Agent, Counters, and AQL (#55)
* Migrate XML counter defs and reader from v1/v2 * Current Working Set * Modified parser * Evaluate AST Start * Update lib/common/xml - move definitions out of class declaration * Update lib/rocprofiler/counters/parser - update build of bison and flex build - reproducible generation - add ROCPROFILER_REGENERATE_COUNTERS_PARSER option - fix namespacing * Update lib/rocprofiler/counters/xml - change location of XML files and install them * Update lib/rocprofiler/counter/tests - normalize the test names - improve test failures (more clear about where failure is) * Update lib/rocprofiler/counters - fix namespace - update to new XML metrics directory * Update lib/rocprofiler/CMakeLists.txt - link to object library * Update lib/rocprofiler/hsa/types.hpp - reorganize includes * Add metric loading class/printers * Agent Implementation * Queue Implementation (#79) * Queue Implementation * API Implementation For Counters (part 1) (#80) * API Implementation For Counters * Bewelton/counter collection 3 (#84) * Added counter sample * More changes * More changes * Update samples/counter_collection - mostly formatting * Update include/rocprofiler/counters.h - formatting * Add lib.common/synchronized.hpp - Synchronized struct * Update lib/rocprofiler/counters/xml/basic_counters.xml - whitespace * Update scripts/patch-parser.cmake - tweaks for consistency * Update lib/rocprofiler/counters/parser/tests/parser_tests.cpp - formatting * Update lib/rocprofiler/counters/parser - improve consistency in rocprofiler-expr-parser-patch - update parser.{h,cpp} and scanner.cpp - formatting + regenerated * Update lib/rocprofiler/aql - formatting - clang-tidy fixes - guard against memory pool access errors * Update lib/rocprofiler/aql/tests - formatting - update use of get_val - normalize test names * Update lib/rocprofiler/counters/tests - formatting - patch basic_counters and derived_counters - normalize test names * Update lib/rocprofiler/aql/tests - set_tests_properties * Update test labels - fix minor issue with gtest labels * Update lib/rocprofiler/counters - formatting - clang-tidy fixes * Update lib/rocprofiler/hsa - fix includes - formatting - clang-tidy fixes - tweak to queue_controller_init interface * Update lib/rocprofiler - include fixes - namespace fixes - clang-tidy fixes - formatting * Update scripts/run-ci.py - exclude counters/parser from code coverage (generated files) * Update include/rocprofiler/counters.h - fix doxygen comment * Update lib/rocprofiler/aql/packet_construct.cpp - guard against HSA_AMD_MEMORY_POOL_ACCESS_DISALLOWED_BY_DEFAULT and HSA_AMD_MEMORY_POOL_ACCESS_NEVER_ALLOWED * Update lib/rocprofiler/counters/parser/raw_ast.hpp - clang-tidy fixes * Update lib/rocprofiler/counters/evaluate_ast.hpp - clang-tidy fixes * Update lib/rocprofiler/aql/tests - disable packet_generation_single and packet_generation_multi tests - the entire implementation rocprofiler::get_ext_table() is incorrect * Minor fixes before cleanup * More changes * More fixes * More fixes * source formatting (clang-format v11) (#99) Co-authored-by: bwelton <bwelton@users.noreply.github.com> * Revert