tritonparse 0.2.4.dev20251008071501__tar.gz → 0.2.4.dev20251010071509__tar.gz

{tritonparse-0.2.4.dev20251008071501 → tritonparse-0.2.4.dev20251010071509}/.github/workflows/test.yml

@@ -142,7 +142,7 @@ jobs:
                   TRITON_COMMIT: ${{ steps.triton-commit.outputs.commit }}
               run: |
                   bash .ci/install-triton.sh
-                  bash .ci/install-triton-kernels.sh
+
             - name: Install project dependencies
               env:
                   CONDA_ENV: tritonparse
@@ -192,12 +192,6 @@ jobs:
               run: |
                   bash .ci/setup.sh
 
-            - name: Install Triton kernels
-              env:
-                  CONDA_ENV: tritonparse-pip
-              run: |
-                  bash .ci/install-triton-kernels.sh
-
             - name: Install project dependencies
               env:
                   CONDA_ENV: tritonparse-pip

{tritonparse-0.2.4.dev20251008071501 → tritonparse-0.2.4.dev20251010071509}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: tritonparse
-Version: 0.2.4.dev20251008071501
+Version: 0.2.4.dev20251010071509
 Summary: TritonParse: A Compiler Tracer, Visualizer, and mini-Reproducer Generator for Triton Kernels
 Author-email: Yueming Hao <yhao@meta.com>
 License-Expression: BSD-3-Clause
@@ -34,13 +34,18 @@ Dynamic: license-file
 - **📝 Multi-format IR Support** - View TTGIR, TTIR, LLIR, PTX, and AMDGCN
 - **🎯 Interactive Code Views** - Click-to-highlight corresponding lines across IR stages
 
+### 🔧 Reproducer & Debugging Tools
+- **🔄 Standalone Script Generation** - Extract any kernel into a self-contained Python script
+- **💾 Tensor Data Reconstruction** - Preserve actual tensor data or use statistical approximation
+- **🎯 Custom Templates** - Flexible reproducer templates for different workflows
+- **🐛 Bug Isolation** - Share reproducible test cases for debugging and collaboration
+
 ### 📊 Structured Logging & Analysis
 - **📝 Compilation & Launch Tracing** - Capture detailed events with source mapping
 - **🔍 Stack Trace Integration** - Full Python stack traces for debugging
 - **📈 Metadata Extraction** - Comprehensive kernel statistics
 
 ### 🛠️ Developer Tools
-- **🔧 Reproducer Generation** - Generate standalone Python scripts to reproduce kernels
 - **🌐 Browser-based Interface** - No installation required, works in your browser
 - **🔒 Privacy-first** - All processing happens locally, no data uploaded
 
@@ -87,6 +92,41 @@ tritonparse.utils.unified_parse("./logs/", out="./parsed_output")
 
 > **🔒 Privacy Note**: Your trace files are processed entirely in your browser - nothing is uploaded to any server!
 
+### 3. Generate Reproducers (Optional)
+
+Extract any kernel into a standalone, executable Python script for debugging or testing:
+
+```bash
+# Generate reproducer from first launch event
+tritonparse reproduce ./parsed_output/trace.ndjson.gz --line 2 --out-dir repro_output
+
+# Run the generated reproducer
+cd repro_output/<kernel_name>/
+python repro_*.py
+```
+
+**Python API:**
+```python
+from tritonparse.reproducer.orchestrator import reproduce
+
+result = reproduce(
+    input_path="./parsed_output/trace.ndjson.gz",
+    line_index=1,           # Which launch event (1-based)
+    out_dir="repro_output"
+)
+```
+
+<details>
+<summary>🎯 Common Reproducer Use Cases (click to expand)</summary>
+
+- **🐛 Bug Isolation**: Extract a failing kernel into a minimal standalone script
+- **⚡ Performance Testing**: Benchmark specific kernels without running the full application
+- **🤝 Team Collaboration**: Share reproducible test cases with colleagues or in bug reports
+- **📊 Regression Testing**: Compare kernel behavior and performance across different versions
+- **🔍 Deep Debugging**: Modify and experiment with kernel parameters in isolation
+
+</details>
+
 ## 🛠️ Installation
 
 **For basic usage (trace generation):**
@@ -117,7 +157,7 @@ pip install triton
 |----------|-------------|
 | **[🏠 Wiki Home](https://github.com/meta-pytorch/tritonparse/wiki)** | Complete documentation and quick navigation |
 | **[📦 Installation](https://github.com/meta-pytorch/tritonparse/wiki/01.-Installation)** | Setup guide for all scenarios |
-| **[📋 Usage Guide](https://github.com/meta-pytorch/tritonparse/wiki/02.-Usage-Guide)** | Complete workflow, examples, and reproducer |
+| **[📋 Usage Guide](https://github.com/meta-pytorch/tritonparse/wiki/02.-Usage-Guide)** | Complete workflow, reproducer generation, and examples |
 | **[🌐 Web Interface](https://github.com/meta-pytorch/tritonparse/wiki/03.-Web-Interface-Guide)** | Master the visualization interface |
 | **[🔧 Developer Guide](https://github.com/meta-pytorch/tritonparse/wiki/04.-Developer-Guide)** | Contributing and architecture overview |
 | **[📝 Code Formatting](https://github.com/meta-pytorch/tritonparse/wiki/05.-Code-Formatting)** | Formatting standards and tools |

{tritonparse-0.2.4.dev20251008071501 → tritonparse-0.2.4.dev20251010071509}/README.md

@@ -16,13 +16,18 @@
 - **📝 Multi-format IR Support** - View TTGIR, TTIR, LLIR, PTX, and AMDGCN
 - **🎯 Interactive Code Views** - Click-to-highlight corresponding lines across IR stages
 
+### 🔧 Reproducer & Debugging Tools
+- **🔄 Standalone Script Generation** - Extract any kernel into a self-contained Python script
+- **💾 Tensor Data Reconstruction** - Preserve actual tensor data or use statistical approximation
+- **🎯 Custom Templates** - Flexible reproducer templates for different workflows
+- **🐛 Bug Isolation** - Share reproducible test cases for debugging and collaboration
+
 ### 📊 Structured Logging & Analysis
 - **📝 Compilation & Launch Tracing** - Capture detailed events with source mapping
 - **🔍 Stack Trace Integration** - Full Python stack traces for debugging
 - **📈 Metadata Extraction** - Comprehensive kernel statistics
 
 ### 🛠️ Developer Tools
-- **🔧 Reproducer Generation** - Generate standalone Python scripts to reproduce kernels
 - **🌐 Browser-based Interface** - No installation required, works in your browser
 - **🔒 Privacy-first** - All processing happens locally, no data uploaded
 
@@ -69,6 +74,41 @@ tritonparse.utils.unified_parse("./logs/", out="./parsed_output")
 
 > **🔒 Privacy Note**: Your trace files are processed entirely in your browser - nothing is uploaded to any server!
 
+### 3. Generate Reproducers (Optional)
+
+Extract any kernel into a standalone, executable Python script for debugging or testing:
+
+```bash
+# Generate reproducer from first launch event
+tritonparse reproduce ./parsed_output/trace.ndjson.gz --line 2 --out-dir repro_output
+
+# Run the generated reproducer
+cd repro_output/<kernel_name>/
+python repro_*.py
+```
+
+**Python API:**
+```python
+from tritonparse.reproducer.orchestrator import reproduce
+
+result = reproduce(
+    input_path="./parsed_output/trace.ndjson.gz",
+    line_index=1,           # Which launch event (1-based)
+    out_dir="repro_output"
+)
+```
+
+<details>
+<summary>🎯 Common Reproducer Use Cases (click to expand)</summary>
+
+- **🐛 Bug Isolation**: Extract a failing kernel into a minimal standalone script
+- **⚡ Performance Testing**: Benchmark specific kernels without running the full application
+- **🤝 Team Collaboration**: Share reproducible test cases with colleagues or in bug reports
+- **📊 Regression Testing**: Compare kernel behavior and performance across different versions
+- **🔍 Deep Debugging**: Modify and experiment with kernel parameters in isolation
+
+</details>
+
 ## 🛠️ Installation
 
