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.

604 lines
21 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 2018 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/debugging/symbolize.h"
#ifndef _WIN32
#include <fcntl.h>
#include <sys/mman.h>
#endif
#include <cstring>
#include <iostream>
#include <memory>
#include "gmock/gmock.h"
#include "gtest/gtest.h"
#include "absl/base/attributes.h"
#include "absl/base/casts.h"
#include "absl/base/config.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/internal/per_thread_tls.h"
#include "absl/base/internal/raw_logging.h"
#include "absl/base/optimization.h"
#include "absl/debugging/internal/stack_consumption.h"
#include "absl/memory/memory.h"
#include "absl/strings/string_view.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
using testing::Contains;
#ifdef _WIN32
#define ABSL_SYMBOLIZE_TEST_NOINLINE __declspec(noinline)
#else
#define ABSL_SYMBOLIZE_TEST_NOINLINE ABSL_ATTRIBUTE_NOINLINE
#endif
// Functions to symbolize. Use C linkage to avoid mangled names.
extern "C" {
ABSL_SYMBOLIZE_TEST_NOINLINE void nonstatic_func() {
// The next line makes this a unique function to prevent the compiler from
// folding identical functions together.
volatile int x = __LINE__;
static_cast<void>(x);
ABSL_BLOCK_TAIL_CALL_OPTIMIZATION();
}
ABSL_SYMBOLIZE_TEST_NOINLINE static void static_func() {
// The next line makes this a unique function to prevent the compiler from
// folding identical functions together.
volatile int x = __LINE__;
static_cast<void>(x);
ABSL_BLOCK_TAIL_CALL_OPTIMIZATION();
}
} // extern "C"
struct Foo {
static void func(int x);
};
// A C++ method that should have a mangled name.
ABSL_SYMBOLIZE_TEST_NOINLINE void Foo::func(int) {
// The next line makes this a unique function to prevent the compiler from
// folding identical functions together.
volatile int x = __LINE__;
static_cast<void>(x);
ABSL_BLOCK_TAIL_CALL_OPTIMIZATION();
}
// Create functions that will remain in different text sections in the
// final binary when linker option "-z,keep-text-section-prefix" is used.
int ABSL_ATTRIBUTE_SECTION_VARIABLE(.text.unlikely) unlikely_func() {
return 0;
}
int ABSL_ATTRIBUTE_SECTION_VARIABLE(.text.hot) hot_func() {
return 0;
}
int ABSL_ATTRIBUTE_SECTION_VARIABLE(.text.startup) startup_func() {
return 0;
}
int ABSL_ATTRIBUTE_SECTION_VARIABLE(.text.exit) exit_func() {
return 0;
}
int /*ABSL_ATTRIBUTE_SECTION_VARIABLE(.text)*/ regular_func() {
return 0;
}
// Thread-local data may confuse the symbolizer, ensure that it does not.
// Variable sizes and order are important.
#if ABSL_PER_THREAD_TLS
static ABSL_PER_THREAD_TLS_KEYWORD char symbolize_test_thread_small[1];
static ABSL_PER_THREAD_TLS_KEYWORD char
symbolize_test_thread_big[2 * 1024 * 1024];
#endif
#if !defined(__EMSCRIPTEN__)
// Used below to hopefully inhibit some compiler/linker optimizations
// that may remove kHpageTextPadding, kPadding0, and kPadding1 from
// the binary.
static volatile bool volatile_bool = false;
// Force the binary to be large enough that a THP .text remap will succeed.
static constexpr size_t kHpageSize = 1 << 21;
const char kHpageTextPadding[kHpageSize * 4] ABSL_ATTRIBUTE_SECTION_VARIABLE(
.text) = "";
#endif // !defined(__EMSCRIPTEN__)
static char try_symbolize_buffer[4096];
// A wrapper function for absl::Symbolize() to make the unit test simple. The
// limit must be < sizeof(try_symbolize_buffer). Returns null if
// absl::Symbolize() returns false, otherwise returns try_symbolize_buffer with
// the result of absl::Symbolize().
static const char *TrySymbolizeWithLimit(void *pc, int limit) {
ABSL_RAW_CHECK(limit <= sizeof(try_symbolize_buffer),
"try_symbolize_buffer is too small");
// Use the heap to facilitate heap and buffer sanitizer tools.
auto heap_buffer = absl::make_unique<char[]>(sizeof(try_symbolize_buffer));
bool found = absl::Symbolize(pc, heap_buffer.get(), limit);
if (found) {
ABSL_RAW_CHECK(strnlen(heap_buffer.get(), limit) < limit,
"absl::Symbolize() did not properly terminate the string");
strncpy(try_symbolize_buffer, heap_buffer.get(),
sizeof(try_symbolize_buffer) - 1);
try_symbolize_buffer[sizeof(try_symbolize_buffer) - 1] = '\0';
}
return found ? try_symbolize_buffer : nullptr;
}
// A wrapper for TrySymbolizeWithLimit(), with a large limit.
