diff --git a/llama/README.md b/llama/README.md index 31c6087d..05c95c90 100644 --- a/llama/README.md +++ b/llama/README.md @@ -93,8 +93,8 @@ make -j ## Syncing with llama.cpp -To update this package to the latest llama.cpp code, use the `sync_llama.sh` script: +To update this package to the latest llama.cpp code, use the `sync.sh` script: ``` -./sync_llama.sh ../../llama.cpp +./sync.sh ../../llama.cpp ``` diff --git a/llama/build-info.cpp b/llama/build-info.cpp index f890ee9c..e6e66949 100644 --- a/llama/build-info.cpp +++ b/llama/build-info.cpp @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/clip.cpp b/llama/clip.cpp index 36d1b3b3..5cbb4532 100644 --- a/llama/clip.cpp +++ b/llama/clip.cpp @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/clip.h b/llama/clip.h index ae2ab6ba..9d95e0b2 100644 --- a/llama/clip.h +++ b/llama/clip.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/common.cpp b/llama/common.cpp index 6596c10d..d87a1dea 100644 --- a/llama/common.cpp +++ b/llama/common.cpp @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/common.h b/llama/common.h index 583451f2..3aba4bfa 100644 --- a/llama/common.h +++ b/llama/common.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/ggml-alloc.c b/llama/ggml-alloc.c index a422c052..268f2234 100644 --- a/llama/ggml-alloc.c +++ b/llama/ggml-alloc.c @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/ggml-alloc.h b/llama/ggml-alloc.h index cc0a1a1c..5311cc17 100644 --- a/llama/ggml-alloc.h +++ b/llama/ggml-alloc.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/ggml-backend-impl.h b/llama/ggml-backend-impl.h index 14babea1..c9f4b7a6 100644 --- a/llama/ggml-backend-impl.h +++ b/llama/ggml-backend-impl.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/ggml-backend.c b/llama/ggml-backend.c index cf698a96..0275b136 100644 --- a/llama/ggml-backend.c +++ b/llama/ggml-backend.c @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/ggml-backend.h b/llama/ggml-backend.h index 85069045..602cd030 100644 --- a/llama/ggml-backend.h +++ b/llama/ggml-backend.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/ggml-common.h b/llama/ggml-common.h index a8520fd2..068516f7 100644 --- a/llama/ggml-common.h +++ b/llama/ggml-common.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/ggml-cuda.cu b/llama/ggml-cuda.cu index 0c602c31..29f6c756 100644 --- a/llama/ggml-cuda.cu +++ b/llama/ggml-cuda.cu @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/ggml-cuda.h b/llama/ggml-cuda.h index f71a9933..9f2b1c88 100644 --- a/llama/ggml-cuda.h +++ b/llama/ggml-cuda.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/ggml-cuda/acc.cu b/llama/ggml-cuda/acc.cu index 96bfe1c9..e6e49958 100644 --- a/llama/ggml-cuda/acc.cu +++ b/llama/ggml-cuda/acc.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "acc.cuh" static __global__ void acc_f32(const float * x, const float * y, float * dst, const int ne, diff --git a/llama/ggml-cuda/acc.cuh b/llama/ggml-cuda/acc.cuh index 1168ea1b..acddd4b8 100644 --- a/llama/ggml-cuda/acc.cuh +++ b/llama/ggml-cuda/acc.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_ACC_BLOCK_SIZE 256 diff --git a/llama/ggml-cuda/alibi.cu b/llama/ggml-cuda/alibi.cu index 6c7f1fd9..0ff2f9f3 100644 --- a/llama/ggml-cuda/alibi.cu +++ b/llama/ggml-cuda/alibi.cu @@ -1,3 +1,107 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "alibi.cuh" static __global__ void alibi_f32(const float * x, float * dst, const int ncols, const int k_rows, diff --git a/llama/ggml-cuda/alibi.cuh b/llama/ggml-cuda/alibi.cuh index 630adfc7..6087d0b0 100644 --- a/llama/ggml-cuda/alibi.cuh +++ b/llama/ggml-cuda/alibi.cuh @@ -1,3 +1,81 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_ALIBI_BLOCK_SIZE 32 diff --git a/llama/ggml-cuda/arange.cu b/llama/ggml-cuda/arange.cu index b5e495a2..598620d0 100644 --- a/llama/ggml-cuda/arange.cu +++ b/llama/ggml-cuda/arange.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "arange.cuh" static __global__ void arange_f32(float * dst, const int ne0, const float start, const float step) { diff --git a/llama/ggml-cuda/arange.cuh b/llama/ggml-cuda/arange.cuh index 41e74fdf..db1650db 100644 --- a/llama/ggml-cuda/arange.cuh +++ b/llama/ggml-cuda/arange.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_ARANGE_BLOCK_SIZE 256 diff --git a/llama/ggml-cuda/argsort.cu b/llama/ggml-cuda/argsort.cu index 16414406..d7972d58 100644 --- a/llama/ggml-cuda/argsort.cu +++ b/llama/ggml-cuda/argsort.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "argsort.cuh" template diff --git a/llama/ggml-cuda/argsort.cuh b/llama/ggml-cuda/argsort.cuh index 68a00154..7ae91c17 100644 --- a/llama/ggml-cuda/argsort.cuh +++ b/llama/ggml-cuda/argsort.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" void ggml_cuda_op_argsort(ggml_backend_cuda_context & ctx, ggml_tensor * dst); diff --git a/llama/ggml-cuda/binbcast.cu b/llama/ggml-cuda/binbcast.cu index 19b08b74..3e1c6585 100644 --- a/llama/ggml-cuda/binbcast.cu +++ b/llama/ggml-cuda/binbcast.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "binbcast.cuh" static __device__ __forceinline__ float op_repeat(const float a, const float b) { diff --git a/llama/ggml-cuda/binbcast.cuh b/llama/ggml-cuda/binbcast.cuh index 4f63d637..b5cbd086 100644 --- a/llama/ggml-cuda/binbcast.cuh +++ b/llama/ggml-cuda/binbcast.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" void ggml_cuda_op_repeat(ggml_backend_cuda_context & ctx, ggml_tensor * dst); diff --git a/llama/ggml-cuda/clamp.cu b/llama/ggml-cuda/clamp.cu index 8009a3e3..97b498be 100644 --- a/llama/ggml-cuda/clamp.cu +++ b/llama/ggml-cuda/clamp.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "clamp.cuh" static __global__ void clamp_f32(const float * x, float * dst, const float min, const float max, const int k) { diff --git a/llama/ggml-cuda/clamp.cuh b/llama/ggml-cuda/clamp.cuh index 7f9559dd..9d94460c 100644 --- a/llama/ggml-cuda/clamp.cuh +++ b/llama/ggml-cuda/clamp.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_CLAMP_BLOCK_SIZE 256 diff --git a/llama/ggml-cuda/common.cuh b/llama/ggml-cuda/common.cuh index 90a0a81e..e465649f 100644 --- a/llama/ggml-cuda/common.cuh +++ b/llama/ggml-cuda/common.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #pragma once #include "ggml.h" diff --git a/llama/ggml-cuda/concat.cu b/llama/ggml-cuda/concat.cu index dac10ec3..351c379e 100644 --- a/llama/ggml-cuda/concat.cu +++ b/llama/ggml-cuda/concat.