llama.cpp
ggml : address integer overflows in binary ops CUDA implementation
#24706

Merged

Commits

ggml : address integer overflows in binary ops CUDA implementation

sszymczy committed 60 days ago
ggml : add size_t casts to avoid integer overflows

sszymczy committed 59 days ago
ggml : add more asserts checking integer overflows in binary ops CUDA implementation

sszymczy committed 52 days ago

Loading