llama.cpp

31288c9c - hexagon: FLASH_ATTN_EXT improvements for longer context and SWA models (#30731)

hexagon: FLASH_ATTN_EXT improvements for longer context and SWA models (#30731) * hex-fa: support for masked block skipping to reduce DMA fetching * hex-fa: finetuning the hmx->hvx crossover thresholds (boosting gemma-4) * hex-fa: update hmx gate to account for mdev.count