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "concat.cuh" // contiguous kernels diff --git a/llama/ggml-cuda/concat.cuh b/llama/ggml-cuda/concat.cuh index aa506a05..b70e5669 100644 --- a/llama/ggml-cuda/concat.cuh +++ b/llama/ggml-cuda/concat.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_CONCAT_BLOCK_SIZE 256 diff --git a/llama/ggml-cuda/convert.cu b/llama/ggml-cuda/convert.cu index c0a44470..b5e56bf5 100644 --- a/llama/ggml-cuda/convert.cu +++ b/llama/ggml-cuda/convert.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "convert.cuh" #include "dequantize.cuh" diff --git a/llama/ggml-cuda/convert.cuh b/llama/ggml-cuda/convert.cuh index 5394be9f..3abba0da 100644 --- a/llama/ggml-cuda/convert.cuh +++ b/llama/ggml-cuda/convert.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_DEQUANTIZE_BLOCK_SIZE 256 diff --git a/llama/ggml-cuda/cpy.cu b/llama/ggml-cuda/cpy.cu index 12d741f0..408ad9c5 100644 --- a/llama/ggml-cuda/cpy.cu +++ b/llama/ggml-cuda/cpy.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "cpy.cuh" typedef void (*cpy_kernel_t)(const char * cx, char * cdst); diff --git a/llama/ggml-cuda/cpy.cuh b/llama/ggml-cuda/cpy.cuh index 79616742..af93b77e 100644 --- a/llama/ggml-cuda/cpy.cuh +++ b/llama/ggml-cuda/cpy.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_CPY_BLOCK_SIZE 32 diff --git a/llama/ggml-cuda/dequantize.cuh b/llama/ggml-cuda/dequantize.cuh index bd3c2d9d..b7e55460 100644 --- a/llama/ggml-cuda/dequantize.cuh +++ b/llama/ggml-cuda/dequantize.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" static __device__ __forceinline__ void dequantize_q4_0(const void * vx, const int64_t ib, const int iqs, dfloat2 & v){ diff --git a/llama/ggml-cuda/diagmask.cu b/llama/ggml-cuda/diagmask.cu index 4b713ba2..16dcfcd5 100644 --- a/llama/ggml-cuda/diagmask.cu +++ b/llama/ggml-cuda/diagmask.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "diagmask.cuh" static __global__ void diag_mask_inf_f32(const float * x, float * dst, const int ncols, const int rows_per_channel, const int n_past) { diff --git a/llama/ggml-cuda/diagmask.cuh b/llama/ggml-cuda/diagmask.cuh index 6cdbef17..62338819 100644 --- a/llama/ggml-cuda/diagmask.cuh +++ b/llama/ggml-cuda/diagmask.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_DIAG_MASK_INF_BLOCK_SIZE 32 diff --git a/llama/ggml-cuda/dmmv.cu b/llama/ggml-cuda/dmmv.cu index 174489e0..88779422 100644 --- a/llama/ggml-cuda/dmmv.cu +++ b/llama/ggml-cuda/dmmv.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "dmmv.cuh" #include "dequantize.cuh" #include "convert.cuh" diff --git a/llama/ggml-cuda/dmmv.cuh b/llama/ggml-cuda/dmmv.cuh index 4c5ebd47..fbb28f4f 100644 --- a/llama/ggml-cuda/dmmv.cuh +++ b/llama/ggml-cuda/dmmv.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" // dmmv = dequantize_mul_mat_vec diff --git a/llama/ggml-cuda/fattn-common.cuh b/llama/ggml-cuda/fattn-common.cuh index c00f8606..bea6c4fd 100644 --- a/llama/ggml-cuda/fattn-common.cuh +++ b/llama/ggml-cuda/fattn-common.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #pragma once #include "common.cuh" diff --git a/llama/ggml-cuda/fattn-tile-f16.cu b/llama/ggml-cuda/fattn-tile-f16.cu index cb11d721..153e248d 100644 --- a/llama/ggml-cuda/fattn-tile-f16.cu +++ b/llama/ggml-cuda/fattn-tile-f16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #include "fattn-common.cuh" #include "fattn-tile-f16.cuh" diff --git a/llama/ggml-cuda/fattn-tile-f16.cuh b/llama/ggml-cuda/fattn-tile-f16.cuh index ffc58784..1ad39a15 100644 --- a/llama/ggml-cuda/fattn-tile-f16.cuh +++ b/llama/ggml-cuda/fattn-tile-f16.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" void ggml_cuda_flash_attn_ext_tile_f16(ggml_backend_cuda_context & ctx, ggml_tensor * dst); diff --git a/llama/ggml-cuda/fattn-tile-f32.cu b/llama/ggml-cuda/fattn-tile-f32.cu index 15e22f49..62a873ef 100644 --- a/llama/ggml-cuda/fattn-tile-f32.cu +++ b/llama/ggml-cuda/fattn-tile-f32.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #include "fattn-common.cuh" #include "fattn-tile-f32.cuh" diff --git a/llama/ggml-cuda/fattn-tile-f32.cuh b/llama/ggml-cuda/fattn-tile-f32.cuh index b1c546c8..74bf3dd9 100644 --- a/llama/ggml-cuda/fattn-tile-f32.cuh +++ b/llama/ggml-cuda/fattn-tile-f32.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" void ggml_cuda_flash_attn_ext_tile_f32(ggml_backend_cuda_context & ctx, ggml_tensor * dst); diff --git a/llama/ggml-cuda/fattn-vec-f16.cuh b/llama/ggml-cuda/fattn-vec-f16.cuh index 9e1aa2c6..a702edd2 100644 --- a/llama/ggml-cuda/fattn-vec-f16.cuh +++ b/llama/ggml-cuda/fattn-vec-f16.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #include "fattn-common.cuh" diff --git a/llama/ggml-cuda/fattn-vec-f32.cuh b/llama/ggml-cuda/fattn-vec-f32.cuh index ddf0c837..d2bd51f3 100644 --- a/llama/ggml-cuda/fattn-vec-f32.cuh +++ b/llama/ggml-cuda/fattn-vec-f32.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #include "fattn-common.cuh" diff --git a/llama/ggml-cuda/fattn-wmma-f16.cuh b/llama/ggml-cuda/fattn-wmma-f16.cuh index 59cd30d7..ada2a966 100644 --- a/llama/ggml-cuda/fattn-wmma-f16.cuh +++ b/llama/ggml-cuda/fattn-wmma-f16.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #include "fattn-common.cuh" diff --git a/llama/ggml-cuda/fattn.cu b/llama/ggml-cuda/fattn.cu index 38d30b21..9dfb824d 100644 --- a/llama/ggml-cuda/fattn.cu +++ b/llama/ggml-cuda/fattn.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #include "fattn-common.cuh" #include "fattn-tile-f16.cuh" diff --git a/llama/ggml-cuda/fattn.cuh b/llama/ggml-cuda/fattn.cuh index ad3ca7a8..abb65afc 100644 --- a/llama/ggml-cuda/fattn.cuh +++ b/llama/ggml-cuda/fattn.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" void ggml_cuda_flash_attn_ext(ggml_backend_cuda_context & ctx, ggml_tensor * dst); diff --git a/llama/ggml-cuda/getrows.cu b/llama/ggml-cuda/getrows.cu index 55af195f..d5276dc6 100644 --- a/llama/ggml-cuda/getrows.cu +++ b/llama/ggml-cuda/getrows.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "getrows.cuh" #include "dequantize.cuh" diff --git a/llama/ggml-cuda/getrows.cuh b/llama/ggml-cuda/getrows.cuh index bbf13023..07e92cc9 100644 --- a/llama/ggml-cuda/getrows.cuh +++ b/llama/ggml-cuda/getrows.