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.

616 lines
20 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 2019 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.
#include "absl/flags/internal/flag.h"
Export of internal Abseil changes -- b2b94b9f533e4f9ae2a2df9de56ccb3b18f31d0d by Gennadiy Rozental <rogeeff@google.com>: Internal change PiperOrigin-RevId: 314366138 -- 7a5ac6be82741aec5b327d7b67efd14d69deed6a by Gennadiy Rozental <rogeeff@google.com>: Introduce Abseil prefixed dynamic annotation macros. PiperOrigin-RevId: 314228914 -- 33e7c605cb1be9bd48bd3590b1eedccd68d3bd1b by Gennadiy Rozental <rogeeff@google.com>: Fixing missing includes and remove unnecessary ones. PiperOrigin-RevId: 314217909 -- 6a3c5c26bfa13317bf0a880f13d0e4be0b971b76 by Gennadiy Rozental <rogeeff@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 314209576 -- f7795aa68020af4a6b4905531ba951e04b88966a by Abseil Team <absl-team@google.com>: absl::FormatTime() and absl::ParseTime() format specifiers: - %EZ now accepts 'z' in addition to 'Z' as a synonym for +00:00. - %ET is introduced, producing 'T' on output, and accepting 'T' or 't' on input. This is for the RFC3339 date-time separator. PiperOrigin-RevId: 313945137 -- 87c437ce3aab3f59a7546e44a28cd1c8aaa152c3 by Laramie Leavitt <lar@google.com>: Rollback PiperOrigin-RevId: 313868206 -- 8049b74349486a0026932b86d29c380b195e1cba by Laramie Leavitt <lar@google.com>: Remove the MockingBitGenBase base class in favor of type-erasure in BitGenRef. In Abseil random, mocking was split across two different classes, MockingBitGenBase and MockingBitGen. This split existed because Google Mock is a test-only library that we don't link into production, so MockingBitGenBase provided a low-overhead scaffold used to lookup mocks when in test code, but which is unused in production code. That has been replaced by type-erasure which looks for a method named CallImpl with the correct signature. Weaken the coupling between MockingBitGen, DistributionCaller, and MockOverloadSet. Rename CallImpl to InvokeMock() Previously, the implementation of DistributionCaller was also split across different files using explicit instantiation of the DistributionCaller struct and some details in the Mocking classes. Now Distribution caller uses the presence of the InvokeMock() method to choose whether to use the mockable call path or the default call path. PiperOrigin-RevId: 313848695 -- 1741d80e08050e1939605f70ca6ff64809785c85 by Gennadiy Rozental <rogeeff@google.com>: Introduce public interface to access reflection handle corresponding the flag. This interface will be the only official way to access flag reflection information from the flag object. The other internal methods will eventually disappear. PiperOrigin-RevId: 313734006 -- c375bead457de29d9c29595d16c66d3e5125b585 by Abseil Team <absl-team@google.com>: Improve documentation of absl::c_partial_sort_copy. This function takes no middle parameter, instead using the size of the result container to determine the size of the partial sort. PiperOrigin-RevId: 313656062 -- bbc759d43656b1b996ad558f23c852a9f14129d2 by Gennadiy Rozental <rogeeff@google.com>: Eliminate dynamic annotation symbols in the library. PiperOrigin-RevId: 313650817 GitOrigin-RevId: b2b94b9f533e4f9ae2a2df9de56ccb3b18f31d0d Change-Id: Ic7a11bbcb723f3ff6a7e2f214bff0a92c6f8ab4d
5 years ago
#include <assert.h>
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
#include <stddef.h>
#include <stdint.h>
#include <string.h>
Export of internal Abseil changes -- b2b94b9f533e4f9ae2a2df9de56ccb3b18f31d0d by Gennadiy Rozental <rogeeff@google.com>: Internal change PiperOrigin-RevId: 314366138 -- 7a5ac6be82741aec5b327d7b67efd14d69deed6a by Gennadiy Rozental <rogeeff@google.com>: Introduce Abseil prefixed dynamic annotation macros. PiperOrigin-RevId: 314228914 -- 33e7c605cb1be9bd48bd3590b1eedccd68d3bd1b by Gennadiy Rozental <rogeeff@google.com>: Fixing missing includes and remove unnecessary ones. PiperOrigin-RevId: 314217909 -- 6a3c5c26bfa13317bf0a880f13d0e4be0b971b76 by Gennadiy Rozental <rogeeff@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 314209576 -- f7795aa68020af4a6b4905531ba951e04b88966a by Abseil Team <absl-team@google.com>: absl::FormatTime() and absl::ParseTime() format specifiers: - %EZ now accepts 'z' in addition to 'Z' as a synonym for +00:00. - %ET is introduced, producing 'T' on output, and accepting 'T' or 't' on input. This is for the RFC3339 date-time separator. PiperOrigin-RevId: 313945137 -- 87c437ce3aab3f59a7546e44a28cd1c8aaa152c3 by Laramie Leavitt <lar@google.com>: Rollback PiperOrigin-RevId: 313868206 -- 8049b74349486a0026932b86d29c380b195e1cba by Laramie Leavitt <lar@google.com>: Remove the MockingBitGenBase base class in favor of type-erasure in BitGenRef. In Abseil random, mocking was split across two different classes, MockingBitGenBase and MockingBitGen. This split existed because Google Mock is a test-only library that we don't link into production, so MockingBitGenBase provided a low-overhead scaffold used to lookup mocks when in test code, but which is unused in production code. That has been replaced by type-erasure which looks for a method named CallImpl with the correct signature. Weaken the coupling between MockingBitGen, DistributionCaller, and MockOverloadSet. Rename CallImpl to InvokeMock() Previously, the implementation of DistributionCaller was also split across different files using explicit instantiation of the DistributionCaller struct and some details in the Mocking classes. Now Distribution caller uses the presence of the InvokeMock() method to choose whether to use the mockable call path or the default call path. PiperOrigin-RevId: 313848695 -- 1741d80e08050e1939605f70ca6ff64809785c85 by Gennadiy Rozental <rogeeff@google.com>: Introduce public interface to access reflection handle corresponding the flag. This interface will be the only official way to access flag reflection information from the flag object. The other internal methods will eventually disappear. PiperOrigin-RevId: 313734006 -- c375bead457de29d9c29595d16c66d3e5125b585 by Abseil Team <absl-team@google.com>: Improve documentation of absl::c_partial_sort_copy. This function takes no middle parameter, instead using the size of the result container to determine the size of the partial sort. PiperOrigin-RevId: 313656062 -- bbc759d43656b1b996ad558f23c852a9f14129d2 by Gennadiy Rozental <rogeeff@google.com>: Eliminate dynamic annotation symbols in the library. PiperOrigin-RevId: 313650817 GitOrigin-RevId: b2b94b9f533e4f9ae2a2df9de56ccb3b18f31d0d Change-Id: Ic7a11bbcb723f3ff6a7e2f214bff0a92c6f8ab4d
5 years ago
#include <array>
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
#include <atomic>
#include <memory>
Export of internal Abseil changes -- b2b94b9f533e4f9ae2a2df9de56ccb3b18f31d0d by Gennadiy Rozental <rogeeff@google.com>: Internal change PiperOrigin-RevId: 314366138 -- 7a5ac6be82741aec5b327d7b67efd14d69deed6a by Gennadiy Rozental <rogeeff@google.com>: Introduce Abseil prefixed dynamic annotation macros. PiperOrigin-RevId: 314228914 -- 33e7c605cb1be9bd48bd3590b1eedccd68d3bd1b by Gennadiy Rozental <rogeeff@google.com>: Fixing missing includes and remove unnecessary ones. PiperOrigin-RevId: 314217909 -- 6a3c5c26bfa13317bf0a880f13d0e4be0b971b76 by Gennadiy Rozental <rogeeff@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 314209576 -- f7795aa68020af4a6b4905531ba951e04b88966a by Abseil Team <absl-team@google.com>: absl::FormatTime() and absl::ParseTime() format specifiers: - %EZ now accepts 'z' in addition to 'Z' as a synonym for +00:00. - %ET is introduced, producing 'T' on output, and accepting 'T' or 't' on input. This is for the RFC3339 date-time separator. PiperOrigin-RevId: 313945137 -- 87c437ce3aab3f59a7546e44a28cd1c8aaa152c3 by Laramie Leavitt <lar@google.com>: Rollback PiperOrigin-RevId: 313868206 -- 8049b74349486a0026932b86d29c380b195e1cba by Laramie Leavitt <lar@google.com>: Remove the MockingBitGenBase base class in favor of type-erasure in BitGenRef. In Abseil random, mocking was split across two different classes, MockingBitGenBase and MockingBitGen. This split existed because Google Mock is a test-only library that we don't link into production, so MockingBitGenBase provided a low-overhead scaffold used to lookup mocks when in test code, but which is unused in production code. That has been replaced by type-erasure which looks for a method named CallImpl with the correct signature. Weaken the coupling between MockingBitGen, DistributionCaller, and MockOverloadSet. Rename CallImpl to InvokeMock() Previously, the implementation of DistributionCaller was also split across different files using explicit instantiation of the DistributionCaller struct and some details in the Mocking classes. Now Distribution caller uses the presence of the InvokeMock() method to choose whether to use the mockable call path or the default call path. PiperOrigin-RevId: 313848695 -- 1741d80e08050e1939605f70ca6ff64809785c85 by Gennadiy Rozental <rogeeff@google.com>: Introduce public interface to access reflection handle corresponding the flag. This interface will be the only official way to access flag reflection information from the flag object. The other internal methods will eventually disappear. PiperOrigin-RevId: 313734006 -- c375bead457de29d9c29595d16c66d3e5125b585 by Abseil Team <absl-team@google.com>: Improve documentation of absl::c_partial_sort_copy. This function takes no middle parameter, instead using the size of the result container to determine the size of the partial sort. PiperOrigin-RevId: 313656062 -- bbc759d43656b1b996ad558f23c852a9f14129d2 by Gennadiy Rozental <rogeeff@google.com>: Eliminate dynamic annotation symbols in the library. PiperOrigin-RevId: 313650817 GitOrigin-RevId: b2b94b9f533e4f9ae2a2df9de56ccb3b18f31d0d Change-Id: Ic7a11bbcb723f3ff6a7e2f214bff0a92c6f8ab4d
5 years ago
#include <new>
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
#include <string>
Export of internal Abseil changes -- b2b94b9f533e4f9ae2a2df9de56ccb3b18f31d0d by Gennadiy Rozental <rogeeff@google.com>: Internal change PiperOrigin-RevId: 314366138 -- 7a5ac6be82741aec5b327d7b67efd14d69deed6a by Gennadiy Rozental <rogeeff@google.com>: Introduce Abseil prefixed dynamic annotation macros. PiperOrigin-RevId: 314228914 -- 33e7c605cb1be9bd48bd3590b1eedccd68d3bd1b by Gennadiy Rozental <rogeeff@google.com>: Fixing missing includes and remove unnecessary ones. PiperOrigin-RevId: 314217909 -- 6a3c5c26bfa13317bf0a880f13d0e4be0b971b76 by Gennadiy Rozental <rogeeff@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 314209576 -- f7795aa68020af4a6b4905531ba951e04b88966a by Abseil Team <absl-team@google.com>: absl::FormatTime() and absl::ParseTime() format specifiers: - %EZ now accepts 'z' in addition to 'Z' as a synonym for +00:00. - %ET is introduced, producing 'T' on output, and accepting 'T' or 't' on input. This is for the RFC3339 date-time separator. PiperOrigin-RevId: 313945137 -- 87c437ce3aab3f59a7546e44a28cd1c8aaa152c3 by Laramie Leavitt <lar@google.com>: Rollback PiperOrigin-RevId: 313868206 -- 8049b74349486a0026932b86d29c380b195e1cba by Laramie Leavitt <lar@google.com>: Remove the MockingBitGenBase base class in favor of type-erasure in BitGenRef. In Abseil random, mocking was split across two different classes, MockingBitGenBase and MockingBitGen. This split existed because Google Mock is a test-only library that we don't link into production, so MockingBitGenBase provided a low-overhead scaffold used to lookup mocks when in test code, but which is unused in production code. That has been replaced by type-erasure which looks for a method named CallImpl with the correct signature. Weaken the coupling between MockingBitGen, DistributionCaller, and MockOverloadSet. Rename CallImpl to InvokeMock() Previously, the implementation of DistributionCaller was also split across different files using explicit instantiation of the DistributionCaller struct and some details in the Mocking classes. Now Distribution caller uses the presence of the InvokeMock() method to choose whether to use the mockable call path or the default call path. PiperOrigin-RevId: 313848695 -- 1741d80e08050e1939605f70ca6ff64809785c85 by Gennadiy Rozental <rogeeff@google.com>: Introduce public interface to access reflection handle corresponding the flag. This interface will be the only official way to access flag reflection information from the flag object. The other internal methods will eventually disappear. PiperOrigin-RevId: 313734006 -- c375bead457de29d9c29595d16c66d3e5125b585 by Abseil Team <absl-team@google.com>: Improve documentation of absl::c_partial_sort_copy. This function takes no middle parameter, instead using the size of the result container to determine the size of the partial sort. PiperOrigin-RevId: 313656062 -- bbc759d43656b1b996ad558f23c852a9f14129d2 by Gennadiy Rozental <rogeeff@google.com>: Eliminate dynamic annotation symbols in the library. PiperOrigin-RevId: 313650817 GitOrigin-RevId: b2b94b9f533e4f9ae2a2df9de56ccb3b18f31d0d Change-Id: Ic7a11bbcb723f3ff6a7e2f214bff0a92c6f8ab4d
5 years ago
#include <typeinfo>
