SYCLomatic

SYCLomatic copied to clipboard

Signed-off-by: chenwei.sun

tomflinda

ziranzha

Signed-off-by: Jiang, Zhiwei

zhiweij1

ziranzha

Signed-off-by: Daiyaan Ahmed

daiyaan-ahmed6

Signed-off-by: Tang, Jiajun [email protected]

tangjj11

[SYCLomatic] Block Load headers core

4

In Progress PR for Load/Store header functions for Block API (related later to #1305 ) cc @yihanwg @danhoeflinger @mmichel11

abhilash1910

performance issue in the dpct program for stream create, copy, destroy

10

There is some performance gap between the cuda and dpct programs when they run on an NVIDIA GPU (e.g. 3090). Thanks for your review. DPCT program:https://github.com/zjin-lcf/HeCBench/blob/master/src/streamCreateCopyDestroy-sycl/main.cpp Create+Copy+Synchronize+Destroy time for 1...

jinz2014

ShengchenJ

[SYCLomatic]Create separate util file for groups

3

From email discussion on SYCLcompat for cooperative group apis, this is created to add all header algorithms related to block/warp for sycl. This PR only replaces the group apis -...

abhilash1910