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_GET_ROWS_BLOCK_SIZE 256 diff --git a/llama/ggml-cuda/im2col.cu b/llama/ggml-cuda/im2col.cu index 3d0d8d4e..81dd9733 100644 --- a/llama/ggml-cuda/im2col.cu +++ b/llama/ggml-cuda/im2col.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "im2col.cuh" template diff --git a/llama/ggml-cuda/im2col.cuh b/llama/ggml-cuda/im2col.cuh index 1ce8fae4..1e9aa07f 100644 --- a/llama/ggml-cuda/im2col.cuh +++ b/llama/ggml-cuda/im2col.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_IM2COL_BLOCK_SIZE 256 diff --git a/llama/ggml-cuda/mmq.cu b/llama/ggml-cuda/mmq.cu index 58799e4c..346a4c72 100644 --- a/llama/ggml-cuda/mmq.cu +++ b/llama/ggml-cuda/mmq.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "mmq.cuh" void ggml_cuda_op_mul_mat_q( diff --git a/llama/ggml-cuda/mmq.cuh b/llama/ggml-cuda/mmq.cuh index 6744cce6..4a04f4a2 100644 --- a/llama/ggml-cuda/mmq.cuh +++ b/llama/ggml-cuda/mmq.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #include "vecdotq.cuh" diff --git a/llama/ggml-cuda/mmvq.cu b/llama/ggml-cuda/mmvq.cu index 5f056e91..ca0c824e 100644 --- a/llama/ggml-cuda/mmvq.cu +++ b/llama/ggml-cuda/mmvq.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "mmvq.cuh" #include "vecdotq.cuh" diff --git a/llama/ggml-cuda/mmvq.cuh b/llama/ggml-cuda/mmvq.cuh index 88c42c4b..e2e138bd 100644 --- a/llama/ggml-cuda/mmvq.cuh +++ b/llama/ggml-cuda/mmvq.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" void ggml_cuda_op_mul_mat_vec_q( diff --git a/llama/ggml-cuda/norm.cu b/llama/ggml-cuda/norm.cu index 30866d51..1770b32d 100644 --- a/llama/ggml-cuda/norm.cu +++ b/llama/ggml-cuda/norm.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "norm.cuh" template diff --git a/llama/ggml-cuda/norm.cuh b/llama/ggml-cuda/norm.cuh index 431a8f74..cbb30be9 100644 --- a/llama/ggml-cuda/norm.cuh +++ b/llama/ggml-cuda/norm.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" void ggml_cuda_op_norm(ggml_backend_cuda_context & ctx, ggml_tensor * dst); diff --git a/llama/ggml-cuda/pad.cu b/llama/ggml-cuda/pad.cu index aba539e8..7bbb79a2 100644 --- a/llama/ggml-cuda/pad.cu +++ b/llama/ggml-cuda/pad.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "pad.cuh" static __global__ void pad_f32(const float * x, float * dst, const int ne0, const int ne00, const int ne01, const int ne02, const int ne03) { diff --git a/llama/ggml-cuda/pad.cuh b/llama/ggml-cuda/pad.cuh index 8fd386b0..b0ab5d3f 100644 --- a/llama/ggml-cuda/pad.cuh +++ b/llama/ggml-cuda/pad.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_PAD_BLOCK_SIZE 256 diff --git a/llama/ggml-cuda/pool2d.cu b/llama/ggml-cuda/pool2d.cu index c6d51e4d..acb837ca 100644 --- a/llama/ggml-cuda/pool2d.cu +++ b/llama/ggml-cuda/pool2d.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "pool2d.cuh" template diff --git a/llama/ggml-cuda/pool2d.cuh b/llama/ggml-cuda/pool2d.cuh index 7841292b..2773cad9 100644 --- a/llama/ggml-cuda/pool2d.cuh +++ b/llama/ggml-cuda/pool2d.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_POOL2D_BLOCK_SIZE 256 diff --git a/llama/ggml-cuda/quantize.cu b/llama/ggml-cuda/quantize.cu index 7578c4b6..ad7e2ef3 100644 --- a/llama/ggml-cuda/quantize.cu +++ b/llama/ggml-cuda/quantize.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "quantize.cuh" static __global__ void quantize_q8_1(const float * __restrict__ x, void * __restrict__ vy, const int64_t kx, const int64_t kx_padded) { diff --git a/llama/ggml-cuda/quantize.cuh b/llama/ggml-cuda/quantize.cuh index b37a4752..82656f24 100644 --- a/llama/ggml-cuda/quantize.cuh +++ b/llama/ggml-cuda/quantize.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_QUANTIZE_BLOCK_SIZE 256 diff --git a/llama/ggml-cuda/rope.cu b/llama/ggml-cuda/rope.cu index 596fb7c1..6e7327a3 100644 --- a/llama/ggml-cuda/rope.cu +++ b/llama/ggml-cuda/rope.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "rope.cuh" struct rope_corr_dims { diff --git a/llama/ggml-cuda/rope.cuh b/llama/ggml-cuda/rope.cuh index 0f787a0b..d06466b5 100644 --- a/llama/ggml-cuda/rope.cuh +++ b/llama/ggml-cuda/rope.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_ROPE_BLOCK_SIZE 256 diff --git a/llama/ggml-cuda/scale.cu b/llama/ggml-cuda/scale.cu index 1405e066..84eb76fe 100644 --- a/llama/ggml-cuda/scale.cu +++ b/llama/ggml-cuda/scale.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "scale.cuh" static __global__ void scale_f32(const float * x, float * dst, const float scale, const int k) { diff --git a/llama/ggml-cuda/scale.cuh b/llama/ggml-cuda/scale.cuh index 8ff75c82..69e29111 100644 --- a/llama/ggml-cuda/scale.cuh +++ b/llama/ggml-cuda/scale.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_SCALE_BLOCK_SIZE 256 diff --git a/llama/ggml-cuda/softmax.cu b/llama/ggml-cuda/softmax.cu index ce64f2f2..d3fc1044 100644 --- a/llama/ggml-cuda/softmax.cu +++ b/llama/ggml-cuda/softmax.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #include "softmax.cuh" diff --git a/llama/ggml-cuda/softmax.cuh b/llama/ggml-cuda/softmax.cuh index 4ef4ff86..60df821a 100644 --- a/llama/ggml-cuda/softmax.cuh +++ b/llama/ggml-cuda/softmax.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_SOFT_MAX_BLOCK_SIZE 1024 diff --git a/llama/ggml-cuda/sumrows.cu b/llama/ggml-cuda/sumrows.cu index 82e8e875..241bfb71 100644 --- a/llama/ggml-cuda/sumrows.cu +++ b/llama/ggml-cuda/sumrows.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "sumrows.cuh" static __global__ void k_sum_rows_f32(const float * x, float * dst, const int ncols) { diff --git a/llama/ggml-cuda/sumrows.cuh b/llama/ggml-cuda/sumrows.cuh index e7545f83..69eb3994 100644 --- a/llama/ggml-cuda/sumrows.cuh +++ b/llama/ggml-cuda/sumrows.