root@c0mp1lr:/root# python tf-vgg16.py
2018-04-18 20:06:12.784559: W tensorflow/stream_executor/rocm/rocm_driver.cc:405] creating context when one is currently active; existing: 0x7f5f3af477c0
2018-04-18 20:06:12.784677: I tensorflow/core/common_runtime/gpu/gpu_device.cc:907] Found device 0 with properties:
name: Device 6863
AMDGPU ISA: gfx900
memoryClockRate (GHz) 1.6
pciBusID 0000:2a:00.0
Total memory: 15.98GiB
Free memory: 15.73GiB
2018-04-18 20:06:12.784715: W tensorflow/stream_executor/rocm/rocm_driver.cc:405] creating context when one is currently active; existing: 0x7f5f3af4e4b0
2018-04-18 20:06:12.784750: I tensorflow/core/common_runtime/gpu/gpu_device.cc:907] Found device 1 with properties:
name: Device 687f
AMDGPU ISA: gfx900
memoryClockRate (GHz) 1.63
pciBusID 0000:2d:00.0
Total memory: 7.98GiB
Free memory: 7.73GiB
2018-04-18 20:06:12.784781: I tensorflow/core/common_runtime/gpu/gpu_device.cc:801] Peer access not supported between device ordinals 0 and 1
2018-04-18 20:06:12.784791: I tensorflow/core/common_runtime/gpu/gpu_device.cc:801] Peer access not supported between device ordinals 1 and 0
2018-04-18 20:06:12.784801: I tensorflow/core/common_runtime/gpu/gpu_device.cc:929] DMA: 0 1
2018-04-18 20:06:12.784807: I tensorflow/core/common_runtime/gpu/gpu_device.cc:939] 0:   Y N
2018-04-18 20:06:12.784813: I tensorflow/core/common_runtime/gpu/gpu_device.cc:939] 1:   N Y
2018-04-18 20:06:12.784820: I tensorflow/core/common_runtime/gpu/gpu_device.cc:997] Creating TensorFlow device (/gpu:0) -> (device: 0, name: Device 6863, pci bus id: 0000:2a:00.0)
2018-04-18 20:06:12.784826: I tensorflow/core/common_runtime/gpu/gpu_device.cc:997] Creating TensorFlow device (/gpu:1) -> (device: 1, name: Device 687f, pci bus id: 0000:2d:00.0)
2018-04-18 20:06:15.412789: I tensorflow/core/common_runtime/gpu/gpu_device.cc:997] Creating TensorFlow device (/gpu:0) -> (device: 0, name: Device 6863, pci bus id: 0000:2a:00.0)
2018-04-18 20:06:15.412825: I tensorflow/core/common_runtime/gpu/gpu_device.cc:997] Creating TensorFlow device (/gpu:1) -> (device: 1, name: Device 687f, pci bus id: 0000:2d:00.0)
2018-04-18 20:06:15.465626: I tensorflow/core/kernels/conv_ops.cc:670] running auto-tune for Convolve
2018-04-18 20:06:15.737537: I tensorflow/core/kernels/conv_ops.cc:670] running auto-tune for Convolve
2018-04-18 20:06:15.740450: I tensorflow/core/kernels/conv_ops.cc:670] running auto-tune for Convolve
2018-04-18 20:06:15.742302: I tensorflow/core/kernels/conv_ops.cc:670] running auto-tune for Convolve
2018-04-18 20:06:15.744173: I tensorflow/core/kernels/conv_ops.cc:670] running auto-tune for Convolve
2018-04-18 20:06:15.745706: I tensorflow/core/kernels/conv_ops.cc:670] running auto-tune for Convolve
2018-04-18 20:06:15.747433: I tensorflow/core/kernels/conv_ops.cc:670] running auto-tune for Convolve
2018-04-18 20:06:15.749587: I tensorflow/core/kernels/conv_ops.cc:670] running auto-tune for Convolve
2018-04-18 20:06:15.751555: I tensorflow/core/kernels/conv_ops.cc:670] running auto-tune for Convolve
The time taken by the kernel is 0.022826