]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
cuda : fix copy of large tensors (ggml_nbytes <= INT_MAX assertion) (#18433)
authorMeeMin <redacted>
Thu, 1 Jan 2026 23:24:20 +0000 (04:54 +0530)
committerGitHub <redacted>
Thu, 1 Jan 2026 23:24:20 +0000 (00:24 +0100)
commite86f3c22211d9b5c3842e2961a022aac9cdbacad
tree94f696fba4a8561570047876cf13a7e991c1689a
parent169ee68ffb7281e7af632419513f55b6f14a8892
cuda : fix copy of large tensors (ggml_nbytes <= INT_MAX assertion) (#18433)

* ggml-cuda: fixed assertion in ggml_cuda_cpy (#18140)

* ggml-cuda: changes in data types to int64_t

* ggml-cuda: added asserts for CUDA block numbers

* ggml-cuda: changed the condition for y and z dimension
ggml/src/ggml-cuda/cpy.cu