Thank you for your reply!
The example you described is a NDRange Kernel in OpenCL. To better fit the
design philosophy of FPGA accelerator, both Xilnix and Intel supported another
mode of OpenCL kernel called Single Work-Item Kernel. In fact, both Xilinx and
Intel recommends single work-item
[quote="remotego, post:5, topic:6676"]
Could you explain a bit more on this multi-core parallelism requirement of
OpenCL?
[/quote]
A typical opencl kernel looks like
```c
__kernel void helloworld(__global char* in, __global char* out)
{
int num = get_global_id(0);
out[num] = in
> 
I'm still curious what will happen if we have conv2d(5, 3, 224, 224)? We'll use
conv2d(8, 3, 224, 224)? Do we need to do some padding to use the kernel
conv2d(8, 3, 224, 224)?
T
any progress update about this feature? Thanks
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/apache/incubator-tvm/issues/4118#issuecomment-636596566