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 -- b2b94b9f533e4f9ae2a2df9de56ccb3b18f31d0d by Gennadiy Rozental <rogeeff@google.com>: Internal change PiperOrigin-RevId: 314366138 -- 7a5ac6be82741aec5b327d7b67efd14d69deed6a by Gennadiy Rozental <rogeeff@google.com>: Introduce Abseil prefixed dynamic annotation macros. PiperOrigin-RevId: 314228914 -- 33e7c605cb1be9bd48bd3590b1eedccd68d3bd1b by Gennadiy Rozental <rogeeff@google.com>: Fixing missing includes and remove unnecessary ones. PiperOrigin-RevId: 314217909 -- 6a3c5c26bfa13317bf0a880f13d0e4be0b971b76 by Gennadiy Rozental <rogeeff@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 314209576 -- f7795aa68020af4a6b4905531ba951e04b88966a by Abseil Team <absl-team@google.com>: absl::FormatTime() and absl::ParseTime() format specifiers: - %EZ now accepts 'z' in addition to 'Z' as a synonym for +00:00. - %ET is introduced, producing 'T' on output, and accepting 'T' or 't' on input. This is for the RFC3339 date-time separator. PiperOrigin-RevId: 313945137 -- 87c437ce3aab3f59a7546e44a28cd1c8aaa152c3 by Laramie Leavitt <lar@google.com>: Rollback PiperOrigin-RevId: 313868206 -- 8049b74349486a0026932b86d29c380b195e1cba by Laramie Leavitt <lar@google.com>: Remove the MockingBitGenBase base class in favor of type-erasure in BitGenRef. In Abseil random, mocking was split across two different classes, MockingBitGenBase and MockingBitGen. This split existed because Google Mock is a test-only library that we don't link into production, so MockingBitGenBase provided a low-overhead scaffold used to lookup mocks when in test code, but which is unused in production code. That has been replaced by type-erasure which looks for a method named CallImpl with the correct signature. Weaken the coupling between MockingBitGen, DistributionCaller, and MockOverloadSet. Rename CallImpl to InvokeMock() Previously, the implementation of DistributionCaller was also split across different files using explicit instantiation of the DistributionCaller struct and some details in the Mocking classes. Now Distribution caller uses the presence of the InvokeMock() method to choose whether to use the mockable call path or the default call path. PiperOrigin-RevId: 313848695 -- 1741d80e08050e1939605f70ca6ff64809785c85 by Gennadiy Rozental <rogeeff@google.com>: Introduce public interface to access reflection handle corresponding the flag. This interface will be the only official way to access flag reflection information from the flag object. The other internal methods will eventually disappear. PiperOrigin-RevId: 313734006 -- c375bead457de29d9c29595d16c66d3e5125b585 by Abseil Team <absl-team@google.com>: Improve documentation of absl::c_partial_sort_copy. This function takes no middle parameter, instead using the size of the result container to determine the size of the partial sort. PiperOrigin-RevId: 313656062 -- bbc759d43656b1b996ad558f23c852a9f14129d2 by Gennadiy Rozental <rogeeff@google.com>: Eliminate dynamic annotation symbols in the library. PiperOrigin-RevId: 313650817 GitOrigin-RevId: b2b94b9f533e4f9ae2a2df9de56ccb3b18f31d0d Change-Id: Ic7a11bbcb723f3ff6a7e2f214bff0a92c6f8ab4d
5 years ago
#include "absl/base/call_once.h"
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
#include "absl/base/casts.h"
#include "absl/base/config.h"
#include "absl/base/dynamic_annotations.h"
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
#include "absl/base/optimization.h"
Export of internal Abseil changes -- b2b94b9f533e4f9ae2a2df9de56ccb3b18f31d0d by Gennadiy Rozental <rogeeff@google.com>: Internal change PiperOrigin-RevId: 314366138 -- 7a5ac6be82741aec5b327d7b67efd14d69deed6a by Gennadiy Rozental <rogeeff@google.com>: Introduce Abseil prefixed dynamic annotation macros. PiperOrigin-RevId: 314228914 -- 33e7c605cb1be9bd48bd3590b1eedccd68d3bd1b by Gennadiy Rozental <rogeeff@google.com>: Fixing missing includes and remove unnecessary ones. PiperOrigin-RevId: 314217909 -- 6a3c5c26bfa13317bf0a880f13d0e4be0b971b76 by Gennadiy Rozental <rogeeff@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 314209576 -- f7795aa68020af4a6b4905531ba951e04b88966a by Abseil Team <absl-team@google.com>: absl::FormatTime() and absl::ParseTime() format specifiers: - %EZ now accepts 'z' in addition to 'Z' as a synonym for +00:00. - %ET is introduced, producing 'T' on output, and accepting 'T' or 't' on input. This is for the RFC3339 date-time separator. PiperOrigin-RevId: 313945137 -- 87c437ce3aab3f59a7546e44a28cd1c8aaa152c3 by Laramie Leavitt <lar@google.com>: Rollback PiperOrigin-RevId: 313868206 -- 8049b74349486a0026932b86d29c380b195e1cba by Laramie Leavitt <lar@google.com>: Remove the MockingBitGenBase base class in favor of type-erasure in BitGenRef. In Abseil random, mocking was split across two different classes, MockingBitGenBase and MockingBitGen. This split existed because Google Mock is a test-only library that we don't link into production, so MockingBitGenBase provided a low-overhead scaffold used to lookup mocks when in test code, but which is unused in production code. That has been replaced by type-erasure which looks for a method named CallImpl with the correct signature. Weaken the coupling between MockingBitGen, DistributionCaller, and MockOverloadSet. Rename CallImpl to InvokeMock() Previously, the implementation of DistributionCaller was also split across different files using explicit instantiation of the DistributionCaller struct and some details in the Mocking classes. Now Distribution caller uses the presence of the InvokeMock() method to choose whether to use the mockable call path or the default call path. PiperOrigin-RevId: 313848695 -- 1741d80e08050e1939605f70ca6ff64809785c85 by Gennadiy Rozental <rogeeff@google.com>: Introduce public interface to access reflection handle corresponding the flag. This interface will be the only official way to access flag reflection information from the flag object. The other internal methods will eventually disappear. PiperOrigin-RevId: 313734006 -- c375bead457de29d9c29595d16c66d3e5125b585 by Abseil Team <absl-team@google.com>: Improve documentation of absl::c_partial_sort_copy. This function takes no middle parameter, instead using the size of the result container to determine the size of the partial sort. PiperOrigin-RevId: 313656062 -- bbc759d43656b1b996ad558f23c852a9f14129d2 by Gennadiy Rozental <rogeeff@google.com>: Eliminate dynamic annotation symbols in the library. PiperOrigin-RevId: 313650817 GitOrigin-RevId: b2b94b9f533e4f9ae2a2df9de56ccb3b18f31d0d Change-Id: Ic7a11bbcb723f3ff6a7e2f214bff0a92c6f8ab4d
5 years ago
#include "absl/flags/config.h"
#include "absl/flags/internal/commandlineflag.h"
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
#include "absl/flags/usage_config.h"
Export of internal Abseil changes -- b2b94b9f533e4f9ae2a2df9de56ccb3b18f31d0d by Gennadiy Rozental <rogeeff@google.com>: Internal change PiperOrigin-RevId: 314366138 -- 7a5ac6be82741aec5b327d7b67efd14d69deed6a by Gennadiy Rozental <rogeeff@google.com>: Introduce Abseil prefixed dynamic annotation macros. PiperOrigin-RevId: 314228914 -- 33e7c605cb1be9bd48bd3590b1eedccd68d3bd1b by Gennadiy Rozental <rogeeff@google.com>: Fixing missing includes and remove unnecessary ones. PiperOrigin-RevId: 314217909 -- 6a3c5c26bfa13317bf0a880f13d0e4be0b971b76 by Gennadiy Rozental <rogeeff@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 314209576 -- f7795aa68020af4a6b4905531ba951e04b88966a by Abseil Team <absl-team@google.com>: absl::FormatTime() and absl::ParseTime() format specifiers: - %EZ now accepts 'z' in addition to 'Z' as a synonym for +00:00. - %ET is introduced, producing 'T' on output, and accepting 'T' or 't' on input. This is for the RFC3339 date-time separator. PiperOrigin-RevId: 313945137 -- 87c437ce3aab3f59a7546e44a28cd1c8aaa152c3 by Laramie Leavitt <lar@google.com>: Rollback PiperOrigin-RevId: 313868206 -- 8049b74349486a0026932b86d29c380b195e1cba by Laramie Leavitt <lar@google.com>: Remove the MockingBitGenBase base class in favor of type-erasure in BitGenRef. In Abseil random, mocking was split across two different classes, MockingBitGenBase and MockingBitGen. This split existed because Google Mock is a test-only library that we don't link into production, so MockingBitGenBase provided a low-overhead scaffold used to lookup mocks when in test code, but which is unused in production code. That has been replaced by type-erasure which looks for a method named CallImpl with the correct signature. Weaken the coupling between MockingBitGen, DistributionCaller, and MockOverloadSet. Rename CallImpl to InvokeMock() Previously, the implementation of DistributionCaller was also split across different files using explicit instantiation of the DistributionCaller struct and some details in the Mocking classes. Now Distribution caller uses the presence of the InvokeMock() method to choose whether to use the mockable call path or the default call path. PiperOrigin-RevId: 313848695 -- 1741d80e08050e1939605f70ca6ff64809785c85 by Gennadiy Rozental <rogeeff@google.com>: Introduce public interface to access reflection handle corresponding the flag. This interface will be the only official way to access flag reflection information from the flag object. The other internal methods will eventually disappear. PiperOrigin-RevId: 313734006 -- c375bead457de29d9c29595d16c66d3e5125b585 by Abseil Team <absl-team@google.com>: Improve documentation of absl::c_partial_sort_copy. This function takes no middle parameter, instead using the size of the result container to determine the size of the partial sort. PiperOrigin-RevId: 313656062 -- bbc759d43656b1b996ad558f23c852a9f14129d2 by Gennadiy Rozental <rogeeff@google.com>: Eliminate dynamic annotation symbols in the library. PiperOrigin-RevId: 313650817 GitOrigin-RevId: b2b94b9f533e4f9ae2a2df9de56ccb3b18f31d0d Change-Id: Ic7a11bbcb723f3ff6a7e2f214bff0a92c6f8ab4d
5 years ago
#include "absl/memory/memory.h"
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
#include "absl/strings/str_cat.h"
#include "absl/strings/string_view.h"
#include "absl/synchronization/mutex.h"
namespace absl {
ABSL_NAMESPACE_BEGIN
namespace flags_internal {
// The help message indicating that the commandline flag has been
// 'stripped'. It will not show up when doing "-help" and its
// variants. The flag is stripped if ABSL_FLAGS_STRIP_HELP is set to 1
// before including absl/flags/flag.h
const char kStrippedFlagHelp[] = "\001\002\003\004 (unknown) \004\003\002\001";
namespace {
// Currently we only validate flag values for user-defined flag types.
bool ShouldValidateFlagValue(FlagFastTypeId flag_type_id) {
#define DONT_VALIDATE(T, _) \
if (flag_type_id == base_internal::FastTypeId<T>()) return false;
ABSL_FLAGS_INTERNAL_SUPPORTED_TYPES(DONT_VALIDATE)
#undef DONT_VALIDATE
return true;
}
// RAII helper used to temporarily unlock and relock `absl::Mutex`.
// This is used when we need to ensure that locks are released while
// invoking user supplied callbacks and then reacquired, since callbacks may
// need to acquire these locks themselves.
class MutexRelock {
public:
Export of internal Abseil changes -- 7d0468a6610ed85586d5c87fd65de8dac5118923 by Derek Mauro <dmauro@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 313226473 -- 1131ef6d116f5ce7d46537a82f300ea06dcaaa53 by Gennadiy Rozental <rogeeff@google.com>: Migrate internal interface to use mutable references. PiperOrigin-RevId: 312931131 -- 96225212a9f5fbd0b38c71fe65539164992c7c3b by Laramie Leavitt <lar@google.com>: Remove random/internal/distributions.h This file was something of an historical artifact. All of the related code has either been removed or migraged, and so the only remaining type belongs with uniform_helper.h, as it is used to infer the return type of the absl::Uniform method in a few cases. PiperOrigin-RevId: 312878173 -- 6dcbd5be58ad425e08740ff64088373ee7fe4a72 by Mark Barolak <mbar@google.com>: Release the StrFormat test case for Cords to open source. PiperOrigin-RevId: 312707974 -- 34484d18dfb63a0a7ad6e2aaeb570e33592968be by Abseil Team <absl-team@google.com>: Let Cord::Cord(string&&), Cord::operator=(string&&), Cord::Append(string&&), and Cord::Prepend(string&&) steal string data and embed it into the Cord as a single external chunk, instead of copying it into flat chunks (at most 4083-byte each). Stealing string data is faster, but it creates a long chunk, which leads to a higher more memory usage if its subcords are created and outlive the whole Cord. These functions revert to copying the data if any of the following conditions holds: - string size is at most kMaxBytesToCopy (511), to avoid the overhead of an external chunk for short strings; - less than half of string capacity is used, to avoid pinning to much unused memory. PiperOrigin-RevId: 312683785 GitOrigin-RevId: 7d0468a6610ed85586d5c87fd65de8dac5118923 Change-Id: If79b5a1dfe6d53a8ddddbc7da84338f11fc4cfa3
5 years ago
explicit MutexRelock(absl::Mutex& mu) : mu_(mu) { mu_.Unlock(); }
~MutexRelock() { mu_.Lock(); }
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
MutexRelock(const MutexRelock&) = delete;
MutexRelock& operator=(const MutexRelock&) = delete;
private:
Export of internal Abseil changes -- 7d0468a6610ed85586d5c87fd65de8dac5118923 by Derek Mauro <dmauro@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 313226473 -- 1131ef6d116f5ce7d46537a82f300ea06dcaaa53 by Gennadiy Rozental <rogeeff@google.com>: Migrate internal interface to use mutable references. PiperOrigin-RevId: 312931131 -- 96225212a9f5fbd0b38c71fe65539164992c7c3b by Laramie Leavitt <lar@google.com>: Remove random/internal/distributions.h This file was something of an historical artifact. All of the related code has either been removed or migraged, and so the only remaining type belongs with uniform_helper.h, as it is used to infer the return type of the absl::Uniform method in a few cases. PiperOrigin-RevId: 312878173 -- 6dcbd5be58ad425e08740ff64088373ee7fe4a72 by Mark Barolak <mbar@google.com>: Release the StrFormat test case for Cords to open source. PiperOrigin-RevId: 312707974 -- 34484d18dfb63a0a7ad6e2aaeb570e33592968be by Abseil Team <absl-team@google.com>: Let Cord::Cord(string&&), Cord::operator=(string&&), Cord::Append(string&&), and Cord::Prepend(string&&) steal string data and embed it into the Cord as a single external chunk, instead of copying it into flat chunks (at most 4083-byte each). Stealing string data is faster, but it creates a long chunk, which leads to a higher more memory usage if its subcords are created and outlive the whole Cord. These functions revert to copying the data if any of the following conditions holds: - string size is at most kMaxBytesToCopy (511), to avoid the overhead of an external chunk for short strings; - less than half of string capacity is used, to avoid pinning to much unused memory. PiperOrigin-RevId: 312683785 GitOrigin-RevId: 7d0468a6610ed85586d5c87fd65de8dac5118923 Change-Id: If79b5a1dfe6d53a8ddddbc7da84338f11fc4cfa3
5 years ago
absl::Mutex& mu_;
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
};
} // namespace
///////////////////////////////////////////////////////////////////////////////
// Persistent state of the flag data.