static const char *TrySymbolize(void *pc) {
return TrySymbolizeWithLimit(pc, sizeof(try_symbolize_buffer));
}
Export of internal Abseil changes -- 69b5d0b2a5adb49a53e51f9da6848eaa484242fe by Derek Mauro <dmauro@google.com>: Changes the absl::Duration factory functions to disallow types that are convertible to int or double, instead requiring that the argument itself is indeed an integer or a floating-point number. This will prevent callers from passing arguments, such as std::atomic<T>. This change is an API break. Information and a tool to fix issues can be found at https://abseil.io/docs/cpp/tools/upgrades/duration-conversions PiperOrigin-RevId: 387153494 -- 786063e438ab6a55ac4baa88ad4d20a8293be52a by Evan Brown <ezb@google.com>: Make ctrl_t be an enum class. This adds type safety, and also when strict aliasing is enabled, the compiler will know that control bytes can't alias non-control bytes. Also make H2() return h2_t. PiperOrigin-RevId: 387120717 -- 7e537aabec1c255d6e7c9d21232c597c1c2077bf by Evan Brown <ezb@google.com>: Add some missing `const` keywords to ctrl_t* function parameters. PiperOrigin-RevId: 386976062 -- da53ac6d91cabd951e81dd0a145e1e52b918955f by Martijn Vels <mvels@google.com>: Change Seek and InitOffset to return nullptr instead of assert / fail. This makes it consistent with the rest of the API (Next, Previous, Skip) and hardens it against invariants that are harder (or less likey) to be upheld correctly by the caller. PiperOrigin-RevId: 386963283 -- a4d1faac020d5025edf53ce81808e5db68da7d89 by Abseil Team <absl-team@google.com>: PC / Backtrace / Symbolization for Emscripten. PiperOrigin-RevId: 386957724 -- 97f2c47d83ba9d3ac89e1f55bd06897686ffd063 by Martijn Vels <mvels@google.com>: Fix static casts ([-Wimplicit-int-conversion]) PiperOrigin-RevId: 386951646 -- 9530c795248543817cbc4013953baa09c35f5e1a by Abseil Team <absl-team@google.com>: Fix incorrect header guard in cord_rep_btree_navigator.h PiperOrigin-RevId: 386907904 -- 90ce5872406df2b7f4c428683741dc13a572267e by Abseil Team <absl-team@google.com>: Small grammar fixes for some StatusCode descriptions. PiperOrigin-RevId: 386906217 -- b30a2fd777f12a04a4d512f37a34614b0d05ce99 by Derek Mauro <dmauro@google.com>: Skip length checking when constructing absl::string_view from std::string. The length check causes unnecessary code bloat. PiperOrigin-RevId: 386857974 -- fa171536c359bfa2a1b80297e844519bb9ee7791 by Martijn Vels <mvels@google.com>: Introduce CordRepBtreeNavigator CordRepBtreeNavigator implements bi-directional navigation over all data edges stored inside a Cord Btree. PiperOrigin-RevId: 386519102 GitOrigin-RevId: 69b5d0b2a5adb49a53e51f9da6848eaa484242fe Change-Id: I1b35188d66133f8cb73d346bc5564aac4e0b3e80
3 years ago
#if defined(ABSL_INTERNAL_HAVE_ELF_SYMBOLIZE) || \
defined(ABSL_INTERNAL_HAVE_DARWIN_SYMBOLIZE) || \
defined(ABSL_INTERNAL_HAVE_EMSCRIPTEN_SYMBOLIZE)
// Test with a return address.
void ABSL_ATTRIBUTE_NOINLINE TestWithReturnAddress() {
#if defined(ABSL_HAVE_ATTRIBUTE_NOINLINE)
void *return_address = __builtin_return_address(0);
const char *symbol = TrySymbolize(return_address);
ABSL_RAW_CHECK(symbol != nullptr, "TestWithReturnAddress failed");
ABSL_RAW_CHECK(strcmp(symbol, "main") == 0, "TestWithReturnAddress failed");
std::cout << "TestWithReturnAddress passed" << std::endl;
#endif
}
#ifndef ABSL_INTERNAL_HAVE_EMSCRIPTEN_SYMBOLIZE
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
TEST(Symbolize, Cached) {
// Compilers should give us pointers to them.
EXPECT_STREQ("nonstatic_func", TrySymbolize((void *)(&nonstatic_func)));
// The name of an internal linkage symbol is not specified; allow either a
// mangled or an unmangled name here.
const char *static_func_symbol = TrySymbolize((void *)(&static_func));
EXPECT_TRUE(strcmp("static_func", static_func_symbol) == 0 ||
strcmp("static_func()", static_func_symbol) == 0);
EXPECT_TRUE(nullptr == TrySymbolize(nullptr));
}
TEST(Symbolize, Truncation) {
constexpr char kNonStaticFunc[] = "nonstatic_func";
EXPECT_STREQ("nonstatic_func",
TrySymbolizeWithLimit((void *)(&nonstatic_func),
strlen(kNonStaticFunc) + 1));
EXPECT_STREQ("nonstatic_...",
TrySymbolizeWithLimit((void *)(&nonstatic_func),
strlen(kNonStaticFunc) + 0));
EXPECT_STREQ("nonstatic...",
TrySymbolizeWithLimit((void *)(&nonstatic_func),
strlen(kNonStaticFunc) - 1));
EXPECT_STREQ("n...", TrySymbolizeWithLimit((void *)(&nonstatic_func), 5));
EXPECT_STREQ("...", TrySymbolizeWithLimit((void *)(&nonstatic_func), 4));
EXPECT_STREQ("..", TrySymbolizeWithLimit((void *)(&nonstatic_func), 3));
EXPECT_STREQ(".", TrySymbolizeWithLimit((void *)(&nonstatic_func), 2));
EXPECT_STREQ("", TrySymbolizeWithLimit((void *)(&nonstatic_func), 1));
EXPECT_EQ(nullptr, TrySymbolizeWithLimit((void *)(&nonstatic_func), 0));
}
TEST(Symbolize, SymbolizeWithDemangling) {
Foo::func(100);
EXPECT_STREQ("Foo::func()", TrySymbolize((void *)(&Foo::func)));
}
TEST(Symbolize, SymbolizeSplitTextSections) {
EXPECT_STREQ("unlikely_func()", TrySymbolize((void *)(&unlikely_func)));
EXPECT_STREQ("hot_func()", TrySymbolize((void *)(&hot_func)));
EXPECT_STREQ("startup_func()", TrySymbolize((void *)(&startup_func)));
EXPECT_STREQ("exit_func()", TrySymbolize((void *)(&exit_func)));
EXPECT_STREQ("regular_func()", TrySymbolize((void *)(&regular_func)));
}
// Tests that verify that Symbolize stack footprint is within some limit.
#ifdef ABSL_INTERNAL_HAVE_DEBUGGING_STACK_CONSUMPTION
static void *g_pc_to_symbolize;
static char g_symbolize_buffer[4096];
static char *g_symbolize_result;
static void SymbolizeSignalHandler(int signo) {
if (absl::Symbolize(g_pc_to_symbolize, g_symbolize_buffer,
sizeof(g_symbolize_buffer))) {
g_symbolize_result = g_symbolize_buffer;
} else {
g_symbolize_result = nullptr;
}
}
// Call Symbolize and figure out the stack footprint of this call.
static const char *SymbolizeStackConsumption(void *pc, int *stack_consumed) {
g_pc_to_symbolize = pc;
*stack_consumed = absl::debugging_internal::GetSignalHandlerStackConsumption(
SymbolizeSignalHandler);
return g_symbolize_result;
}
static int GetStackConsumptionUpperLimit() {
// Symbolize stack consumption should be within 2kB.
int stack_consumption_upper_limit = 2048;
#if defined(ABSL_HAVE_ADDRESS_SANITIZER) || \
defined(ABSL_HAVE_MEMORY_SANITIZER) || defined(ABSL_HAVE_THREAD_SANITIZER)
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
// Account for sanitizer instrumentation requiring additional stack space.
stack_consumption_upper_limit *= 5;
#endif
return stack_consumption_upper_limit;
}
TEST(Symbolize, SymbolizeStackConsumption) {
int stack_consumed = 0;
const char *symbol =
SymbolizeStackConsumption((void *)(&nonstatic_func), &stack_consumed);
EXPECT_STREQ("nonstatic_func", symbol);
EXPECT_GT(stack_consumed, 0);
EXPECT_LT(stack_consumed, GetStackConsumptionUpperLimit());
// The name of an internal linkage symbol is not specified; allow either a
// mangled or an unmangled name here.
symbol = SymbolizeStackConsumption((void *)(&static_func), &stack_consumed);
EXPECT_TRUE(strcmp("static_func", symbol) == 0 ||
strcmp("static_func()", symbol) == 0);
EXPECT_GT(stack_consumed, 0);
EXPECT_LT(stack_consumed, GetStackConsumptionUpperLimit());
}
TEST(Symbolize, SymbolizeWithDemanglingStackConsumption) {
Foo::func(100);
int stack_consumed = 0;
const char *symbol =
SymbolizeStackConsumption((void *)(&Foo::func), &stack_consumed);
EXPECT_STREQ("Foo::func()", symbol);
EXPECT_GT(stack_consumed, 0);
EXPECT_LT(stack_consumed, GetStackConsumptionUpperLimit());
}
#endif // ABSL_INTERNAL_HAVE_DEBUGGING_STACK_CONSUMPTION
Export of internal Abseil changes -- 9e8b4a286d70df9487bff080816bd07ae38af5f8 by Evan Brown <ezb@google.com>: Add btree_node::transfer_n/transfer_n_backward and replace usage of uninitialized_move_n and value_destroy_n. PiperOrigin-RevId: 314600027 -- 6c452aa1ee7e46ab941ba7d1fa636da8ea3d7370 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: 314584095 -- 07853c47dc98698d67d65a3b9b662a65ab9def0a by Abseil Team <absl-team@google.com>: Add PC / backtrace / symbolization support for Apple platforms. Full backtrace support requires iOS 9+ PiperOrigin-RevId: 314415072 -- 43889f17a132b31f6558c6482721cbbc776128fd by Gennadiy Rozental <rogeeff@google.com>: Consolidate all reflection interface in the new module 'reflection' and expose interface to locate reflection handle by name. PiperOrigin-RevId: 314390358 GitOrigin-RevId: 9e8b4a286d70df9487bff080816bd07ae38af5f8 Change-Id: I8e0910437740cf9ea9da5000adddfcef127e1158
5 years ago
#ifndef ABSL_INTERNAL_HAVE_DARWIN_SYMBOLIZE
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
// Use a 64K page size for PPC.
const size_t kPageSize = 64 << 10;
// We place a read-only symbols into the .text section and verify that we can
// symbolize them and other symbols after remapping them.
