[llvm-bugs] [Bug 39302] New: [AMDGPU][MC][GFX9] Several VOP3 opcodes do not support op_sel

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=39302

            Bug ID: 39302
           Summary: [AMDGPU][MC][GFX9] Several VOP3 opcodes do not support
                    op_sel
           Product: libraries
           Version: trunk
          Hardware: PC
                OS: All
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: Backend: AMDGPU
          Assignee: unassignedbugs at nondot.org
          Reporter: dpreobrazhensky at luxoft.com
                CC: llvm-bugs at lists.llvm.org

According with AMD docs, the following opcodes should support op_sel:

- v_alignbit_b32
- v_alignbyte_b32
- v_interp_p2_f16

Also there are 'legacy' opcodes:

- v_mad_legacy_f16
- v_mad_legacy_i16
- v_mad_legacy_u16
- v_fma_legacy_f16
- v_interp_p2_legacy_f16
- v_div_fixup_legacy_f16

These are labelled in SP3 as accepting op_sel; AMD docs says nothing about
that. I believe these legacy opcodes should not support op_sel because this
feature is not in GFX8.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20181015/41fd96cb/attachment.html>