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" void ggml_cuda_op_sum_rows(ggml_backend_cuda_context & ctx, ggml_tensor * dst); diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu index 6696a238..cfd7e8a7 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q4_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q4_0.cu index dd070db2..817efe2d 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q4_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q4_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q4_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q4_1.cu index 54dcde6f..277e8fc2 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q4_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q4_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q5_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q5_0.cu index 4ec22f79..47722adf 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q5_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q5_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q5_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q5_1.cu index 3c15bf7f..af548e34 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q5_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q5_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q8_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q8_0.cu index 7e61b5fd..01a268ae 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q8_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q8_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-f16.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-f16.cu index fdb15b58..cbde6106 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-f16.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-f16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu index 0f7c417d..7d8734c5 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_1.cu index 851f33c4..e7cc120f 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q5_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q5_0.cu index 763809cb..a6b29bea 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q5_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q5_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q5_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q5_1.cu index f2a276e5..115a8f69 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q5_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q5_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q8_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q8_0.cu index cb227f6f..b6e9b64e 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q8_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q8_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-f16.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-f16.cu index 97ac0520..01ae3deb 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-f16.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-f16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q4_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q4_0.cu index c772b426..bb761ecf 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q4_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q4_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q4_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q4_1.cu index 5cb74308..bd4f2035 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q4_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q4_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q5_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q5_0.cu index 98a709d1..8b797364 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q5_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q5_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q5_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q5_1.cu index 4f2f947a..1324b33a 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q5_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q5_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q8_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q8_0.cu index 11f96b6f..432ce759 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q8_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q8_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-f16.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-f16.cu index b39bdc06..45a99d8d 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-f16.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-f16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q4_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q4_0.cu index bbd6a2c7..a6549a73 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q4_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q4_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q4_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q4_1.cu index 9d84ff2b..8abaa17a 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q4_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q4_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q5_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q5_0.cu index bc8a5bff..7ae05c52 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q5_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q5_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q5_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q5_1.cu index a679100c..cd6c7980 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q5_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q5_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q8_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q8_0.cu index 8f21bccf..362b035e 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q8_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q8_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-f16.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-f16.cu index 858b00fd..fffe9344 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-f16.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-f16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q4_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q4_0.cu index 0fc8011f..0ad3077b 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q4_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q4_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q4_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q4_1.cu index 261fdf62..03a1a375 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q4_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q4_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q5_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q5_0.cu index 0fb82473..cd27a825 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q5_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q5_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q5_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q5_1.cu index a9d9d089..2adc1a0b 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q5_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q5_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q8_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q8_0.cu index 7d7b2792..118bfa56 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q8_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q8_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-f16.