 **For basic usage (trace generation):**
@@ -99,7 +139,7 @@ pip install triton
 |----------|-------------|
 | **[🏠 Wiki Home](https://github.com/meta-pytorch/tritonparse/wiki)** | Complete documentation and quick navigation |
 | **[📦 Installation](https://github.com/meta-pytorch/tritonparse/wiki/01.-Installation)** | Setup guide for all scenarios |
-| **[📋 Usage Guide](https://github.com/meta-pytorch/tritonparse/wiki/02.-Usage-Guide)** | Complete workflow, examples, and reproducer |
+| **[📋 Usage Guide](https://github.com/meta-pytorch/tritonparse/wiki/02.-Usage-Guide)** | Complete workflow, reproducer generation, and examples |
 | **[🌐 Web Interface](https://github.com/meta-pytorch/tritonparse/wiki/03.-Web-Interface-Guide)** | Master the visualization interface |
 | **[🔧 Developer Guide](https://github.com/meta-pytorch/tritonparse/wiki/04.-Developer-Guide)** | Contributing and architecture overview |
 | **[📝 Code Formatting](https://github.com/meta-pytorch/tritonparse/wiki/05.-Code-Formatting)** | Formatting standards and tools |

{tritonparse-0.2.4.dev20251008071501 → tritonparse-0.2.4.dev20251010071509}/tests/test_tritonparse.py

@@ -7,7 +7,6 @@ TORCHINDUCTOR_FX_GRAPH_CACHE=0 TRITONPARSE_DEBUG=1 python -m unittest tests.test
 """
 
 import gzip
-import importlib.util
 import json
 import os
 import shutil
@@ -33,8 +32,6 @@ from tritonparse.shared_vars import TEST_KEEP_OUTPUT
 from tritonparse.structured_logging import convert, extract_python_source_info
 from tritonparse.tools.disasm import is_nvdisasm_available
 
-HAS_TRITON_KERNELS = importlib.util.find_spec("triton_kernels") is not None
-
 
 def create_fresh_triton_cache():
     """Create a fresh Triton cache directory and return cache management context"""
@@ -108,7 +105,6 @@ def clear_all_caches(*kernels):
     # Reset torch compiler state
     torch.compiler.reset()
     torch._dynamo.reset()
-    torch._inductor.metrics.reset()
     print("✓ Reset torch compiler, dynamo, and inductor state")
 
     # Clear Triton kernel device caches for all provided kernels
@@ -1053,99 +1049,6 @@ class TestTritonparseCUDA(unittest.TestCase):
                 print("✓ Cleaned up temporary directory")
             tritonparse.structured_logging.clear_logging_config()
 
