Announcement

Collapse
No announcement yet.

GFX1013 Target Added To LLVM 13.0 For RDNA2 APUs

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #21
    Originally posted by Spacefish View Post
    B.t.w. ROCm uses hand optimized assembly code for the specific ISA / card for a lot of functions like BLAS and machine learning like the winograd convolution kernel, as compilers donĀ“t deliver the fastest code possible
    Yep... our newer CDNA parts also include matrix operations which don't lend themselves to being compiler targets - so the libraries hard-code those instructions.
    Test signature

    Comment

    Working...
    X