tenstorrent / tt-metal

:metal: TT-NN operator library, and TT-Metalium low level kernel programming model.
Apache License 2.0
471 stars 74 forks source link

Failing Silu in Yolov7 model #12786

Open HariniMohan0102 opened 1 month ago

HariniMohan0102 commented 1 month ago

Describe the bug On unit testing the SiLU ops of Yolov7 model, few input configurations failed with the error:

E       RuntimeError: TT_THROW @ ../tt_metal/impl/allocator/allocator.cpp:143: tt::exception
E       info:
E       Out of Memory: Not enough space to allocate 13107200 B L1 buffer across 20 banks, where each bank needs to store 655360 B

To Reproduce Steps to reproduce the behavior:

  1. Checkout to the branch harini/yolov7_failing_silu
  2. Run the command: pytest tests/ttnn/unit_tests/operations/test_silu_row_major.py::test_failing_silu_yolov7

Expected behavior Expected to pass the failing configurations.

Please complete the following environment information:

dvartaniansTT commented 1 month ago

@HariniMohan0102 is this being impletented data parallel or using a single chip on N300?

HariniMohan0102 commented 1 month ago

@dvartaniansTT Unit test is done on n300 in single chip.