const char kPadding0[kPageSize * 4] ABSL_ATTRIBUTE_SECTION_VARIABLE(.text) =
"";
const char kPadding1[kPageSize * 4] ABSL_ATTRIBUTE_SECTION_VARIABLE(.text) =
"";
static int FilterElfHeader(struct dl_phdr_info *info, size_t size, void *data) {
for (int i = 0; i < info->dlpi_phnum; i++) {
if (info->dlpi_phdr[i].p_type == PT_LOAD &&
info->dlpi_phdr[i].p_flags == (PF_R | PF_X)) {
const void *const vaddr =
absl::bit_cast<void *>(info->dlpi_addr + info->dlpi_phdr[i].p_vaddr);
const auto segsize = info->dlpi_phdr[i].p_memsz;
const char *self_exe;
if (info->dlpi_name != nullptr && info->dlpi_name[0] != '\0') {
self_exe = info->dlpi_name;
} else {
self_exe = "/proc/self/exe";
}
absl::debugging_internal::RegisterFileMappingHint(
vaddr, reinterpret_cast<const char *>(vaddr) + segsize,
info->dlpi_phdr[i].p_offset, self_exe);
return 1;
}
}
return 1;
}
TEST(Symbolize, SymbolizeWithMultipleMaps) {
// Force kPadding0 and kPadding1 to be linked in.
if (volatile_bool) {
ABSL_RAW_LOG(INFO, "%s", kPadding0);
ABSL_RAW_LOG(INFO, "%s", kPadding1);
}
// Verify we can symbolize everything.
char buf[512];
memset(buf, 0, sizeof(buf));
absl::Symbolize(kPadding0, buf, sizeof(buf));
EXPECT_STREQ("kPadding0", buf);
memset(buf, 0, sizeof(buf));
absl::Symbolize(kPadding1, buf, sizeof(buf));
EXPECT_STREQ("kPadding1", buf);
// Specify a hint for the executable segment.
dl_iterate_phdr(FilterElfHeader, nullptr);
// Reload at least one page out of kPadding0, kPadding1
const char *ptrs[] = {kPadding0, kPadding1};
for (const char *ptr : ptrs) {
const int kMapFlags = MAP_ANONYMOUS | MAP_PRIVATE;
void *addr = mmap(nullptr, kPageSize, PROT_READ, kMapFlags, 0, 0);
ASSERT_NE(addr, MAP_FAILED);
// kPadding[0-1] is full of zeroes, so we can remap anywhere within it, but
// we ensure there is at least a full page of padding.
void *remapped = reinterpret_cast<void *>(
reinterpret_cast<uintptr_t>(ptr + kPageSize) & ~(kPageSize - 1ULL));
const int kMremapFlags = (MREMAP_MAYMOVE | MREMAP_FIXED);
void *ret = mremap(addr, kPageSize, kPageSize, kMremapFlags, remapped);
ASSERT_NE(ret, MAP_FAILED);
}
// Invalidate the symbolization cache so we are forced to rely on the hint.
absl::Symbolize(nullptr, buf, sizeof(buf));
// Verify we can still symbolize.
const char *expected[] = {"kPadding0", "kPadding1"};
const size_t offsets[] = {0, kPageSize, 2 * kPageSize, 3 * kPageSize};
for (int i = 0; i < 2; i++) {
for (size_t offset : offsets) {
memset(buf, 0, sizeof(buf));
absl::Symbolize(ptrs[i] + offset, buf, sizeof(buf));
EXPECT_STREQ(expected[i], buf);
}
}
}
// Appends string(*args->arg) to args->symbol_buf.
static void DummySymbolDecorator(
const absl::debugging_internal::SymbolDecoratorArgs *args) {
std::string *message = static_cast<std::string *>(args->arg);
strncat(args->symbol_buf, message->c_str(),
args->symbol_buf_size - strlen(args->symbol_buf) - 1);
}
TEST(Symbolize, InstallAndRemoveSymbolDecorators) {
int ticket_a;
std::string a_message("a");
EXPECT_GE(ticket_a = absl::debugging_internal::InstallSymbolDecorator(
DummySymbolDecorator, &a_message),
0);
int ticket_b;
std::string b_message("b");
EXPECT_GE(ticket_b = absl::debugging_internal::InstallSymbolDecorator(
DummySymbolDecorator, &b_message),
0);
int ticket_c;
std::string c_message("c");
EXPECT_GE(ticket_c = absl::debugging_internal::InstallSymbolDecorator(
DummySymbolDecorator, &c_message),
0);
char *address = reinterpret_cast<char *>(1);
EXPECT_STREQ("abc", TrySymbolize(address++));
EXPECT_TRUE(absl::debugging_internal::RemoveSymbolDecorator(ticket_b));
EXPECT_STREQ("ac", TrySymbolize(address++));
// Cleanup: remove all remaining decorators so other stack traces don't
// get mystery "ac" decoration.
EXPECT_TRUE(absl::debugging_internal::RemoveSymbolDecorator(ticket_a));
EXPECT_TRUE(absl::debugging_internal::RemoveSymbolDecorator(ticket_c));
}
// Some versions of Clang with optimizations enabled seem to be able
// to optimize away the .data section if no variables live in the
// section. This variable should get placed in the .data section, and
// the test below checks for the existence of a .data section.
static int in_data_section = 1;
TEST(Symbolize, ForEachSection) {
int fd = TEMP_FAILURE_RETRY(open("/proc/self/exe", O_RDONLY));
ASSERT_NE(fd, -1);
std::vector<std::string> sections;
ASSERT_TRUE(absl::debugging_internal::ForEachSection(
fd, [&sections](const absl::string_view name, const ElfW(Shdr) &) {
sections.emplace_back(name);
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 true;
}));
// Check for the presence of common section names.
