Mirrors/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2025-01-11 21:10:24 +01:00
Code Issues Packages Projects Releases Wiki Activity

cuda : build only necessary templates

Browse Source
This commit is contained in:
Georgi Gerganov 2024-08-14 10:29:23 +03:00
parent ae41fd2e65
commit 503983a69a
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
102 changed files with 60 additions and 567 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
ggml/src/ggml-cuda/fattn.cu
View File

@ -208,13 +208,7 @@ static void ggml_cuda_flash_attn_ext_vec_f16(ggml_backend_cuda_context & ctx, gg
FATTN_VEC_F16_CASE(256, GGML_TYPE_F16, GGML_TYPE_F16)
#else
FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0)
FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0)
FATTN_VEC_F16_CASE( 64, GGML_TYPE_F16, GGML_TYPE_F16)
FATTN_VEC_F16_CASE(128, GGML_TYPE_F16, GGML_TYPE_F16)
FATTN_VEC_F16_CASE(256, GGML_TYPE_F16, GGML_TYPE_F16)
#endif // GGML_CUDA_FA_ALL_QUANTS
on_no_fattn_vec_case(Q->ne[0]);
@ -283,13 +277,7 @@ static void ggml_cuda_flash_attn_ext_vec_f32(ggml_backend_cuda_context & ctx, gg
FATTN_VEC_F32_CASE(256, GGML_TYPE_F16, GGML_TYPE_F16)
#else
FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0)
FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0)
FATTN_VEC_F32_CASE( 64, GGML_TYPE_F16, GGML_TYPE_F16)
FATTN_VEC_F32_CASE(128, GGML_TYPE_F16, GGML_TYPE_F16)
FATTN_VEC_F32_CASE(256, GGML_TYPE_F16, GGML_TYPE_F16)
#endif // GGML_CUDA_FA_ALL_QUANTS
on_no_fattn_vec_case(Q->ne[0]);

90
ggml/src/ggml-cuda/mmq.cu
View File

@ -29,60 +29,60 @@ void ggml_cuda_op_mul_mat_q(
const mmq_args args = {src0_dd_i, src1_ddq_i, dst_dd_i, ne00, row_diff, stride00, src1_padded_row_size, src1_ncols, ne11, nrows_dst};
switch (src0->type) {
case GGML_TYPE_Q4_0:
mul_mat_q_case<GGML_TYPE_Q4_0>(ctx, args, stream);
break;
case GGML_TYPE_Q4_1:
mul_mat_q_case<GGML_TYPE_Q4_1>(ctx, args, stream);
break;
case GGML_TYPE_Q5_0:
mul_mat_q_case<GGML_TYPE_Q5_0>(ctx, args, stream);
break;
case GGML_TYPE_Q5_1:
mul_mat_q_case<GGML_TYPE_Q5_1>(ctx, args, stream);
break;
// case GGML_TYPE_Q4_0:
// mul_mat_q_case<GGML_TYPE_Q4_0>(ctx, args, stream);
// break;
// case GGML_TYPE_Q4_1:
// mul_mat_q_case<GGML_TYPE_Q4_1>(ctx, args, stream);
// break;
// case GGML_TYPE_Q5_0:
// mul_mat_q_case<GGML_TYPE_Q5_0>(ctx, args, stream);
// break;
// case GGML_TYPE_Q5_1:
// mul_mat_q_case<GGML_TYPE_Q5_1>(ctx, args, stream);
// break;
case GGML_TYPE_Q8_0:
mul_mat_q_case<GGML_TYPE_Q8_0>(ctx, args, stream);
break;
case GGML_TYPE_Q2_K:
mul_mat_q_case<GGML_TYPE_Q2_K>(ctx, args, stream);
break;
case GGML_TYPE_Q3_K:
mul_mat_q_case<GGML_TYPE_Q3_K>(ctx, args, stream);
break;
// case GGML_TYPE_Q2_K:
// mul_mat_q_case<GGML_TYPE_Q2_K>(ctx, args, stream);
// break;
// case GGML_TYPE_Q3_K:
// mul_mat_q_case<GGML_TYPE_Q3_K>(ctx, args, stream);
// break;
case GGML_TYPE_Q4_K:
mul_mat_q_case<GGML_TYPE_Q4_K>(ctx, args, stream);
break;
case GGML_TYPE_Q5_K:
mul_mat_q_case<GGML_TYPE_Q5_K>(ctx, args, stream);
break;
// case GGML_TYPE_Q5_K:
// mul_mat_q_case<GGML_TYPE_Q5_K>(ctx, args, stream);
// break;
case GGML_TYPE_Q6_K:
mul_mat_q_case<GGML_TYPE_Q6_K>(ctx, args, stream);
break;
case GGML_TYPE_IQ2_XXS:
mul_mat_q_case<GGML_TYPE_IQ2_XXS>(ctx, args, stream);
break;
case GGML_TYPE_IQ2_XS:
mul_mat_q_case<GGML_TYPE_IQ2_XS>(ctx, args, stream);
break;
case GGML_TYPE_IQ2_S:
mul_mat_q_case<GGML_TYPE_IQ2_S>(ctx, args, stream);
break;
case GGML_TYPE_IQ3_XXS:
mul_mat_q_case<GGML_TYPE_IQ3_XXS>(ctx, args, stream);
break;
case GGML_TYPE_IQ3_S:
mul_mat_q_case<GGML_TYPE_IQ3_S>(ctx, args, stream);
break;
case GGML_TYPE_IQ1_S:
mul_mat_q_case<GGML_TYPE_IQ1_S>(ctx, args, stream);
break;
case GGML_TYPE_IQ4_XS:
mul_mat_q_case<GGML_TYPE_IQ4_XS>(ctx, args, stream);
break;
case GGML_TYPE_IQ4_NL:
mul_mat_q_case<GGML_TYPE_IQ4_NL>(ctx, args, stream);
break;
// case GGML_TYPE_IQ2_XXS:
// mul_mat_q_case<GGML_TYPE_IQ2_XXS>(ctx, args, stream);
// break;
// case GGML_TYPE_IQ2_XS:
// mul_mat_q_case<GGML_TYPE_IQ2_XS>(ctx, args, stream);
// break;
// case GGML_TYPE_IQ2_S:
// mul_mat_q_case<GGML_TYPE_IQ2_S>(ctx, args, stream);
// break;
// case GGML_TYPE_IQ3_XXS:
// mul_mat_q_case<GGML_TYPE_IQ3_XXS>(ctx, args, stream);
// break;
// case GGML_TYPE_IQ3_S:
// mul_mat_q_case<GGML_TYPE_IQ3_S>(ctx, args, stream);
// break;
// case GGML_TYPE_IQ1_S:
// mul_mat_q_case<GGML_TYPE_IQ1_S>(ctx, args, stream);
// break;
// case GGML_TYPE_IQ4_XS:
// mul_mat_q_case<GGML_TYPE_IQ4_XS>(ctx, args, stream);
// break;
// case GGML_TYPE_IQ4_NL:
// mul_mat_q_case<GGML_TYPE_IQ4_NL>(ctx, args, stream);
// break;
default:
GGML_ABORT("fatal error");
break;

