Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

TTNN asserts with 1-d arrays #788

Open
uazizTT opened this issue Sep 20, 2024 · 0 comments
Open

TTNN asserts with 1-d arrays #788

uazizTT opened this issue Sep 20, 2024 · 0 comments

Comments

@uazizTT
Copy link
Contributor

uazizTT commented Sep 20, 2024

I had a simple test to add two 1-d arrays:

import pytest
import jax
import jax.numpy as jnp
from infrastructure import verify_module

def test_1d_array_add():
  def module_add(a, b):
    return a + b

  verify_module(module_add, [(1,), (1,)])

It has started failing with this error:

Always | FATAL    | Can only unpad to row major tensor of even width
terminate called after throwing an instance of 'std::runtime_error'
  what():  TT_FATAL @ /home/uaziz/pjrt/third_party/tt-mlir/src/tt-mlir/third_party/tt-metal/src/tt-metal/ttnn/cpp/ttnn/operations/data_movement/untilize_with_unpadding/device/untilize_with_unpadding_op.cpp:25: ((this->output_tensor_end[-1] + 1) % 2 == 0)
info:
Can only unpad to row major tensor of even width
nsmithtt added a commit that referenced this issue Sep 20, 2024
The to_layout/from_device was causing the following fallout:
- #779
- #788
- All zeros output, separately addressed #776

For now just reverting back to to_memory_config.
nsmithtt added a commit that referenced this issue Sep 20, 2024
The to_layout/from_device was causing the following fallout:
- #779
- #788
- All zeros output, separately addressed #776

For now just reverting back to to_memory_config.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant