Skip to content
Discussion options

You must be logged in to vote

可能你对这部分有一点误解。sch.annotate 并不能“替代” tvm.tir.TensorIntrin.register("tmm16", tmm16_desc, tmm16_impl)。这两者是需要同时作用才能工作的。
这里的sch.annotate是给定程序一个tmm16的实现(这个函数在是自己实现的,在编译器中找不到所以需要给定程序),配合Tensorize一起使用

Replies: 1 comment 7 replies

Comment options

You must be logged in to vote
7 replies
@hbsun2113
Comment options

@Hzfengsy
Comment options

@hbsun2113
Comment options

@Hzfengsy
Comment options

@hbsun2113
Comment options

Answer selected by hbsun2113
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
None yet
2 participants