llama: fix RPC for -fit on (#18233)
This commit is contained in:
parent
5e25ddebff
commit
0e1ccf15c7
|
|
@ -583,7 +583,7 @@ static rpc_tensor serialize_tensor(const ggml_tensor * tensor) {
|
|||
if (tensor->buffer) {
|
||||
ggml_backend_buffer_t buffer = tensor->buffer;
|
||||
ggml_backend_rpc_buffer_context * ctx = (ggml_backend_rpc_buffer_context *)buffer->context;
|
||||
result.buffer = ctx->remote_ptr;
|
||||
result.buffer = ctx != nullptr ? ctx->remote_ptr : 0;
|
||||
} else {
|
||||
result.buffer = 0;
|
||||
}
|
||||
|
|
|
|||
Loading…
Reference in New Issue