NVIDIA/Fuser

Add more features to the pre-segmentation pass to move take_along_axis ahead of certain unary ops

Open

#4491 aperta il 21 mag 2025

Vedi su GitHub
 (0 commenti) (0 reazioni) (1 assegnatario)C++ (80 fork)github user discovery
New Opsgood first issue

Metriche repository

Star
 (392 star)
Metriche merge PR
 (Metriche PR in attesa)

Descrizione

The PR add a pre-segmentation pass that moved a take_along_axis operator ahead of certain unary pointwise operators such as castOp and the squeeze op.

The implementation had certain limitations. We should extend the current implementation to remove these.

  • We only consider fusions with a single take_along_axis op. We should extend our implementation to support the case where there are multiple take_along_axis ops.
  • We move a take_along_axis ahead of a squeeze op. Currently we do so only if a single dimension is being squeezed. We should remove this limitation. This will mean that we have to add multiple unsqueeze ops to the indexTv of the take_along_axis.

Guida contributor