llama.cpp
cuda : replace remaining shfl_xor with calls to warp_reduce functions
#5744
Merged
Merged
cuda : replace remaining shfl_xor with calls to warp_reduce functions #5744
slaren
merged
1 commit into
ggml-org:master
from
Engininja2:call-warp-reduce-functions
cuda : replace remaining shfl_xor with calls to warp_reduce functions
b8ca8176
slaren
approved these changes
on 2024-02-27
JohannesGaessler
approved these changes
on 2024-02-27
Login to write a write a comment.
Login via GitHub