EXPECT_THAT(sections, Contains(".text"));
EXPECT_THAT(sections, Contains(".rodata"));
EXPECT_THAT(sections, Contains(".bss"));
++in_data_section;
EXPECT_THAT(sections, Contains(".data"));
close(fd);
}
Export of internal Abseil changes -- 9e8b4a286d70df9487bff080816bd07ae38af5f8 by Evan Brown <ezb@google.com>: Add btree_node::transfer_n/transfer_n_backward and replace usage of uninitialized_move_n and value_destroy_n. PiperOrigin-RevId: 314600027 -- 6c452aa1ee7e46ab941ba7d1fa636da8ea3d7370 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: 314584095 -- 07853c47dc98698d67d65a3b9b662a65ab9def0a by Abseil Team <absl-team@google.com>: Add PC / backtrace / symbolization support for Apple platforms. Full backtrace support requires iOS 9+ PiperOrigin-RevId: 314415072 -- 43889f17a132b31f6558c6482721cbbc776128fd by Gennadiy Rozental <rogeeff@google.com>: Consolidate all reflection interface in the new module 'reflection' and expose interface to locate reflection handle by name. PiperOrigin-RevId: 314390358 GitOrigin-RevId: 9e8b4a286d70df9487bff080816bd07ae38af5f8 Change-Id: I8e0910437740cf9ea9da5000adddfcef127e1158
5 years ago
#endif // !ABSL_INTERNAL_HAVE_DARWIN_SYMBOLIZE
Export of internal Abseil changes -- 69b5d0b2a5adb49a53e51f9da6848eaa484242fe by Derek Mauro <dmauro@google.com>: Changes the absl::Duration factory functions to disallow types that are convertible to int or double, instead requiring that the argument itself is indeed an integer or a floating-point number. This will prevent callers from passing arguments, such as std::atomic<T>. This change is an API break. Information and a tool to fix issues can be found at https://abseil.io/docs/cpp/tools/upgrades/duration-conversions PiperOrigin-RevId: 387153494 -- 786063e438ab6a55ac4baa88ad4d20a8293be52a by Evan Brown <ezb@google.com>: Make ctrl_t be an enum class. This adds type safety, and also when strict aliasing is enabled, the compiler will know that control bytes can't alias non-control bytes. Also make H2() return h2_t. PiperOrigin-RevId: 387120717 -- 7e537aabec1c255d6e7c9d21232c597c1c2077bf by Evan Brown <ezb@google.com>: Add some missing `const` keywords to ctrl_t* function parameters. PiperOrigin-RevId: 386976062 -- da53ac6d91cabd951e81dd0a145e1e52b918955f by Martijn Vels <mvels@google.com>: Change Seek and InitOffset to return nullptr instead of assert / fail. This makes it consistent with the rest of the API (Next, Previous, Skip) and hardens it against invariants that are harder (or less likey) to be upheld correctly by the caller. PiperOrigin-RevId: 386963283 -- a4d1faac020d5025edf53ce81808e5db68da7d89 by Abseil Team <absl-team@google.com>: PC / Backtrace / Symbolization for Emscripten. PiperOrigin-RevId: 386957724 -- 97f2c47d83ba9d3ac89e1f55bd06897686ffd063 by Martijn Vels <mvels@google.com>: Fix static casts ([-Wimplicit-int-conversion]) PiperOrigin-RevId: 386951646 -- 9530c795248543817cbc4013953baa09c35f5e1a by Abseil Team <absl-team@google.com>: Fix incorrect header guard in cord_rep_btree_navigator.h PiperOrigin-RevId: 386907904 -- 90ce5872406df2b7f4c428683741dc13a572267e by Abseil Team <absl-team@google.com>: Small grammar fixes for some StatusCode descriptions. PiperOrigin-RevId: 386906217 -- b30a2fd777f12a04a4d512f37a34614b0d05ce99 by Derek Mauro <dmauro@google.com>: Skip length checking when constructing absl::string_view from std::string. The length check causes unnecessary code bloat. PiperOrigin-RevId: 386857974 -- fa171536c359bfa2a1b80297e844519bb9ee7791 by Martijn Vels <mvels@google.com>: Introduce CordRepBtreeNavigator CordRepBtreeNavigator implements bi-directional navigation over all data edges stored inside a Cord Btree. PiperOrigin-RevId: 386519102 GitOrigin-RevId: 69b5d0b2a5adb49a53e51f9da6848eaa484242fe Change-Id: I1b35188d66133f8cb73d346bc5564aac4e0b3e80
3 years ago
#endif // !ABSL_INTERNAL_HAVE_EMSCRIPTEN_SYMBOLIZE
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
// x86 specific tests. Uses some inline assembler.
