[llvm] [NVPTX] Unify and extend barrier{.cta} intrinsic support (PR #140615)

Artem Belevich via llvm-commits llvm-commits at lists.llvm.org
Mon May 19 14:53:08 PDT 2025


================
@@ -240,6 +240,34 @@ def BF16RT   : RegTyInfo<bf16, Int16Regs, bf16imm, fpimm, supports_imm = 0>;
 def F16X2RT  : RegTyInfo<v2f16, Int32Regs, ?, ?, supports_imm = 0>;
 def BF16X2RT : RegTyInfo<v2bf16, Int32Regs, ?, ?, supports_imm = 0>;
 
+// This class provides a basic wrapper around an NVPTXInst that abstracts the
+// specific syntax of most PTX instructions. It automatically handles the
+// construction of the asm string based on the provided dag arguments.
----------------
Artem-B wrote:

An example or two  would be useful.

https://github.com/llvm/llvm-project/pull/140615


More information about the llvm-commits mailing list