ggml : adjust mul_mat_f16 work memory (#1226)

* llama : minor - remove explicity int64_t cast

* ggml : reduce memory buffer for F16 mul_mat when not using cuBLAS

* ggml : add asserts to guard for incorrect wsize

1 files changed, 1 insertions, 1 deletions

diff --git a/llama.cpp b/llama.cpp
index dc4bdc53..f8b4c8e4 100644
--- a/llama.cpp
+++ b/llama.cpp
@@ -780,7 +780,7 @@ static bool kv_cache_init(
     const int n_embd  = hparams.n_embd;
     const int n_layer = hparams.n_layer;
 
-    const int64_t n_mem      = (int64_t)n_layer*n_ctx;
+    const int64_t n_mem      = n_layer*n_ctx;
     const int64_t n_elements = n_embd*n_mem;
 
     cache.buf.resize(2u*n_elements*ggml_type_size(wtype) + 2u*MB);