiperOrigin-RevId: 312190313 -- dc8b42e054046741e9ed65335bfdface997c6063 by Abseil Team <absl-team@google.com>: Internal change. PiperOrigin-RevId: 312167304 -- f13d248fafaf206492c1362c3574031aea3abaf7 by Matthew Brown <matthewbr@google.com>: Cleanup StrFormat extensions a little. PiperOrigin-RevId: 312166336 -- 9d9117589667afe2332bb7ad42bc967ca7c54502 by Derek Mauro <dmauro@google.com>: Internal change PiperOrigin-RevId: 312105213 -- 9a12b9b3aa0e59b8ee6cf9408ed0029045543a9b by Abseil Team <absl-team@google.com>: Complete IGNORE_TYPE macro renaming. PiperOrigin-RevId: 311999699 -- 64756f20d61021d999bd0d4c15e9ad3857382f57 by Gennadiy Rozental <rogeeff@google.com>: Switch to fixed bytes specific default value. This fixes the Abseil Flags for big endian platforms. PiperOrigin-RevId: 311844448 -- bdbe6b5b29791dbc3816ada1828458b3010ff1e9 by Laramie Leavitt <lar@google.com>: Change many distribution tests to use pcg_engine as a deterministic source of entropy. It's reasonable to test that the BitGen itself has good entropy, however when testing the cross product of all random distributions x all the architecture variations x all submitted changes results in a large number of tests. In order to account for these failures while still using good entropy requires that our allowed sigma need to account for all of these independent tests. Our current sigma values are too restrictive, and we see a lot of failures, so we have to either relax the sigma values or convert some of the statistical tests to use deterministic values. This changelist does the latter. PiperOrigin-RevId: 311840096 GitOrigin-RevId: f012012ef78234a6a4585321b67d7b7c92ebc266 Change-Id: Ic84886f38ff30d7d72c126e9b63c9a61eb729a1a
5 years ago
template <
typename Dest, typename Source,
typename std::enable_if<
!internal_casts::is_bitcastable<Dest, Source>::value,
int>::type = 0>
Export of internal Abseil changes -- f012012ef78234a6a4585321b67d7b7c92ebc266 by Laramie Leavitt <lar@google.com>: Slight restructuring of absl/random/internal randen implementation. Convert round-keys.inc into randen_round_keys.cc file. Consistently use a 128-bit pointer type for internal method parameters. This allows simpler pointer arithmetic in C++ & permits removal of some constants and casts. Remove some redundancy in comments & constexpr variables. Specifically, all references to Randen algorithm parameters use RandenTraits; duplication in RandenSlow removed. PiperOrigin-RevId: 312190313 -- dc8b42e054046741e9ed65335bfdface997c6063 by Abseil Team <absl-team@google.com>: Internal change. PiperOrigin-RevId: 312167304 -- f13d248fafaf206492c1362c3574031aea3abaf7 by Matthew Brown <matthewbr@google.com>: Cleanup StrFormat extensions a little. PiperOrigin-RevId: 312166336 -- 9d9117589667afe2332bb7ad42bc967ca7c54502 by Derek Mauro <dmauro@google.com>: Internal change PiperOrigin-RevId: 312105213 -- 9a12b9b3aa0e59b8ee6cf9408ed0029045543a9b by Abseil Team <absl-team@google.com>: Complete IGNORE_TYPE macro renaming. PiperOrigin-RevId: 311999699 -- 64756f20d61021d999bd0d4c15e9ad3857382f57 by Gennadiy Rozental <rogeeff@google.com>: Switch to fixed bytes specific default value. This fixes the Abseil Flags for big endian platforms. PiperOrigin-RevId: 311844448 -- bdbe6b5b29791dbc3816ada1828458b3010ff1e9 by Laramie Leavitt <lar@google.com>: Change many distribution tests to use pcg_engine as a deterministic source of entropy. It's reasonable to test that the BitGen itself has good entropy, however when testing the cross product of all random distributions x all the architecture variations x all submitted changes results in a large number of tests. In order to account for these failures while still using good entropy requires that our allowed sigma need to account for all of these independent tests. Our current sigma values are too restrictive, and we see a lot of failures, so we have to either relax the sigma values or convert some of the statistical tests to use deterministic values. This changelist does the latter. PiperOrigin-RevId: 311840096 GitOrigin-RevId: f012012ef78234a6a4585321b67d7b7c92ebc266 Change-Id: Ic84886f38ff30d7d72c126e9b63c9a61eb729a1a
5 years ago
ABSL_DEPRECATED(
"absl::bit_cast type requirements were violated. Update the types "
"being used such that they are the same size and are both "
"TriviallyCopyable.")
Export of internal Abseil changes -- f012012ef78234a6a4585321b67d7b7c92ebc266 by Laramie Leavitt <lar@google.com>: Slight restructuring of absl/random/internal randen implementation. Convert round-keys.inc into randen_round_keys.cc file. Consistently use a 128-bit pointer type for internal method parameters. This allows simpler pointer arithmetic in C++ & permits removal of some constants and casts. Remove some redundancy in comments & constexpr variables. Specifically, all references to Randen algorithm parameters use RandenTraits; duplication in RandenSlow removed. PiperOrigin-RevId: 312190313 -- dc8b42e054046741e9ed65335bfdface997c6063 by Abseil Team <absl-team@google.com>: Internal change. PiperOrigin-RevId: 312167304 -- f13d248fafaf206492c1362c3574031aea3abaf7 by Matthew Brown <matthewbr@google.com>: Cleanup StrFormat extensions a little. PiperOrigin-RevId: 312166336 -- 9d9117589667afe2332bb7ad42bc967ca7c54502 by Derek Mauro <dmauro@google.com>: Internal change PiperOrigin-RevId: 312105213 -- 9a12b9b3aa0e59b8ee6cf9408ed0029045543a9b by Abseil Team <absl-team@google.com>: Complete IGNORE_TYPE macro renaming. PiperOrigin-RevId: 311999699 -- 64756f20d61021d999bd0d4c15e9ad3857382f57 by Gennadiy Rozental <rogeeff@google.com>: Switch to fixed bytes specific default value. This fixes the Abseil Flags for big endian platforms. PiperOrigin-RevId: 311844448 -- bdbe6b5b29791dbc3816ada1828458b3010ff1e9 by Laramie Leavitt <lar@google.com>: Change many distribution tests to use pcg_engine as a deterministic source of entropy. It's reasonable to test that the BitGen itself has good entropy, however when testing the cross product of all random distributions x all the architecture variations x all submitted changes results in a large number of tests. In order to account for these failures while still using good entropy requires that our allowed sigma need to account for all of these independent tests. Our current sigma values are too restrictive, and we see a lot of failures, so we have to either relax the sigma values or convert some of the statistical tests to use deterministic values. This changelist does the latter. PiperOrigin-RevId: 311840096 GitOrigin-RevId: f012012ef78234a6a4585321b67d7b7c92ebc266 Change-Id: Ic84886f38ff30d7d72c126e9b63c9a61eb729a1a
5 years ago
inline Dest bit_cast(const Source& source) {
static_assert(sizeof(Dest) == sizeof(Source),
"Source and destination types should have equal sizes.");
Dest dest;
memcpy(&dest, &source, sizeof(dest));
return dest;
}
ABSL_NAMESPACE_END
} // namespace absl
#endif // ABSL_BASE_CASTS_H_