未验证 提交 efa34534 编写于 作者: R Ruibiao Chen 提交者: GitHub

Fix typos of managed_memory_msg when OOM (#49228)

上级 a9fd0807
......@@ -75,7 +75,7 @@ phi::Allocation* CUDAAllocator::AllocateImpl(size_t size) {
managed_memory_msg = string::Sprintf(
"If the above ways do not solve the out of memory problem, you can try "
"to use CUDA managed memory. The command is `export "
"FLAGS_use_cuda_managed_memory=false`.");
"FLAGS_use_cuda_managed_memory=true`.");
}
PADDLE_THROW_BAD_ALLOC(platform::errors::ResourceExhausted(
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册