30
ggml/src/ggml-cuda/mmq.cuh
View File

@ -2906,24 +2906,24 @@ void mul_mat_q_case(ggml_backend_cuda_context & ctx, const mmq_args & args, cuda
#define DECL_MMQ_CASE(type) \
template void mul_mat_q_case<type>(ggml_backend_cuda_context & ctx, const mmq_args & args, cudaStream_t stream) \
extern DECL_MMQ_CASE(GGML_TYPE_Q4_0);
extern DECL_MMQ_CASE(GGML_TYPE_Q4_1);
extern DECL_MMQ_CASE(GGML_TYPE_Q5_0);
extern DECL_MMQ_CASE(GGML_TYPE_Q5_1);
//extern DECL_MMQ_CASE(GGML_TYPE_Q4_0);
//extern DECL_MMQ_CASE(GGML_TYPE_Q4_1);
//extern DECL_MMQ_CASE(GGML_TYPE_Q5_0);
//extern DECL_MMQ_CASE(GGML_TYPE_Q5_1);
extern DECL_MMQ_CASE(GGML_TYPE_Q8_0);
extern DECL_MMQ_CASE(GGML_TYPE_Q2_K);
extern DECL_MMQ_CASE(GGML_TYPE_Q3_K);
//extern DECL_MMQ_CASE(GGML_TYPE_Q2_K);
//extern DECL_MMQ_CASE(GGML_TYPE_Q3_K);
extern DECL_MMQ_CASE(GGML_TYPE_Q4_K);
extern DECL_MMQ_CASE(GGML_TYPE_Q5_K);
//extern DECL_MMQ_CASE(GGML_TYPE_Q5_K);
extern DECL_MMQ_CASE(GGML_TYPE_Q6_K);
extern DECL_MMQ_CASE(GGML_TYPE_IQ2_XXS);
extern DECL_MMQ_CASE(GGML_TYPE_IQ2_XS);
extern DECL_MMQ_CASE(GGML_TYPE_IQ2_S);
extern DECL_MMQ_CASE(GGML_TYPE_IQ3_XXS);
extern DECL_MMQ_CASE(GGML_TYPE_IQ3_S);
extern DECL_MMQ_CASE(GGML_TYPE_IQ1_S);
extern DECL_MMQ_CASE(GGML_TYPE_IQ4_NL);
extern DECL_MMQ_CASE(GGML_TYPE_IQ4_XS);
//extern DECL_MMQ_CASE(GGML_TYPE_IQ2_XXS);
//extern DECL_MMQ_CASE(GGML_TYPE_IQ2_XS);
//extern DECL_MMQ_CASE(GGML_TYPE_IQ2_S);
//extern DECL_MMQ_CASE(GGML_TYPE_IQ3_XXS);
//extern DECL_MMQ_CASE(GGML_TYPE_IQ3_S);
//extern DECL_MMQ_CASE(GGML_TYPE_IQ1_S);
//extern DECL_MMQ_CASE(GGML_TYPE_IQ4_NL);
//extern DECL_MMQ_CASE(GGML_TYPE_IQ4_XS);
// -------------------------------------------------------------------------------------------------------------------------

