[Performance] Avoid cuda sync in postprocess of LLM decoding#9011

Open

lszxb wants to merge 3 commits intoPaddlePaddle:developfrom lszxb:fix_avoid_cuda_sync

+16-8

Commits on Aug 26, 2024

avoid cuda sync in postprocess of LLM decoding
lszxb
committed

Commits on Sep 10, 2024

Revert workarounds about full_with_tensor as it will fix by paddle framework
lszxb
committed

Commits on Oct 28, 2024

Merge branch 'develop' into fix_avoid_cuda_sync
lszxb
committed