HIP Streams to Queues Translation (#235)

* rocprofiler_stream_id_t: opaque handle for a stream - e.g. HIP stream - the same HIP stream may map to different HSA queues at different points in the application - added to: - rocprofiler_buffer_tracing_hip_api_record_t - rocprofiler_buffer_tracing_memory_copy_record_t - rocprofiler_callback_tracing_hip_api_data_t - rocprofiler_callback_tracing_memory_copy_data_t --------- Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com> Co-authored-by: Mark Meserve <mark.meserve@amd.com> Co-authored-by: Elwazir, Ammar <Ammar.Elwazir@amd.com> Co-authored-by: Ammar ELWazir <aelwazir@amd.com> Co-authored-by: Jakaraddi, Manjunath <Manjunath.Jakaraddi@amd.com> Co-authored-by: Bhardwaj, Gopesh <Gopesh.Bhardwaj@amd.com> Co-authored-by: Nagaraj, Sriraksha <Sriraksha.Nagaraj@amd.com> Co-authored-by: U, Srihari <Srihari.U@amd.com> Co-authored-by: Madsen, Jonathan <Jonathan.Madsen@amd.com> Co-authored-by: Welton, Benjamin <Benjamin.Welton@amd.com> Co-authored-by: Benjamin Welton <ben@amd.com> Co-authored-by: Indic, Vladimir <Vladimir.Indic@amd.com> Co-authored-by: Benjamin Welton <bewelton@amd.com> [ROCm/rocprofiler-sdk commit: ccd1e54293]
2025-03-14 02:45:13 -07:00
1 parent c08db2daa1
commit 7aeaffd871
@@ -41,4 +41,5 @@ add_subdirectory(rocjpeg-trace)
 if(TARGET att_decoder_testing)
    add_subdirectory(advanced-thread-trace)
 endif()
+add_subdirectory(hip-stream-display)
 add_subdirectory(agent-index)
@@ -0,0 +1,59 @@
+#
+#
+#
+cmake_minimum_required(VERSION 3.21.0 FATAL_ERROR)
+
+project(
+    rocprofiler-tests-rocprofv3-hip-stream-display
+    LANGUAGES CXX
+    VERSION 0.0.0)
+
+find_package(rocprofiler-sdk REQUIRED)
+find_package(rocDecode)
+
+rocprofiler_configure_pytest_files(CONFIG pytest.ini COPY validate.py conftest.py)
+
+string(REPLACE "LD_PRELOAD=" "ROCPROF_PRELOAD=" PRELOAD_ENV
+               "${ROCPROFILER_MEMCHECK_PRELOAD_ENV}")
+
+set(hip-stream-display-env "${PRELOAD_ENV}")
+
+add_test(
+    NAME rocprofv3-test-hip-stream-display-execute
+    COMMAND
+        $<TARGET_FILE:rocprofiler-sdk::rocprofv3> --kernel-rename -d
+        ${CMAKE_CURRENT_BINARY_DIR}/%tag%-trace -o out --output-format json pftrace
+        --log-level env -- $<TARGET_FILE:transpose>)
+
+set_tests_properties(
+    rocprofv3-test-hip-stream-display-execute
+    PROPERTIES TIMEOUT
+               45
+               LABELS
+               "integration-tests"
+               ENVIRONMENT
+               "${hip-stream-display-env}"
+               FAIL_REGULAR_EXPRESSION
+               "threw an exception"
+               DISABLED
+               $<NOT:$<TARGET_EXISTS:transpose>>)
+
+add_test(
+    NAME rocprofv3-test-hip-stream-display-validate
+    COMMAND
+        ${Python3_EXECUTABLE} ${CMAKE_CURRENT_BINARY_DIR}/validate.py --json-input
+        ${CMAKE_CURRENT_BINARY_DIR}/hip-stream-display/out_results.json --pftrace-input
+        ${CMAKE_CURRENT_BINARY_DIR}/hip-stream-display/out_results.pftrace)
+
+set_tests_properties(
+    rocprofv3-test-hip-stream-display-validate
+    PROPERTIES TIMEOUT
+               45
+               LABELS
+               "integration-tests"
+               DEPENDS
+               rocprofv3-test-hip-stream-display-execute
+               FAIL_REGULAR_EXPRESSION
+               "AssertionError"
+               DISABLED
+               $<NOT:$<TARGET_EXISTS:transpose>>)
@@ -0,0 +1,42 @@
+#!/usr/bin/env python3
+
+import csv
+import json
+import os
+import pytest
+
+from rocprofiler_sdk.pytest_utils.dotdict import dotdict
+from rocprofiler_sdk.pytest_utils import collapse_dict_list
+from rocprofiler_sdk.pytest_utils.perfetto_reader import PerfettoReader
+
+
+def pytest_addoption(parser):
+    parser.addoption(
+        "--json-input",
+        action="store",
+        default="hip-stream-display/out_results.json",
+        help="Input JSON",
+    )
+    parser.addoption(
+        "--pftrace-input",
+        action="store",
+        default="hip-stream-display/out_results.pftrace",
+        help="Input pftrace file",
+    )
+
+
+@pytest.fixture
+def json_data(request):
