Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦
C
fix(qwen36): heap overflow on a container whose two config files disagree
JustVugga5777a4aea4d
diff --git a/c/qwen36.c b/c/qwen36.cindex 9307c731..46dcaf83 100644--- a/c/qwen36.c+++ b/c/qwen36.c@@ -1145,6 +1145,9 @@ static void validate_cfg(const Cfg *c, int n_layers_from_config) {}static void load_meta(Cfg *c, const char *snap) {+ /* is_attn was sized by load_cfg from config.json and is NOT resized here, so+ * every write below is bounded by this count, not by whatever the meta says. */+ const int is_attn_len = c->n_layers;char path[2048]; snprintf(path, sizeof(path), "%s/qwen36_meta.json", snap);FILE *f = fopen(path, "rb"); if (!f) { printf("[meta] %s not found; using i%%4==3 + defaults\n", path); return; }fseek(f,0,SEEK_END); long n=ftell(f); fseek(f,0,SEEK_SET);@@ -1165,6 +1168,16 @@ static void load_meta(Cfg *c, const char *snap) {G("dn_vheads", dn_vheads); G("dn_kheads", dn_kheads); G("dn_kdim", dn_kdim);G("dn_vdim", dn_vdim); G("dn_convk", dn_convk); G("dn_conv_dim", dn_conv_dim);#undef G+ /* validate_cfg makes exactly this check, and used to be the only one --+ * but it runs AFTER load_meta returns, one line too late to stop the+ * is_attn writes below. A container whose config.json said 4 layers and+ * whose meta said 8 therefore wrote past a 4-byte allocation before+ * anything refused it: ASan heap-buffer-overflow at the layer_types+ * loop, reached from an ordinary model directory. Refuse here, while+ * is_attn is still the only thing that has been sized. */+ CFG_NEED(c->n_layers == is_attn_len,+ "config.json says %d layers, qwen36_meta.json says %d",+ is_attn_len, c->n_layers);if((v=json_get(r,"partial_rotary_factor"))&&v->t==J_NUM) c->partial_rotary_factor=(float)v->num;if((v=json_get(r,"rope_theta"))&&v->t==J_NUM) c->theta=(float)v->num;if((v=json_get(r,"rms_eps"))&&v->t==J_NUM) c->eps=(float)v->num;