Skip to content

metal : reduce command encoding overhead #9698

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Oct 1, 2024
Merged

metal : reduce command encoding overhead #9698

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
43b9d69
metal : reduce command encoding overhead
ggerganov Oct 1, 2024
5273e59
metal : add comments
ggerganov Oct 1, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 0 additions & 7 deletions examples/cvector-generator/pca.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -204,13 +204,6 @@ static ggml_status compute_piter(
ggml_backend_cpu_set_n_threads(model.backend, params.n_threads);
}

// TODO: enable GPU support when support for GGML_OP_SQRT is added
//#ifdef GGML_USE_METAL
// if (ggml_backend_is_metal(model.backend)) {
// ggml_backend_metal_set_n_cb(model.backend, params.n_threads);
// }
//#endif

ggml_status res = ggml_backend_graph_compute(model.backend, gf);
if (res == GGML_STATUS_SUCCESS) {
auto extract_i = [](std::string prefix, std::string str) -> int {
Expand Down
6 changes: 0 additions & 6 deletions examples/llava/clip.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2444,12 +2444,6 @@ bool clip_image_batch_encode(clip_ctx * ctx, const int n_threads, const clip_ima
ggml_backend_cpu_set_n_threads(ctx->backend, n_threads);
}

#ifdef GGML_USE_METAL
if (ggml_backend_is_metal(ctx->backend)) {
ggml_backend_metal_set_n_cb(ctx->backend, n_threads);
}
#endif

ggml_backend_graph_compute(ctx->backend, gf);

// the last node is the embedding tensor
Expand Down
5 changes: 0 additions & 5 deletions ggml/include/ggml-metal.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,9 +25,6 @@
#include <stddef.h>
#include <stdbool.h>

// max memory buffers that can be mapped to the device
#define GGML_METAL_MAX_BUFFERS 64

struct ggml_tensor;
struct ggml_cgraph;

Expand All @@ -48,8 +45,6 @@ GGML_API bool ggml_backend_is_metal(ggml_backend_t backend);

GGML_API GGML_CALL ggml_backend_buffer_t ggml_backend_metal_buffer_from_ptr(void * data, size_t size, size_t max_size);

GGML_API void ggml_backend_metal_set_n_cb(ggml_backend_t backend, int n_cb);

GGML_API void ggml_backend_metal_set_abort_callback(ggml_backend_t backend, ggml_abort_callback abort_callback, void * user_data);

GGML_API GGML_CALL ggml_backend_buffer_type_t ggml_backend_metal_buffer_type(void);
Expand Down
Loading
Loading