-    @unittest.skipUnless(torch.cuda.is_available(), "CUDA not available")
-    @unittest.skipUnless(HAS_TRITON_KERNELS, "triton_kernels not installed")
-    def test_triton_kernels_Tensor(self):
-        from triton_kernels.topk_details._topk_forward import _topk_forward
-        from tritonparse.reproducer import utils as reproducer_utils
-
-        input_json_path = os.path.join(
-            os.path.dirname(__file__),
-            "example_output",
-            "repro",
-            "repro_context_20250816192455.json",
-        )
-        grid, args_dict = reproducer_utils.create_args_from_json(input_json_path)
-        print("Generated kernel arguments dictionary:")
-        for name, arg in args_dict.items():
-            print(f"  {name}: {arg}")
-        print(f"Grid: {grid}")
-
-        # Set up test environment
-        temp_dir = tempfile.mkdtemp()
-        temp_dir_logs = os.path.join(temp_dir, "logs")
-        temp_dir_parsed = os.path.join(temp_dir, "parsed_output")
-        os.makedirs(temp_dir_logs, exist_ok=True)
-        os.makedirs(temp_dir_parsed, exist_ok=True)
-        print(f"Temporary directory: {temp_dir}")
-
-        # Initialize logging
-        tritonparse.structured_logging.init(temp_dir_logs, enable_trace_launch=True)
-        try:
-            _topk_forward[tuple(grid)](
-                args_dict["X"],
-                args_dict["stride_xm"],
-                args_dict["Yv"],
-                args_dict["Yi"],
-                args_dict["stride_ym"],
-                args_dict["USE_PROVIDED_INDX"],
-                args_dict["Bits"],
-                args_dict["stride_rm"],
-                args_dict["stride_rn"],
-                args_dict["n_rows"],
-                args_dict["n_expts_tot"],
-                args_dict["S"],
-                args_dict["BLOCK_S"],
-                args_dict["s_blocks"],
-                args_dict["APPLY_SOFTMAX"],
-                args_dict["BLOCK_M"],
-                args_dict["N_EXPTS_PAD"],
-                args_dict["N_EXPTS_ACT"],
-                args_dict["BLOCK_N"],
-            )
-            torch.cuda.synchronize()
-
-            # Read the only ndjson file under temp_dir_logs/logs/
-            ndjson_files = [
-                f for f in os.listdir(temp_dir_logs) if f.endswith(".ndjson")
-            ]
-            assert (
-                len(ndjson_files) == 1
-            ), f"Expected exactly one ndjson in {temp_dir_logs}, found {len(ndjson_files)}: {ndjson_files}"
-
-            ndjson_path = os.path.join(temp_dir_logs, ndjson_files[0])
-            launches = []
-            with open(ndjson_path, "r") as f:
-                for line_num, line in enumerate(f, 1):
-                    try:
-                        event = json.loads(line.strip())
-                        if event.get("event_type") == "launch":
-                            launches.append(event)
-                    except json.JSONDecodeError as e:
-                        self.fail(f"JSON decode error at line {line_num}: {e}")
-
-            assert (
-                len(launches) == 1
-            ), f"Expected exactly one launch event, found {len(launches)}"
-
-            launch = launches[0]
-            extracted_args = launch.get("extracted_args", {})
-            assert "X" in extracted_args, "Missing 'X' in extracted_args"
-            assert (
-                extracted_args["X"].get("type") == "triton_kernels.tensor.Tensor"
-            ), f"X.type is {extracted_args['X'].get('type')}"
-
-        finally:
-            # Clean up
-            if TEST_KEEP_OUTPUT:
-                print(
-                    f"✓ Preserving temporary directory (TEST_KEEP_OUTPUT=1): {temp_dir}"
-                )
-            else:
-                shutil.rmtree(temp_dir)
-                print("✓ Cleaned up temporary directory")
-            tritonparse.structured_logging.clear_logging_config()
-
     @unittest.skipIf(is_fbcode(), "Skip in internal FB environment")
     @unittest.skipUnless(torch.cuda.is_available(), "CUDA not available")
     def test_reproducer_end_to_end(self):
@@ -1264,6 +1167,281 @@ class TestTritonparseCUDA(unittest.TestCase):
             shutil.rmtree(temp_dir)
             print("✓ Cleaned up temporary directory")
 