extern "C" {
inline void *ABSL_ATTRIBUTE_ALWAYS_INLINE inline_func() {
void *pc = nullptr;
#if defined(__i386__)
__asm__ __volatile__("call 1f;\n 1: pop %[PC]" : [ PC ] "=r"(pc));
#elif defined(__x86_64__)
__asm__ __volatile__("leaq 0(%%rip),%[PC];\n" : [ PC ] "=r"(pc));
#endif
return pc;
}
void *ABSL_ATTRIBUTE_NOINLINE non_inline_func() {
void *pc = nullptr;
#if defined(__i386__)
__asm__ __volatile__("call 1f;\n 1: pop %[PC]" : [ PC ] "=r"(pc));
#elif defined(__x86_64__)
__asm__ __volatile__("leaq 0(%%rip),%[PC];\n" : [ PC ] "=r"(pc));
#endif
return pc;
}
void ABSL_ATTRIBUTE_NOINLINE TestWithPCInsideNonInlineFunction() {
#if defined(ABSL_HAVE_ATTRIBUTE_NOINLINE) && \
(defined(__i386__) || defined(__x86_64__))
void *pc = non_inline_func();
const char *symbol = TrySymbolize(pc);
ABSL_RAW_CHECK(symbol != nullptr, "TestWithPCInsideNonInlineFunction failed");
ABSL_RAW_CHECK(strcmp(symbol, "non_inline_func") == 0,
"TestWithPCInsideNonInlineFunction failed");
std::cout << "TestWithPCInsideNonInlineFunction passed" << std::endl;
#endif
}
void ABSL_ATTRIBUTE_NOINLINE TestWithPCInsideInlineFunction() {
#if defined(ABSL_HAVE_ATTRIBUTE_ALWAYS_INLINE) && \
(defined(__i386__) || defined(__x86_64__))
void *pc = inline_func(); // Must be inlined.
const char *symbol = TrySymbolize(pc);
ABSL_RAW_CHECK(symbol != nullptr, "TestWithPCInsideInlineFunction failed");
ABSL_RAW_CHECK(strcmp(symbol, __FUNCTION__) == 0,
"TestWithPCInsideInlineFunction failed");
std::cout << "TestWithPCInsideInlineFunction passed" << std::endl;
#endif
}
}
Export of internal Abseil changes -- ac1df60490c9583e475e22de7adfc40023196fbf by Martijn Vels <mvels@google.com>: Change Cord constructor(string_view) to explicit make_tree and Cordz tracking This CL changes the ctor to use an easier to maintain model where Cord code explicitly invokes Cordz update or new / tree logic, which avoids the ambiguity of the 'branched' InlineRep::set_tree code. This removes the need to equip InlineRep with 'MethodIdentifier' or other necessary call info, and also is a cleaner model: InlineRep is carrying too much code now that should plainly sit in Cord, especially with all internal abstractions having moved to InlineData. See child CL(s) for desired state PiperOrigin-RevId: 369433619 -- b665af7f586e6c679a8b27d4f78d5a1d2b596058 by Abseil Team <absl-team@google.com>: Rename the 'Compare' template type to 'LessThan', as the passed-in function is expected to act like operator<. It is worth avoiding confusion with std::compare, which returns an int (-1/0/1), as due to implicit casting this can lead to hard-to-spot bugs. PiperOrigin-RevId: 369391118 -- c3c775269cad0f4982ec63f3616dd78bb9e52dca by Martijn Vels <mvels@google.com>: Integrate CordzUpdateTracker into CordzInfo PiperOrigin-RevId: 369348824 -- 771d81ed357496c117179e1daec76eba5155932d by Martijn Vels <mvels@google.com>: Replace mutex() with Lock() / Unlock() function Mini design future tracking of CordzInfo sampled cords: CordzInfo holds a CordRep* reference without a reference count. Cord is responsible for synchronizing updates for sampled cords such that the CordRep* contained in CordzInfo is at all times valid. This is done by scoping Lock() and Unlock() calls around the code modifying the code of a sampled cord. For example (using the future CL CordzUpdateScope()): CordzInfo* cordz_info = get_cordz_info(); CordzUpdateScope scope(cordz_info, CordzUpdateTracker::kRemovePrefix); CordRep* rep = RemovePrefixImpl(root); set_tree(rep); if (cordz_info) { cordz_info->SetCordRep(rep); } On CordzInfo::Unlock(), if the internal rep is null, the cord is no longer sampled, and CordzInfo will be deleted. Thus any update resulting in the Cord being inlined will automatically no longer be sampled. PiperOrigin-RevId: 369338802 -- 5563c12df04a1e965a03b50bdd032739c55c0706 by Martijn Vels <mvels@google.com>: Add UpdateTracker to CordzStatistics PiperOrigin-RevId: 369318178 -- 6b4d8463722a3e55a3e8f6cb3741a41055e7f83e by Martijn Vels <mvels@google.com>: Add kClear, kConstructor* and kUnknown values and fix typo PiperOrigin-RevId: 369297163 -- 041adcbc929789d6d53371a8236840fc350e1eeb by Derek Mauro <dmauro@google.com>: Switch from malloc to operator new in pool_urbg.cc so it can only fail by throwing/aborting PiperOrigin-RevId: 369274087 -- 5d97a5f43e3f2d02d0a5bbe586d93b5751812981 by Benjamin Barenblat <bbaren@google.com>: Correct Thumb function bound computation in the symbolizer On 32-bit ARM, all functions are aligned to multiples of two bytes, and the lowest-order bit in a function’s address is ignored by the CPU when computing branch targets. That bit is still present in instructions and ELF symbol tables, though; it’s repurposed to indicate whether the function contains ARM or Thumb code. If the symbolizer doesn’t ignore that bit, it will believe Thumb functions have boundaries that are off by one byte, so instruct the symbolizer to null out the lowest-order bit after retrieving it from the symbol table. PiperOrigin-RevId: 369254082 -- 462bb307c6cc332c1e2c3adb5f0cad51804bf937 by Derek Mauro <dmauro@google.com>: Add a check for malloc failure in pool_urbg.cc GitHub #940 PiperOrigin-RevId: 369238100 GitOrigin-RevId: ac1df60490c9583e475e22de7adfc40023196fbf Change-Id: Ic6ec91c62cd3a0031f6a75a43a83da959ece2d25
4 years ago
#if defined(__arm__) && ABSL_HAVE_ATTRIBUTE(target)
// Test that we correctly identify bounds of Thumb functions on ARM.
