pitrou commented on code in PR #40699:
URL: https://github.com/apache/arrow/pull/40699#discussion_r1537905137
##########
cpp/src/arrow/gpu/cuda_memory.cc:
##########
@@ -501,5 +502,27 @@ Result<std::shared_ptr<MemoryManager>>
DefaultMemoryMapper(ArrowDeviceType devic
}
}
+Result<std::shared_ptr<MemoryManager>> DefaultGPUMemoryMapper(int64_t
device_id) {
+ ARROW_ASSIGN_OR_RAISE(auto device, arrow::cuda::CudaDevice::Make(device_id));
+ return device->default_memory_manager();
+}
+
+Status RegisterCUDADeviceInternal() {
+ RETURN_NOT_OK(
+ RegisterDeviceMemoryManager(DeviceAllocationType::kCUDA,
DefaultGPUMemoryMapper));
+ RETURN_NOT_OK(RegisterDeviceMemoryManager(DeviceAllocationType::kCUDA_HOST,
+ DefaultGPUMemoryMapper));
+
RETURN_NOT_OK(RegisterDeviceMemoryManager(DeviceAllocationType::kCUDA_MANAGED,
+ DefaultGPUMemoryMapper));
+ return Status::OK();
+}
+
+std::once_flag cuda_registered;
+
+Status RegisterCUDADevice() {
Review Comment:
Well, if they have a need to a dedicated CUDA mapper, they can just pass
their own `DeviceMemoryMapper` when importing, AFAICT. What do you think
@zeroshade ?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
