aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorGeorgi Gerganov <ggerganov@gmail.com>2023-07-27 11:00:54 +0300
committerGeorgi Gerganov <ggerganov@gmail.com>2023-07-27 11:00:54 +0300
commit1a941869cbef8e9cc351a6c6987e4ae3b0f021f7 (patch)
tree6152089801c80810412f441a73604802265b8e4f
parentb5472ea0ada081a6e1c06998ebbc9a24aa2cd4a4 (diff)
metal : disable graph concurrency optimization due to bug (#2413)
-rw-r--r--llama.cpp7
1 files changed, 4 insertions, 3 deletions
diff --git a/llama.cpp b/llama.cpp
index 024af99..9a8ecdc 100644
--- a/llama.cpp
+++ b/llama.cpp
@@ -1722,9 +1722,10 @@ static bool llama_eval_internal(
#ifdef GGML_USE_METAL
if (lctx.ctx_metal && N == 1) {
- if (!ggml_metal_if_optimized(lctx.ctx_metal)) {
- ggml_metal_graph_find_concurrency(lctx.ctx_metal, gf);
- }
+ // TODO: disabled until #2413 is resolved
+ //if (!ggml_metal_if_optimized(lctx.ctx_metal)) {
+ // ggml_metal_graph_find_concurrency(lctx.ctx_metal, gf);
+ //}
ggml_metal_set_n_cb (lctx.ctx_metal, n_threads);
ggml_metal_graph_compute(lctx.ctx_metal, gf);
ggml_metal_get_tensor (lctx.ctx_metal, cur);