class FlagImpl;
class FlagState : public flags_internal::FlagStateInterface {
public:
template <typename V>
Export of internal Abseil changes -- 7d0468a6610ed85586d5c87fd65de8dac5118923 by Derek Mauro <dmauro@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 313226473 -- 1131ef6d116f5ce7d46537a82f300ea06dcaaa53 by Gennadiy Rozental <rogeeff@google.com>: Migrate internal interface to use mutable references. PiperOrigin-RevId: 312931131 -- 96225212a9f5fbd0b38c71fe65539164992c7c3b by Laramie Leavitt <lar@google.com>: Remove random/internal/distributions.h This file was something of an historical artifact. All of the related code has either been removed or migraged, and so the only remaining type belongs with uniform_helper.h, as it is used to infer the return type of the absl::Uniform method in a few cases. PiperOrigin-RevId: 312878173 -- 6dcbd5be58ad425e08740ff64088373ee7fe4a72 by Mark Barolak <mbar@google.com>: Release the StrFormat test case for Cords to open source. PiperOrigin-RevId: 312707974 -- 34484d18dfb63a0a7ad6e2aaeb570e33592968be by Abseil Team <absl-team@google.com>: Let Cord::Cord(string&&), Cord::operator=(string&&), Cord::Append(string&&), and Cord::Prepend(string&&) steal string data and embed it into the Cord as a single external chunk, instead of copying it into flat chunks (at most 4083-byte each). Stealing string data is faster, but it creates a long chunk, which leads to a higher more memory usage if its subcords are created and outlive the whole Cord. These functions revert to copying the data if any of the following conditions holds: - string size is at most kMaxBytesToCopy (511), to avoid the overhead of an external chunk for short strings; - less than half of string capacity is used, to avoid pinning to much unused memory. PiperOrigin-RevId: 312683785 GitOrigin-RevId: 7d0468a6610ed85586d5c87fd65de8dac5118923 Change-Id: If79b5a1dfe6d53a8ddddbc7da84338f11fc4cfa3
5 years ago
FlagState(FlagImpl& flag_impl, const V& v, bool modified,
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
bool on_command_line, int64_t counter)
: flag_impl_(flag_impl),
value_(v),
modified_(modified),
on_command_line_(on_command_line),
counter_(counter) {}
~FlagState() override {
if (flag_impl_.ValueStorageKind() != FlagValueStorageKind::kAlignedBuffer &&
flag_impl_.ValueStorageKind() != FlagValueStorageKind::kSequenceLocked)
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
return;
Export of internal Abseil changes -- 7d0468a6610ed85586d5c87fd65de8dac5118923 by Derek Mauro <dmauro@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 313226473 -- 1131ef6d116f5ce7d46537a82f300ea06dcaaa53 by Gennadiy Rozental <rogeeff@google.com>: Migrate internal interface to use mutable references. PiperOrigin-RevId: 312931131 -- 96225212a9f5fbd0b38c71fe65539164992c7c3b by Laramie Leavitt <lar@google.com>: Remove random/internal/distributions.h This file was something of an historical artifact. All of the related code has either been removed or migraged, and so the only remaining type belongs with uniform_helper.h, as it is used to infer the return type of the absl::Uniform method in a few cases. PiperOrigin-RevId: 312878173 -- 6dcbd5be58ad425e08740ff64088373ee7fe4a72 by Mark Barolak <mbar@google.com>: Release the StrFormat test case for Cords to open source. PiperOrigin-RevId: 312707974 -- 34484d18dfb63a0a7ad6e2aaeb570e33592968be by Abseil Team <absl-team@google.com>: Let Cord::Cord(string&&), Cord::operator=(string&&), Cord::Append(string&&), and Cord::Prepend(string&&) steal string data and embed it into the Cord as a single external chunk, instead of copying it into flat chunks (at most 4083-byte each). Stealing string data is faster, but it creates a long chunk, which leads to a higher more memory usage if its subcords are created and outlive the whole Cord. These functions revert to copying the data if any of the following conditions holds: - string size is at most kMaxBytesToCopy (511), to avoid the overhead of an external chunk for short strings; - less than half of string capacity is used, to avoid pinning to much unused memory. PiperOrigin-RevId: 312683785 GitOrigin-RevId: 7d0468a6610ed85586d5c87fd65de8dac5118923 Change-Id: If79b5a1dfe6d53a8ddddbc7da84338f11fc4cfa3
5 years ago
flags_internal::Delete(flag_impl_.op_, value_.heap_allocated);
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
}
private:
friend class FlagImpl;
// Restores the flag to the saved state.
void Restore() const override {
Export of internal Abseil changes -- 7d0468a6610ed85586d5c87fd65de8dac5118923 by Derek Mauro <dmauro@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 313226473 -- 1131ef6d116f5ce7d46537a82f300ea06dcaaa53 by Gennadiy Rozental <rogeeff@google.com>: Migrate internal interface to use mutable references. PiperOrigin-RevId: 312931131 -- 96225212a9f5fbd0b38c71fe65539164992c7c3b by Laramie Leavitt <lar@google.com>: Remove random/internal/distributions.h This file was something of an historical artifact. All of the related code has either been removed or migraged, and so the only remaining type belongs with uniform_helper.h, as it is used to infer the return type of the absl::Uniform method in a few cases. PiperOrigin-RevId: 312878173 -- 6dcbd5be58ad425e08740ff64088373ee7fe4a72 by Mark Barolak <mbar@google.com>: Release the StrFormat test case for Cords to open source. PiperOrigin-RevId: 312707974 -- 34484d18dfb63a0a7ad6e2aaeb570e33592968be by Abseil Team <absl-team@google.com>: Let Cord::Cord(string&&), Cord::operator=(string&&), Cord::Append(string&&), and Cord::Prepend(string&&) steal string data and embed it into the Cord as a single external chunk, instead of copying it into flat chunks (at most 4083-byte each). Stealing string data is faster, but it creates a long chunk, which leads to a higher more memory usage if its subcords are created and outlive the whole Cord. These functions revert to copying the data if any of the following conditions holds: - string size is at most kMaxBytesToCopy (511), to avoid the overhead of an external chunk for short strings; - less than half of string capacity is used, to avoid pinning to much unused memory. PiperOrigin-RevId: 312683785 GitOrigin-RevId: 7d0468a6610ed85586d5c87fd65de8dac5118923 Change-Id: If79b5a1dfe6d53a8ddddbc7da84338f11fc4cfa3
5 years ago
if (!flag_impl_.RestoreState(*this)) return;
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 -- 7d0468a6610ed85586d5c87fd65de8dac5118923 by Derek Mauro <dmauro@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 313226473 -- 1131ef6d116f5ce7d46537a82f300ea06dcaaa53 by Gennadiy Rozental <rogeeff@google.com>: Migrate internal interface to use mutable references. PiperOrigin-RevId: 312931131 -- 96225212a9f5fbd0b38c71fe65539164992c7c3b by Laramie Leavitt <lar@google.com>: Remove random/internal/distributions.h This file was something of an historical artifact. All of the related code has either been removed or migraged, and so the only remaining type belongs with uniform_helper.h, as it is used to infer the return type of the absl::Uniform method in a few cases. PiperOrigin-RevId: 312878173 -- 6dcbd5be58ad425e08740ff64088373ee7fe4a72 by Mark Barolak <mbar@google.com>: Release the StrFormat test case for Cords to open source. PiperOrigin-RevId: 312707974 -- 34484d18dfb63a0a7ad6e2aaeb570e33592968be by Abseil Team <absl-team@google.com>: Let Cord::Cord(string&&), Cord::operator=(string&&), Cord::Append(string&&), and Cord::Prepend(string&&) steal string data and embed it into the Cord as a single external chunk, instead of copying it into flat chunks (at most 4083-byte each). Stealing string data is faster, but it creates a long chunk, which leads to a higher more memory usage if its subcords are created and outlive the whole Cord. These functions revert to copying the data if any of the following conditions holds: - string size is at most kMaxBytesToCopy (511), to avoid the overhead of an external chunk for short strings; - less than half of string capacity is used, to avoid pinning to much unused memory. PiperOrigin-RevId: 312683785 GitOrigin-RevId: 7d0468a6610ed85586d5c87fd65de8dac5118923 Change-Id: If79b5a1dfe6d53a8ddddbc7da84338f11fc4cfa3
5 years ago
ABSL_INTERNAL_LOG(INFO,
absl::StrCat("Restore saved value of ", flag_impl_.Name(),
" to: ", flag_impl_.CurrentValue()));
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
}
// Flag and saved flag data.
Export of internal Abseil changes -- 7d0468a6610ed85586d5c87fd65de8dac5118923 by Derek Mauro <dmauro@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 313226473 -- 1131ef6d116f5ce7d46537a82f300ea06dcaaa53 by Gennadiy Rozental <rogeeff@google.com>: Migrate internal interface to use mutable references. PiperOrigin-RevId: 312931131 -- 96225212a9f5fbd0b38c71fe65539164992c7c3b by Laramie Leavitt <lar@google.com>: Remove random/internal/distributions.h This file was something of an historical artifact. All of the related code has either been removed or migraged, and so the only remaining type belongs with uniform_helper.h, as it is used to infer the return type of the absl::Uniform method in a few cases. PiperOrigin-RevId: 312878173 -- 6dcbd5be58ad425e08740ff64088373ee7fe4a72 by Mark Barolak <mbar@google.com>: Release the StrFormat test case for Cords to open source. PiperOrigin-RevId: 312707974 -- 34484d18dfb63a0a7ad6e2aaeb570e33592968be by Abseil Team <absl-team@google.com>: Let Cord::Cord(string&&), Cord::operator=(string&&), Cord::Append(string&&), and Cord::Prepend(string&&) steal string data and embed it into the Cord as a single external chunk, instead of copying it into flat chunks (at most 4083-byte each). Stealing string data is faster, but it creates a long chunk, which leads to a higher more memory usage if its subcords are created and outlive the whole Cord. These functions revert to copying the data if any of the following conditions holds: - string size is at most kMaxBytesToCopy (511), to avoid the overhead of an external chunk for short strings; - less than half of string capacity is used, to avoid pinning to much unused memory. PiperOrigin-RevId: 312683785 GitOrigin-RevId: 7d0468a6610ed85586d5c87fd65de8dac5118923 Change-Id: If79b5a1dfe6d53a8ddddbc7da84338f11fc4cfa3
5 years ago
FlagImpl& flag_impl_;
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
union SavedValue {
explicit SavedValue(void* v) : heap_allocated(v) {}
explicit SavedValue(int64_t v) : one_word(v) {}
void* heap_allocated;
int64_t one_word;
} value_;
bool modified_;
bool on_command_line_;
int64_t counter_;
};
///////////////////////////////////////////////////////////////////////////////
// Flag implementation, which does not depend on flag value type.