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-f16.cu index a092ee2d..7da915b9 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-f16.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-f16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q4_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q4_0.cu index db55927a..66d49c32 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q4_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q4_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q4_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q4_1.cu index c3c21cef..a17736df 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q4_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q4_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q5_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q5_0.cu index 35dd9f52..41d8c6d0 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q5_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q5_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q5_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q5_1.cu index 050c22ac..4800bf50 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q5_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q5_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu index de4866c5..63dc37ea 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu index 57a10bc4..de960397 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu index e0f08b46..c5333b6c 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q4_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q4_0.cu index 1c8e8a46..74d561bc 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q4_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q4_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q4_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q4_1.cu index cefed83f..d1435f15 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q4_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q4_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q5_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q5_0.cu index aede6e35..34ed5b18 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q5_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q5_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q5_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q5_1.cu index 1a1a92c7..cffcaad2 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q5_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q5_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q8_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q8_0.cu index ad667473..319529d9 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q8_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q8_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu index c499f455..02f05ac4 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q4_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q4_0.cu index 8286ebf3..f449f7d1 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q4_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q4_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q4_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q4_1.cu index 45878688..c77e21e8 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q4_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q4_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q5_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q5_0.cu index d89103ce..fff43be6 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q5_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q5_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q5_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q5_1.cu index bb75fd42..c0fff458 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q5_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q5_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q8_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q8_0.cu index b1629817..ec4b3fb7 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q8_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q8_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-f16.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-f16.cu index d8657604..9117b831 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-f16.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-f16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu index 2e5bd2f1..ce9a8515 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_1.cu index be5f302d..ceceb334 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q5_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q5_0.cu index 8dd91cd7..6ea92d98 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q5_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q5_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q5_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q5_1.cu index 4cb79150..6342bdd9 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q5_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q5_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q8_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q8_0.cu index 09dea426..5a621d3a 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q8_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q8_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-f16.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-f16.cu index 0fbb6076..5637c3ac 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-f16.