zhangyiss/mlx
1
0
Fork 0
mirror of https://github.com/ml-explore/mlx.git synced 2025-12-16 01:49:05 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

redesign for faster cpu/gpu synch (#1869)

Browse Source 
* redesign for faster cpu/gpu synch

* load + more async CPU

* use command encoder API and move more ops to use it

* make fence back-end generic + CPU only fence

* faster build

* fix async eval

* fixes + handle temporaries

* fix / improve cpu conv

* remove unused status, fix siblings

* fix extensions

* fix

* fix no cpu build

* format

* comments

* fix perf regression, remove unecessary abort

* fix events, task limit cpu

* fix waiting

* fix donation / temporaries in normalization
This commit is contained in:
Awni Hannun
2025-03-06 19:23:38 -08:00
committed by GitHub
parent 5245f12a46
commit c4230747a1
103 changed files with 5013 additions and 3873 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
mlx/backend/metal/ternary.cpp
View File

@@ -71,12 +71,9 @@ void ternary_op_gpu_inplace(
auto& compute_encoder = d.get_command_encoder(s.index);
compute_encoder.set_compute_pipeline_state(kernel);
bool donate_a = a.data_shared_ptr() == nullptr;
bool donate_b = b.data_shared_ptr() == nullptr;
bool donate_c = c.data_shared_ptr() == nullptr;
compute_encoder.set_input_array(donate_a ? out : a, 0);
compute_encoder.set_input_array(donate_b ? out : b, 1);
compute_encoder.set_input_array(donate_c ? out : c, 2);
compute_encoder.set_input_array(a, 0);
compute_encoder.set_input_array(b, 1);
compute_encoder.set_input_array(c, 2);
compute_encoder.set_output_array(out, 3);
auto thread_group_size = kernel->maxTotalThreadsPerThreadgroup();
@@ -129,7 +126,7 @@ void ternary_op_gpu(
auto& b = inputs[1];
auto& c = inputs[2];
TernaryOpType topt = get_ternary_op_type(a, b, c);
set_ternary_op_output_data(a, b, c, out, topt, true /* donate_with_move */);
set_ternary_op_output_data(a, b, c, out, topt);
ternary_op_gpu_inplace(inputs, out, op, s);
}