DynValueDeleter::DynValueDeleter(FlagOpFn op_arg) : op(op_arg) {}
void DynValueDeleter::operator()(void* ptr) const {
if (op == nullptr) return;
Delete(op, ptr);
}
void FlagImpl::Init() {
new (&data_guard_) absl::Mutex;
auto def_kind = static_cast<FlagDefaultKind>(def_kind_);
switch (ValueStorageKind()) {
Export of internal Abseil changes -- 04cb3b22497190170aa5b774e98080c5de2ba60b by Abseil Team <absl-team@google.com>: Alternative bit mixer for LowLevelHash on ARM LowLevelHash's bit-mixer is inefficient on ARM because it calculates a 128-bit product of two 64-bit numbers. On ARM, this requires a sequence of two instructions with a high combined latency and poor throughput. This change provides alternative bit-mixing code for ARM that uses only 64-bit arithmetic (multiplication, xor, and left-shifts) and speeds things up considerably. The bit-mixing code for ARM was inspired by by Woothash[1] and xxh3[1]. Once I landed on a sequence of operations that provided good mixing, I used a test harness to search for the combination of shift / rotate factors that provided the best mixing, as indicated by SMHasher hash quality tests. The new mixing code passes 13 out of 15 of the hash quality test suites in SMHasher, with the two failures being in the noise range: e.g. 1 collision vs. zero expected in a keyset of ~8m keys. [1]: https://github.com/tommyettinger/waterhash/blob/49f5cf0b63b9/woothash.h#L16-L20 [2]: https://github.com/Cyan4973/xxHash/blob/6853ddc36e46/xxhash.h#L3240-L3265 PiperOrigin-RevId: 391833008 -- 17a4de1f9d623155c75b19285d414cd55a487cd6 by Saleem Abdulrasool <abdulras@google.com>: debugging: add support for unwinding on RISCV Linux This adds partial support for unwinding the RISCV call stack. It is largely duplicated from the AArch64 support with alterations for the ELF RISCV psABI. This covers RISCV64 and RISCV32, though not the ILP32E calling convention. PiperOrigin-RevId: 391818522 -- 32c93e449327b2cea32b32f6365e84b420fe1ed3 by Gennadiy Rozental <rogeeff@google.com>: New storage for types smaller than 8 bytes. Also adding new read interface for types smaller than or rqual to 8 bytes to avoid passing the pointer. PiperOrigin-RevId: 391726822 -- e987ac08a7787801cbfc7d7c96649e97fa8cff1a by Abseil Team <absl-team@google.com>: Extern template `find_first_non_full` to reduce linkage size for TU with single not inlined function. PiperOrigin-RevId: 391718862 -- 73af9bfcb5bf045089133e18bbd20eb5bb699172 by Gennadiy Rozental <rogeeff@google.com>: Make most non-mutable most int128 methods and friend free functions constexpr. Some functions are implemented offline (at least in some configurations) and can't be made constexpr. Mutable methods can't be made constexpr until we drop c++11 support. Fixes #978 PiperOrigin-RevId: 391706535 GitOrigin-RevId: 04cb3b22497190170aa5b774e98080c5de2ba60b Change-Id: If051fad5ff004e2e82fa53618fc04a6fe3d2d4be
3 years ago
case FlagValueStorageKind::kValueAndInitBit:
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
case FlagValueStorageKind::kOneWordAtomic: {
alignas(int64_t) std::array<char, sizeof(int64_t)> buf{};
if (def_kind == FlagDefaultKind::kGenFunc) {
(*default_value_.gen_func)(buf.data());
} else {
assert(def_kind != FlagDefaultKind::kDynamicValue);
std::memcpy(buf.data(), &default_value_, Sizeof(op_));
}
Export of internal Abseil changes -- 04cb3b22497190170aa5b774e98080c5de2ba60b by Abseil Team <absl-team@google.com>: Alternative bit mixer for LowLevelHash on ARM LowLevelHash's bit-mixer is inefficient on ARM because it calculates a 128-bit product of two 64-bit numbers. On ARM, this requires a sequence of two instructions with a high combined latency and poor throughput. This change provides alternative bit-mixing code for ARM that uses only 64-bit arithmetic (multiplication, xor, and left-shifts) and speeds things up considerably. The bit-mixing code for ARM was inspired by by Woothash[1] and xxh3[1]. Once I landed on a sequence of operations that provided good mixing, I used a test harness to search for the combination of shift / rotate factors that provided the best mixing, as indicated by SMHasher hash quality tests. The new mixing code passes 13 out of 15 of the hash quality test suites in SMHasher, with the two failures being in the noise range: e.g. 1 collision vs. zero expected in a keyset of ~8m keys. [1]: https://github.com/tommyettinger/waterhash/blob/49f5cf0b63b9/woothash.h#L16-L20 [2]: https://github.com/Cyan4973/xxHash/blob/6853ddc36e46/xxhash.h#L3240-L3265 PiperOrigin-RevId: 391833008 -- 17a4de1f9d623155c75b19285d414cd55a487cd6 by Saleem Abdulrasool <abdulras@google.com>: debugging: add support for unwinding on RISCV Linux This adds partial support for unwinding the RISCV call stack. It is largely duplicated from the AArch64 support with alterations for the ELF RISCV psABI. This covers RISCV64 and RISCV32, though not the ILP32E calling convention. PiperOrigin-RevId: 391818522 -- 32c93e449327b2cea32b32f6365e84b420fe1ed3 by Gennadiy Rozental <rogeeff@google.com>: New storage for types smaller than 8 bytes. Also adding new read interface for types smaller than or rqual to 8 bytes to avoid passing the pointer. PiperOrigin-RevId: 391726822 -- e987ac08a7787801cbfc7d7c96649e97fa8cff1a by Abseil Team <absl-team@google.com>: Extern template `find_first_non_full` to reduce linkage size for TU with single not inlined function. PiperOrigin-RevId: 391718862 -- 73af9bfcb5bf045089133e18bbd20eb5bb699172 by Gennadiy Rozental <rogeeff@google.com>: Make most non-mutable most int128 methods and friend free functions constexpr. Some functions are implemented offline (at least in some configurations) and can't be made constexpr. Mutable methods can't be made constexpr until we drop c++11 support. Fixes #978 PiperOrigin-RevId: 391706535 GitOrigin-RevId: 04cb3b22497190170aa5b774e98080c5de2ba60b Change-Id: If051fad5ff004e2e82fa53618fc04a6fe3d2d4be
3 years ago
if (ValueStorageKind() == FlagValueStorageKind::kValueAndInitBit) {
// We presume here the memory layout of FlagValueAndInitBit struct.
uint8_t initialized = 1;
std::memcpy(buf.data() + Sizeof(op_), &initialized,
sizeof(initialized));
}
// Type can contain valid uninitialized bits, e.g. padding.
ABSL_ANNOTATE_MEMORY_IS_INITIALIZED(buf.data(), buf.size());
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
OneWordValue().store(absl::bit_cast<int64_t>(buf),
std::memory_order_release);
break;
}
case FlagValueStorageKind::kSequenceLocked: {
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
// For this storage kind the default_value_ always points to gen_func
// during initialization.
assert(def_kind == FlagDefaultKind::kGenFunc);
(*default_value_.gen_func)(AtomicBufferValue());
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
break;
}
Export of internal Abseil changes -- 04cb3b22497190170aa5b774e98080c5de2ba60b by Abseil Team <absl-team@google.com>: Alternative bit mixer for LowLevelHash on ARM LowLevelHash's bit-mixer is inefficient on ARM because it calculates a 128-bit product of two 64-bit numbers. On ARM, this requires a sequence of two instructions with a high combined latency and poor throughput. This change provides alternative bit-mixing code for ARM that uses only 64-bit arithmetic (multiplication, xor, and left-shifts) and speeds things up considerably. The bit-mixing code for ARM was inspired by by Woothash[1] and xxh3[1]. Once I landed on a sequence of operations that provided good mixing, I used a test harness to search for the combination of shift / rotate factors that provided the best mixing, as indicated by SMHasher hash quality tests. The new mixing code passes 13 out of 15 of the hash quality test suites in SMHasher, with the two failures being in the noise range: e.g. 1 collision vs. zero expected in a keyset of ~8m keys. [1]: https://github.com/tommyettinger/waterhash/blob/49f5cf0b63b9/woothash.h#L16-L20 [2]: https://github.com/Cyan4973/xxHash/blob/6853ddc36e46/xxhash.h#L3240-L3265 PiperOrigin-RevId: 391833008 -- 17a4de1f9d623155c75b19285d414cd55a487cd6 by Saleem Abdulrasool <abdulras@google.com>: debugging: add support for unwinding on RISCV Linux This adds partial support for unwinding the RISCV call stack. It is largely duplicated from the AArch64 support with alterations for the ELF RISCV psABI. This covers RISCV64 and RISCV32, though not the ILP32E calling convention. PiperOrigin-RevId: 391818522 -- 32c93e449327b2cea32b32f6365e84b420fe1ed3 by Gennadiy Rozental <rogeeff@google.com>: New storage for types smaller than 8 bytes. Also adding new read interface for types smaller than or rqual to 8 bytes to avoid passing the pointer. PiperOrigin-RevId: 391726822 -- e987ac08a7787801cbfc7d7c96649e97fa8cff1a by Abseil Team <absl-team@google.com>: Extern template `find_first_non_full` to reduce linkage size for TU with single not inlined function. PiperOrigin-RevId: 391718862 -- 73af9bfcb5bf045089133e18bbd20eb5bb699172 by Gennadiy Rozental <rogeeff@google.com>: Make most non-mutable most int128 methods and friend free functions constexpr. Some functions are implemented offline (at least in some configurations) and can't be made constexpr. Mutable methods can't be made constexpr until we drop c++11 support. Fixes #978 PiperOrigin-RevId: 391706535 GitOrigin-RevId: 04cb3b22497190170aa5b774e98080c5de2ba60b Change-Id: If051fad5ff004e2e82fa53618fc04a6fe3d2d4be
3 years ago
case FlagValueStorageKind::kAlignedBuffer:
// For this storage kind the default_value_ always points to gen_func
// during initialization.
assert(def_kind == FlagDefaultKind::kGenFunc);
(*default_value_.gen_func)(AlignedBufferValue());
break;
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
}
seq_lock_.MarkInitialized();
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::Mutex* FlagImpl::DataGuard() const {
absl::call_once(const_cast<FlagImpl*>(this)->init_control_, &FlagImpl::Init,
const_cast<FlagImpl*>(this));
// data_guard_ is initialized inside Init.
return reinterpret_cast<absl::Mutex*>(&data_guard_);
}
void FlagImpl::AssertValidType(FlagFastTypeId rhs_type_id,
const std::type_info* (*gen_rtti)()) const {
FlagFastTypeId lhs_type_id = flags_internal::FastTypeId(op_);
// `rhs_type_id` is the fast type id corresponding to the declaration
// visibile at the call site. `lhs_type_id` is the fast type id
// corresponding to the type specified in flag definition. They must match
// for this operation to be well-defined.
if (ABSL_PREDICT_TRUE(lhs_type_id == rhs_type_id)) return;
const std::type_info* lhs_runtime_type_id =
flags_internal::RuntimeTypeId(op_);
const std::type_info* rhs_runtime_type_id = (*gen_rtti)();
if (lhs_runtime_type_id == rhs_runtime_type_id) return;
#ifdef ABSL_INTERNAL_HAS_RTTI
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
if (*lhs_runtime_type_id == *rhs_runtime_type_id) return;
#endif
ABSL_INTERNAL_LOG(
FATAL, absl::StrCat("Flag '", Name(),
"' is defined as one type and declared as another"));
}
std::unique_ptr<void, DynValueDeleter> FlagImpl::MakeInitValue() const {
void* res = nullptr;
switch (DefaultKind()) {
case FlagDefaultKind::kDynamicValue:
res = flags_internal::Clone(op_, default_value_.dynamic_value);
break;
case FlagDefaultKind::kGenFunc:
res = flags_internal::Alloc(op_);
(*default_value_.gen_func)(res);
break;
default:
res = flags_internal::Clone(op_, &default_value_);
break;
}
return {res, DynValueDeleter{op_}};
}
void FlagImpl::StoreValue(const void* src) {
switch (ValueStorageKind()) {
Export of internal Abseil changes -- 04cb3b22497190170aa5b774e98080c5de2ba60b by Abseil Team <absl-team@google.com>: Alternative bit mixer for LowLevelHash on ARM LowLevelHash's bit-mixer is inefficient on ARM because it calculates a 128-bit product of two 64-bit numbers. On ARM, this requires a sequence of two instructions with a high combined latency and poor throughput. This change provides alternative bit-mixing code for ARM that uses only 64-bit arithmetic (multiplication, xor, and left-shifts) and speeds things up considerably. The bit-mixing code for ARM was inspired by by Woothash[1] and xxh3[1]. Once I landed on a sequence of operations that provided good mixing, I used a test harness to search for the combination of shift / rotate factors that provided the best mixing, as indicated by SMHasher hash quality tests. The new mixing code passes 13 out of 15 of the hash quality test suites in SMHasher, with the two failures being in the noise range: e.g. 1 collision vs. zero expected in a keyset of ~8m keys. [1]: https://github.com/tommyettinger/waterhash/blob/49f5cf0b63b9/woothash.h#L16-L20 [2]: https://github.com/Cyan4973/xxHash/blob/6853ddc36e46/xxhash.h#L3240-L3265 PiperOrigin-RevId: 391833008 -- 17a4de1f9d623155c75b19285d414cd55a487cd6 by Saleem Abdulrasool <abdulras@google.com>: debugging: add support for unwinding on RISCV Linux This adds partial support for unwinding the RISCV call stack. It is largely duplicated from the AArch64 support with alterations for the ELF RISCV psABI. This covers RISCV64 and RISCV32, though not the ILP32E calling convention. PiperOrigin-RevId: 391818522 -- 32c93e449327b2cea32b32f6365e84b420fe1ed3 by Gennadiy Rozental <rogeeff@google.com>: New storage for types smaller than 8 bytes. Also adding new read interface for types smaller than or rqual to 8 bytes to avoid passing the pointer. PiperOrigin-RevId: 391726822 -- e987ac08a7787801cbfc7d7c96649e97fa8cff1a by Abseil Team <absl-team@google.com>: Extern template `find_first_non_full` to reduce linkage size for TU with single not inlined function. PiperOrigin-RevId: 391718862 -- 73af9bfcb5bf045089133e18bbd20eb5bb699172 by Gennadiy Rozental <rogeeff@google.com>: Make most non-mutable most int128 methods and friend free functions constexpr. Some functions are implemented offline (at least in some configurations) and can't be made constexpr. Mutable methods can't be made constexpr until we drop c++11 support. Fixes #978 PiperOrigin-RevId: 391706535 GitOrigin-RevId: 04cb3b22497190170aa5b774e98080c5de2ba60b Change-Id: If051fad5ff004e2e82fa53618fc04a6fe3d2d4be
3 years ago
case FlagValueStorageKind::kValueAndInitBit:
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
case FlagValueStorageKind::kOneWordAtomic: {
Export of internal Abseil changes -- 04cb3b22497190170aa5b774e98080c5de2ba60b by Abseil Team <absl-team@google.com>: Alternative bit mixer for LowLevelHash on ARM LowLevelHash's bit-mixer is inefficient on ARM because it calculates a 128-bit product of two 64-bit numbers. On ARM, this requires a sequence of two instructions with a high combined latency and poor throughput. This change provides alternative bit-mixing code for ARM that uses only 64-bit arithmetic (multiplication, xor, and left-shifts) and speeds things up considerably. The bit-mixing code for ARM was inspired by by Woothash[1] and xxh3[1]. Once I landed on a sequence of operations that provided good mixing, I used a test harness to search for the combination of shift / rotate factors that provided the best mixing, as indicated by SMHasher hash quality tests. The new mixing code passes 13 out of 15 of the hash quality test suites in SMHasher, with the two failures being in the noise range: e.g. 1 collision vs. zero expected in a keyset of ~8m keys. [1]: https://github.com/tommyettinger/waterhash/blob/49f5cf0b63b9/woothash.h#L16-L20 [2]: https://github.com/Cyan4973/xxHash/blob/6853ddc36e46/xxhash.h#L3240-L3265 PiperOrigin-RevId: 391833008 -- 17a4de1f9d623155c75b19285d414cd55a487cd6 by Saleem Abdulrasool <abdulras@google.com>: debugging: add support for unwinding on RISCV Linux This adds partial support for unwinding the RISCV call stack. It is largely duplicated from the AArch64 support with alterations for the ELF RISCV psABI. This covers RISCV64 and RISCV32, though not the ILP32E calling convention. PiperOrigin-RevId: 391818522 -- 32c93e449327b2cea32b32f6365e84b420fe1ed3 by Gennadiy Rozental <rogeeff@google.com>: New storage for types smaller than 8 bytes. Also adding new read interface for types smaller than or rqual to 8 bytes to avoid passing the pointer. PiperOrigin-RevId: 391726822 -- e987ac08a7787801cbfc7d7c96649e97fa8cff1a by Abseil Team <absl-team@google.com>: Extern template `find_first_non_full` to reduce linkage size for TU with single not inlined function. PiperOrigin-RevId: 391718862 -- 73af9bfcb5bf045089133e18bbd20eb5bb699172 by Gennadiy Rozental <rogeeff@google.com>: Make most non-mutable most int128 methods and friend free functions constexpr. Some functions are implemented offline (at least in some configurations) and can't be made constexpr. Mutable methods can't be made constexpr until we drop c++11 support. Fixes #978 PiperOrigin-RevId: 391706535 GitOrigin-RevId: 04cb3b22497190170aa5b774e98080c5de2ba60b Change-Id: If051fad5ff004e2e82fa53618fc04a6fe3d2d4be
3 years ago
// Load the current value to avoid setting 'init' bit manualy.
