Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

ggml-webgpu: add support for f16 repeat (#26307)

Browse Source
This commit is contained in:
Masashi Yoshimura
2026-08-02 08:28:31 +02:00
committed by GitHub
parent 11924d4c17
commit 7a2db1a0cf
4 changed files with 10 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-webgpu/ggml-webgpu-shader-lib.hpp
+4
View File
@@ -2774,6 +2774,10 @@ class ggml_webgpu_shader_lib {
defines.push_back("TYPE_F32"); defines.push_back("TYPE_F32");
variant += "_f32"; variant += "_f32";
break; break;
case GGML_TYPE_F16:
defines.push_back("TYPE_F16");
variant += "_f16";
break;
case GGML_TYPE_I32: case GGML_TYPE_I32:
defines.push_back("TYPE_I32"); defines.push_back("TYPE_I32");
variant += "_i32"; variant += "_i32";
ggml/src/ggml-webgpu/ggml-webgpu.cpp
+2 -1
View File
@@ -4290,7 +4290,8 @@ static bool ggml_backend_webgpu_device_supports_op(ggml_backend_dev_t dev, const
supports_op = (src0->type == GGML_TYPE_F32 || src0->type == GGML_TYPE_I32); supports_op = (src0->type == GGML_TYPE_F32 || src0->type == GGML_TYPE_I32);
break; break;
case GGML_OP_REPEAT: case GGML_OP_REPEAT:
supports_op = (src0->type == GGML_TYPE_F32 || src0->type == GGML_TYPE_I32 || src0->type == GGML_TYPE_I16); supports_op = (src0->type == GGML_TYPE_F32 || src0->type == GGML_TYPE_F16 || src0->type == GGML_TYPE_I32 ||
src0->type == GGML_TYPE_I16);
break; break;
case GGML_OP_CPY: case GGML_OP_CPY:
case GGML_OP_CONT: case GGML_OP_CONT:
ggml/src/ggml-webgpu/wgsl-shaders/repeat.wgsl
+3
View File
@@ -27,6 +27,9 @@ struct Params {
#ifdef TYPE_I32 #ifdef TYPE_I32
#define DataType i32 #define DataType i32
#endif #endif
#ifdef TYPE_F16
#define DataType f16
#endif
#ifdef TYPE_I16 #ifdef TYPE_I16
// same size (16-bit) is sufficient for repeat // same size (16-bit) is sufficient for repeat
#define DataType f16 #define DataType f16
tests/test-backend-ops.cpp
+1
View File
@@ -8511,6 +8511,7 @@ static std::vector<std::unique_ptr<test_case>> make_test_cases_eval() {
test_cases.emplace_back(new test_repeat(GGML_TYPE_F32, {10, 5, 4, ne3}, {1, 2, 1, 1})); test_cases.emplace_back(new test_repeat(GGML_TYPE_F32, {10, 5, 4, ne3}, {1, 2, 1, 1}));
test_cases.emplace_back(new test_repeat(GGML_TYPE_F32, {10, 5, 4, ne3}, {1, 1, 2, 1})); test_cases.emplace_back(new test_repeat(GGML_TYPE_F32, {10, 5, 4, ne3}, {1, 1, 2, 1}));
test_cases.emplace_back(new test_repeat(GGML_TYPE_F32, {10, 5, 4, ne3}, {1, 1, 1, 2})); test_cases.emplace_back(new test_repeat(GGML_TYPE_F32, {10, 5, 4, ne3}, {1, 1, 1, 2}));
test_cases.emplace_back(new test_repeat(GGML_TYPE_F16, {10, 5, 4, ne3}, {2, 1, 1, 1}));
test_cases.emplace_back(new test_repeat(GGML_TYPE_I32, {10, 5, 4, ne3}, {2, 1, 1, 1})); test_cases.emplace_back(new test_repeat(GGML_TYPE_I32, {10, 5, 4, ne3}, {2, 1, 1, 1}));
test_cases.emplace_back(new test_repeat(GGML_TYPE_I16, {10, 5, 4, ne3}, {1, 1, 1, 2})); test_cases.emplace_back(new test_repeat(GGML_TYPE_I16, {10, 5, 4, ne3}, {1, 1, 1, 2}));
test_cases.emplace_back(new test_repeat(GGML_TYPE_BF16, {10, 5, 4, ne3}, {2, 1, 1, 1})); test_cases.emplace_back(new test_repeat(GGML_TYPE_BF16, {10, 5, 4, ne3}, {2, 1, 1, 1}));