Skip to content

add max buffer sizes to opencl and metal backends #5181

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 1 commit into from
Jan 29, 2024
Merged

add max buffer sizes to opencl and metal backends #5181

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
12 changes: 11 additions & 1 deletion ggml-metal.m
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2375,6 +2375,16 @@ GGML_CALL static size_t ggml_backend_metal_buffer_type_get_alignment(ggml_backen
UNUSED(buft);
}

GGML_CALL static size_t ggml_backend_metal_buffer_type_get_max_size(ggml_backend_buffer_type_t buft) {
id<MTLDevice> device = ggml_backend_metal_get_device();
size_t max_size = device.maxBufferLength;
ggml_backend_metal_free_device();

return max_size;

UNUSED(buft);
}

GGML_CALL static bool ggml_backend_metal_buffer_type_supports_backend(ggml_backend_buffer_type_t buft, ggml_backend_t backend) {
return ggml_backend_is_metal(backend) || ggml_backend_is_cpu(backend);

Expand All @@ -2393,7 +2403,7 @@ GGML_CALL ggml_backend_buffer_type_t ggml_backend_metal_buffer_type(void) {
/* .get_name = */ ggml_backend_metal_buffer_type_get_name,
/* .alloc_buffer = */ ggml_backend_metal_buffer_type_alloc_buffer,
/* .get_alignment = */ ggml_backend_metal_buffer_type_get_alignment,
/* .get_max_size = */ NULL, // TODO: return device.maxBufferLength
/* .get_max_size = */ ggml_backend_metal_buffer_type_get_max_size,
/* .get_alloc_size = */ NULL, // defaults to ggml_nbytes
/* .supports_backend = */ ggml_backend_metal_buffer_type_supports_backend,
/* .is_host = */ ggml_backend_metal_buffer_type_is_host,
Expand Down
11 changes: 10 additions & 1 deletion ggml-opencl.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2125,6 +2125,15 @@ static size_t ggml_backend_opencl_buffer_type_get_alignment(ggml_backend_buffer_
GGML_UNUSED(buffer_type);
}

static size_t ggml_backend_opencl_buffer_type_get_max_size(ggml_backend_buffer_type_t buffer_type) {
static size_t max_size = -1;
if (max_size == (size_t)-1) {
ggml_cl_init();
clGetDeviceInfo(device, CL_DEVICE_MAX_MEM_ALLOC_SIZE, sizeof(size_t), &max_size, NULL);
}
return max_size;
}

static bool ggml_backend_opencl_buffer_type_supports_backend(ggml_backend_buffer_type_t buffer_type, ggml_backend_t backend) {
//return ggml_backend_is_opencl(backend); // opencl must be used through the cpu backend
return ggml_backend_is_cpu(backend);
Expand All @@ -2136,7 +2145,7 @@ static ggml_backend_buffer_type_i ggml_backend_opencl_buffer_type_interface = {
/* .get_name = */ ggml_backend_opencl_buffer_type_name,
/* .alloc_buffer = */ ggml_backend_opencl_buffer_type_alloc_buffer,
/* .get_alignment = */ ggml_backend_opencl_buffer_type_get_alignment,
/* .get_max_size = */ NULL, // TODO: return from device info
/* .get_max_size = */ ggml_backend_opencl_buffer_type_get_max_size,
/* .get_alloc_size = */ NULL,
/* .supports_backend = */ ggml_backend_opencl_buffer_type_supports_backend,
/* .is_host = */ NULL,
Expand Down