PTL submodule * Update scripts/run-ci.py - exclude counters/parser from code coverage (generated files) * Migrating counters state to context * Linting * source formatting (clang-format v11) (#101) Co-authored-by: bwelton <bwelton@users.noreply.github.com> * revert run-ci * Testing fixes * More test changes * Fix minor typo * Small queue change * Small queue change * source formatting (clang-format v11) (#102) Co-authored-by: bwelton <bwelton@users.noreply.github.com> * source formatting (clang-format v11) (#105) Co-authored-by: bwelton <bwelton@users.noreply.github.com> * Documentation Change * More documentation fixes * source formatting (clang-format v11) (#106) Co-authored-by: bwelton <bwelton@users.noreply.github.com> * Threading fixes * Threading fixes * source formatting (clang-format v11) (#107) Co-authored-by: bwelton <bwelton@users.noreply.github.com> * Threading fixes * More test fixes * More agent fixes * More build fixes * source formatting (clang-format v11) (#109) Co-authored-by: bwelton <bwelton@users.noreply.github.com> * changed test timeouts * Build fix * Build fix * Updates to agent * source formatting (clang-format v11) (#114) Co-authored-by: bwelton <bwelton@users.noreply.github.com> * cmake formatting (cmake-format) (#113) Co-authored-by: bwelton <bwelton@users.noreply.github.com> * remove git worktree folder * Doc update * testing fix * Another test fix * More test changes * Rebase * source formatting (clang-format v11) (#116) Co-authored-by: bwelton <bwelton@users.noreply.github.com> * Documentation * source formatting (clang-format v11) (#119) Co-authored-by: bwelton <bwelton@users.noreply.github.com> * PTL Changes * Minor agent fix for empty labels * source formatting (clang-format v11) (#120) Co-authored-by: bwelton <bwelton@users.noreply.github.com> * Minor agent fix for empty labels * Refactor read_map * source formatting (clang-format v11) (#121) Co-authored-by: bwelton <bwelton@users.noreply.github.com> * Refactor read_map * Cache fixes * source formatting (clang-format v11) (#122) Co-authored-by: bwelton <bwelton@users.noreply.github.com> --------- Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com> Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: bwelton <bwelton@users.noreply.github.com>
This commit is contained in:
Benjamin Welton
committed by
GitHub
GitHub
orang tua
a798a26227
melakukan
010693b795
@@ -18,3 +18,4 @@ endif()
|
||||
add_subdirectory(pc_sampling)
|
||||
add_subdirectory(api_callback_tracing)
|
||||
add_subdirectory(api_buffered_tracing)
|
||||
add_subdirectory(counter_collection)
|
||||
|
||||
@@ -0,0 +1,55 @@
|
||||
#
|
||||
#
|
||||
#
|
||||
cmake_minimum_required(VERSION 3.21.0 FATAL_ERROR)
|
||||
|
||||
if(NOT CMAKE_HIP_COMPILER)
|
||||
find_program(
|
||||
amdclangpp_EXECUTABLE
|
||||
NAMES amdclang++
|
||||
HINTS ${ROCM_PATH} ENV ROCM_PATH /opt/rocm
|
||||
PATHS ${ROCM_PATH} ENV ROCM_PATH /opt/rocm
|
||||
PATH_SUFFIXES bin llvm/bin NO_CACHE)
|
||||
mark_as_advanced(amdclangpp_EXECUTABLE)
|
||||
|
||||
if(amdclangpp_EXECUTABLE)
|
||||
set(CMAKE_HIP_COMPILER "${amdclangpp_EXECUTABLE}")
|
||||
endif()
|
||||
endif()
|
