Shangyan Zhou 9eb2f84b3e Optimize intranode combine. (#247) ... * Increase the test round. * Add warp synchronization. * Shuffle the send warps. * Add time elapsed into bench result.		2025-06-24 09:10:23 +08:00
..
api.cuh	Remove the low-latency usage flag (#214)	2025-06-16 13:30:14 +08:00
buffer.cuh	Initial commit	2025-02-25 09:07:53 +08:00
CMakeLists.txt	Support UE8M0 data format. (#206)	2025-06-12 09:38:19 +08:00
configs.cuh	Support Ampere architecture (#204)	2025-06-11 15:48:18 +08:00
exception.cuh	Initial commit	2025-02-25 09:07:53 +08:00
ibgda_device.cuh	Fully remove barrier FIFO designs (#200)	2025-06-10 16:23:20 +08:00
internode_ll.cu	Update internode_ll.cu (#246)	2025-06-23 15:18:10 +08:00
internode.cu	Fix the tail loading issue. (#219)	2025-06-18 09:23:25 +08:00
intranode.cu	Optimize intranode combine. (#247)	2025-06-24 09:10:23 +08:00
launch.cuh	Support more hidden size	2025-06-20 16:37:28 +08:00
layout.cu	Support Ampere architecture (#204)	2025-06-11 15:48:18 +08:00
runtime.cu	Support Ampere architecture (#204)	2025-06-11 15:48:18 +08:00
utils.cuh	Add automatic warp count control for low-latency kernels (#213)	2025-06-16 11:56:43 +08:00