Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Revert "sched : reintroduce less synchronizations during split compute (#20793)" (#25138)

Browse Source
This commit is contained in:
Oliver Simons
2026-06-30 08:41:45 +08:00
committed by GitHub
parent 6f4f53f2b7
commit 86b94708f2
2 changed files with 7 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-backend.cpp
+3 -7
View File
@@ -1551,8 +1551,6 @@ static enum ggml_status ggml_backend_sched_compute_splits(ggml_backend_sched_t s
int split_backend_id = split->backend_id;
ggml_backend_t split_backend = sched->backends[split_backend_id];
ggml_backend_synchronize(split_backend);
// copy the input tensors to the split backend
for (int input_id = 0; input_id < split->n_inputs; input_id++) {
ggml_backend_t input_backend = ggml_backend_sched_get_tensor_backend(sched, split->inputs[input_id]);
@@ -1563,15 +1561,15 @@ static enum ggml_status ggml_backend_sched_compute_splits(ggml_backend_sched_t s
// inputs from the user must be copied immediately to prevent the user overwriting the data before the copy is done
if (sched->events[split_backend_id][sched->cur_copy] != NULL) {
ggml_backend_event_synchronize(sched->events[split_backend_id][sched->cur_copy]);
} else if (!split_backend->iface.cpy_tensor_async) {
} else {
ggml_backend_synchronize(split_backend);
}
ggml_backend_tensor_copy_async(input_backend, split_backend, input, input_cpy);
ggml_backend_tensor_copy(input, input_cpy);
} else {
// wait for the split backend to finish using the input before overwriting it
if (sched->events[split_backend_id][sched->cur_copy] != NULL) {
ggml_backend_event_wait(split_backend, sched->events[split_backend_id][sched->cur_copy]);
} else if (!split_backend->iface.cpy_tensor_async) {
} else {
ggml_backend_synchronize(split_backend);
}
@@ -1676,8 +1674,6 @@ static enum ggml_status ggml_backend_sched_compute_splits(ggml_backend_sched_t s
}
}
ggml_backend_synchronize(split_backend);
if (!sched->callback_eval) {
enum ggml_status ec = ggml_backend_graph_compute_async(split_backend, &split->graph);
if (ec != GGML_STATUS_SUCCESS) {