[libc] [llvm] [clang] [lldb] [libcxx] [compiler-rt] [lld] [clang-tools-extra] [flang] [AMDGPU][GFX12] VOP encoding and codegen - add support for v_cvt fp8/… (PR #78414)

Mirko Brkušanin via flang-commits flang-commits at lists.llvm.org
Mon Jan 22 08:59:55 PST 2024


================
@@ -305,6 +305,11 @@ class VOP3OpSel_gfx10<bits<10> op, VOPProfile p> : VOP3e_gfx10<op, p> {
 
 class VOP3OpSel_gfx11_gfx12<bits<10> op, VOPProfile p> : VOP3OpSel_gfx10<op, p>;
 
+class VOP3FP8OpSel_gfx11_gfx12<bits<10> op, VOPProfile p> : VOP3e_gfx10<op, p> {
+  let Inst{11} = !if(p.HasSrc0, src0_modifiers{2}, 0);
+  let Inst{12} = !if(p.HasSrc0, src0_modifiers{3}, 0);
----------------
mbrkusanin wrote:

@kosarev Is this what you had in mind? This uses OP_SEL_1 from src0_modifiers as opsel[1] so we can avoid using src1/src1_modifiers.

https://github.com/llvm/llvm-project/pull/78414


More information about the flang-commits mailing list