//
// Thumb functions have the lowest-order bit set in their addresses in the ELF
// symbol table. This requires some extra logic to properly compute function
// bounds. To test this logic, nudge a Thumb function right up against an ARM
// function and try to symbolize the ARM function.
//
// A naive implementation will simply use the Thumb function's entry point as
// written in the symbol table and will therefore treat the Thumb function as
// extending one byte further in the instruction stream than it actually does.
// When asked to symbolize the start of the ARM function, it will identify an
// overlap between the Thumb and ARM functions, and it will return the name of
// the Thumb function.
//
// A correct implementation, on the other hand, will null out the lowest-order
// bit in the Thumb function's entry point. It will correctly compute the end of
// the Thumb function, it will find no overlap between the Thumb and ARM
// functions, and it will return the name of the ARM function.
__attribute__((target("thumb"))) int ArmThumbOverlapThumb(int x) {
return x * x * x;
}
__attribute__((target("arm"))) int ArmThumbOverlapArm(int x) {
return x * x * x;
}
void ABSL_ATTRIBUTE_NOINLINE TestArmThumbOverlap() {
#if defined(ABSL_HAVE_ATTRIBUTE_NOINLINE)
const char *symbol = TrySymbolize((void *)&ArmThumbOverlapArm);
ABSL_RAW_CHECK(symbol != nullptr, "TestArmThumbOverlap failed");
ABSL_RAW_CHECK(strcmp("ArmThumbOverlapArm()", symbol) == 0,
"TestArmThumbOverlap failed");
std::cout << "TestArmThumbOverlap passed" << std::endl;
#endif
}
#endif // defined(__arm__) && ABSL_HAVE_ATTRIBUTE(target)
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
#elif defined(_WIN32)
#if !defined(ABSL_CONSUME_DLL)
TEST(Symbolize, Basics) {
EXPECT_STREQ("nonstatic_func", TrySymbolize((void *)(&nonstatic_func)));
// The name of an internal linkage symbol is not specified; allow either a
// mangled or an unmangled name here.
const char *static_func_symbol = TrySymbolize((void *)(&static_func));
ASSERT_TRUE(static_func_symbol != nullptr);
EXPECT_TRUE(strstr(static_func_symbol, "static_func") != nullptr);
EXPECT_TRUE(nullptr == TrySymbolize(nullptr));
}
TEST(Symbolize, Truncation) {
constexpr char kNonStaticFunc[] = "nonstatic_func";
EXPECT_STREQ("nonstatic_func",
TrySymbolizeWithLimit((void *)(&nonstatic_func),
strlen(kNonStaticFunc) + 1));
EXPECT_STREQ("nonstatic_...",
TrySymbolizeWithLimit((void *)(&nonstatic_func),
strlen(kNonStaticFunc) + 0));
EXPECT_STREQ("nonstatic...",
TrySymbolizeWithLimit((void *)(&nonstatic_func),
strlen(kNonStaticFunc) - 1));
EXPECT_STREQ("n...", TrySymbolizeWithLimit((void *)(&nonstatic_func), 5));
EXPECT_STREQ("...", TrySymbolizeWithLimit((void *)(&nonstatic_func), 4));
EXPECT_STREQ("..", TrySymbolizeWithLimit((void *)(&nonstatic_func), 3));
EXPECT_STREQ(".", TrySymbolizeWithLimit((void *)(&nonstatic_func), 2));
EXPECT_STREQ("", TrySymbolizeWithLimit((void *)(&nonstatic_func), 1));
EXPECT_EQ(nullptr, TrySymbolizeWithLimit((void *)(&nonstatic_func), 0));
}
TEST(Symbolize, SymbolizeWithDemangling) {
const char *result = TrySymbolize((void *)(&Foo::func));
ASSERT_TRUE(result != nullptr);
EXPECT_TRUE(strstr(result, "Foo::func") != nullptr) << result;
}
#endif // !defined(ABSL_CONSUME_DLL)
#else // Symbolizer unimplemented
TEST(Symbolize, Unimplemented) {
char buf[64];
EXPECT_FALSE(absl::Symbolize((void *)(&nonstatic_func), buf, sizeof(buf)));
EXPECT_FALSE(absl::Symbolize((void *)(&static_func), buf, sizeof(buf)));
EXPECT_FALSE(absl::Symbolize((void *)(&Foo::func), buf, sizeof(buf)));
}
#endif
int main(int argc, char **argv) {
#if !defined(__EMSCRIPTEN__)
// Make sure kHpageTextPadding is linked into the binary.
if (volatile_bool) {
ABSL_RAW_LOG(INFO, "%s", kHpageTextPadding);
}
#endif // !defined(__EMSCRIPTEN__)
#if ABSL_PER_THREAD_TLS
// Touch the per-thread variables.
