gguf : harden loader against malformed tensor dims and metadata types (llama/25596)

* gguf : harden loader against malformed tensor dims and metadata types

* gguf: address review on malformed-metadata hardening

- report the expected vs. actual type when general.alignment is not u32
- use ggml_nelements() > 0 for the zero-element guard and keep the
  representability checks visually aligned
- add test-gguf cases for a wrong-typed alignment key and a zero-dim
  tensor (both used to crash: assert-abort and SIGFPE respectively)

Ran tests/test-gguf: 164/164 pass. Used an AI assistant to help draft
these edits; reviewed and verified by me.

* cont : less comments

Co-authored-by: Georgi Gerganov <redacted>
---------

Co-authored-by: Georgi Gerganov <redacted>

diff --git a/ggml/src/gguf.cpp b/ggml/src/gguf.cpp
index 9f9e4fe5d104fe753e9fbc646efb231c3d78ee76..6c7b5817812b3246ecb9172e20b3dcf83dc216eb 100644 (file)
--- a/ggml/src/gguf.cpp
+++ b/ggml/src/gguf.cpp
@@ -611,6 +611,13 @@ static struct gguf_context * gguf_init_from_reader(const struct gguf_reader & gr
         GGML_ASSERT(int64_t(ctx->kv.size()) == n_kv);
 
         const int alignment_idx = gguf_find_key(ctx, GGUF_KEY_GENERAL_ALIGNMENT);
+        if (alignment_idx != -1 && gguf_get_kv_type(ctx, alignment_idx) != GGUF_TYPE_UINT32) {
+            GGML_LOG_ERROR("%s: key '%s' must be of type %s but is %s\n",
+                __func__, GGUF_KEY_GENERAL_ALIGNMENT, gguf_type_name(GGUF_TYPE_UINT32),
+                gguf_type_name(gguf_get_kv_type(ctx, alignment_idx)));
+            gguf_free(ctx);
+            return nullptr;
+        }
         ctx->alignment = alignment_idx == -1 ? GGUF_DEFAULT_ALIGNMENT : gguf_get_val_u32(ctx, alignment_idx);
 
         if (ctx->alignment == 0 || (ctx->alignment & (ctx->alignment - 1)) != 0) {
@@ -682,9 +689,11 @@ static struct gguf_context * gguf_init_from_reader(const struct gguf_reader & gr
             }
 
             // check that the total number of elements is representable
-            if (ok && ((INT64_MAX/info.t.ne[1] <= info.t.ne[0]) ||
-                       (INT64_MAX/info.t.ne[2] <= info.t.ne[0]*info.t.ne[1]) ||
-                       (INT64_MAX/info.t.ne[3] <= info.t.ne[0]*info.t.ne[1]*info.t.ne[2]))) {
+            // (a zero-element tensor is trivially representable; the guard also avoids a division by zero below)
+            if (ok && ggml_nelements(&info.t) > 0 &&
+                ((INT64_MAX/info.t.ne[1] <= info.t.ne[0]) ||
+                 (INT64_MAX/info.t.ne[2] <= info.t.ne[0]*info.t.ne[1]) ||
+                 (INT64_MAX/info.t.ne[3] <= info.t.ne[0]*info.t.ne[1]*info.t.ne[2]))) {
 
                 GGML_LOG_ERROR("%s: total number of elements in tensor '%s' with shape "
                     "(%" PRIi64 ", %" PRIi64 ", %" PRIi64 ", %" PRIi64 ") is >= %" PRIi64 "\n",