happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

fix int overflow and remove view op to pass unit test

This commit is contained in:
hongruichen 2024-12-03 10:52:49 +08:00
parent cf91253c41
commit 0d02ee09ed
2 changed files with 1 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
ggml/src/ggml-qnn/backend-ops.cpp
View File

@ -666,8 +666,6 @@ bool ggml_qnn_supports_op(ggml_backend_qnn_device_context *ctx, const ggml_tenso
case GGML_OP_MUL_MAT:
return ggml_qnn_supports_matmul_op(ctx, op);
case GGML_OP_VIEW:
return true;
default:
return false;
}

2
ggml/src/ggml-qnn/ggml-qnn.cpp
View File

@ -227,7 +227,7 @@ size_t ggml_backend_qnn_buffer_type_get_alignment(ggml_backend_buffer_type_t buf
size_t ggml_backend_qnn_buffer_type_get_max_size(ggml_backend_buffer_type_t buft) {
GGML_UNUSED(buft);
// TODO: get the max size from device
return (2 * 1024 * 1024 * 1024);
return 1024 * 1024 * 1024;
}
bool ggml_backend_qnn_buffer_is_host(ggml_backend_buffer_type_t buft) {