Darshvino commented on issue #13850:
URL: https://github.com/apache/tvm/issues/13850#issuecomment-1405922131

   Hi @vinx13,
   
   Thanks a lot for your reply.
   
   Actually, I am trying to write this operation: **((w_ * x_) & mask_value) >> 
shift_value** in a Tensorize schedule everything using intrinsics just to make 
sure that it’s fully optimized. And when I am trying to use the LLVM 
intrinsics(llvm.aarch64.neon equivalent) for the multiplication, there are no 
llvm.aarch64.neon equivalent for the type of multiplication which I am using. 
And just to give you an info: I need an equivalent for this: vmulq_u16: 
[Intrinsics – Arm 
Developer](https://developer.arm.com/architectures/instruction-sets/intrinsics/#q=vmulq_u16)
 not sure how I can implement it with LLVM intrinsics.
   
   So that is the reason I got thought to directly use the instruction in the 
Tensorize schedule. [Compiler Explorer](https://godbolt.org/z/h43cPMfTK)


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to