viable/strict/1780997283: Add oneDNN LSTM primitive support for XPU inference (#185531)
🔒
https://github.com
«Summary
Use dnnl::lstm_forward primitive on XPU for LSTM inference, replacing the per-timestep fused cell approach. The oneDNN primitive processes the entire sequence internally, eliminating O(T) kernel launches.
Changes...»
Automatische Weiterleitung...
1.5s