szha commented on a change in pull request #11041: gpu mem pool strategy
URL: https://github.com/apache/incubator-mxnet/pull/11041#discussion_r191633830
##########
File path: src/storage/pooled_storage_manager.h
##########
@@ -71,7 +78,7 @@ class GPUPooledStorageManager final : public StorageManager {
private:
void DirectFreeNoLock(Storage::Handle handle) {
cudaError_t err = cudaFree(handle.dptr);
- size_t size = handle.size + NDEV;
Review comment:
cc'd @ptrendx. My understanding on this was that there needs to be enough
bytes to make sure that for 32 devices at least each device has 1 byte, for
nccl scattering. Could you confirm, @ptrendx?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