int64_t one_word_val = OneWordValue().load(std::memory_order_acquire);
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
std::memcpy(&one_word_val, src, Sizeof(op_));
OneWordValue().store(one_word_val, std::memory_order_release);
seq_lock_.IncrementModificationCount();
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
break;
}
case FlagValueStorageKind::kSequenceLocked: {
seq_lock_.Write(AtomicBufferValue(), src, Sizeof(op_));
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
break;
}
Export of internal Abseil changes -- 04cb3b22497190170aa5b774e98080c5de2ba60b by Abseil Team <absl-team@google.com>: Alternative bit mixer for LowLevelHash on ARM LowLevelHash's bit-mixer is inefficient on ARM because it calculates a 128-bit product of two 64-bit numbers. On ARM, this requires a sequence of two instructions with a high combined latency and poor throughput. This change provides alternative bit-mixing code for ARM that uses only 64-bit arithmetic (multiplication, xor, and left-shifts) and speeds things up considerably. The bit-mixing code for ARM was inspired by by Woothash[1] and xxh3[1]. Once I landed on a sequence of operations that provided good mixing, I used a test harness to search for the combination of shift / rotate factors that provided the best mixing, as indicated by SMHasher hash quality tests. The new mixing code passes 13 out of 15 of the hash quality test suites in SMHasher, with the two failures being in the noise range: e.g. 1 collision vs. zero expected in a keyset of ~8m keys. [1]: https://github.com/tommyettinger/waterhash/blob/49f5cf0b63b9/woothash.h#L16-L20 [2]: https://github.com/Cyan4973/xxHash/blob/6853ddc36e46/xxhash.h#L3240-L3265 PiperOrigin-RevId: 391833008 -- 17a4de1f9d623155c75b19285d414cd55a487cd6 by Saleem Abdulrasool <abdulras@google.com>: debugging: add support for unwinding on RISCV Linux This adds partial support for unwinding the RISCV call stack. It is largely duplicated from the AArch64 support with alterations for the ELF RISCV psABI. This covers RISCV64 and RISCV32, though not the ILP32E calling convention. PiperOrigin-RevId: 391818522 -- 32c93e449327b2cea32b32f6365e84b420fe1ed3 by Gennadiy Rozental <rogeeff@google.com>: New storage for types smaller than 8 bytes. Also adding new read interface for types smaller than or rqual to 8 bytes to avoid passing the pointer. PiperOrigin-RevId: 391726822 -- e987ac08a7787801cbfc7d7c96649e97fa8cff1a by Abseil Team <absl-team@google.com>: Extern template `find_first_non_full` to reduce linkage size for TU with single not inlined function. PiperOrigin-RevId: 391718862 -- 73af9bfcb5bf045089133e18bbd20eb5bb699172 by Gennadiy Rozental <rogeeff@google.com>: Make most non-mutable most int128 methods and friend free functions constexpr. Some functions are implemented offline (at least in some configurations) and can't be made constexpr. Mutable methods can't be made constexpr until we drop c++11 support. Fixes #978 PiperOrigin-RevId: 391706535 GitOrigin-RevId: 04cb3b22497190170aa5b774e98080c5de2ba60b Change-Id: If051fad5ff004e2e82fa53618fc04a6fe3d2d4be
3 years ago
case FlagValueStorageKind::kAlignedBuffer:
Copy(op_, src, AlignedBufferValue());
seq_lock_.IncrementModificationCount();
break;
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
}
modified_ = true;
InvokeCallback();
}
absl::string_view FlagImpl::Name() const { return name_; }
std::string FlagImpl::Filename() const {
return flags_internal::GetUsageConfig().normalize_filename(filename_);
}
std::string FlagImpl::Help() const {
return HelpSourceKind() == FlagHelpKind::kLiteral ? help_.literal
: help_.gen_func();
}
FlagFastTypeId FlagImpl::TypeId() const {
return flags_internal::FastTypeId(op_);
}
int64_t FlagImpl::ModificationCount() const {
return seq_lock_.ModificationCount();
}
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
bool FlagImpl::IsSpecifiedOnCommandLine() const {
absl::MutexLock l(DataGuard());
return on_command_line_;
}
std::string FlagImpl::DefaultValue() const {
absl::MutexLock l(DataGuard());
auto obj = MakeInitValue();
return flags_internal::Unparse(op_, obj.get());
}
std::string FlagImpl::CurrentValue() const {
auto* guard = DataGuard(); // Make sure flag initialized
switch (ValueStorageKind()) {
Export of internal Abseil changes -- 04cb3b22497190170aa5b774e98080c5de2ba60b by Abseil Team <absl-team@google.com>: Alternative bit mixer for LowLevelHash on ARM LowLevelHash's bit-mixer is inefficient on ARM because it calculates a 128-bit product of two 64-bit numbers. On ARM, this requires a sequence of two instructions with a high combined latency and poor throughput. This change provides alternative bit-mixing code for ARM that uses only 64-bit arithmetic (multiplication, xor, and left-shifts) and speeds things up considerably. The bit-mixing code for ARM was inspired by by Woothash[1] and xxh3[1]. Once I landed on a sequence of operations that provided good mixing, I used a test harness to search for the combination of shift / rotate factors that provided the best mixing, as indicated by SMHasher hash quality tests. The new mixing code passes 13 out of 15 of the hash quality test suites in SMHasher, with the two failures being in the noise range: e.g. 1 collision vs. zero expected in a keyset of ~8m keys. [1]: https://github.com/tommyettinger/waterhash/blob/49f5cf0b63b9/woothash.h#L16-L20 [2]: https://github.com/Cyan4973/xxHash/blob/6853ddc36e46/xxhash.h#L3240-L3265 PiperOrigin-RevId: 391833008 -- 17a4de1f9d623155c75b19285d414cd55a487cd6 by Saleem Abdulrasool <abdulras@google.com>: debugging: add support for unwinding on RISCV Linux This adds partial support for unwinding the RISCV call stack. It is largely duplicated from the AArch64 support with alterations for the ELF RISCV psABI. This covers RISCV64 and RISCV32, though not the ILP32E calling convention. PiperOrigin-RevId: 391818522 -- 32c93e449327b2cea32b32f6365e84b420fe1ed3 by Gennadiy Rozental <rogeeff@google.com>: New storage for types smaller than 8 bytes. Also adding new read interface for types smaller than or rqual to 8 bytes to avoid passing the pointer. PiperOrigin-RevId: 391726822 -- e987ac08a7787801cbfc7d7c96649e97fa8cff1a by Abseil Team <absl-team@google.com>: Extern template `find_first_non_full` to reduce linkage size for TU with single not inlined function. PiperOrigin-RevId: 391718862 -- 73af9bfcb5bf045089133e18bbd20eb5bb699172 by Gennadiy Rozental <rogeeff@google.com>: Make most non-mutable most int128 methods and friend free functions constexpr. Some functions are implemented offline (at least in some configurations) and can't be made constexpr. Mutable methods can't be made constexpr until we drop c++11 support. Fixes #978 PiperOrigin-RevId: 391706535 GitOrigin-RevId: 04cb3b22497190170aa5b774e98080c5de2ba60b Change-Id: If051fad5ff004e2e82fa53618fc04a6fe3d2d4be
3 years ago
case FlagValueStorageKind::kValueAndInitBit:
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
case FlagValueStorageKind::kOneWordAtomic: {
const auto one_word_val =
absl::bit_cast<std::array<char, sizeof(int64_t)>>(
OneWordValue().load(std::memory_order_acquire));
return flags_internal::Unparse(op_, one_word_val.data());
}
case FlagValueStorageKind::kSequenceLocked: {
std::unique_ptr<void, DynValueDeleter> cloned(flags_internal::Alloc(op_),
DynValueDeleter{op_});
ReadSequenceLockedData(cloned.get());
return flags_internal::Unparse(op_, cloned.get());
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 -- 04cb3b22497190170aa5b774e98080c5de2ba60b by Abseil Team <absl-team@google.com>: Alternative bit mixer for LowLevelHash on ARM LowLevelHash's bit-mixer is inefficient on ARM because it calculates a 128-bit product of two 64-bit numbers. On ARM, this requires a sequence of two instructions with a high combined latency and poor throughput. This change provides alternative bit-mixing code for ARM that uses only 64-bit arithmetic (multiplication, xor, and left-shifts) and speeds things up considerably. The bit-mixing code for ARM was inspired by by Woothash[1] and xxh3[1]. Once I landed on a sequence of operations that provided good mixing, I used a test harness to search for the combination of shift / rotate factors that provided the best mixing, as indicated by SMHasher hash quality tests. The new mixing code passes 13 out of 15 of the hash quality test suites in SMHasher, with the two failures being in the noise range: e.g. 1 collision vs. zero expected in a keyset of ~8m keys. [1]: https://github.com/tommyettinger/waterhash/blob/49f5cf0b63b9/woothash.h#L16-L20 [2]: https://github.com/Cyan4973/xxHash/blob/6853ddc36e46/xxhash.h#L3240-L3265 PiperOrigin-RevId: 391833008 -- 17a4de1f9d623155c75b19285d414cd55a487cd6 by Saleem Abdulrasool <abdulras@google.com>: debugging: add support for unwinding on RISCV Linux This adds partial support for unwinding the RISCV call stack. It is largely duplicated from the AArch64 support with alterations for the ELF RISCV psABI. This covers RISCV64 and RISCV32, though not the ILP32E calling convention. PiperOrigin-RevId: 391818522 -- 32c93e449327b2cea32b32f6365e84b420fe1ed3 by Gennadiy Rozental <rogeeff@google.com>: New storage for types smaller than 8 bytes. Also adding new read interface for types smaller than or rqual to 8 bytes to avoid passing the pointer. PiperOrigin-RevId: 391726822 -- e987ac08a7787801cbfc7d7c96649e97fa8cff1a by Abseil Team <absl-team@google.com>: Extern template `find_first_non_full` to reduce linkage size for TU with single not inlined function. PiperOrigin-RevId: 391718862 -- 73af9bfcb5bf045089133e18bbd20eb5bb699172 by Gennadiy Rozental <rogeeff@google.com>: Make most non-mutable most int128 methods and friend free functions constexpr. Some functions are implemented offline (at least in some configurations) and can't be made constexpr. Mutable methods can't be made constexpr until we drop c++11 support. Fixes #978 PiperOrigin-RevId: 391706535 GitOrigin-RevId: 04cb3b22497190170aa5b774e98080c5de2ba60b Change-Id: If051fad5ff004e2e82fa53618fc04a6fe3d2d4be
3 years ago
case FlagValueStorageKind::kAlignedBuffer: {
absl::MutexLock l(guard);
return flags_internal::Unparse(op_, AlignedBufferValue());
}
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
}
return "";
}
void FlagImpl::SetCallback(const FlagCallbackFunc mutation_callback) {
absl::MutexLock l(DataGuard());
if (callback_ == nullptr) {
callback_ = new FlagCallback;
}
callback_->func = mutation_callback;
InvokeCallback();
}
void FlagImpl::InvokeCallback() const {
if (!callback_) return;
// Make a copy of the C-style function pointer that we are about to invoke
// before we release the lock guarding it.
