ROCmSoftwarePlatform
Exhaustive tune reduce operators
This will tune the block size and algorithm chosen. It also fixes the benchmarking.
Codecov Report
All modified and coverable lines are covered by tests :white_check_mark:
Additional details and impacted files
@@ Coverage Diff @@
## develop #3751 +/- ##
========================================
Coverage 92.11% 92.11%
========================================
Files 525 526 +1
Lines 24119 24119
========================================
Hits 22216 22216
Misses 1903 1903
| Files with missing lines | Coverage Δ | |
|---|---|---|
| src/include/migraphx/bit.hpp | 100.00% <100.00%> (ø) |
|
| src/include/migraphx/generic_float.hpp | 99.14% <ø> (-0.02%) |
:arrow_down: |
:rocket: New features to boost your workflow:
- :snowflake: Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
![codecov[bot] avatar](https://avatars.githubusercontent.com/in/254?v=4 "Published by a pub.dev verified publisher"){kind=small}
![azure-pipelines[bot] avatar](https://avatars.githubusercontent.com/in/9426?v=4 "Published by a pub.dev verified publisher"){kind=small}
| Test | Batch | Rate new 073c3e |
Rate old b4ba1c |
Diff | Compare |
|---|---|---|---|---|---|
| torchvision-resnet50 | 64 | 3,258.31 | 3,235.95 | 0.69% | :white_check_mark: |
| torchvision-resnet50_fp16 | 64 | 6,934.90 | 6,908.41 | 0.38% | :white_check_mark: |
| torchvision-densenet121 | 32 | 2,457.17 | 2,445.21 | 0.49% | :white_check_mark: |
| torchvision-densenet121_fp16 | 32 | 4,220.96 | 4,219.44 | 0.04% | :white_check_mark: |
| torchvision-inceptionv3 | 32 | 1,627.71 | 1,618.24 | 0.59% | :white_check_mark: |
| torchvision-inceptionv3_fp16 | 32 | 2,721.76 | 2,714.76 | 0.26% | :white_check_mark: |
| cadene-inceptionv4 | 16 | 761.13 | 756.47 | 0.62% | :white_check_mark: |
| cadene-resnext64x4 | 16 | 818.72 | 814.62 | 0.50% | :white_check_mark: |
| slim-mobilenet | 64 | 7,475.60 | 7,433.86 | 0.56% | :white_check_mark: |
| slim-nasnetalarge | 64 | 217.81 | 216.86 | 0.44% | :white_check_mark: |
| slim-resnet50v2 | 64 | 3,460.26 | 3,440.36 | 0.58% | :white_check_mark: |
| bert-mrpc-onnx | 8 | 1,150.66 | 1,138.32 | 1.08% | :white_check_mark: |
| bert-mrpc-tf | 1 | 457.66 | 454.52 | 0.69% | :white_check_mark: |
| pytorch-examples-wlang-gru | 1 | 485.95 | 475.40 | 2.22% | :white_check_mark: |
| pytorch-examples-wlang-lstm | 1 | 448.87 | 443.05 | 1.31% | :white_check_mark: |
| torchvision-resnet50_1 | 1 | 810.86 | 818.79 | -0.97% | :white_check_mark: |
| cadene-dpn92_1 | 1 | 431.87 | 425.66 | 1.46% | :white_check_mark: |
| cadene-resnext101_1 | 1 | 393.54 | 391.85 | 0.43% | :white_check_mark: |
| onnx-taau-downsample | 1 | 396.86 | 394.70 | 0.55% | :white_check_mark: |
| dlrm-criteoterabyte | 1 | 32.34 | 32.18 | 0.49% | :white_check_mark: |
| dlrm-criteoterabyte_fp16 | 1 | 51.21 | 51.12 | 0.17% | :white_check_mark: |
| agentmodel | 1 | 10,462.72 | 9,553.93 | 9.51% | :high_brightness: |
| unet_fp16 | 2 | 58.61 | 58.49 | 0.21% | :white_check_mark: |
| resnet50v1_fp16 | 1 | 1,080.35 | 1,088.59 | -0.76% | :white_check_mark: |
| resnet50v1_int8 | 1 | 1,049.93 | 1,077.39 | -2.55% | :white_check_mark: |
| bert_base_cased_fp16 | 64 | 1,170.94 | 1,163.40 | 0.65% | :white_check_mark: |
| bert_large_uncased_fp16 | 32 | 356.30 | 354.51 | 0.50% | :white_check_mark: |
| bert_large_fp16 | 1 | 196.31 | 193.67 | 1.36% | :white_check_mark: |
| distilgpt2_fp16 | 16 | 2,232.85 | 2,217.48 | 0.69% | :white_check_mark: |
| yolov5s | 1 | 540.21 | 547.68 | -1.36% | :white_check_mark: |
| tinyllama | 1 | 43.89 | 43.63 | 0.61% | :white_check_mark: |
| vicuna-fastchat | 1 | 45.03 | 43.89 | 2.60% | :white_check_mark: |
| whisper-tiny-encoder | 1 | 421.89 | 420.34 | 0.37% | :white_check_mark: |