symbolize_test_thread_small[0] = 0;
symbolize_test_thread_big[0] = 0;
#endif
absl::InitializeSymbolizer(argv[0]);
testing::InitGoogleTest(&argc, argv);
Export of internal Abseil changes -- 9e8b4a286d70df9487bff080816bd07ae38af5f8 by Evan Brown <ezb@google.com>: Add btree_node::transfer_n/transfer_n_backward and replace usage of uninitialized_move_n and value_destroy_n. PiperOrigin-RevId: 314600027 -- 6c452aa1ee7e46ab941ba7d1fa636da8ea3d7370 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: 314584095 -- 07853c47dc98698d67d65a3b9b662a65ab9def0a by Abseil Team <absl-team@google.com>: Add PC / backtrace / symbolization support for Apple platforms. Full backtrace support requires iOS 9+ PiperOrigin-RevId: 314415072 -- 43889f17a132b31f6558c6482721cbbc776128fd by Gennadiy Rozental <rogeeff@google.com>: Consolidate all reflection interface in the new module 'reflection' and expose interface to locate reflection handle by name. PiperOrigin-RevId: 314390358 GitOrigin-RevId: 9e8b4a286d70df9487bff080816bd07ae38af5f8 Change-Id: I8e0910437740cf9ea9da5000adddfcef127e1158
5 years ago
#if defined(ABSL_INTERNAL_HAVE_ELF_SYMBOLIZE) || \
defined(ABSL_INTERNAL_HAVE_DARWIN_SYMBOLIZE)
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
TestWithPCInsideInlineFunction();
TestWithPCInsideNonInlineFunction();
TestWithReturnAddress();
Export of internal Abseil changes -- ac1df60490c9583e475e22de7adfc40023196fbf by Martijn Vels <mvels@google.com>: Change Cord constructor(string_view) to explicit make_tree and Cordz tracking This CL changes the ctor to use an easier to maintain model where Cord code explicitly invokes Cordz update or new / tree logic, which avoids the ambiguity of the 'branched' InlineRep::set_tree code. This removes the need to equip InlineRep with 'MethodIdentifier' or other necessary call info, and also is a cleaner model: InlineRep is carrying too much code now that should plainly sit in Cord, especially with all internal abstractions having moved to InlineData. See child CL(s) for desired state PiperOrigin-RevId: 369433619 -- b665af7f586e6c679a8b27d4f78d5a1d2b596058 by Abseil Team <absl-team@google.com>: Rename the 'Compare' template type to 'LessThan', as the passed-in function is expected to act like operator<. It is worth avoiding confusion with std::compare, which returns an int (-1/0/1), as due to implicit casting this can lead to hard-to-spot bugs. PiperOrigin-RevId: 369391118 -- c3c775269cad0f4982ec63f3616dd78bb9e52dca by Martijn Vels <mvels@google.com>: Integrate CordzUpdateTracker into CordzInfo PiperOrigin-RevId: 369348824 -- 771d81ed357496c117179e1daec76eba5155932d by Martijn Vels <mvels@google.com>: Replace mutex() with Lock() / Unlock() function Mini design future tracking of CordzInfo sampled cords: CordzInfo holds a CordRep* reference without a reference count. Cord is responsible for synchronizing updates for sampled cords such that the CordRep* contained in CordzInfo is at all times valid. This is done by scoping Lock() and Unlock() calls around the code modifying the code of a sampled cord. For example (using the future CL CordzUpdateScope()): CordzInfo* cordz_info = get_cordz_info(); CordzUpdateScope scope(cordz_info, CordzUpdateTracker::kRemovePrefix); CordRep* rep = RemovePrefixImpl(root); set_tree(rep); if (cordz_info) { cordz_info->SetCordRep(rep); } On CordzInfo::Unlock(), if the internal rep is null, the cord is no longer sampled, and CordzInfo will be deleted. Thus any update resulting in the Cord being inlined will automatically no longer be sampled. PiperOrigin-RevId: 369338802 -- 5563c12df04a1e965a03b50bdd032739c55c0706 by Martijn Vels <mvels@google.com>: Add UpdateTracker to CordzStatistics PiperOrigin-RevId: 369318178 -- 6b4d8463722a3e55a3e8f6cb3741a41055e7f83e by Martijn Vels <mvels@google.com>: Add kClear, kConstructor* and kUnknown values and fix typo PiperOrigin-RevId: 369297163 -- 041adcbc929789d6d53371a8236840fc350e1eeb by Derek Mauro <dmauro@google.com>: Switch from malloc to operator new in pool_urbg.cc so it can only fail by throwing/aborting PiperOrigin-RevId: 369274087 -- 5d97a5f43e3f2d02d0a5bbe586d93b5751812981 by Benjamin Barenblat <bbaren@google.com>: Correct Thumb function bound computation in the symbolizer On 32-bit ARM, all functions are aligned to multiples of two bytes, and the lowest-order bit in a function’s address is ignored by the CPU when computing branch targets. That bit is still present in instructions and ELF symbol tables, though; it’s repurposed to indicate whether the function contains ARM or Thumb code. If the symbolizer doesn’t ignore that bit, it will believe Thumb functions have boundaries that are off by one byte, so instruct the symbolizer to null out the lowest-order bit after retrieving it from the symbol table. PiperOrigin-RevId: 369254082 -- 462bb307c6cc332c1e2c3adb5f0cad51804bf937 by Derek Mauro <dmauro@google.com>: Add a check for malloc failure in pool_urbg.cc GitHub #940 PiperOrigin-RevId: 369238100 GitOrigin-RevId: ac1df60490c9583e475e22de7adfc40023196fbf Change-Id: Ic6ec91c62cd3a0031f6a75a43a83da959ece2d25
4 years ago
#if defined(__arm__) && ABSL_HAVE_ATTRIBUTE(target)
TestArmThumbOverlap();
#endif
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
#endif
return RUN_ALL_TESTS();
}