+    filename = request.config.getoption("--json-input")
+    if not os.path.isfile(filename):
+        return pytest.skip("stream tracing unavailable")
+    with open(filename, "r") as inp:
+        return dotdict(collapse_dict_list(json.load(inp)))
+
+
+@pytest.fixture
+def pftrace_data(request):
+    filename = request.config.getoption("--pftrace-input")
+    if not os.path.isfile(filename):
+        return pytest.skip("stream tracing unavailable")
+    return PerfettoReader(filename).read()[0]
@@ -0,0 +1,5 @@
+
+[pytest]
+addopts = --durations=20 -rA -s -vv
+testpaths = validate.py
+pythonpath = @ROCPROFILER_SDK_TESTS_BINARY_DIR@/pytest-packages
@@ -0,0 +1,75 @@
+#!/usr/bin/env python3
+
+import sys
+import pytest
+import json
+
+from collections import defaultdict
+
+
+# helper function
+def node_exists(name, data, min_len=1):
+    assert name in data
+    assert data[name] is not None
+    if isinstance(data[name], (list, tuple, dict, set)):
+        assert len(data[name]) >= min_len
+
+
+def get_operation(record, kind_name, op_name=None):
+    for idx, itr in enumerate(record["strings"]["buffer_records"]):
+        if kind_name == itr["kind"]:
+            if op_name is None:
+                return idx, itr["operations"]
+            else:
+                for oidx, oname in enumerate(itr["operations"]):
+                    if op_name == oname:
+                        return oidx
+    return None
+
+
+def test_stream_trace(json_data):
+    data = json_data["rocprofiler-sdk-tool"]
+    buffer_records = data["buffer_records"]
+
+    kernel_dispatch_data = buffer_records["kernel_dispatch"]
+    memory_copies_data = buffer_records["memory_copies"]
+    assert len(kernel_dispatch_data) > 0
+    assert len(memory_copies_data) > 0
+
+    # Expect stream ids to be set to 1 or 2 for transpose executable
+    expected_stream_ids = set((1, 2))
+    # check buffering data
+    for titr in (kernel_dispatch_data, memory_copies_data):
+        for node in rocdecode_data:
+            assert "size" in node
+            assert "kind" in node
+            assert "operation" in node
+            assert "correlation_id" in node
+            assert "end_timestamp" in node
+            assert "start_timestamp" in node
+            assert "thread_id" in node
+            assert "_stream_id" in node
+
+            assert node.size > 0
+            assert node.thread_id > 0
+            assert node.start_timestamp > 0
+            assert node.end_timestamp > 0
+            assert node.start_timestamp < node.end_timestamp
+
+            assert node._stream_id.handle in expected_stream_ids
+
+
+def test_perfetto_data(pftrace_data, json_data):
+    import rocprofiler_sdk.tests.rocprofv3 as rocprofv3
+
+    assert pftrace_data != None
+    rocprofv3.test_perfetto_data(
+        pftrace_data,
+        json_data,
+        ("kernel", "memory_copy"),
+    )
+
+
+if __name__ == "__main__":
+    exit_code = pytest.main(["-x", __file__] + sys.argv[1:])
+    sys.exit(exit_code)
@@ -137,7 +137,7 @@ def test_kernel_trace(json_data):
        external_corr_id = dispatch["correlation_id"]["external"]
        assert external_corr_id > 0

-        kernel_rename = get_kernel_rename(external_corr_id)
+        kernel_rename = get_kernel_rename(dispatch.kernel_rename_val)
        assert kernel_rename is not None, f"{dispatch}"
        assert kernel_rename != kernel_name, f"{dispatch}"
        assert (
@@ -75,6 +75,7 @@ def test_kernel_trace(json_data):

    def get_kernel_rename(corr_id):
        for itr in data.strings.correlation_id.external:
+            print(itr)
            if itr.key == corr_id:
                return itr.value
        return None
@@ -113,7 +114,7 @@ def test_kernel_trace(json_data):
        external_corr_id = dispatch["correlation_id"]["external"]
        assert external_corr_id > 0

-        kernel_rename = get_kernel_rename(external_corr_id)
+        kernel_rename = get_kernel_rename(dispatch.kernel_rename_val)
        assert kernel_rename is not None, f"{dispatch}"
        assert kernel_rename != kernel_name, f"{dispatch}"
        assert (