mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2025-10-30 08:42:00 +00:00
Fix uploading tensor data to device, including 3D, 4D, and non-contiguous tensors. Use correct offsets into data that is already in VRAM. Correct handling of OpenCL events when multiple commands are queued.
70 KiB
70 KiB