mikhail/llama.cpp-mtp-turboquant
1
0
Fork 0
Code Issues Pull Requests Actions 33 Packages Projects Releases Wiki Activity

CUDA: refactor and deduplicate vector FA kernels (#16208)

Browse Source 
* CUDA: refactor and deduplicate vector FA kernels
This commit is contained in:
Johannes Gäßler
2025-09-27 18:45:07 +02:00
committed by GitHub
parent 0499b29c6f
commit 75a3a6c2cd
129 changed files with 1396 additions and 1989 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu
-5
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_F16);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q4_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q4_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q5_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q5_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q8_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-f16.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q5_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q5_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q8_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-f16.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q4_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q4_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q5_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q5_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q8_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-f16.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q4_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q4_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q5_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q5_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q8_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-f16.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q4_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q4_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q5_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q5_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q8_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-f16.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q4_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q4_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q5_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q5_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q4_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q4_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q5_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q5_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q8_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu
-5
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_F16);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q4_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q4_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q5_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q5_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q8_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-f16.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q5_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q5_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q8_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-f16.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q4_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q4_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q5_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q5_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q8_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-f16.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q4_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q4_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q5_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q5_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q8_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-f16.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q4_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q4_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q5_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q5_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_1-q8_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-f16.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q4_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q4_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q5_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q5_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q4_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q4_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q5_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q5_1.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q8_0.cu
-5
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);
ggml/src/ggml-cuda/template-instances/fattn-vec-instance-f16-f16.cu
+7
View File
@@ -0,0 +1,7 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec.cuh"
DECL_FATTN_VEC_CASE( 64, GGML_TYPE_F16, GGML_TYPE_F16);
DECL_FATTN_VEC_CASE(128, GGML_TYPE_F16, GGML_TYPE_F16);
DECL_FATTN_VEC_CASE(256, GGML_TYPE_F16, GGML_TYPE_F16);
ggml/src/ggml-cuda/template-instances/fattn-vec-instance-f16-q4_0.cu
+7
View File
@@ -0,0 +1,7 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec.cuh"
DECL_FATTN_VEC_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q4_0);
DECL_FATTN_VEC_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q4_0);
DECL_FATTN_VEC_CASE(256, GGML_TYPE_F16, GGML_TYPE_Q4_0);
ggml/src/ggml-cuda/template-instances/fattn-vec-instance-f16-q4_1.cu
+7
View File
@@ -0,0 +1,7 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec.cuh"
DECL_FATTN_VEC_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q4_1);
DECL_FATTN_VEC_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q4_1);
DECL_FATTN_VEC_CASE(256, GGML_TYPE_F16, GGML_TYPE_Q4_1);
ggml/src/ggml-cuda/template-instances/fattn-vec-instance-f16-q5_0.cu
+7
View File
@@ -0,0 +1,7 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec.cuh"
DECL_FATTN_VEC_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q5_0);
DECL_FATTN_VEC_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q5_0);
DECL_FATTN_VEC_CASE(256, GGML_TYPE_F16, GGML_TYPE_Q5_0);
ggml/src/ggml-cuda/template-instances/fattn-vec-instance-f16-q5_1.cu
+7
View File
@@ -0,0 +1,7 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec.cuh"
DECL_FATTN_VEC_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q5_1);
DECL_FATTN_VEC_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q5_1);
DECL_FATTN_VEC_CASE(256, GGML_TYPE_F16, GGML_TYPE_Q5_1);
ggml/src/ggml-cuda/template-instances/fattn-vec-instance-f16-q8_0.cu
+7
View File
@@ -0,0 +1,7 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec.cuh"
DECL_FATTN_VEC_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q8_0);
DECL_FATTN_VEC_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q8_0);
DECL_FATTN_VEC_CASE(256, GGML_TYPE_F16, GGML_TYPE_Q8_0);
ggml/src/ggml-cuda/template-instances/fattn-vec-instance-q4_0-f16.cu
+7
View File
@@ -0,0 +1,7 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec.cuh"
DECL_FATTN_VEC_CASE( 64, GGML_TYPE_Q4_0, GGML_TYPE_F16);
DECL_FATTN_VEC_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_F16);
DECL_FATTN_VEC_CASE(256, GGML_TYPE_Q4_0, GGML_TYPE_F16);
ggml/src/ggml-cuda/template-instances/fattn-vec-instance-q4_0-q4_0.cu
+7
View File
@@ -0,0 +1,7 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec.cuh"
DECL_FATTN_VEC_CASE( 64, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0);
DECL_FATTN_VEC_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0);
DECL_FATTN_VEC_CASE(256, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0);
ggml/src/ggml-cuda/template-instances/fattn-vec-instance-q4_0-q4_1.cu
+7
View File
@@ -0,0 +1,7 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec.cuh"
DECL_FATTN_VEC_CASE( 64, GGML_TYPE_Q4_0, GGML_TYPE_Q4_1);
DECL_FATTN_VEC_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_1);
DECL_FATTN_VEC_CASE(256, GGML_TYPE_Q4_0, GGML_TYPE_Q4_1);
ggml/src/ggml-cuda/template-instances/fattn-vec-instance-q4_0-q5_0.cu
+7
View File
@@ -0,0 +1,7 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec.cuh"
DECL_FATTN_VEC_CASE( 64, GGML_TYPE_Q4_0, GGML_TYPE_Q5_0);
DECL_FATTN_VEC_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_0);
DECL_FATTN_VEC_CASE(256, GGML_TYPE_Q4_0, GGML_TYPE_Q5_0);
ggml/src/ggml-cuda/template-instances/fattn-vec-instance-q4_0-q5_1.cu
+7
View File
@@ -0,0 +1,7 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec.cuh"
DECL_FATTN_VEC_CASE( 64, GGML_TYPE_Q4_0, GGML_TYPE_Q5_1);
DECL_FATTN_VEC_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_1);
DECL_FATTN_VEC_CASE(256, GGML_TYPE_Q4_0, GGML_TYPE_Q5_1);
ggml/src/ggml-cuda/template-instances/fattn-vec-instance-q4_0-q8_0.cu
+7
View File
@@ -0,0 +1,7 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec.cuh"
DECL_FATTN_VEC_CASE( 64, GGML_TYPE_Q4_0, GGML_TYPE_Q8_0);
DECL_FATTN_VEC_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q8_0);
DECL_FATTN_VEC_CASE(256, GGML_TYPE_Q4_0, GGML_TYPE_Q8_0);
ggml/src/ggml-cuda/template-instances/fattn-vec-instance-q4_1-f16.cu
+7
View File
@@ -0,0 +1,7 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec.cuh"
DECL_FATTN_VEC_CASE( 64, GGML_TYPE_Q4_1, GGML_TYPE_F16);
DECL_FATTN_VEC_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_F16);
DECL_FATTN_VEC_CASE(256, GGML_TYPE_Q4_1, GGML_TYPE_F16);
ggml/src/ggml-cuda/template-instances/fattn-vec-instance-q4_1-q4_0.cu
+7
View File
@@ -0,0 +1,7 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec.cuh"
DECL_FATTN_VEC_CASE( 64, GGML_TYPE_Q4_1, GGML_TYPE_Q4_0);
DECL_FATTN_VEC_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q4_0);
DECL_FATTN_VEC_CASE(256, GGML_TYPE_Q4_1, GGML_TYPE_Q4_0);

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