Give the user the option to override where model weights are stored (#232)

* Give the user the option to override where model weights are stored

* Fix ggml_nbytes() problem and cleanup

For a tensor with zero elements ggml_nbytes() was returning
uint64_t::max, and this was causing graph allocation failure.

* Add timing info to CUDA graph evaluation

* Add more timing info

---------

Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>

1 files changed, 2 insertions, 0 deletions

diff --git a/ggml/src/ggml-alloc.c b/ggml/src/ggml-alloc.c
index e485326a..d811dee6 100644
--- a/ggml/src/ggml-alloc.c
+++ b/ggml/src/ggml-alloc.c
@@ -174,6 +174,8 @@ static size_t ggml_dyn_tallocr_alloc(struct ggml_dyn_tallocr * alloc, size_t siz
             // this should never happen
             fprintf(stderr, "%s: not enough space in the buffer to allocate %zu bytes, largest block available %zu bytes\n",
                     __func__, size, max_avail);
+            fprintf(stderr, "%s: tensor was %s with %zu elements and %zu bytes\n", __func__, tensor->name,
+                    ggml_nelements(tensor), ggml_nbytes(tensor));
             GGML_ABORT("not enough space in the buffer");
         }
     }