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-f16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q4_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q4_0.cu index 2aeab83b..337ebc56 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q4_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q4_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q4_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q4_1.cu index 599415b4..d2316647 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q4_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q4_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q5_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q5_0.cu index e4f8e308..c57f31bb 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q5_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q5_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q5_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q5_1.cu index 34d16652..5693bbd3 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q5_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q5_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q8_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q8_0.cu index 4bebef45..7278d4b7 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q8_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q8_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-f16.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-f16.cu index 326468da..ff0ea22e 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-f16.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-f16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q4_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q4_0.cu index 511b58f4..f5d9254c 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q4_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q4_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q4_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q4_1.cu index d9906d14..3501943d 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q4_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q4_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q5_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q5_0.cu index f61c183a..cf0b7d78 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q5_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q5_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q5_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q5_1.cu index c10450fd..157abfd3 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q5_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q5_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q8_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q8_0.cu index 2d5cb195..96dbf49c 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q8_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q8_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-f16.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-f16.cu index b384f34d..c9a7d79d 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-f16.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-f16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q4_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q4_0.cu index 446e293b..948392db 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q4_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q4_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q4_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q4_1.cu index 6f430298..28db7404 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q4_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q4_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q5_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q5_0.cu index 1cd8ba88..12b2795a 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q5_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q5_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q5_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q5_1.cu index 1ee2eab6..5f428d13 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q5_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q5_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q8_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q8_0.cu index 2bc77816..d4774fc5 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q8_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q8_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-f16.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-f16.cu index d55ced08..39180299 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-f16.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-f16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q4_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q4_0.cu index 8361e99c..be1d5bb5 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q4_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q4_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q4_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q4_1.cu index 7507a67c..63c564bd 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q4_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q4_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q5_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q5_0.cu index 61f050b2..244ff284 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q5_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q5_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q5_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q5_1.cu index d4a49d9c..7d829023 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q5_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q5_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu index d1462789..f8e90769 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu index e73f917a..7d935c2c 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu index d40825df..de917434 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q4_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q4_0.cu index b5c6869f..1ec95ad5 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q4_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q4_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q4_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q4_1.cu index 4e21b0cc..6463247e 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q4_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q4_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q5_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q5_0.cu index 2eac321b..423a6432 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q5_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q5_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q5_1.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q5_1.cu index f7d2c3b4..9af97fe1 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q5_1.