||||
|
||||
project(rocprofiler-samples-counter-collection LANGUAGES CXX HIP)
|
||||
|
||||
foreach(_TYPE DEBUG MINSIZEREL RELEASE RELWITHDEBINFO)
|
||||
if("${CMAKE_HIP_FLAGS_${_TYPE}}" STREQUAL "")
|
||||
set(CMAKE_HIP_FLAGS_${_TYPE} "${CMAKE_CXX_FLAGS_${_TYPE}}")
|
||||
endif()
|
||||
endforeach()
|
||||
|
||||
if(NOT TARGET rocprofiler::rocprofiler)
|
||||
find_package(rocprofiler REQUIRED)
|
||||
endif()
|
||||
|
||||
add_library(counter-collection-client SHARED)
|
||||
target_sources(counter-collection-client PRIVATE client.cpp client.hpp)
|
||||
target_link_libraries(counter-collection-client PRIVATE rocprofiler::rocprofiler)
|
||||
|
||||
set_source_files_properties(main.cpp PROPERTIES LANGUAGE HIP)
|
||||
find_package(Threads REQUIRED)
|
||||
|
||||
add_executable(counter-collection)
|
||||
target_sources(counter-collection PRIVATE main.cpp)
|
||||
target_link_libraries(counter-collection PRIVATE counter-collection-client
|
||||
Threads::Threads)
|
||||
|
||||
add_test(NAME counter-collection COMMAND $<TARGET_FILE:counter-collection>)
|
||||
|
||||
set_tests_properties(
|
||||
counter-collection
|
||||
PROPERTIES
|
||||
TIMEOUT
|
||||
45
|
||||
LABELS
|
||||
"samples"
|
||||
ENVIRONMENT
|
||||
"${ROCPROFILER_MEMCHECK_PRELOAD_ENV};HSA_TOOLS_LIB=$<TARGET_FILE:rocprofiler::rocprofiler-library>"
|
||||
)
|
||||
@@ -0,0 +1,164 @@
|
||||
#include "client.hpp"
|
||||
|
||||
#include <set>
|
||||
#include <sstream> // std::stringstream
|
||||
#include <vector>
|
||||
|
||||
#include <rocprofiler/registration.h>
|
||||
#include <rocprofiler/rocprofiler.h>
|
||||
|
||||
#define ROCPROFILER_CALL(result, msg) \
|
||||
{ \
|
||||
rocprofiler_status_t CHECKSTATUS = result; \
|
||||
if(CHECKSTATUS != ROCPROFILER_STATUS_SUCCESS) \
|
||||
{ \
|
||||
std::cerr << #result << " failed with error code " << CHECKSTATUS << std::endl; \
|
||||
throw std::runtime_error(#result " failure"); \
|
||||
} \
|
||||
}
|
||||
|
||||
int
|
||||
start()
|
||||
{
|
||||
return 1;
|
||||
}
|
||||
|
||||
namespace
|
||||
{
|
||||
rocprofiler_context_id_t&
|
||||
get_client_ctx()
|
||||
{
|
||||
static rocprofiler_context_id_t ctx;
|
||||
return ctx;
|
||||
}
|
||||
|
||||
void
|
||||
test_callback(rocprofiler_queue_id_t queue_id,
|
||||
rocprofiler_agent_t agent_id,
|
||||
rocprofiler_correlation_id_t corr_id,
|
||||
const hsa_kernel_dispatch_packet_t*,
|
||||
void*,
|
||||
rocprofiler_dispatch_profile_counting_record_t**,
|
||||
size_t,
|
||||
rocprofiler_profile_config_id_t)
|
||||
{
|
||||
// Callback containing counter data.
|
||||
std::clog << "[" << __FUNCTION__ << "] " << queue_id.handle << " | " << agent_id.id.handle
|
||||
<< " | " << corr_id.id << "\n";
|
||||
}
|
||||
|
||||
int
|
||||
tool_init(rocprofiler_client_finalize_t, void*)
|
||||
{
|
||||
std::set<std::string> counters_to_collect = {"SQ_WAVES"};
|
||||
|
||||
std::vector<rocprofiler_agent_t> gpu_agents;
|
||||
auto agent_query = [](const rocprofiler_agent_t** agents, size_t num_agents, void* user_data) {
|
||||
std::vector<rocprofiler_agent_t>* vec =
|
||||
static_cast<std::vector<rocprofiler_agent_t>*>(user_data);
|
||||
for(size_t i = 0; i < num_agents; i++)
