OpenCL* for CPU
Ask questions and share information on Intel® SDK for OpenCL™ Applications and OpenCL™ implementations for Intel® CPU.
公告
This forum covers OpenCL* for CPU only. OpenCL* for GPU questions can be asked in the GPU Compute Software forum. Intel® FPGA SDK for OpenCL™ questions can be ask in the FPGA Intel® High Level Design forum.
1745 讨论

device fission on gen9 GPU

Biren_Doshi
初学者
3,585 次查看

Is device fission is supported on Gen 9 GPU?

I want to run multiple OpenCL kernel concurrently on GPU. Is there any way i can run?

Thanks,

Biren Doshi


 

 

0 项奖励
4 回复数
Jeffrey_M_Intel1
3,585 次查看

Device fission is implemented for Intel CPUs only.  https://software.intel.com/en-us/articles/opencl-device-fission-for-cpu-performance.  On the GPU the kernel will use all EUs.   However, you may be able to increase the efficiency of your GPU utilization for small workloads by enabling  out of order execution mode when you create your queue.

0 项奖励
Huseyin_Tugrul_B_
3,585 次查看

Jeffrey M. (Intel) wrote:

Device fission is implemented for Intel CPUs only.  https://software.intel.com/en-us/articles/opencl-device-fission-for-cpu-performance.  On the GPU the kernel will use all EUs.   However, you may be able to increase the efficiency of your GPU utilization for small workloads by enabling  out of order execution mode when you create your queue.

 

Does Intel gpu have context switching on compute-unit level? When I enqueue 100 different kernels using different number of compute units(such as one using only 2 units, some 5, some all,..), does the gpu set unused units idle or do they get their own work from another kernel? 

0 项奖励
Jeffrey_M_Intel1
3,585 次查看

Out of order execution may be a step in the direction you want.  It isn't the same thing as context switching on CPUs but it can allow more efficient use of the hardware for small jobs.  More info here: http://www.iwocl.org/wp-content/uploads/IWOCL2016-employing-ooqueues.pdf

 

0 项奖励
Biren_Doshi
初学者
3,585 次查看

Thanks Jeffrey for explanation.

0 项奖励
回复