mirror of
https://github.com/ml-explore/mlx.git
synced 2025-12-16 01:49:05 +08:00
[CUDA] Switch to CUDA graphs (#2317)
* cuda graph prototype fix signal bug + start to add dependencies capture more capture more ops remaining ops fix reduce and rope deps add concurrent context try update, but not working cosistent topology order use node api use node api directly to reduce overhead fix bug use kernels in unary cache graph format fix synchronization format * comment
This commit is contained in:
@@ -37,22 +37,20 @@ void eval(array& arr) {
|
||||
}
|
||||
|
||||
auto& encoder = cu::get_command_encoder(arr.primitive().stream());
|
||||
if (encoder.has_gpu_work()) {
|
||||
// Keep used buffers alive until kernel finishes running.
|
||||
std::unordered_set<std::shared_ptr<array::Data>> buffers;
|
||||
for (auto& in : arr.inputs()) {
|
||||
buffers.insert(in.data_shared_ptr());
|
||||
}
|
||||
for (auto& s : arr.siblings()) {
|
||||
buffers.insert(s.data_shared_ptr());
|
||||
}
|
||||
// Remove the output if it was donated to by an input.
|
||||
if (auto it = buffers.find(arr.data_shared_ptr()); it != buffers.end()) {
|
||||
buffers.erase(it);
|
||||
}
|
||||
encoder.add_completed_handler([buffers = std::move(buffers)]() {});
|
||||
// Keep used buffers alive until kernel finishes running.
|
||||
std::unordered_set<std::shared_ptr<array::Data>> buffers;
|
||||
for (auto& in : arr.inputs()) {
|
||||
buffers.insert(in.data_shared_ptr());
|
||||
}
|
||||
encoder.end_encoding();
|
||||
for (auto& s : arr.siblings()) {
|
||||
buffers.insert(s.data_shared_ptr());
|
||||
}
|
||||
// Remove the output if it was donated to by an input.
|
||||
if (auto it = buffers.find(arr.data_shared_ptr()); it != buffers.end()) {
|
||||
buffers.erase(it);
|
||||
}
|
||||
encoder.add_completed_handler([buffers = std::move(buffers)]() {});
|
||||
encoder.maybe_commit();
|
||||
}
|
||||
|
||||
void finalize(Stream s) {
|
||||
|
||||
Reference in New Issue
Block a user