fix : cuda order of synchronization when setting a buffer (ggml/679)

* fix : cuda order of synchronization when setting a buffer

* also sync before memcpy

---------

Co-authored-by: slaren <slarengh@gmail.com>
This commit is contained in:
Erik Scholz 2024-01-05 16:00:00 +01:00 committed by Georgi Gerganov
parent 0e26a6c92e
commit 11b1b63b14
No known key found for this signature in database
GPG Key ID: BF970631944C16B7
2 changed files with 2 additions and 2 deletions

View File

@ -1 +1 @@
f96711108d55bdbbd277e6be07204dce6a94fb93
3eace58911ea8d2cf35defdc59848d99b91a57f5

View File

@ -9689,8 +9689,8 @@ static void ggml_backend_cuda_buffer_set_tensor(ggml_backend_buffer_t buffer, gg
ggml_cuda_set_device(ctx->device);
CUDA_CHECK(cudaDeviceSynchronize());
CUDA_CHECK(cudaMemcpy((char *)tensor->data + offset, data, size, cudaMemcpyHostToDevice));
CUDA_CHECK(cudaDeviceSynchronize());
}
static void ggml_backend_cuda_buffer_get_tensor(ggml_backend_buffer_t buffer, const ggml_tensor * tensor, void * data, size_t offset, size_t size) {