Shared Library Constructor (rocprofv3 deadlock fix) (#599)

* Moved tests/apps to tests/bin

* Renamed cmake project in tests/bin

* Update samples

- Use ROCPROFILER_DEFAULT_FAIL_REGEX
- tweaks to stdout messages

* Update tests

- Use ROCPROFILER_DEFAULT_FAIL_REGEX

* Add tests/lib

- libraries with HIP code

* Update PTL submodule

- remove atexit delete of thread_id_map

* Update cmake/rocprofiler_options.cmake

- Set ROCPROFILER_DEFAULT_FAIL_REGEX

* Update common lib: env + logging

- improved customization of logging settings
- default to disabling logging to files
- install failure handler for rocprofv3
- set_env support in environment.*

* Add lib/rocprofiler-sdk/shared_library.cpp

- shared library constructor

* Update lib/rocprofiler-sdk-tool/tool.cpp

- destructor thread safety
- convert callback_name_info and buffered_name_info to pointers
- install failure handler for logging

* Add tests/bin/hip-in-libraries

- hip-in-libraries is an exe which uses two shared libraries where each shared library contains HIP kernels
  - used for testing deadlocking within __hipRegisterFatBinary

* Update bin/rocprofv3

- reorganized the env variables
- use exec to launch command
- set ROCPROFILER_LIBRARY_CTOR=1

* Add tests/rocprofv3/tracing-hip-in-libraries

- uses hip-in-libraries exe for exe which uses shared libraries to launch HIP kernels

* Update bin/rocprofv3

- fix counter collection (no exec)

* Update lib/rocprofiler-sdk-tool/tool.cpp

- replace "Kernel-Name" with "Kernel_Name"

* Update lib/rocprofiler-sdk/registration.cpp

Use RTLD_LOCAL instead of RTLD_GLOBAL for env libraries

* Update tests/rocprofv3

- replace "Kernel-Name" with "Kernel_Name"

* Update tests

- vector-ops (bin) stream syncs + runs with 4 queues per device
- improve counter-collection/input1 validation
- rocprofv3/tracing-hip-in-libraries does not do sys-trace
- improved validation script for tracing-hip-in-libraries
- updated dispatch_callback in json-tool.cpp following reworking of prototypes for counter collection

* Update samples/counter_collection

- updated dispatch_callback(s) and record_callback(s) following reworking of prototypes

* Update bin/rocprofv3

- reorganized help menu
- added options for sub-HSA tables
- added --hip-runtime-trace
- changed --hip-trace to include --hip-compiler-trace

* Update lib/rocprofiler-sdk-tool

- improved kernel filtering
- removed arch_vgpr, accum_vgpr, sgpr code (in rocprofiler-sdk)
- fixed issue with counter-collection w/o tracing
- added support for fine grained HSA API tracing
- removed directly linking to HSA-runtime

* Update lib/rocprofiler-sdk/agent.cpp

- rocp_agents != hsa_agents is non-fatal when ROCPROFILER_BUILD_CI=OFF (CMake option)

* GPR (vector and scalar) info in kernel symbol data

- rocprofiler_callback_tracing_code_object_kernel_symbol_register_data_t contains general purpose register info

* Header include order fix

- Include repo headers first
- Third party library headers next
- standard library headers last

* Update dispatch profiling public API

- introduce rocprofiler_profile_counting_dispatch_data_t
- change signature of rocprofiler_profile_counting_dispatch_callback_t and rocprofiler_profile_counting_record_callback_t
- provide rocprofiler_user_data_t pointer in dispatch callback
- provide rocprofiler_user_data_t value (from dispatch cb) in record callback

* Update tests/bin/CMakeLists.txt

- fix add_subdirectory(hip-in-libraries) order

* Update VERSION

- bump to 0.2.0 in prep for AFAR
Этот коммит содержится в:
Jonathan R. Madsen
2024-03-07 22:21:26 -06:00
коммит произвёл GitHub
родитель 665c546e65
Коммит 7b6d3c70bd
85 изменённых файлов: 2497 добавлений и 856 удалений
+48 -10
Просмотреть файл
@@ -28,17 +28,28 @@
#include <fstream>
#include <mutex>
#include <string>
#include <unordered_map>
namespace rocprofiler
{
namespace common
{
namespace
{
void
init_logging(std::string_view env_var)
install_failure_signal_handler()
{
static auto _once = std::once_flag{};
std::call_once(_once, [env_var]() {
std::call_once(_once, []() { google::InstallFailureSignalHandler(); });
}
} // namespace
void
init_logging(std::string_view env_var, logging_config cfg)
{
static auto _once = std::once_flag{};
std::call_once(_once, [env_var, &cfg]() {
auto get_argv0 = []() {
auto ifs = std::ifstream{"/proc/self/cmdline"};
auto sarg = std::string{};
@@ -50,18 +61,16 @@ init_logging(std::string_view env_var)
return sarg;
};
static auto argv0 = get_argv0();
google::InitGoogleLogging(argv0.c_str());
auto loglvl = common::get_env(env_var, "error");
auto loglvl = common::get_env(env_var, "");
for(auto& itr : loglvl)
itr = tolower(itr);
// default to warning
auto loglvl_v = google::WARNING;
if(loglvl.find_first_not_of("0123456789") == std::string::npos)
auto& loglvl_v = cfg.loglevel;
if(!loglvl.empty() && loglvl.find_first_not_of("0123456789") == std::string::npos)
{
loglvl_v = std::stoul(loglvl);
}
else
else if(!loglvl.empty())
{
const auto opts =
std::unordered_map<std::string_view, uint32_t>{{"info", google::INFO},
@@ -77,10 +86,39 @@ init_logging(std::string_view env_var)
loglvl_v = opts.at(loglvl);
}
FLAGS_minloglevel = loglvl_v;
FLAGS_stderrthreshold = loglvl_v;
update_logging(cfg, true);
if(!google::IsGoogleLoggingInitialized())
{
static auto argv0 = get_argv0();
google::InitGoogleLogging(argv0.c_str());
}
update_logging(cfg);
LOG(INFO) << "logging initialized via " << env_var;
});
}
void
update_logging(const logging_config& cfg, bool setup_env, int env_override)
{
static auto _mtx = std::mutex{};
auto _lk = std::unique_lock<std::mutex>{_mtx};
FLAGS_timestamp_in_logfile_name = false;
FLAGS_minloglevel = cfg.loglevel;
FLAGS_stderrthreshold = cfg.loglevel;
FLAGS_logtostderr = cfg.logtostderr;
FLAGS_alsologtostderr = cfg.alsologtostderr;
if(cfg.install_failure_handler) install_failure_signal_handler();
if(setup_env)
{
common::set_env("GOOGLE_LOG_DIR", get_env("PWD", ""), env_override);
common::set_env("GOOGLE_LOGTOSTDERR", cfg.loglevel, env_override);
common::set_env("GOOGLE_ALSOLOGTOSTDERR", cfg.alsologtostderr, env_override);
}
}
} // namespace common
} // namespace rocprofiler