|
||||
{
|
||||
const rocprofiler_agent_t* agent = agents[i];
|
||||
if(agent->type == ROCPROFILER_AGENT_TYPE_GPU)
|
||||
{
|
||||
vec->push_back(*agent);
|
||||
}
|
||||
}
|
||||
return ROCPROFILER_STATUS_SUCCESS;
|
||||
};
|
||||
|
||||
ROCPROFILER_CALL(rocprofiler_query_available_agents(
|
||||
agent_query, sizeof(rocprofiler_agent_t), static_cast<void*>(&gpu_agents)),
|
||||
"Could not query agents");
|
||||
|
||||
ROCPROFILER_CALL(rocprofiler_create_context(&get_client_ctx()), "context creation failed");
|
||||
|
||||
std::vector<rocprofiler_profile_config_id_t> profile_configs;
|
||||
|
||||
for(auto& agent : gpu_agents)
|
||||
{
|
||||
std::vector<rocprofiler_counter_id_t> collect_counters;
|
||||
std::vector<rocprofiler_counter_id_t> gpu_counters;
|
||||
std::clog << agent.name << "\n";
|
||||
ROCPROFILER_CALL(
|
||||
rocprofiler_iterate_agent_supported_counters(
|
||||
agent,
|
||||
[](rocprofiler_counter_id_t* counters, size_t num_counters, void* user_data) {
|
||||
std::vector<rocprofiler_counter_id_t>* vec =
|
||||
static_cast<std::vector<rocprofiler_counter_id_t>*>(user_data);
|
||||
for(size_t i = 0; i < num_counters; i++)
|
||||
{
|
||||
vec->push_back(counters[i]);
|
||||
}
|
||||
return ROCPROFILER_STATUS_SUCCESS;
|
||||
},
|
||||
static_cast<void*>(&gpu_counters)),
|
||||
"Could not fetch supported counters");
|
||||
|
||||
for(auto& counter : gpu_counters)
|
||||
{
|
||||
const char* name;
|
||||
size_t size;
|
||||
ROCPROFILER_CALL(rocprofiler_query_counter_name(counter, &name, &size),
|
||||
"Could not query name");
|
||||
if(counters_to_collect.count(std::string(name)) > 0)
|
||||
{
|
||||
std::clog << "Counter: " << counter.handle << " " << name << "\n";
|
||||
collect_counters.push_back(counter);
|
||||
}
|
||||
}
|
||||
rocprofiler_profile_config_id_t profile;
|
||||
ROCPROFILER_CALL(rocprofiler_create_profile_config(
|
||||
agent, collect_counters.data(), collect_counters.size(), &profile),
|
||||
"Could not construct profile cfg");
|
||||
ROCPROFILER_CALL(rocprofiler_configure_dispatch_profile_counting_service(
|
||||
get_client_ctx(), profile, test_callback, nullptr),
|
||||
"Could not setup dispatch service");
|
||||
profile_configs.push_back(profile);
|
||||
}
|
||||
rocprofiler_start_context(get_client_ctx());
|
||||
|
||||
// no errors
|
||||
return 0;
|
||||
}
|
||||
|
||||
void
|
||||
tool_fini(void*)
|
||||
{
|
||||
rocprofiler_stop_context(get_client_ctx());
|
||||
std::clog << "In tool fini\n";
|
||||
}
|
||||
|
||||
} // namespace
|
||||
|
||||
extern "C" rocprofiler_tool_configure_result_t*
|
||||
rocprofiler_configure(uint32_t version,
|
||||
const char* runtime_version,
|
||||
uint32_t,
|
||||
rocprofiler_client_id_t* id)
|
||||
{
|
||||
// set the client name
|
||||
id->name = "CounterClientSample";
|
||||
|
||||
// compute major/minor/patch version info
|
||||
uint32_t major = version / 10000;
|
||||
uint32_t minor = (version % 10000) / 100;
|
||||
uint32_t patch = version % 100;
|
||||
|
||||
// generate info string
|
||||
auto info = std::stringstream{};
|
||||
info << id->name << " is using rocprofiler v" << major << "." << minor << "." << patch << " ("