FlagCallbackFunc cb = callback_->func;
// If the flag has a mutation callback this function invokes it. While the
// callback is being invoked the primary flag's mutex is unlocked and it is
// re-locked back after call to callback is completed. Callback invocation is
// guarded by flag's secondary mutex instead which prevents concurrent
// callback invocation. Note that it is possible for other thread to grab the
// primary lock and update flag's value at any time during the callback
// invocation. This is by design. Callback can get a value of the flag if
// necessary, but it might be different from the value initiated the callback
// and it also can be different by the time the callback invocation is
// completed. Requires that *primary_lock be held in exclusive mode; it may be
// released and reacquired by the implementation.
Export of internal Abseil changes -- 7d0468a6610ed85586d5c87fd65de8dac5118923 by Derek Mauro <dmauro@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 313226473 -- 1131ef6d116f5ce7d46537a82f300ea06dcaaa53 by Gennadiy Rozental <rogeeff@google.com>: Migrate internal interface to use mutable references. PiperOrigin-RevId: 312931131 -- 96225212a9f5fbd0b38c71fe65539164992c7c3b by Laramie Leavitt <lar@google.com>: Remove random/internal/distributions.h This file was something of an historical artifact. All of the related code has either been removed or migraged, and so the only remaining type belongs with uniform_helper.h, as it is used to infer the return type of the absl::Uniform method in a few cases. PiperOrigin-RevId: 312878173 -- 6dcbd5be58ad425e08740ff64088373ee7fe4a72 by Mark Barolak <mbar@google.com>: Release the StrFormat test case for Cords to open source. PiperOrigin-RevId: 312707974 -- 34484d18dfb63a0a7ad6e2aaeb570e33592968be by Abseil Team <absl-team@google.com>: Let Cord::Cord(string&&), Cord::operator=(string&&), Cord::Append(string&&), and Cord::Prepend(string&&) steal string data and embed it into the Cord as a single external chunk, instead of copying it into flat chunks (at most 4083-byte each). Stealing string data is faster, but it creates a long chunk, which leads to a higher more memory usage if its subcords are created and outlive the whole Cord. These functions revert to copying the data if any of the following conditions holds: - string size is at most kMaxBytesToCopy (511), to avoid the overhead of an external chunk for short strings; - less than half of string capacity is used, to avoid pinning to much unused memory. PiperOrigin-RevId: 312683785 GitOrigin-RevId: 7d0468a6610ed85586d5c87fd65de8dac5118923 Change-Id: If79b5a1dfe6d53a8ddddbc7da84338f11fc4cfa3
5 years ago
MutexRelock relock(*DataGuard());
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::MutexLock lock(&callback_->guard);
cb();
}
std::unique_ptr<FlagStateInterface> FlagImpl::SaveState() {
absl::MutexLock l(DataGuard());
bool modified = modified_;
bool on_command_line = on_command_line_;
switch (ValueStorageKind()) {
Export of internal Abseil changes -- 04cb3b22497190170aa5b774e98080c5de2ba60b by Abseil Team <absl-team@google.com>: Alternative bit mixer for LowLevelHash on ARM LowLevelHash's bit-mixer is inefficient on ARM because it calculates a 128-bit product of two 64-bit numbers. On ARM, this requires a sequence of two instructions with a high combined latency and poor throughput. This change provides alternative bit-mixing code for ARM that uses only 64-bit arithmetic (multiplication, xor, and left-shifts) and speeds things up considerably. The bit-mixing code for ARM was inspired by by Woothash[1] and xxh3[1]. Once I landed on a sequence of operations that provided good mixing, I used a test harness to search for the combination of shift / rotate factors that provided the best mixing, as indicated by SMHasher hash quality tests. The new mixing code passes 13 out of 15 of the hash quality test suites in SMHasher, with the two failures being in the noise range: e.g. 1 collision vs. zero expected in a keyset of ~8m keys. [1]: https://github.com/tommyettinger/waterhash/blob/49f5cf0b63b9/woothash.h#L16-L20 [2]: https://github.com/Cyan4973/xxHash/blob/6853ddc36e46/xxhash.h#L3240-L3265 PiperOrigin-RevId: 391833008 -- 17a4de1f9d623155c75b19285d414cd55a487cd6 by Saleem Abdulrasool <abdulras@google.com>: debugging: add support for unwinding on RISCV Linux This adds partial support for unwinding the RISCV call stack. It is largely duplicated from the AArch64 support with alterations for the ELF RISCV psABI. This covers RISCV64 and RISCV32, though not the ILP32E calling convention. PiperOrigin-RevId: 391818522 -- 32c93e449327b2cea32b32f6365e84b420fe1ed3 by Gennadiy Rozental <rogeeff@google.com>: New storage for types smaller than 8 bytes. Also adding new read interface for types smaller than or rqual to 8 bytes to avoid passing the pointer. PiperOrigin-RevId: 391726822 -- e987ac08a7787801cbfc7d7c96649e97fa8cff1a by Abseil Team <absl-team@google.com>: Extern template `find_first_non_full` to reduce linkage size for TU with single not inlined function. PiperOrigin-RevId: 391718862 -- 73af9bfcb5bf045089133e18bbd20eb5bb699172 by Gennadiy Rozental <rogeeff@google.com>: Make most non-mutable most int128 methods and friend free functions constexpr. Some functions are implemented offline (at least in some configurations) and can't be made constexpr. Mutable methods can't be made constexpr until we drop c++11 support. Fixes #978 PiperOrigin-RevId: 391706535 GitOrigin-RevId: 04cb3b22497190170aa5b774e98080c5de2ba60b Change-Id: If051fad5ff004e2e82fa53618fc04a6fe3d2d4be
3 years ago
case FlagValueStorageKind::kValueAndInitBit:
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
case FlagValueStorageKind::kOneWordAtomic: {
return absl::make_unique<FlagState>(
Export of internal Abseil changes -- 7d0468a6610ed85586d5c87fd65de8dac5118923 by Derek Mauro <dmauro@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 313226473 -- 1131ef6d116f5ce7d46537a82f300ea06dcaaa53 by Gennadiy Rozental <rogeeff@google.com>: Migrate internal interface to use mutable references. PiperOrigin-RevId: 312931131 -- 96225212a9f5fbd0b38c71fe65539164992c7c3b by Laramie Leavitt <lar@google.com>: Remove random/internal/distributions.h This file was something of an historical artifact. All of the related code has either been removed or migraged, and so the only remaining type belongs with uniform_helper.h, as it is used to infer the return type of the absl::Uniform method in a few cases. PiperOrigin-RevId: 312878173 -- 6dcbd5be58ad425e08740ff64088373ee7fe4a72 by Mark Barolak <mbar@google.com>: Release the StrFormat test case for Cords to open source. PiperOrigin-RevId: 312707974 -- 34484d18dfb63a0a7ad6e2aaeb570e33592968be by Abseil Team <absl-team@google.com>: Let Cord::Cord(string&&), Cord::operator=(string&&), Cord::Append(string&&), and Cord::Prepend(string&&) steal string data and embed it into the Cord as a single external chunk, instead of copying it into flat chunks (at most 4083-byte each). Stealing string data is faster, but it creates a long chunk, which leads to a higher more memory usage if its subcords are created and outlive the whole Cord. These functions revert to copying the data if any of the following conditions holds: - string size is at most kMaxBytesToCopy (511), to avoid the overhead of an external chunk for short strings; - less than half of string capacity is used, to avoid pinning to much unused memory. PiperOrigin-RevId: 312683785 GitOrigin-RevId: 7d0468a6610ed85586d5c87fd65de8dac5118923 Change-Id: If79b5a1dfe6d53a8ddddbc7da84338f11fc4cfa3
5 years ago
*this, OneWordValue().load(std::memory_order_acquire), modified,
on_command_line, ModificationCount());
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
}
case FlagValueStorageKind::kSequenceLocked: {
void* cloned = flags_internal::Alloc(op_);
// Read is guaranteed to be successful because we hold the lock.
bool success =
seq_lock_.TryRead(cloned, AtomicBufferValue(), Sizeof(op_));
assert(success);
static_cast<void>(success);
return absl::make_unique<FlagState>(*this, cloned, modified,
on_command_line, ModificationCount());
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 -- 04cb3b22497190170aa5b774e98080c5de2ba60b by Abseil Team <absl-team@google.com>: Alternative bit mixer for LowLevelHash on ARM LowLevelHash's bit-mixer is inefficient on ARM because it calculates a 128-bit product of two 64-bit numbers. On ARM, this requires a sequence of two instructions with a high combined latency and poor throughput. This change provides alternative bit-mixing code for ARM that uses only 64-bit arithmetic (multiplication, xor, and left-shifts) and speeds things up considerably. The bit-mixing code for ARM was inspired by by Woothash[1] and xxh3[1]. Once I landed on a sequence of operations that provided good mixing, I used a test harness to search for the combination of shift / rotate factors that provided the best mixing, as indicated by SMHasher hash quality tests. The new mixing code passes 13 out of 15 of the hash quality test suites in SMHasher, with the two failures being in the noise range: e.g. 1 collision vs. zero expected in a keyset of ~8m keys. [1]: https://github.com/tommyettinger/waterhash/blob/49f5cf0b63b9/woothash.h#L16-L20 [2]: https://github.com/Cyan4973/xxHash/blob/6853ddc36e46/xxhash.h#L3240-L3265 PiperOrigin-RevId: 391833008 -- 17a4de1f9d623155c75b19285d414cd55a487cd6 by Saleem Abdulrasool <abdulras@google.com>: debugging: add support for unwinding on RISCV Linux This adds partial support for unwinding the RISCV call stack. It is largely duplicated from the AArch64 support with alterations for the ELF RISCV psABI. This covers RISCV64 and RISCV32, though not the ILP32E calling convention. PiperOrigin-RevId: 391818522 -- 32c93e449327b2cea32b32f6365e84b420fe1ed3 by Gennadiy Rozental <rogeeff@google.com>: New storage for types smaller than 8 bytes. Also adding new read interface for types smaller than or rqual to 8 bytes to avoid passing the pointer. PiperOrigin-RevId: 391726822 -- e987ac08a7787801cbfc7d7c96649e97fa8cff1a by Abseil Team <absl-team@google.com>: Extern template `find_first_non_full` to reduce linkage size for TU with single not inlined function. PiperOrigin-RevId: 391718862 -- 73af9bfcb5bf045089133e18bbd20eb5bb699172 by Gennadiy Rozental <rogeeff@google.com>: Make most non-mutable most int128 methods and friend free functions constexpr. Some functions are implemented offline (at least in some configurations) and can't be made constexpr. Mutable methods can't be made constexpr until we drop c++11 support. Fixes #978 PiperOrigin-RevId: 391706535 GitOrigin-RevId: 04cb3b22497190170aa5b774e98080c5de2ba60b Change-Id: If051fad5ff004e2e82fa53618fc04a6fe3d2d4be
3 years ago
case FlagValueStorageKind::kAlignedBuffer: {
return absl::make_unique<FlagState>(
*this, flags_internal::Clone(op_, AlignedBufferValue()), modified,
on_command_line, ModificationCount());
}
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
}
return nullptr;
}
bool FlagImpl::RestoreState(const FlagState& flag_state) {
absl::MutexLock l(DataGuard());
if (flag_state.counter_ == ModificationCount()) {
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
return false;
}
switch (ValueStorageKind()) {
Export of internal Abseil changes -- 04cb3b22497190170aa5b774e98080c5de2ba60b by Abseil Team <absl-team@google.com>: Alternative bit mixer for LowLevelHash on ARM LowLevelHash's bit-mixer is inefficient on ARM because it calculates a 128-bit product of two 64-bit numbers. On ARM, this requires a sequence of two instructions with a high combined latency and poor throughput. This change provides alternative bit-mixing code for ARM that uses only 64-bit arithmetic (multiplication, xor, and left-shifts) and speeds things up considerably. The bit-mixing code for ARM was inspired by by Woothash[1] and xxh3[1]. Once I landed on a sequence of operations that provided good mixing, I used a test harness to search for the combination of shift / rotate factors that provided the best mixing, as indicated by SMHasher hash quality tests. The new mixing code passes 13 out of 15 of the hash quality test suites in SMHasher, with the two failures being in the noise range: e.g. 1 collision vs. zero expected in a keyset of ~8m keys. [1]: https://github.com/tommyettinger/waterhash/blob/49f5cf0b63b9/woothash.h#L16-L20 [2]: https://github.com/Cyan4973/xxHash/blob/6853ddc36e46/xxhash.h#L3240-L3265 PiperOrigin-RevId: 391833008 -- 17a4de1f9d623155c75b19285d414cd55a487cd6 by Saleem Abdulrasool <abdulras@google.com>: debugging: add support for unwinding on RISCV Linux This adds partial support for unwinding the RISCV call stack. It is largely duplicated from the AArch64 support with alterations for the ELF RISCV psABI. This covers RISCV64 and RISCV32, though not the ILP32E calling convention. PiperOrigin-RevId: 391818522 -- 32c93e449327b2cea32b32f6365e84b420fe1ed3 by Gennadiy Rozental <rogeeff@google.com>: New storage for types smaller than 8 bytes. Also adding new read interface for types smaller than or rqual to 8 bytes to avoid passing the pointer. PiperOrigin-RevId: 391726822 -- e987ac08a7787801cbfc7d7c96649e97fa8cff1a by Abseil Team <absl-team@google.com>: Extern template `find_first_non_full` to reduce linkage size for TU with single not inlined function. PiperOrigin-RevId: 391718862 -- 73af9bfcb5bf045089133e18bbd20eb5bb699172 by Gennadiy Rozental <rogeeff@google.com>: Make most non-mutable most int128 methods and friend free functions constexpr. Some functions are implemented offline (at least in some configurations) and can't be made constexpr. Mutable methods can't be made constexpr until we drop c++11 support. Fixes #978 PiperOrigin-RevId: 391706535 GitOrigin-RevId: 04cb3b22497190170aa5b774e98080c5de2ba60b Change-Id: If051fad5ff004e2e82fa53618fc04a6fe3d2d4be
3 years ago
case FlagValueStorageKind::kValueAndInitBit:
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
case FlagValueStorageKind::kOneWordAtomic:
StoreValue(&flag_state.value_.one_word);
break;
Export of internal Abseil changes -- 04cb3b22497190170aa5b774e98080c5de2ba60b by Abseil Team <absl-team@google.com>: Alternative bit mixer for LowLevelHash on ARM LowLevelHash's bit-mixer is inefficient on ARM because it calculates a 128-bit product of two 64-bit numbers. On ARM, this requires a sequence of two instructions with a high combined latency and poor throughput. This change provides alternative bit-mixing code for ARM that uses only 64-bit arithmetic (multiplication, xor, and left-shifts) and speeds things up considerably. The bit-mixing code for ARM was inspired by by Woothash[1] and xxh3[1]. Once I landed on a sequence of operations that provided good mixing, I used a test harness to search for the combination of shift / rotate factors that provided the best mixing, as indicated by SMHasher hash quality tests. The new mixing code passes 13 out of 15 of the hash quality test suites in SMHasher, with the two failures being in the noise range: e.g. 1 collision vs. zero expected in a keyset of ~8m keys. [1]: https://github.com/tommyettinger/waterhash/blob/49f5cf0b63b9/woothash.h#L16-L20 [2]: https://github.com/Cyan4973/xxHash/blob/6853ddc36e46/xxhash.h#L3240-L3265 PiperOrigin-RevId: 391833008 -- 17a4de1f9d623155c75b19285d414cd55a487cd6 by Saleem Abdulrasool <abdulras@google.com>: debugging: add support for unwinding on RISCV Linux This adds partial support for unwinding the RISCV call stack. It is largely duplicated from the AArch64 support with alterations for the ELF RISCV psABI. This covers RISCV64 and RISCV32, though not the ILP32E calling convention. PiperOrigin-RevId: 391818522 -- 32c93e449327b2cea32b32f6365e84b420fe1ed3 by Gennadiy Rozental <rogeeff@google.com>: New storage for types smaller than 8 bytes. Also adding new read interface for types smaller than or rqual to 8 bytes to avoid passing the pointer. PiperOrigin-RevId: 391726822 -- e987ac08a7787801cbfc7d7c96649e97fa8cff1a by Abseil Team <absl-team@google.com>: Extern template `find_first_non_full` to reduce linkage size for TU with single not inlined function. PiperOrigin-RevId: 391718862 -- 73af9bfcb5bf045089133e18bbd20eb5bb699172 by Gennadiy Rozental <rogeeff@google.com>: Make most non-mutable most int128 methods and friend free functions constexpr. Some functions are implemented offline (at least in some configurations) and can't be made constexpr. Mutable methods can't be made constexpr until we drop c++11 support. Fixes #978 PiperOrigin-RevId: 391706535 GitOrigin-RevId: 04cb3b22497190170aa5b774e98080c5de2ba60b Change-Id: If051fad5ff004e2e82fa53618fc04a6fe3d2d4be
3 years ago
case FlagValueStorageKind::kSequenceLocked:
case FlagValueStorageKind::kAlignedBuffer:
StoreValue(flag_state.value_.heap_allocated);
break;
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
}
modified_ = flag_state.modified_;
on_command_line_ = flag_state.on_command_line_;
return true;
}
template <typename StorageT>
StorageT* FlagImpl::OffsetValue() const {
char* p = reinterpret_cast<char*>(const_cast<FlagImpl*>(this));
// The offset is deduced via Flag value type specific op_.