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q5_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q8_0.cu b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q8_0.cu index a013f400..064bb392 100644 --- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q8_0.cu +++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q8_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-vec-f32.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu b/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu index 2d94e65c..076d4233 100644 --- a/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu +++ b/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-wmma-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu b/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu index c3d9df3c..5029593a 100644 --- a/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu +++ b/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-wmma-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu b/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu index bb680e40..b5d008eb 100644 --- a/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu +++ b/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-wmma-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu b/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu index 073f71b1..fb148383 100644 --- a/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu +++ b/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-wmma-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu b/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu index d30710c5..ee8af8f0 100644 --- a/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu +++ b/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../fattn-wmma-f16.cuh" diff --git a/llama/ggml-cuda/template-instances/mmq-instance-q2_k.cu b/llama/ggml-cuda/template-instances/mmq-instance-q2_k.cu index 6415369d..ed3c331a 100644 --- a/llama/ggml-cuda/template-instances/mmq-instance-q2_k.cu +++ b/llama/ggml-cuda/template-instances/mmq-instance-q2_k.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../mmq.cuh" diff --git a/llama/ggml-cuda/template-instances/mmq-instance-q3_k.cu b/llama/ggml-cuda/template-instances/mmq-instance-q3_k.cu index ffb6213a..1dbfe57d 100644 --- a/llama/ggml-cuda/template-instances/mmq-instance-q3_k.cu +++ b/llama/ggml-cuda/template-instances/mmq-instance-q3_k.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../mmq.cuh" diff --git a/llama/ggml-cuda/template-instances/mmq-instance-q4_0.cu b/llama/ggml-cuda/template-instances/mmq-instance-q4_0.cu index 0c0b0c8a..221485a5 100644 --- a/llama/ggml-cuda/template-instances/mmq-instance-q4_0.cu +++ b/llama/ggml-cuda/template-instances/mmq-instance-q4_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../mmq.cuh" diff --git a/llama/ggml-cuda/template-instances/mmq-instance-q4_1.cu b/llama/ggml-cuda/template-instances/mmq-instance-q4_1.cu index ee67f694..be424bd2 100644 --- a/llama/ggml-cuda/template-instances/mmq-instance-q4_1.cu +++ b/llama/ggml-cuda/template-instances/mmq-instance-q4_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../mmq.cuh" diff --git a/llama/ggml-cuda/template-instances/mmq-instance-q4_k.cu b/llama/ggml-cuda/template-instances/mmq-instance-q4_k.cu index 9eeb3cd7..aeb1071e 100644 --- a/llama/ggml-cuda/template-instances/mmq-instance-q4_k.cu +++ b/llama/ggml-cuda/template-instances/mmq-instance-q4_k.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../mmq.cuh" diff --git a/llama/ggml-cuda/template-instances/mmq-instance-q5_0.cu b/llama/ggml-cuda/template-instances/mmq-instance-q5_0.cu index cc57fb97..278e3003 100644 --- a/llama/ggml-cuda/template-instances/mmq-instance-q5_0.cu +++ b/llama/ggml-cuda/template-instances/mmq-instance-q5_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../mmq.cuh" diff --git a/llama/ggml-cuda/template-instances/mmq-instance-q5_1.cu b/llama/ggml-cuda/template-instances/mmq-instance-q5_1.cu index 721ac790..e8f1d0e0 100644 --- a/llama/ggml-cuda/template-instances/mmq-instance-q5_1.cu +++ b/llama/ggml-cuda/template-instances/mmq-instance-q5_1.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../mmq.cuh" diff --git a/llama/ggml-cuda/template-instances/mmq-instance-q5_k.cu b/llama/ggml-cuda/template-instances/mmq-instance-q5_k.cu index a2e90ffd..ee9d03ff 100644 --- a/llama/ggml-cuda/template-instances/mmq-instance-q5_k.cu +++ b/llama/ggml-cuda/template-instances/mmq-instance-q5_k.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../mmq.cuh" diff --git a/llama/ggml-cuda/template-instances/mmq-instance-q6_k.cu b/llama/ggml-cuda/template-instances/mmq-instance-q6_k.cu index 470938fe..4df022f6 100644 --- a/llama/ggml-cuda/template-instances/mmq-instance-q6_k.cu +++ b/llama/ggml-cuda/template-instances/mmq-instance-q6_k.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../mmq.cuh" diff --git a/llama/ggml-cuda/template-instances/mmq-instance-q8_0.cu b/llama/ggml-cuda/template-instances/mmq-instance-q8_0.cu index 974477bb..7034a4bf 100644 --- a/llama/ggml-cuda/template-instances/mmq-instance-q8_0.cu +++ b/llama/ggml-cuda/template-instances/mmq-instance-q8_0.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + // This file has been autogenerated by generate_cu_files.py, do not edit manually. #include "../mmq.cuh" diff --git a/llama/ggml-cuda/tsembd.cu b/llama/ggml-cuda/tsembd.cu index 153ddbcd..452dfd63 100644 --- a/llama/ggml-cuda/tsembd.cu +++ b/llama/ggml-cuda/tsembd.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "tsembd.cuh" static __global__ void timestep_embedding_f32(const float * timesteps, float * dst, const int nb1, const int dim, const int max_period) { diff --git a/llama/ggml-cuda/tsembd.cuh b/llama/ggml-cuda/tsembd.cuh index 84340e3d..6d423c1f 100644 --- a/llama/ggml-cuda/tsembd.cuh +++ b/llama/ggml-cuda/tsembd.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_TIMESTEP_EMBEDDING_BLOCK_SIZE 256 diff --git a/llama/ggml-cuda/unary.cu b/llama/ggml-cuda/unary.cu index ac03d5c6..38e5f1b4 100644 --- a/llama/ggml-cuda/unary.cu +++ b/llama/ggml-cuda/unary.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "unary.cuh" static __global__ void gelu_f32(const float * x, float * dst, const int k) { diff --git a/llama/ggml-cuda/unary.cuh b/llama/ggml-cuda/unary.cuh index a1d07c04..75fd718b 100644 --- a/llama/ggml-cuda/unary.cuh +++ b/llama/ggml-cuda/unary.