extern/whisper.cpp
1
0
Fork 1
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-08-17 00:11:10 +02:00
Code Issues Packages Projects Releases Wiki Activity

kv-cache : use ggml_set_rows (llama/14285)

Browse Source 
* kv-cache : use ggml_set_rows

ggml-ci

* graph : separate k and v indices

ggml-ci

* cont : remove redundant ifs

ggml-ci

* kv-cache : improve find_slot impl

* kv-cache : bounds-check when accessing slot_info indices

* kv-cache : add comments

ggml-ci

* ggml : add TODOs for adding GGML_OP_SET_ROWS support in the backends

ggml-ci
This commit is contained in:
Georgi Gerganov
2025-07-03 10:53:35 +03:00
parent 0abd0660e1
commit ed639c7f22
4 changed files with 24 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
ggml/src/ggml-cann/ggml-cann.cpp
View File

@ -2086,6 +2086,12 @@ static bool ggml_backend_cann_supports_op(ggml_backend_dev_t dev,
return false;
}
} break;
case GGML_OP_SET_ROWS:
{
// TODO: add support
// ref: https://github.com/ggml-org/llama.cpp/pull/14274
return false;
} break;
case GGML_OP_CPY: {
ggml_tensor *src = op->src[0];
if ((op->type != GGML_TYPE_F32 && op->type != GGML_TYPE_F16) ||

6
ggml/src/ggml-opencl/ggml-opencl.cpp
View File

@ -2222,6 +2222,12 @@ static bool ggml_opencl_supports_op(ggml_backend_dev_t dev, const struct ggml_te
default:
return false;
}
case GGML_OP_SET_ROWS:
{
// TODO: add support
// ref: https://github.com/ggml-org/llama.cpp/pull/14274
return false;
} break;
case GGML_OP_CPY:
case GGML_OP_DUP:
case GGML_OP_CONT:

6
ggml/src/ggml-sycl/ggml-sycl.cpp
View File

@ -4285,6 +4285,12 @@ static bool ggml_backend_sycl_device_supports_op(ggml_backend_dev_t dev, const g
return false;
}
}
case GGML_OP_SET_ROWS:
{
// TODO: add support
// ref: https://github.com/ggml-org/llama.cpp/pull/14274
return false;
} break;
case GGML_OP_CPY:
{
ggml_type src0_type = op->src[0]->type;

6
ggml/src/ggml-vulkan/ggml-vulkan.cpp
View File

@ -10339,6 +10339,12 @@ static bool ggml_backend_vk_device_supports_op(ggml_backend_dev_t dev, const ggm
return false;
}
} break;
case GGML_OP_SET_ROWS:
{
// TODO: add support
// ref: https://github.com/ggml-org/llama.cpp/pull/14274
return false;
} break;
case GGML_OP_CONT:
case GGML_OP_CPY:
case GGML_OP_DUP: