Hi,
I was trying to offload some portion of my code and I set OFFLOAD_REPORT =3. The report is like this.
[Offload] [HOST] [Tag 172] [CPU Time] 0.120382(seconds)
[Offload] [MIC 0] [Tag 172] [CPU->MIC Data] 58720296 (bytes)
[Offload] [MIC 0] [Tag 172] [MIC Time] 0.032349(seconds)
[Offload] [MIC 0] [Tag 172] [MIC->CPU Data] 58720256 (bytes)
[Offload] [MIC 0] [Tag 170] [State] Gather copyout data
[Offload] [MIC 0] [Tag 170] [State] MIC->CPU copyout data 0
[Offload] [HOST] [Tag 170] [State] Scatter copyout data
[Offload] [HOST] [Tag 170] [CPU Time] 0.171873(seconds)
[Offload] [MIC 0] [Tag 170] [CPU->MIC Data] 75579486 (bytes)
[Offload] [MIC 0] [Tag 170] [MIC Time] 0.134238(seconds)
[Offload] [MIC 0] [Tag 170] [MIC->CPU Data] 28 (bytes)
For same amount of data MIC-CPU data transfer speed is much higher(40 times). PCI express speed is 8GB/s. But these timings are no where near to my expectation. Can some one tell me how can i increase data transfer speed.
Does this MIC Time include computation on MIC as well or just data transfer time?
Also, I want to know is there any concept like pinned memory (like in GPU) in case of xeon phi.
Thanks
sivaramakrishna