| whisper-tiny-decoder | 1 | 413.78 | 412.38 | 0.34% | :white_check_mark: |
| llama2_7b | 1 | nan | nan | nan% | :x: |
| qwen1.5-7b | 1 | 23.54 | 23.38 | 0.65% | :white_check_mark: |
| phi3-3.8b | 1 | nan | nan | nan% | :x: |
| mask-rcnn | 1 | 18.72 | 18.31 | 2.26% | :white_check_mark: |
| llama3-8b | 1 | 21.73 | 21.64 | 0.42% | :white_check_mark: |
| whisper-large-encoder | 1 | 10.21 | 10.17 | 0.43% | :white_check_mark: |
| whisper-large-decoder | 1 | 99.51 | 97.86 | 1.69% | :white_check_mark: |
| mistral-7b | 1 | 23.78 | 23.65 | 0.52% | :white_check_mark: |
| FLUX.1-schnell | 1 | 879.22 | 895.71 | -1.84% | :white_check_mark: |
| nan | nan | nan | nan | nan% | :x: |
This build is not recommended to merge :red_circle:
:x:bert-mrpc-tf: ERROR - check error output
2025-04-18 21:54:01.854436: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: SSE3 SSE4.1 SSE4.2 AVX AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
I0000 00:00:1745031247.299536 163665 gpu_device.cc:2022] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 62973 MB memory: -> device: 0, name: AMD Instinct MI250X/MI250, pci bus id: 0000:b3:00.0
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
I0000 00:00:1745031248.165054 163665 mlir_graph_optimization_pass.cc:401] MLIR V1 optimization pass is not enabled
2025-04-18 21:54:17.873914: E external/local_xla/xla/service/gpu/llvm_gpu_backend/gpu_backend_lib.cc:250] bitcode module is required by this HLO module but was not found at ./opencl.bc
2025-04-18 21:54:17.874193: E external/local_xla/xla/service/gpu/llvm_gpu_backend/gpu_backend_lib.cc:250] bitcode module is required by this HLO module but was not found at ./opencl.bc
2025-04-18 21:54:17.874239: E external/local_xla/xla/service/gpu/llvm_gpu_backend/gpu_backend_lib.cc:250] bitcode module is required by this HLO module but was not found at ./opencl.bc
2025-04-18 21:54:17.874287: E external/local_xla/xla/service/gpu/llvm_gpu_backend/gpu_backend_lib.cc:250] bitcode module is required by this HLO module but was not found at ./opencl.bc
2025-04-18 21:54:17.874315: E external/local_xla/xla/service/gpu/llvm_gpu_backend/gpu_backend_lib.cc:250] bitcode module is required by this HLO module but was not found at ./opencl.bc
2025-04-18 21:54:17.874362: E external/local_xla/xla/service/gpu/llvm_gpu_backend/gpu_backend_lib.cc:250] bitcode module is required by this HLO module but was not found at ./opencl.bc
2025-04-18 21:54:17.874405: E external/local_xla/xla/service/gpu/llvm_gpu_backend/gpu_backend_lib.cc:250] bitcode module is required by this HLO module but was not found at ./opencl.bc
2025-04-18 21:54:17.874454: E external/local_xla/xla/service/gpu/llvm_gpu_backend/gpu_backend_lib.cc:250] bitcode module is required by this HLO module but was not found at ./opencl.bc
error: Failure when generating HSACO
error: Failure when generating HSACO
error: Failure when generating HSACO
error: Failure when generating HSACO
error: Failure when generating HSACO
error: Failure when generating HSACO
error: Failure when generating HSACO
error: Failure when generating HSACO
2025-04-18 21:54:17.875536: E tensorflow/compiler/mlir/tools/kernel_gen/tf_framework_c_interface.cc:228] INTERNAL: Generating device code failed.
2025-04-18 21:54:17.876711: W tensorflow/core/framework/op_kernel.cc:1829] UNKNOWN: JIT compilation failed.
2025-04-18 21:54:17.876733: I tensorflow/core/framework/local_rendezvous.cc:405] Local rendezvous is aborting with status: UNKNOWN: JIT compilation failed.
[[{{node import/bert/embeddings/LayerNorm/moments/SquaredDifference}}]]
2025-04-18 21:54:17.876745: I tensorflow/core/framework/local_rendezvous.cc:405] Local rendezvous is aborting with status: UNKNOWN: JIT compilation failed.