size_t offset = flags_internal::ValueOffset(op_);
return reinterpret_cast<StorageT*>(p + offset);
}
void* FlagImpl::AlignedBufferValue() const {
assert(ValueStorageKind() == FlagValueStorageKind::kAlignedBuffer);
return OffsetValue<void>();
}
std::atomic<uint64_t>* FlagImpl::AtomicBufferValue() const {
assert(ValueStorageKind() == FlagValueStorageKind::kSequenceLocked);
return OffsetValue<std::atomic<uint64_t>>();
}
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
std::atomic<int64_t>& FlagImpl::OneWordValue() const {
Export of internal Abseil changes -- 04cb3b22497190170aa5b774e98080c5de2ba60b by Abseil Team <absl-team@google.com>: Alternative bit mixer for LowLevelHash on ARM LowLevelHash's bit-mixer is inefficient on ARM because it calculates a 128-bit product of two 64-bit numbers. On ARM, this requires a sequence of two instructions with a high combined latency and poor throughput. This change provides alternative bit-mixing code for ARM that uses only 64-bit arithmetic (multiplication, xor, and left-shifts) and speeds things up considerably. The bit-mixing code for ARM was inspired by by Woothash[1] and xxh3[1]. Once I landed on a sequence of operations that provided good mixing, I used a test harness to search for the combination of shift / rotate factors that provided the best mixing, as indicated by SMHasher hash quality tests. The new mixing code passes 13 out of 15 of the hash quality test suites in SMHasher, with the two failures being in the noise range: e.g. 1 collision vs. zero expected in a keyset of ~8m keys. [1]: https://github.com/tommyettinger/waterhash/blob/49f5cf0b63b9/woothash.h#L16-L20 [2]: https://github.com/Cyan4973/xxHash/blob/6853ddc36e46/xxhash.h#L3240-L3265 PiperOrigin-RevId: 391833008 -- 17a4de1f9d623155c75b19285d414cd55a487cd6 by Saleem Abdulrasool <abdulras@google.com>: debugging: add support for unwinding on RISCV Linux This adds partial support for unwinding the RISCV call stack. It is largely duplicated from the AArch64 support with alterations for the ELF RISCV psABI. This covers RISCV64 and RISCV32, though not the ILP32E calling convention. PiperOrigin-RevId: 391818522 -- 32c93e449327b2cea32b32f6365e84b420fe1ed3 by Gennadiy Rozental <rogeeff@google.com>: New storage for types smaller than 8 bytes. Also adding new read interface for types smaller than or rqual to 8 bytes to avoid passing the pointer. PiperOrigin-RevId: 391726822 -- e987ac08a7787801cbfc7d7c96649e97fa8cff1a by Abseil Team <absl-team@google.com>: Extern template `find_first_non_full` to reduce linkage size for TU with single not inlined function. PiperOrigin-RevId: 391718862 -- 73af9bfcb5bf045089133e18bbd20eb5bb699172 by Gennadiy Rozental <rogeeff@google.com>: Make most non-mutable most int128 methods and friend free functions constexpr. Some functions are implemented offline (at least in some configurations) and can't be made constexpr. Mutable methods can't be made constexpr until we drop c++11 support. Fixes #978 PiperOrigin-RevId: 391706535 GitOrigin-RevId: 04cb3b22497190170aa5b774e98080c5de2ba60b Change-Id: If051fad5ff004e2e82fa53618fc04a6fe3d2d4be
3 years ago
assert(ValueStorageKind() == FlagValueStorageKind::kOneWordAtomic ||
ValueStorageKind() == FlagValueStorageKind::kValueAndInitBit);
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
return OffsetValue<FlagOneWordValue>()->value;
}
// Attempts to parse supplied `value` string using parsing routine in the `flag`
// argument. If parsing successful, this function replaces the dst with newly
// parsed value. In case if any error is encountered in either step, the error
// message is stored in 'err'
std::unique_ptr<void, DynValueDeleter> FlagImpl::TryParse(
Export of internal Abseil changes -- 7d0468a6610ed85586d5c87fd65de8dac5118923 by Derek Mauro <dmauro@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 313226473 -- 1131ef6d116f5ce7d46537a82f300ea06dcaaa53 by Gennadiy Rozental <rogeeff@google.com>: Migrate internal interface to use mutable references. PiperOrigin-RevId: 312931131 -- 96225212a9f5fbd0b38c71fe65539164992c7c3b by Laramie Leavitt <lar@google.com>: Remove random/internal/distributions.h This file was something of an historical artifact. All of the related code has either been removed or migraged, and so the only remaining type belongs with uniform_helper.h, as it is used to infer the return type of the absl::Uniform method in a few cases. PiperOrigin-RevId: 312878173 -- 6dcbd5be58ad425e08740ff64088373ee7fe4a72 by Mark Barolak <mbar@google.com>: Release the StrFormat test case for Cords to open source. PiperOrigin-RevId: 312707974 -- 34484d18dfb63a0a7ad6e2aaeb570e33592968be by Abseil Team <absl-team@google.com>: Let Cord::Cord(string&&), Cord::operator=(string&&), Cord::Append(string&&), and Cord::Prepend(string&&) steal string data and embed it into the Cord as a single external chunk, instead of copying it into flat chunks (at most 4083-byte each). Stealing string data is faster, but it creates a long chunk, which leads to a higher more memory usage if its subcords are created and outlive the whole Cord. These functions revert to copying the data if any of the following conditions holds: - string size is at most kMaxBytesToCopy (511), to avoid the overhead of an external chunk for short strings; - less than half of string capacity is used, to avoid pinning to much unused memory. PiperOrigin-RevId: 312683785 GitOrigin-RevId: 7d0468a6610ed85586d5c87fd65de8dac5118923 Change-Id: If79b5a1dfe6d53a8ddddbc7da84338f11fc4cfa3
5 years ago
absl::string_view value, std::string& err) const {
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
std::unique_ptr<void, DynValueDeleter> tentative_value = MakeInitValue();
std::string parse_err;
if (!flags_internal::Parse(op_, value, tentative_value.get(), &parse_err)) {
absl::string_view err_sep = parse_err.empty() ? "" : "; ";
Export of internal Abseil changes -- 7d0468a6610ed85586d5c87fd65de8dac5118923 by Derek Mauro <dmauro@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 313226473 -- 1131ef6d116f5ce7d46537a82f300ea06dcaaa53 by Gennadiy Rozental <rogeeff@google.com>: Migrate internal interface to use mutable references. PiperOrigin-RevId: 312931131 -- 96225212a9f5fbd0b38c71fe65539164992c7c3b by Laramie Leavitt <lar@google.com>: Remove random/internal/distributions.h This file was something of an historical artifact. All of the related code has either been removed or migraged, and so the only remaining type belongs with uniform_helper.h, as it is used to infer the return type of the absl::Uniform method in a few cases. PiperOrigin-RevId: 312878173 -- 6dcbd5be58ad425e08740ff64088373ee7fe4a72 by Mark Barolak <mbar@google.com>: Release the StrFormat test case for Cords to open source. PiperOrigin-RevId: 312707974 -- 34484d18dfb63a0a7ad6e2aaeb570e33592968be by Abseil Team <absl-team@google.com>: Let Cord::Cord(string&&), Cord::operator=(string&&), Cord::Append(string&&), and Cord::Prepend(string&&) steal string data and embed it into the Cord as a single external chunk, instead of copying it into flat chunks (at most 4083-byte each). Stealing string data is faster, but it creates a long chunk, which leads to a higher more memory usage if its subcords are created and outlive the whole Cord. These functions revert to copying the data if any of the following conditions holds: - string size is at most kMaxBytesToCopy (511), to avoid the overhead of an external chunk for short strings; - less than half of string capacity is used, to avoid pinning to much unused memory. PiperOrigin-RevId: 312683785 GitOrigin-RevId: 7d0468a6610ed85586d5c87fd65de8dac5118923 Change-Id: If79b5a1dfe6d53a8ddddbc7da84338f11fc4cfa3
5 years ago
err = absl::StrCat("Illegal value '", value, "' specified for flag '",
Name(), "'", err_sep, parse_err);
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
return nullptr;
}
return tentative_value;
}
void FlagImpl::Read(void* dst) const {
auto* guard = DataGuard(); // Make sure flag initialized
switch (ValueStorageKind()) {
Export of internal Abseil changes -- 04cb3b22497190170aa5b774e98080c5de2ba60b by Abseil Team <absl-team@google.com>: Alternative bit mixer for LowLevelHash on ARM LowLevelHash's bit-mixer is inefficient on ARM because it calculates a 128-bit product of two 64-bit numbers. On ARM, this requires a sequence of two instructions with a high combined latency and poor throughput. This change provides alternative bit-mixing code for ARM that uses only 64-bit arithmetic (multiplication, xor, and left-shifts) and speeds things up considerably. The bit-mixing code for ARM was inspired by by Woothash[1] and xxh3[1]. Once I landed on a sequence of operations that provided good mixing, I used a test harness to search for the combination of shift / rotate factors that provided the best mixing, as indicated by SMHasher hash quality tests. The new mixing code passes 13 out of 15 of the hash quality test suites in SMHasher, with the two failures being in the noise range: e.g. 1 collision vs. zero expected in a keyset of ~8m keys. [1]: https://github.com/tommyettinger/waterhash/blob/49f5cf0b63b9/woothash.h#L16-L20 [2]: https://github.com/Cyan4973/xxHash/blob/6853ddc36e46/xxhash.h#L3240-L3265 PiperOrigin-RevId: 391833008 -- 17a4de1f9d623155c75b19285d414cd55a487cd6 by Saleem Abdulrasool <abdulras@google.com>: debugging: add support for unwinding on RISCV Linux This adds partial support for unwinding the RISCV call stack. It is largely duplicated from the AArch64 support with alterations for the ELF RISCV psABI. This covers RISCV64 and RISCV32, though not the ILP32E calling convention. PiperOrigin-RevId: 391818522 -- 32c93e449327b2cea32b32f6365e84b420fe1ed3 by Gennadiy Rozental <rogeeff@google.com>: New storage for types smaller than 8 bytes. Also adding new read interface for types smaller than or rqual to 8 bytes to avoid passing the pointer. PiperOrigin-RevId: 391726822 -- e987ac08a7787801cbfc7d7c96649e97fa8cff1a by Abseil Team <absl-team@google.com>: Extern template `find_first_non_full` to reduce linkage size for TU with single not inlined function. PiperOrigin-RevId: 391718862 -- 73af9bfcb5bf045089133e18bbd20eb5bb699172 by Gennadiy Rozental <rogeeff@google.com>: Make most non-mutable most int128 methods and friend free functions constexpr. Some functions are implemented offline (at least in some configurations) and can't be made constexpr. Mutable methods can't be made constexpr until we drop c++11 support. Fixes #978 PiperOrigin-RevId: 391706535 GitOrigin-RevId: 04cb3b22497190170aa5b774e98080c5de2ba60b Change-Id: If051fad5ff004e2e82fa53618fc04a6fe3d2d4be
3 years ago
case FlagValueStorageKind::kValueAndInitBit:
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
case FlagValueStorageKind::kOneWordAtomic: {
const int64_t one_word_val =
OneWordValue().load(std::memory_order_acquire);
std::memcpy(dst, &one_word_val, Sizeof(op_));
break;
}
case FlagValueStorageKind::kSequenceLocked: {
ReadSequenceLockedData(dst);
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
break;
}