+    @unittest.skipUnless(torch.cuda.is_available(), "CUDA not available")
+    def test_tensor_blob_manager(self):
+        """Test TensorBlobManager functionality with context manager"""
+
+        # Setup fresh cache for this test
+        test_cache_dir, prev_cache_dir = self.setup_test_with_fresh_cache()
+
+        # Define a simple kernel that accepts tensor inputs
+        @triton.jit
+        def tensor_input_kernel(
+            input_ptr,
+            output_ptr,
+            n_elements,
+            BLOCK_SIZE: tl.constexpr,
+        ):
+            pid = tl.program_id(axis=0)
+            block_start = pid * BLOCK_SIZE
+            offsets = block_start + tl.arange(0, BLOCK_SIZE)
+            mask = offsets < n_elements
+
+            x = tl.load(input_ptr + offsets, mask=mask)
+            y = x * 2.0
+            tl.store(output_ptr + offsets, y, mask=mask)
+
+        def run_kernel(input_tensor):
+            n_elements = input_tensor.numel()
+            output = torch.empty_like(input_tensor)
+            BLOCK_SIZE = 256
+            grid = (triton.cdiv(n_elements, BLOCK_SIZE),)
+            tensor_input_kernel[grid](input_tensor, output, n_elements, BLOCK_SIZE)
+            return output
+
+        def collect_blob_files(manager_dir_path):
+            """Collect all .bin and .bin.gz files from saved_tensors directory."""
+            saved_tensors_dir = os.path.join(manager_dir_path, "saved_tensors")
+            bin_files = []
+            gz_files = []
+
+            if not os.path.exists(saved_tensors_dir):
+                return bin_files, gz_files
+
+            for subdir in os.listdir(saved_tensors_dir):
+                subdir_path = os.path.join(saved_tensors_dir, subdir)
+                if os.path.isdir(subdir_path):
+                    for filename in os.listdir(subdir_path):
+                        full_path = os.path.join(subdir_path, filename)
+                        if filename.endswith(".bin.gz"):
+                            gz_files.append(full_path)
+                        elif filename.endswith(".bin"):
+                            bin_files.append(full_path)
+
+            return bin_files, gz_files
+
+        def count_all_blobs(manager_dir_path):
+            """Count total number of blob files (.bin and .bin.gz)."""
+            bin_files, gz_files = collect_blob_files(manager_dir_path)
+            return len(bin_files) + len(gz_files)
+
+        # Prepare test data
+        torch.manual_seed(0)
+
+        # === Test 1: Mixed tensor sizes with compression threshold ===
+        print("\n=== Test 1: Mixed Tensor Sizes with Compression Threshold ===")
+        temp_output_dir_1 = tempfile.mkdtemp()
+
+        with tritonparse.context_manager.TritonParseManager(
+            enable_trace_launch=True,
+            enable_tensor_blob_storage=True,
+            out=temp_output_dir_1,
+        ) as manager:
+            # Test different tensor sizes around the 1MB compression threshold
+            test_cases = [
+                ((512,), "Tiny 2KB"),  # 2KB < 1MB -> .bin
+                ((100 * 1024,), "Medium 400KB"),  # 400KB < 1MB -> .bin
+                ((5 * 1024 * 1024,), "Large 20MB"),  # 20MB > 1MB -> .bin.gz
+                ((100 * 1024 * 1024,), "Very large 400MB"),  # 400MB > 1MB -> .bin.gz
+            ]
+
+            # Create tensors and run kernels
+            for size, desc in test_cases:
+                x = torch.randn(size, device=self.cuda_device, dtype=torch.float32)
+                y = run_kernel(x)
+                y.sum()
+            torch.cuda.synchronize()
+
+            # Collect and verify blob files
+            bin_files, gz_files = collect_blob_files(manager.dir_path)
+            assert len(bin_files) + len(gz_files) > 0, "No blob files found"
+
+            print(f"Found {len(bin_files)} .bin files:")
+            for f in bin_files:
+                print(f"  {f} ({os.path.getsize(f)} bytes)")
+            print(f"Found {len(gz_files)} .bin.gz files:")
+            for f in gz_files:
+                print(f"  {f} ({os.path.getsize(f)} bytes)")
+
+            # Verify correct number of files (2 small uncompressed, 2 large compressed)
+            assert (
+                len(bin_files) == 4
+            ), f"Expected 4 .bin files (2KB, 400KB), got {len(bin_files)}"
+            assert (
+                len(gz_files) == 4
+            ), f"Expected 4 .bin.gz files (20MB, 400MB), got {len(gz_files)}"
+
+            print(
+                f"✓ Mixed sizes: {len(bin_files)} uncompressed (.bin), {len(gz_files)} compressed (.bin.gz)"
+            )
+
+            # Verify both formats can be loaded
+            from tritonparse.tools.load_tensor import load_tensor
+
+            if bin_files:
+                loaded = load_tensor(bin_files[0])
+                assert loaded is not None, "Failed to load .bin file"
+                print("✓ Successfully loaded .bin file")
+
+            if gz_files:
+                loaded = load_tensor(gz_files[0])
+                assert loaded is not None, "Failed to load .bin.gz file"
+                print("✓ Successfully loaded .bin.gz file")
+
+            print("✓ Both formats (.bin and .bin.gz) verified")
+
+        # === Test 2: Deduplication ===
+        print("\n=== Test 2: Deduplication ===")
+        temp_output_dir_2 = tempfile.mkdtemp()
+
+        with tritonparse.context_manager.TritonParseManager(
+            enable_trace_launch=True,
+            enable_tensor_blob_storage=True,
+            out=temp_output_dir_2,
+        ) as manager:
+            # Use the same tensor multiple times
+            x = torch.randn((512,), device=self.cuda_device, dtype=torch.float32)
+
+            # Run kernel 3 times with same input
+            for i in range(3):
+                y = run_kernel(x)
+                y.sum()
+            torch.cuda.synchronize()
+
+            # Count blob files
+            # Note: The system may save both input and output tensors.
+            # - Input tensor x: reused 3 times → should deduplicate to 1 blob