[[{{node import/bert/embeddings/LayerNorm/moments/SquaredDifference}}]]
[[import/loss/output/_21]]
2025-04-18 21:54:17.876761: I tensorflow/core/framework/local_rendezvous.cc:424] Local rendezvous recv item cancelled. Key hash: 11217777527359497193
Traceback (most recent call last):
File "/usr/local/lib/python3.10/dist-packages/tensorflow/python/client/session.py", line 1407, in _do_call
return fn(*args)
File "/usr/local/lib/python3.10/dist-packages/tensorflow/python/client/session.py", line 1390, in _run_fn
return self._call_tf_sessionrun(options, feed_dict, fetch_list,
File "/usr/local/lib/python3.10/dist-packages/tensorflow/python/client/session.py", line 1483, in _call_tf_sessionrun
return tf_session.TF_SessionRun_wrapper(self._session, options, feed_dict,
tensorflow.python.framework.errors_impl.UnknownError: 2 root error(s) found.
(0) UNKNOWN: JIT compilation failed.
[[{{node import/bert/embeddings/LayerNorm/moments/SquaredDifference}}]]
[[import/loss/output/_21]]
(1) UNKNOWN: JIT compilation failed.
[[{{node import/bert/embeddings/LayerNorm/moments/SquaredDifference}}]]
0 successful operations.
0 derived errors ignored.
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/src/AMDMIGraphX/tools/accuracy/accuracy_checker.py", line 340, in
main()
File "/src/AMDMIGraphX/tools/accuracy/accuracy_checker.py", line 324, in main
y_out = sess.run(y, feed_dict=tf_dict)
File "/usr/local/lib/python3.10/dist-packages/tensorflow/python/client/session.py", line 977, in run
result = self._run(None, fetches, feed_dict, options_ptr,
File "/usr/local/lib/python3.10/dist-packages/tensorflow/python/client/session.py", line 1220, in _run
results = self._do_run(handle, final_targets, final_fetches,
File "/usr/local/lib/python3.10/dist-packages/tensorflow/python/client/session.py", line 1400, in _do_run
return self._do_call(_run_fn, feeds, fetches, targets, options,
File "/usr/local/lib/python3.10/dist-packages/tensorflow/python/client/session.py", line 1426, in _do_call
raise type(e)(node_def, op, message) # pylint: disable=no-value-for-parameter
tensorflow.python.framework.errors_impl.UnknownError: Graph execution error:
Detected at node 'import/bert/embeddings/LayerNorm/moments/SquaredDifference' defined at (most recent call last):
Node: 'import/bert/embeddings/LayerNorm/moments/SquaredDifference'
Detected at node 'import/bert/embeddings/LayerNorm/moments/SquaredDifference' defined at (most recent call last):
Node: 'import/bert/embeddings/LayerNorm/moments/SquaredDifference'
2 root error(s) found.
(0) UNKNOWN: JIT compilation failed.
[[{{node import/bert/embeddings/LayerNorm/moments/SquaredDifference}}]]
[[import/loss/output/_21]]
(1) UNKNOWN: JIT compilation failed.
[[{{node import/bert/embeddings/LayerNorm/moments/SquaredDifference}}]]
0 successful operations.
0 derived errors ignored.
Original stack trace for 'import/bert/embeddings/LayerNorm/moments/SquaredDifference'::red_circle:bert_large_uncased_fp16: FAILED: MIGraphX is not within tolerance - check verbose output
:x:llama2_7b: ERROR - check error output
Traceback (most recent call last):
File "/src/AMDMIGraphX/tools/accuracy/accuracy_checker.py", line 340, in
main()
File "/src/AMDMIGraphX/tools/accuracy/accuracy_checker.py", line 205, in main
model = migraphx.parse_onnx(model_name, default_dim_value=batch)
RuntimeError: /src/AMDMIGraphX/src/onnx/onnx_parser.cpp:265: parse_from: PARSE_FROM: Failed reading onnx file: /new-saved-models/llama2_7b/decoder_model.onnx:x:phi3-3.8b: ERROR - check error output
Traceback (most recent call last):
File "/src/AMDMIGraphX/tools/accuracy/accuracy_checker.py", line 340, in
main()
File "/src/AMDMIGraphX/tools/accuracy/accuracy_checker.py", line 205, in main
model = migraphx.parse_onnx(model_name, default_dim_value=batch)
RuntimeError: /src/AMDMIGraphX/src/onnx/onnx_parser.cpp:265: parse_from: PARSE_FROM: Failed reading onnx file: /new-saved-models/phi3-3.8b/model.onnx:red_circle:mask-rcnn: FAILED: MIGraphX is not within tolerance - check verbose output