[llvm] [NVPTX] Add TMA bulk tensor copy intrinsics (PR #96083)

[Durgadoss R via llvm-commits]

================
@@ -0,0 +1,169 @@
+; RUN: llc < %s -march=nvptx64 -mcpu=sm_90 -mattr=+ptx80| FileCheck --check-prefixes=CHECK_PTX64 %s
----------------
durga4github wrote:

Thanks for this suggestion! It would be very beneficial indeed.
Please let me know if the latest revision aligns with what you had in mind.

https://github.com/llvm/llvm-project/pull/96083