tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-07-27 03:33:46 -04:00
Code Issues Packages Projects Releases Wiki Activity

sched : fix multiple evaluations of the same graph with pipeline parallelism (#14855)

Browse Source 
ggml-ci
This commit is contained in:
Diego Devesa
2025-07-25 01:07:26 -07:00
committed by GitHub
parent 3f4fc97f1d
commit c12bbde372
1 changed files with 8 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
ggml/src/ggml-backend.cpp
View File

@ -647,6 +647,7 @@ struct ggml_backend_sched {
// pipeline parallelism support
int n_copies;
int cur_copy;
int next_copy;
ggml_backend_event_t events[GGML_SCHED_MAX_BACKENDS][GGML_SCHED_MAX_COPIES];
struct ggml_tensor * graph_inputs[GGML_SCHED_MAX_SPLIT_INPUTS];
int n_graph_inputs;
@ -1433,8 +1434,6 @@ static enum ggml_status ggml_backend_sched_compute_splits(ggml_backend_sched_t s
}
}
sched->cur_copy = (sched->cur_copy + 1) % sched->n_copies;
return GGML_STATUS_SUCCESS;
}
@ -1535,10 +1534,10 @@ void ggml_backend_sched_reset(ggml_backend_sched_t sched) {
bool ggml_backend_sched_reserve(ggml_backend_sched_t sched, struct ggml_cgraph * measure_graph) {
GGML_ASSERT((int)sched->hash_set.size >= measure_graph->n_nodes + measure_graph->n_leafs);
ggml_backend_sched_split_graph(sched, measure_graph);
ggml_backend_sched_synchronize(sched);
ggml_backend_sched_split_graph(sched, measure_graph);
if (!ggml_gallocr_reserve_n(sched->galloc, &sched->graph, sched->node_backend_ids, sched->leaf_backend_ids)) {
return false;
}
@ -1550,6 +1549,10 @@ bool ggml_backend_sched_reserve(ggml_backend_sched_t sched, struct ggml_cgraph *
bool ggml_backend_sched_alloc_graph(ggml_backend_sched_t sched, struct ggml_cgraph * graph) {
GGML_ASSERT((int)sched->hash_set.size >= graph->n_nodes + graph->n_leafs);
GGML_ASSERT(!sched->is_alloc);
sched->cur_copy = sched->next_copy;
sched->next_copy = (sched->next_copy + 1) % sched->n_copies;
ggml_backend_sched_split_graph(sched, graph);
@ -1590,7 +1593,7 @@ void ggml_backend_sched_synchronize(ggml_backend_sched_t sched) {
// if the graph is not already allocated, always use copy 0 after a synchronization
// this ensures that during generation the same copy is used every time,
// which avoids changes in the graph that could cause CUDA or other graphs to be disabled
sched->cur_copy = 0;
sched->next_copy = 0;
}
}