OpenCL: Fix duplication of layers in VRAM and RAM, add GPU mul kernel (#1653)

* Use events instead of clFinish, where possible

* OpenCL: Don't load gpu layers into RAM, add mul_f32 kernel

* Reduce queueing overhead for contiguous tensors by using single mul kernel call

* Adapt to #1612 cl_mem malloc changes

* Reduce code duplication between cuda and opencl branches

* Improve implementation

1 files changed, 2 insertions, 0 deletions

diff --git a/ggml-opencl.h b/ggml-opencl.h
index 5a1a500..c850bb8 100644
--- a/ggml-opencl.h
+++ b/ggml-opencl.h
@@ -8,6 +8,7 @@ extern "C" {
 
 void ggml_cl_init(void);
 
+void   ggml_cl_mul(const struct ggml_tensor * src0, const struct ggml_tensor * src1, struct ggml_tensor * dst);
 bool   ggml_cl_can_mul_mat(const struct ggml_tensor * src0, const struct ggml_tensor * src1, struct ggml_tensor * dst);
 size_t ggml_cl_mul_mat_get_wsize(const struct ggml_tensor * src0, const struct ggml_tensor * src1, struct ggml_tensor * dst);
 void   ggml_cl_mul_mat(const struct ggml_tensor * src0, const struct ggml_tensor * src1, struct ggml_tensor * dst, void * wdata, size_t wsize);
@@ -16,6 +17,7 @@ void * ggml_cl_host_malloc(size_t size);
 void   ggml_cl_host_free(void * ptr);
 
 void ggml_cl_transform_tensor(struct ggml_tensor * tensor);
+void ggml_cl_load_data(const char * fname, struct ggml_tensor * tensor, size_t offset);
 
 #ifdef  __cplusplus
 }