Export of internal Abseil changes -- 04cb3b22497190170aa5b774e98080c5de2ba60b by Abseil Team <absl-team@google.com>: Alternative bit mixer for LowLevelHash on ARM LowLevelHash's bit-mixer is inefficient on ARM because it calculates a 128-bit product of two 64-bit numbers. On ARM, this requires a sequence of two instructions with a high combined latency and poor throughput. This change provides alternative bit-mixing code for ARM that uses only 64-bit arithmetic (multiplication, xor, and left-shifts) and speeds things up considerably. The bit-mixing code for ARM was inspired by by Woothash[1] and xxh3[1]. Once I landed on a sequence of operations that provided good mixing, I used a test harness to search for the combination of shift / rotate factors that provided the best mixing, as indicated by SMHasher hash quality tests. The new mixing code passes 13 out of 15 of the hash quality test suites in SMHasher, with the two failures being in the noise range: e.g. 1 collision vs. zero expected in a keyset of ~8m keys. [1]: https://github.com/tommyettinger/waterhash/blob/49f5cf0b63b9/woothash.h#L16-L20 [2]: https://github.com/Cyan4973/xxHash/blob/6853ddc36e46/xxhash.h#L3240-L3265 PiperOrigin-RevId: 391833008 -- 17a4de1f9d623155c75b19285d414cd55a487cd6 by Saleem Abdulrasool <abdulras@google.com>: debugging: add support for unwinding on RISCV Linux This adds partial support for unwinding the RISCV call stack. It is largely duplicated from the AArch64 support with alterations for the ELF RISCV psABI. This covers RISCV64 and RISCV32, though not the ILP32E calling convention. PiperOrigin-RevId: 391818522 -- 32c93e449327b2cea32b32f6365e84b420fe1ed3 by Gennadiy Rozental <rogeeff@google.com>: New storage for types smaller than 8 bytes. Also adding new read interface for types smaller than or rqual to 8 bytes to avoid passing the pointer. PiperOrigin-RevId: 391726822 -- e987ac08a7787801cbfc7d7c96649e97fa8cff1a by Abseil Team <absl-team@google.com>: Extern template `find_first_non_full` to reduce linkage size for TU with single not inlined function. PiperOrigin-RevId: 391718862 -- 73af9bfcb5bf045089133e18bbd20eb5bb699172 by Gennadiy Rozental <rogeeff@google.com>: Make most non-mutable most int128 methods and friend free functions constexpr. Some functions are implemented offline (at least in some configurations) and can't be made constexpr. Mutable methods can't be made constexpr until we drop c++11 support. Fixes #978 PiperOrigin-RevId: 391706535 GitOrigin-RevId: 04cb3b22497190170aa5b774e98080c5de2ba60b Change-Id: If051fad5ff004e2e82fa53618fc04a6fe3d2d4be
3 years ago
case FlagValueStorageKind::kAlignedBuffer: {
absl::MutexLock l(guard);
flags_internal::CopyConstruct(op_, AlignedBufferValue(), dst);
break;
}
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 -- 04cb3b22497190170aa5b774e98080c5de2ba60b by Abseil Team <absl-team@google.com>: Alternative bit mixer for LowLevelHash on ARM LowLevelHash's bit-mixer is inefficient on ARM because it calculates a 128-bit product of two 64-bit numbers. On ARM, this requires a sequence of two instructions with a high combined latency and poor throughput. This change provides alternative bit-mixing code for ARM that uses only 64-bit arithmetic (multiplication, xor, and left-shifts) and speeds things up considerably. The bit-mixing code for ARM was inspired by by Woothash[1] and xxh3[1]. Once I landed on a sequence of operations that provided good mixing, I used a test harness to search for the combination of shift / rotate factors that provided the best mixing, as indicated by SMHasher hash quality tests. The new mixing code passes 13 out of 15 of the hash quality test suites in SMHasher, with the two failures being in the noise range: e.g. 1 collision vs. zero expected in a keyset of ~8m keys. [1]: https://github.com/tommyettinger/waterhash/blob/49f5cf0b63b9/woothash.h#L16-L20 [2]: https://github.com/Cyan4973/xxHash/blob/6853ddc36e46/xxhash.h#L3240-L3265 PiperOrigin-RevId: 391833008 -- 17a4de1f9d623155c75b19285d414cd55a487cd6 by Saleem Abdulrasool <abdulras@google.com>: debugging: add support for unwinding on RISCV Linux This adds partial support for unwinding the RISCV call stack. It is largely duplicated from the AArch64 support with alterations for the ELF RISCV psABI. This covers RISCV64 and RISCV32, though not the ILP32E calling convention. PiperOrigin-RevId: 391818522 -- 32c93e449327b2cea32b32f6365e84b420fe1ed3 by Gennadiy Rozental <rogeeff@google.com>: New storage for types smaller than 8 bytes. Also adding new read interface for types smaller than or rqual to 8 bytes to avoid passing the pointer. PiperOrigin-RevId: 391726822 -- e987ac08a7787801cbfc7d7c96649e97fa8cff1a by Abseil Team <absl-team@google.com>: Extern template `find_first_non_full` to reduce linkage size for TU with single not inlined function. PiperOrigin-RevId: 391718862 -- 73af9bfcb5bf045089133e18bbd20eb5bb699172 by Gennadiy Rozental <rogeeff@google.com>: Make most non-mutable most int128 methods and friend free functions constexpr. Some functions are implemented offline (at least in some configurations) and can't be made constexpr. Mutable methods can't be made constexpr until we drop c++11 support. Fixes #978 PiperOrigin-RevId: 391706535 GitOrigin-RevId: 04cb3b22497190170aa5b774e98080c5de2ba60b Change-Id: If051fad5ff004e2e82fa53618fc04a6fe3d2d4be
3 years ago
int64_t FlagImpl::ReadOneWord() const {
assert(ValueStorageKind() == FlagValueStorageKind::kOneWordAtomic ||
ValueStorageKind() == FlagValueStorageKind::kValueAndInitBit);
auto* guard = DataGuard(); // Make sure flag initialized
(void)guard;
return OneWordValue().load(std::memory_order_acquire);
}
bool FlagImpl::ReadOneBool() const {
assert(ValueStorageKind() == FlagValueStorageKind::kValueAndInitBit);
auto* guard = DataGuard(); // Make sure flag initialized
(void)guard;
return absl::bit_cast<FlagValueAndInitBit<bool>>(
OneWordValue().load(std::memory_order_acquire))
.value;
}
void FlagImpl::ReadSequenceLockedData(void* dst) const {
int size = Sizeof(op_);
// Attempt to read using the sequence lock.
if (ABSL_PREDICT_TRUE(seq_lock_.TryRead(dst, AtomicBufferValue(), size))) {
return;
}
// We failed due to contention. Acquire the lock to prevent contention
// and try again.
absl::ReaderMutexLock l(DataGuard());
bool success = seq_lock_.TryRead(dst, AtomicBufferValue(), size);
assert(success);
static_cast<void>(success);
}
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
void FlagImpl::Write(const void* src) {
absl::MutexLock l(DataGuard());
if (ShouldValidateFlagValue(flags_internal::FastTypeId(op_))) {
std::unique_ptr<void, DynValueDeleter> obj{flags_internal::Clone(op_, src),
DynValueDeleter{op_}};
std::string ignored_error;
std::string src_as_str = flags_internal::Unparse(op_, src);
if (!flags_internal::Parse(op_, src_as_str, obj.get(), &ignored_error)) {
ABSL_INTERNAL_LOG(ERROR, absl::StrCat("Attempt to set flag '", Name(),
"' to invalid value ", src_as_str));
}
}
StoreValue(src);
}
// Sets the value of the flag based on specified string `value`. If the flag
// was successfully set to new value, it returns true. Otherwise, sets `err`
// to indicate the error, leaves the flag unchanged, and returns false. There
// are three ways to set the flag's value:
// * Update the current flag value
// * Update the flag's default value
// * Update the current flag value if it was never set before
// The mode is selected based on 'set_mode' parameter.
bool FlagImpl::ParseFrom(absl::string_view value, FlagSettingMode set_mode,
Export of internal Abseil changes -- 7d0468a6610ed85586d5c87fd65de8dac5118923 by Derek Mauro <dmauro@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 313226473 -- 1131ef6d116f5ce7d46537a82f300ea06dcaaa53 by Gennadiy Rozental <rogeeff@google.com>: Migrate internal interface to use mutable references. PiperOrigin-RevId: 312931131 -- 96225212a9f5fbd0b38c71fe65539164992c7c3b by Laramie Leavitt <lar@google.com>: Remove random/internal/distributions.h This file was something of an historical artifact. All of the related code has either been removed or migraged, and so the only remaining type belongs with uniform_helper.h, as it is used to infer the return type of the absl::Uniform method in a few cases. PiperOrigin-RevId: 312878173 -- 6dcbd5be58ad425e08740ff64088373ee7fe4a72 by Mark Barolak <mbar@google.com>: Release the StrFormat test case for Cords to open source. PiperOrigin-RevId: 312707974 -- 34484d18dfb63a0a7ad6e2aaeb570e33592968be by Abseil Team <absl-team@google.com>: Let Cord::Cord(string&&), Cord::operator=(string&&), Cord::Append(string&&), and Cord::Prepend(string&&) steal string data and embed it into the Cord as a single external chunk, instead of copying it into flat chunks (at most 4083-byte each). Stealing string data is faster, but it creates a long chunk, which leads to a higher more memory usage if its subcords are created and outlive the whole Cord. These functions revert to copying the data if any of the following conditions holds: - string size is at most kMaxBytesToCopy (511), to avoid the overhead of an external chunk for short strings; - less than half of string capacity is used, to avoid pinning to much unused memory. PiperOrigin-RevId: 312683785 GitOrigin-RevId: 7d0468a6610ed85586d5c87fd65de8dac5118923 Change-Id: If79b5a1dfe6d53a8ddddbc7da84338f11fc4cfa3
5 years ago
ValueSource source, std::string& err) {
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::MutexLock l(DataGuard());
switch (set_mode) {
case SET_FLAGS_VALUE: {
// set or modify the flag's value
auto tentative_value = TryParse(value, err);
if (!tentative_value) return false;
StoreValue(tentative_value.get());
if (source == kCommandLine) {
on_command_line_ = true;
}
break;
}
case SET_FLAG_IF_DEFAULT: {
// set the flag's value, but only if it hasn't been set by someone else
if (modified_) {
// TODO(rogeeff): review and fix this semantic. Currently we do not fail
// in this case if flag is modified. This is misleading since the flag's
// value is not updated even though we return true.
// *err = absl::StrCat(Name(), " is already set to ",
// CurrentValue(), "\n");
// return false;
return true;
}
auto tentative_value = TryParse(value, err);
if (!tentative_value) return false;
StoreValue(tentative_value.get());
break;
}
case SET_FLAGS_DEFAULT: {
auto tentative_value = TryParse(value, err);
if (!tentative_value) return false;
if (DefaultKind() == FlagDefaultKind::kDynamicValue) {
void* old_value = default_value_.dynamic_value;
default_value_.dynamic_value = tentative_value.release();
tentative_value.reset(old_value);
} else {
default_value_.dynamic_value = tentative_value.release();
def_kind_ = static_cast<uint8_t>(FlagDefaultKind::kDynamicValue);
}
if (!modified_) {
// Need to set both default value *and* current, in this case.
StoreValue(default_value_.dynamic_value);
modified_ = false;
}
break;
}
}
return true;
}
void FlagImpl::CheckDefaultValueParsingRoundtrip() const {
std::string v = DefaultValue();
absl::MutexLock lock(DataGuard());
auto dst = MakeInitValue();
std::string error;
if (!flags_internal::Parse(op_, v, dst.get(), &error)) {
ABSL_INTERNAL_LOG(
FATAL,
absl::StrCat("Flag ", Name(), " (from ", Filename(),
"): string form of default value '", v,
"' could not be parsed; error=", error));
}
// We do not compare dst to def since parsing/unparsing may make
// small changes, e.g., precision loss for floating point types.
}
bool FlagImpl::ValidateInputValue(absl::string_view value) const {
absl::MutexLock l(DataGuard());
auto obj = MakeInitValue();
std::string ignored_error;
return flags_internal::Parse(op_, value, obj.get(), &ignored_error);
}
} // namespace flags_internal
ABSL_NAMESPACE_END
} // namespace absl