+            # - Output tensors y: 3 separate allocations → may be 3 blobs (if different) or 1 blob (if identical)
+            # Expected: fewer blobs than total tensor references due to deduplication
+            blob_count = count_all_blobs(manager.dir_path)
+            # With deduplication, we should have significantly fewer blobs than 6 (3 inputs + 3 outputs)
+            assert (
+                blob_count < 6
+            ), f"Deduplication should reduce blob count, got {blob_count} for 3 launches"
+            # We expect at least 1 blob (the deduplicated input)
+            assert blob_count >= 1, f"Should have at least 1 blob, got {blob_count}"
+            print(
+                f"✓ Deduplication working: {blob_count} unique blob(s) for 3 launches (< 6 without dedup)"
+            )
+
+        # === Test 3: Quota limit ===
+        print("\n=== Test 3: Quota Limit ===")
+        temp_output_dir_3 = tempfile.mkdtemp()
+
+        # Calculate quota to allow exactly one tensor to be saved
+        # A 10000 element float32 tensor = 10000 * 4 bytes = 40KB
+        # After torch.save serialization, it will be larger (includes metadata)
+        # Compressed size will be smaller for random data (but still substantial)
+        # Set quota to ~60KB to allow first tensor but not second
+        # Note: Random data doesn't compress as well as zeros
+        quota_for_one_tensor = 60 * 1024  # 60KB should fit one serialized tensor
+
+        with tritonparse.context_manager.TritonParseManager(
+            enable_trace_launch=True,
+            enable_tensor_blob_storage=True,
+            tensor_storage_quota=quota_for_one_tensor,
+            out=temp_output_dir_3,
+        ) as manager:
+            # Create first tensor - should be saved successfully
+            large_x1 = torch.randn(
+                (10000,), device=self.cuda_device, dtype=torch.float32
+            )
+            y1 = run_kernel(large_x1)
+            y1.sum()
+            torch.cuda.synchronize()
+
+            # Check that first tensor was saved
+            blob_count_after_first = count_all_blobs(manager.dir_path)
+            print(f"  Blobs after first kernel launch: {blob_count_after_first}")
+
+            # Create second tensor - should exceed quota and trigger storage disable
+            large_x2 = torch.randn(
+                (10000,), device=self.cuda_device, dtype=torch.float32
+            )
+            y2 = run_kernel(large_x2)
+            y2.sum()
+            torch.cuda.synchronize()
+
+            # Verify quota enforcement
+            blob_count_final = count_all_blobs(manager.dir_path)
+            print(f"  Blobs after second kernel launch: {blob_count_final}")
+
+            # We expect at least 1 blob was saved (from first launch)
+            assert (
+                blob_count_after_first >= 1
+            ), f"First tensor should be saved, got {blob_count_after_first} blobs"
+
+            # After quota exceeded, no more blobs should be added
+            # (blob_count_final should equal blob_count_after_first or be slightly higher
+            # if some outputs were saved before quota was hit)
+            assert (
+                blob_count_final <= blob_count_after_first + 1
+            ), f"Quota should prevent saving many more blobs: first={blob_count_after_first}, final={blob_count_final}"
+
+            print(
+                f"✓ Quota enforced: {blob_count_after_first} blob(s) saved before quota limit"
+            )
+
+        # The test passes if it doesn't crash - storage should be disabled after quota exceeded
+        print("✓ Quota limit test passed (storage disabled when quota exceeded)")
+
+        # Reset global variables to default after Test 3 to avoid polluting Test 4
+        tritonparse.structured_logging.TRITONPARSE_TENSOR_STORAGE_QUOTA = (
+            100 * 1024 * 1024 * 1024
+        )  # 100GB default
+        tritonparse.structured_logging.TRITONPARSE_SAVE_TENSOR_BLOBS = (
+            False  # Reset to default (disabled)
+        )
+
+        # === Test 4: Disabled storage ===
+        print("\n=== Test 4: Disabled Storage ===")
+        temp_output_dir_4 = tempfile.mkdtemp()
+
+        # When storage is explicitly disabled, don't set quota to avoid confusion
+        with tritonparse.context_manager.TritonParseManager(
+            enable_trace_launch=True,
+            enable_tensor_blob_storage=False,  # Explicitly disabled
+            out=temp_output_dir_4,
+        ) as manager:
+            x = torch.randn((512,), device=self.cuda_device, dtype=torch.float32)
+            y = run_kernel(x)
+            y.sum()
+            torch.cuda.synchronize()
+
+            # Verify no saved_tensors directory or it's empty
+            total_blobs = count_all_blobs(manager.dir_path)
+            assert (
+                total_blobs == 0
+            ), f"Expected no blobs when storage disabled, found {total_blobs}"
+            print("✓ Storage correctly disabled when enable_tensor_blob_storage=False")
+
+        # Clean up all test outputs
+        try:
+            if TEST_KEEP_OUTPUT:
+                print(
+                    f"\n✓ Preserving output directories (TEST_KEEP_OUTPUT=1):\n"
+                    f"  Test 1: {temp_output_dir_1}\n"
+                    f"  Test 2: {temp_output_dir_2}\n"
+                    f"  Test 3: {temp_output_dir_3}\n"
+                    f"  Test 4: {temp_output_dir_4}"
+                )
+            else:
+                for temp_dir in [
+                    temp_output_dir_1,
+                    temp_output_dir_2,
+                    temp_output_dir_3,
+                    temp_output_dir_4,
+                ]:
+                    if os.path.exists(temp_dir):
+                        shutil.rmtree(temp_dir)
+                print("✓ Cleaned up all test output directories")
+        except Exception as e:
+            print(f"Warning: Failed to clean up output directories: {e}")
+
+        finally:
+            # Cleanup test-specific cache
+            self.cleanup_test_cache(test_cache_dir, prev_cache_dir)
+
 
 if __name__ == "__main__":
     unittest.main()

{tritonparse-0.2.4.dev20251008071501 → tritonparse-0.2.4.dev20251010071509}/tritonparse.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: tritonparse
-Version: 0.2.4.dev20251008071501
+Version: 0.2.4.dev20251010071509
 Summary: TritonParse: A Compiler Tracer, Visualizer, and mini-Reproducer Generator for Triton Kernels
 Author-email: Yueming Hao <yhao@meta.com>
 License-Expression: BSD-3-Clause
@@ -34,13 +34,18 @@ Dynamic: license-file
 - **📝 Multi-format IR Support** - View TTGIR, TTIR, LLIR, PTX, and AMDGCN
 - **🎯 Interactive Code Views** - Click-to-highlight corresponding lines across IR stages
 
+### 🔧 Reproducer & Debugging Tools
+- **🔄 Standalone Script Generation** - Extract any kernel into a self-contained Python script
+- **💾 Tensor Data Reconstruction** - Preserve actual tensor data or use statistical approximation
+- **🎯 Custom Templates** - Flexible reproducer templates for different workflows
+- **🐛 Bug Isolation** - Share reproducible test cases for debugging and collaboration
+
 ### 📊 Structured Logging & Analysis
 - **📝 Compilation & Launch Tracing** - Capture detailed events with source mapping
 - **🔍 Stack Trace Integration** - Full Python stack traces for debugging
 - **📈 Metadata Extraction** - Comprehensive kernel statistics
 