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q4_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q4_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q4_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q4_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q5_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q5_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q5_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q5_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q8_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q8_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-f16.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_F16);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q5_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q5_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q8_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q8_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-f16.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_F16);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q4_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q4_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q4_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q4_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q5_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q5_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q5_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q5_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q8_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q8_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-f16.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_F16);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q4_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q4_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q4_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q4_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q5_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q5_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q5_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q5_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q8_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q8_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-f16.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_F16);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q4_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q4_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q4_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q4_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q5_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q5_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q5_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q5_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q8_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q8_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-f16.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_F16);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q4_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q4_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q4_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q4_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q5_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q5_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q5_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q5_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(256, GGML_TYPE_F16, GGML_TYPE_F16);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(64, GGML_TYPE_F16, GGML_TYPE_F16);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q4_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(64, GGML_TYPE_F16, GGML_TYPE_Q4_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q4_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(64, GGML_TYPE_F16, GGML_TYPE_Q4_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q5_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(64, GGML_TYPE_F16, GGML_TYPE_Q5_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q5_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(64, GGML_TYPE_F16, GGML_TYPE_Q5_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q8_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(64, GGML_TYPE_F16, GGML_TYPE_Q8_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q4_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q4_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q4_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q4_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q5_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q5_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q5_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q5_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q8_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q8_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-f16.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_F16);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q5_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q5_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q8_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q8_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-f16.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_F16);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q4_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q4_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q4_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q4_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q5_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q5_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q5_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q5_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q8_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q8_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-f16.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_F16);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q4_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q4_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q4_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q4_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q5_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q5_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q5_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q5_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q8_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q8_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-f16.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_F16);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q4_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q4_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q4_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q4_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q5_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q5_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q5_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q5_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q8_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q8_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-f16.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_F16);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q4_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q4_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q4_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q4_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q5_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q5_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q5_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q5_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(256, GGML_TYPE_F16, GGML_TYPE_F16);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(64, GGML_TYPE_F16, GGML_TYPE_F16);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q4_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(64, GGML_TYPE_F16, GGML_TYPE_Q4_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q4_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(64, GGML_TYPE_F16, GGML_TYPE_Q4_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q5_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(64, GGML_TYPE_F16, GGML_TYPE_Q5_0);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q5_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(64, GGML_TYPE_F16, GGML_TYPE_Q5_1);

5
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q8_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f32.cuh"
DECL_FATTN_VEC_F32_CASE(64, GGML_TYPE_F16, GGML_TYPE_Q8_0);

5
ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../mmq.cuh"
DECL_MMQ_CASE(GGML_TYPE_IQ1_S);

5
ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../mmq.cuh"
DECL_MMQ_CASE(GGML_TYPE_IQ2_S);

5
ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../mmq.cuh"
DECL_MMQ_CASE(GGML_TYPE_IQ2_XS);

5
ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../mmq.cuh"
DECL_MMQ_CASE(GGML_TYPE_IQ2_XXS);

5
ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../mmq.cuh"
DECL_MMQ_CASE(GGML_TYPE_IQ3_S);

5
ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../mmq.cuh"
DECL_MMQ_CASE(GGML_TYPE_IQ3_XXS);

5
ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../mmq.cuh"
DECL_MMQ_CASE(GGML_TYPE_IQ4_NL);

5
ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../mmq.cuh"
DECL_MMQ_CASE(GGML_TYPE_IQ4_XS);

5
ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../mmq.cuh"
DECL_MMQ_CASE(GGML_TYPE_Q2_K);

5
ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../mmq.cuh"
DECL_MMQ_CASE(GGML_TYPE_Q3_K);

5
ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../mmq.cuh"
DECL_MMQ_CASE(GGML_TYPE_Q4_0);

5
ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../mmq.cuh"
DECL_MMQ_CASE(GGML_TYPE_Q4_1);

5
ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu
View File

@ -1,5 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../mmq.cuh"
DECL_MMQ_CASE(GGML_TYPE_Q5_0);

Some files were not shown because too many files have changed in this diff Show More