|
||||
<< runtime_version << ")";
|
||||
|
||||
std::clog << info.str() << std::endl;
|
||||
|
||||
// create configure data
|
||||
static auto cfg =
|
||||
rocprofiler_tool_configure_result_t{sizeof(rocprofiler_tool_configure_result_t),
|
||||
&tool_init,
|
||||
&tool_fini,
|
||||
static_cast<void*>(nullptr)};
|
||||
|
||||
// return pointer to configure data
|
||||
return &cfg;
|
||||
}
|
||||
@@ -0,0 +1,8 @@
|
||||
#pragma once
|
||||
|
||||
#include <iostream>
|
||||
|
||||
#define CLIENT_API __attribute__((visibility("default")))
|
||||
|
||||
int
|
||||
start() CLIENT_API;
|
||||
@@ -0,0 +1,92 @@
|
||||
#include <hip/hip_runtime.h>
|
||||
|
||||
#include "client.hpp"
|
||||
|
||||
#define HIP_CALL(call) \
|
||||
do \
|
||||
{ \
|
||||
hipError_t err = call; \
|
||||
if(err != hipSuccess) \
|
||||
{ \
|
||||
fprintf(stderr, "%s\n", hipGetErrorString(err)); \
|
||||
abort(); \
|
||||
} \
|
||||
} while(0)
|
||||
|
||||
__global__ void
|
||||
kernelA(int x, int y)
|
||||
{
|
||||
x = x + y;
|
||||
}
|
||||
|
||||
__global__ void
|
||||
kernelB(int x, int y)
|
||||
{
|
||||
x = x + y;
|
||||
}
|
||||
|
||||
template <typename T>
|
||||
__global__ void
|
||||
kernelC(T* C_d, const T* A_d, size_t N)
|
||||
{
|
||||
size_t offset = (blockIdx.x * blockDim.x + threadIdx.x);
|
||||
size_t stride = blockDim.x * gridDim.x;
|
||||
for(size_t i = offset; i < N; i += stride)
|
||||
{
|
||||
C_d[i] = A_d[i] * A_d[i];
|
||||
}
|
||||
}
|
||||
|
||||
void
|
||||
launchKernals()
|
||||
{
|
||||
const int NUM_LAUNCH = 1000;
|
||||
// Normal HIP Calls
|
||||
int* gpuMem;
|
||||
[[maybe_unused]] hipDeviceProp_t devProp;
|
||||
HIP_CALL(hipGetDeviceProperties(&devProp, 0));
|
||||
HIP_CALL(hipMalloc((void**) &gpuMem, 1 * sizeof(int)));
|
||||
|
||||
for(int i = 0; i < NUM_LAUNCH; i++)
|
||||
{
|
||||
// KernelA and KernelB to be profiled as part of the session
|
||||
hipLaunchKernelGGL(kernelA, dim3(1), dim3(1), 0, 0, 1, 2);
|
||||
hipLaunchKernelGGL(kernelB, dim3(1), dim3(1), 0, 0, 1, 2);
|
||||
}
|
||||
|
||||
const int NElems = 512 * 512;
|
||||
const int Nbytes = NElems * 2;
|
||||
int * A_d, *C_d;
|
||||
int A_h[NElems], C_h[NElems];
|
||||
|
||||
for(int i = 0; i < NElems; i++)
|
||||
{
|
||||
A_h[i] = i;
|
||||
}
|
||||
|
||||
HIP_CALL(hipDeviceSynchronize());
|
||||
|
||||
HIP_CALL(hipMalloc(&A_d, Nbytes));
|
||||
HIP_CALL(hipMalloc(&C_d, Nbytes));
|
||||
HIP_CALL(hipMemcpy(A_d, A_h, Nbytes, hipMemcpyHostToDevice));
|
||||
HIP_CALL(hipDeviceSynchronize());
|
||||
const unsigned blocks = 512;
|
||||
const unsigned threadsPerBlock = 256;
|
||||
for(int i = 0; i < NUM_LAUNCH; i++)
|
||||
{
|
||||
hipLaunchKernelGGL(kernelC, dim3(blocks), dim3(threadsPerBlock), 0, 0, C_d, A_d, NElems);
|
||||
}
|
||||
HIP_CALL(hipMemcpy(C_h, C_d, Nbytes, hipMemcpyDeviceToHost));
|
||||
HIP_CALL(hipDeviceSynchronize());
|
||||
HIP_CALL(hipFree(gpuMem));
|
||||
HIP_CALL(hipFree(A_d));
|
||||
HIP_CALL(hipFree(C_d));
|
||||
std::cerr << "Run complete\n";
|
||||
}
|
||||
|
||||
int
|
||||
main()
|
||||
{
|
||||
start();
|
||||
launchKernals();
|
||||
}
|
||||
Reference in New Issue
Block a user