 ### 🛠️ Developer Tools
-- **🔧 Reproducer Generation** - Generate standalone Python scripts to reproduce kernels
 - **🌐 Browser-based Interface** - No installation required, works in your browser
 - **🔒 Privacy-first** - All processing happens locally, no data uploaded
 
@@ -87,6 +92,41 @@ tritonparse.utils.unified_parse("./logs/", out="./parsed_output")
 
 > **🔒 Privacy Note**: Your trace files are processed entirely in your browser - nothing is uploaded to any server!
 
+### 3. Generate Reproducers (Optional)
+
+Extract any kernel into a standalone, executable Python script for debugging or testing:
+
+```bash
+# Generate reproducer from first launch event
+tritonparse reproduce ./parsed_output/trace.ndjson.gz --line 2 --out-dir repro_output
+
+# Run the generated reproducer
+cd repro_output/<kernel_name>/
+python repro_*.py
+```
+
+**Python API:**
+```python
+from tritonparse.reproducer.orchestrator import reproduce
+
+result = reproduce(
+    input_path="./parsed_output/trace.ndjson.gz",
+    line_index=1,           # Which launch event (1-based)
+    out_dir="repro_output"
+)
+```
+
+<details>
+<summary>🎯 Common Reproducer Use Cases (click to expand)</summary>
+
+- **🐛 Bug Isolation**: Extract a failing kernel into a minimal standalone script
+- **⚡ Performance Testing**: Benchmark specific kernels without running the full application
+- **🤝 Team Collaboration**: Share reproducible test cases with colleagues or in bug reports
+- **📊 Regression Testing**: Compare kernel behavior and performance across different versions
+- **🔍 Deep Debugging**: Modify and experiment with kernel parameters in isolation
+
+</details>
+
 ## 🛠️ Installation
 
 **For basic usage (trace generation):**
@@ -117,7 +157,7 @@ pip install triton
 |----------|-------------|
 | **[🏠 Wiki Home](https://github.com/meta-pytorch/tritonparse/wiki)** | Complete documentation and quick navigation |
 | **[📦 Installation](https://github.com/meta-pytorch/tritonparse/wiki/01.-Installation)** | Setup guide for all scenarios |
-| **[📋 Usage Guide](https://github.com/meta-pytorch/tritonparse/wiki/02.-Usage-Guide)** | Complete workflow, examples, and reproducer |
+| **[📋 Usage Guide](https://github.com/meta-pytorch/tritonparse/wiki/02.-Usage-Guide)** | Complete workflow, reproducer generation, and examples |
 | **[🌐 Web Interface](https://github.com/meta-pytorch/tritonparse/wiki/03.-Web-Interface-Guide)** | Master the visualization interface |
 | **[🔧 Developer Guide](https://github.com/meta-pytorch/tritonparse/wiki/04.-Developer-Guide)** | Contributing and architecture overview |
 | **[📝 Code Formatting](https://github.com/meta-pytorch/tritonparse/wiki/05.-Code-Formatting)** | Formatting standards and tools |

{tritonparse-0.2.4.dev20251008071501 → tritonparse-0.2.4.dev20251010071509}/tritonparse.egg-info/SOURCES.txt

@@ -10,7 +10,6 @@ pyproject.toml
 run.py
 .ci/README.md
 .ci/install-project.sh
-.ci/install-triton-kernels.sh
 .ci/install-triton.sh
 .ci/run-tests.sh
 .ci/setup.sh
@@ -32,7 +31,6 @@ tests/example_output/parsed_output/f0_fc0_a0_cai-.ndjson.gz
 tests/example_output/parsed_output/log_file_list.json
 tests/example_output/parsed_output_complex/dedicated_log_triton_trace_findhao__mapped.ndjson.gz
 tests/example_output/parsed_output_complex/log_file_list.json
-tests/example_output/repro/repro_context_20250816192455.json
 tritonparse/__init__.py
 tritonparse/__main__.py
 tritonparse/cli.py