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_GELU_BLOCK_SIZE 256 diff --git a/llama/ggml-cuda/upscale.cu b/llama/ggml-cuda/upscale.cu index cf513c3a..e261bc17 100644 --- a/llama/ggml-cuda/upscale.cu +++ b/llama/ggml-cuda/upscale.cu @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "upscale.cuh" static __global__ void upscale_f32(const float * x, float * dst, diff --git a/llama/ggml-cuda/upscale.cuh b/llama/ggml-cuda/upscale.cuh index d4d76523..06d64e6f 100644 --- a/llama/ggml-cuda/upscale.cuh +++ b/llama/ggml-cuda/upscale.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" #define CUDA_UPSCALE_BLOCK_SIZE 256 diff --git a/llama/ggml-cuda/vecdotq.cuh b/llama/ggml-cuda/vecdotq.cuh index b9573a7c..bdac93c8 100644 --- a/llama/ggml-cuda/vecdotq.cuh +++ b/llama/ggml-cuda/vecdotq.cuh @@ -1,3 +1,29 @@ +/** + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * + * MIT License + * + * Copyright (c) 2023-2024 The ggml authors + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + #include "common.cuh" static __device__ __forceinline__ int get_int_from_int8(const int8_t * x8, const int & i32) { diff --git a/llama/ggml-impl.h b/llama/ggml-impl.h index 721c3c28..f16d6674 100644 --- a/llama/ggml-impl.h +++ b/llama/ggml-impl.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/ggml-metal-darwin_arm64.m b/llama/ggml-metal-darwin_arm64.m index 400d43f4..a5c7d37b 100644 --- a/llama/ggml-metal-darwin_arm64.m +++ b/llama/ggml-metal-darwin_arm64.m @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/ggml-metal.h b/llama/ggml-metal.h index b6c88a2a..a44dd32a 100644 --- a/llama/ggml-metal.h +++ b/llama/ggml-metal.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/ggml-metal.metal b/llama/ggml-metal.metal index 5ee578c4..e081496c 100644 --- a/llama/ggml-metal.metal +++ b/llama/ggml-metal.metal @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/ggml-quants.c b/llama/ggml-quants.c index 796b2a32..cc7d6f16 100644 --- a/llama/ggml-quants.c +++ b/llama/ggml-quants.c @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/ggml-quants.h b/llama/ggml-quants.h index 99b67dca..c9a9b732 100644 --- a/llama/ggml-quants.h +++ b/llama/ggml-quants.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/ggml.c b/llama/ggml.c index 7fadd6e2..e1d2f674 100644 --- a/llama/ggml.c +++ b/llama/ggml.c @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/ggml.h b/llama/ggml.h index b7a073fd..03ffc2cf 100644 --- a/llama/ggml.h +++ b/llama/ggml.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/grammar-parser.cpp b/llama/grammar-parser.cpp index 6ebc78e9..f5d4404e 100644 --- a/llama/grammar-parser.cpp +++ b/llama/grammar-parser.cpp @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/grammar-parser.h b/llama/grammar-parser.h index 5c16f71c..4c7f1e8a 100644 --- a/llama/grammar-parser.h +++ b/llama/grammar-parser.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/json-schema-to-grammar.cpp b/llama/json-schema-to-grammar.cpp index 45fcd14b..737f05b7 100644 --- a/llama/json-schema-to-grammar.cpp +++ b/llama/json-schema-to-grammar.cpp @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/json-schema-to-grammar.h b/llama/json-schema-to-grammar.h index 5544b894..053f509d 100644 --- a/llama/json-schema-to-grammar.h +++ b/llama/json-schema-to-grammar.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/llama.cpp b/llama/llama.cpp index fc77dd99..b995e497 100644 --- a/llama/llama.cpp +++ b/llama/llama.cpp @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/llama.h b/llama/llama.h index 908a57ab..3e0355b2 100644 --- a/llama/llama.h +++ b/llama/llama.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/llava.cpp b/llama/llava.cpp index 6fd9cba2..fc55207b 100644 --- a/llama/llava.cpp +++ b/llama/llava.cpp @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/llava.h b/llama/llava.h index 3214c0e0..0d120283 100644 --- a/llama/llava.h +++ b/llama/llava.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/log.h b/llama/log.h index 434519b7..03855df6 100644 --- a/llama/log.h +++ b/llama/log.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/sampling.cpp b/llama/sampling.cpp index 89fe9876..07f6219d 100644 --- a/llama/sampling.cpp +++ b/llama/sampling.cpp @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/sampling.h b/llama/sampling.h index 6fd4dc95..cc6524b3 100644 --- a/llama/sampling.h +++ b/llama/sampling.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/stb_image.h b/llama/stb_image.h index b007891f..a2f5386a 100644 --- a/llama/stb_image.h +++ b/llama/stb_image.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/sync_llama.sh b/llama/sync.sh similarity index 87% rename from llama/sync_llama.sh rename to llama/sync.sh index 5c11491c..f2103352 100755 --- a/llama/sync_llama.sh +++ b/llama/sync.sh @@ -90,17 +90,22 @@ trap cleanup 0 cat <$TEMP_LICENSE /** - * llama.cpp - git $sha1 - do not edit this file + * llama.cpp - commit $sha1 - do not edit this file * $(sed 's/^/ * /' <$src_dir/LICENSE) */ EOF -for IN in $dst_dir/*.{c,h,cpp,m,metal,cu}; do - if [[ "$IN" == *"sgemm.cpp" || "$IN" == *"sgemm.h" || "$IN" == *"sampling_ext.cpp" || "$IN" == *"sampling_ext.h" ]]; then - continue - fi +LICENSE_FILES=$(find $dst_dir -type f \( -name "*.c" -o -name "*.h" -o -name "*.cpp" -o -name "*.m" -o -name "*.metal" -o -name "*.cu" -o -name "*.cuh" \)) +EXCLUDED_FILES=("sgemm.cpp" "sgemm.h" "sampling_ext.cpp" "sampling_ext.h" "metal.c") + +for IN in $LICENSE_FILES; do + for EXCLUDED in "${EXCLUDED_FILES[@]}"; do + if [[ "$IN" == *"$EXCLUDED" ]]; then + continue 2 + fi + done TMP=$(mktemp) cat $TEMP_LICENSE $IN >$TMP mv $TMP $IN diff --git a/llama/unicode-data.cpp b/llama/unicode-data.cpp index 102c362d..33efff9f 100644 --- a/llama/unicode-data.cpp +++ b/llama/unicode-data.cpp @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/unicode-data.h b/llama/unicode-data.h index a123af3a..458e5bf1 100644 --- a/llama/unicode-data.h +++ b/llama/unicode-data.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/unicode.cpp b/llama/unicode.cpp index 72453d00..4e0ff2aa 100644 --- a/llama/unicode.cpp +++ b/llama/unicode.cpp @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License * diff --git a/llama/unicode.h b/llama/unicode.h index 8cb52e94..729b28e8 100644 --- a/llama/unicode.h +++ b/llama/unicode.h @@ -1,5 +1,5 @@ /** - * llama.cpp - git ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file + * llama.cpp - commit ee459f40f65810a810151b24eba5b8bd174ceffe - do not edit this file * * MIT License *