context : fix worst-case reserve outputs (ggml-org#12545)

ggml-ci

Author: ggerganov

src/llama-context.cpp

@@ -294,10 +294,7 @@ llama_context::llama_context(
         // TODO: something cleaner
         const auto n_outputs_save = n_outputs;
 
-        // max number of outputs
-        n_outputs = n_tokens;
-
-        LLAMA_LOG_DEBUG("%s: n_tokens = %d, n_seqs = %d, n_outputs = %d\n", __func__, n_tokens, n_seqs, n_outputs);
+        LLAMA_LOG_DEBUG("%s: worst-case: n_tokens = %d, n_seqs = %d, n_outputs = %d\n", __func__, n_tokens, n_seqs, n_outputs);
 
         int n_splits_pp = -1;
         int n_nodes_pp  = -1;
@@ -313,8 +310,15 @@ llama_context::llama_context(
         // reserve pp graph first so that buffers are only allocated once
         {
             llama_ubatch ubatch_pp = { true, n_tokens, n_tokens / n_seqs, n_seqs, &token, nullptr, nullptr, nullptr, nullptr, nullptr};
+
+            // max number of outputs
+            n_outputs = ubatch_pp.n_tokens;
+
+            LLAMA_LOG_DEBUG("%s: reserving graph for n_tokens = %d, n_seqs = %d\n", __func__, ubatch_pp.n_tokens, ubatch_pp.n_seqs);
+
             auto * gf = graph_init();
             graph_build(ctx_compute.get(), gf, ubatch_pp, LLM_GRAPH_TYPE_DEFAULT);
+
             if (!ggml_backend_sched_reserve(sched.get(), gf)) {
                 throw std::runtime_error("failed to allocate compute pp buffers");
             }
@@ -326,20 +330,33 @@ llama_context::llama_context(
         // reserve with tg graph to get the number of splits and nodes
         {
             llama_ubatch ubatch_tg = { true, 1, 1, n_seqs, &token, nullptr, nullptr, nullptr, nullptr, nullptr};
+
+            n_outputs = ubatch_tg.n_tokens;
+
+            LLAMA_LOG_DEBUG("%s: reserving graph for n_tokens = %d, n_seqs = %d\n", __func__, ubatch_tg.n_tokens, ubatch_tg.n_seqs);
+
             auto * gf = graph_init();
             graph_build(ctx_compute.get(), gf, ubatch_tg, LLM_GRAPH_TYPE_DEFAULT);
+
             if (!ggml_backend_sched_reserve(sched.get(), gf)) {
                 throw std::runtime_error("failed to allocate compute tg buffers");
             }
+
             n_splits_tg = ggml_backend_sched_get_n_splits(sched.get());
             n_nodes_tg  = ggml_graph_n_nodes(gf);
         }
 
         // reserve again with pp graph to avoid ggml-alloc reallocations during inference
         {
             llama_ubatch ubatch_pp = { true, n_tokens, n_tokens / n_seqs, n_seqs, &token, nullptr, nullptr, nullptr, nullptr, nullptr};
+
+            n_outputs = ubatch_pp.n_tokens;
+
+            LLAMA_LOG_DEBUG("%s: reserving graph for n_tokens = %d, n_seqs = %d\n", __func__, ubatch_pp.n_tokens, ubatch_pp.n_seqs);
+
             auto * gf = graph_init();
             graph_build(ctx_compute.get(), gf, ubatch_pp, LLM_GRAPH_TYPE_DEFAULT);
+
             if (!ggml_backend_sched_reserve(sched.get(), gf)) {
                 throw std::runtime_error("failed to allocate compute pp buffers");
             }