whisper.cpp/ggml
Jeff Bolz eb982dd786
vulkan: support arbitrary KV dimension in flash attention (llama/16160)
The "Clamp" spec constant is already based on whether KV is a multiple of Bc,
so use that to control whether bounds checking is performed. Add bounds checking
to the scalar and coopmat1 paths. Coopmat2 didn't need any changes (the K/V
tensors are already optionally clamped, nothing else needed to be changed).
2025-09-29 15:18:12 +03:00
..
cmake ggml: Skip backend library linking code when GGML_BACKEND_DL=ON (llama/15094) 2025-08-18 20:30:45 +03:00
include llama: print memory breakdown on exit (llama/15860) 2025-09-29 15:18:10 +03:00
src vulkan: support arbitrary KV dimension in flash attention (llama/16160) 2025-09-29 15:18:12 +03:00
.gitignore whisper : reorganize source code + improve CMake (#2256) 2024-06-26 19:34:09 +03:00
CMakeLists.txt common : use cpp-httplib as a cURL alternative for downloads (llama/16185) 2025-09-29 15:18:11 +03:00