mtmd : fix padding of n_tokens (#19930)

tools/mtmd/mtmd-helper.cpp

@@ -248,7 +248,7 @@ int32_t mtmd_helper_decode_image_chunk(
 
     int32_t n_tokens = mtmd_input_chunk_get_n_tokens(chunk);
     int32_t i_batch = 0;
-    int32_t n_img_batches = GGML_PAD(n_tokens, n_batch) / n_batch;
+    int32_t n_img_batches = (n_tokens + n_batch - 1) / n_batch;
     decode_embd_batch batch_embd(encoded_embd, n_tokens, n_pos_per_embd, n_mmproj_embd);
 
     if (mtmd_decode_use_mrope(ctx)) {

tools/server/server-context.cpp

@@ -2363,7 +2363,7 @@ private:
                                             //printf("[DEBUG] `do_reset` was set to `true` after failing to restore a checkpoint");
                                         } else {
                                             pos_next = std::min(pos_next, std::max(it->pos_min + 1, it->pos_max));
-                                            n_past = slot.prompt.tokens.size_up_to_pos(pos_next);
+                                            n_past = std::min(slot.prompt.tokens.size_up_to_pos(pos_next), (size_t) it->n_tokens);
                                             SLT_WRN(slot, "restored context checkpoint (pos_min = %d, pos_max = %d, n_tokens = %" PRId64 ", size = %.3f MiB)\n", it->pos_min, it->pos_max, it->n_tokens, (float) checkpoint_size / 1024 / 1024);
                                         }
                                     }