Hi liangfu,
Thanks for your reply! > OpenCL requires multi-core parallelism, and we don’t have multi-core support > in VTA for now. Could you explain a bit more on this multi-core parallelism requirement of OpenCL? We intend to implement the proposed VTA core using single-work-item OpenCL kernels. > How to reuse current VTA hardware to communicate with TVM runtime through > PCI-e interface? A thin middle-layer driver is proposed to facilitate the communication between TVM runtime and the accelerator through PCIe. Thanks! --- [Visit Topic](https://discuss.tvm.ai/t/rfc-vta-support-for-cloud-devices-opencl-compatible/6676/5) to respond. You are receiving this because you enabled mailing list mode. To unsubscribe from these emails, [click here](https://discuss.tvm.ai/email/unsubscribe/1656654e359e0e16a6e7bb557b9832af7d19227ab96014d406b651d8f5612eff).
