[PATCH] D126475: [AMDGPU] gfx11 vop3 instructions

[Joe Nash via Phabricator via llvm-commits]

Joe_Nash added a comment.

In D126475#3542688 <https://reviews.llvm.org/D126475#3542688>, @dp wrote:

> 



> This change also defines VOP1 and VOP2 opcodes so the name of the change is a bit misleading.

This is a side effect of 1) the same encoding for an instruction on GFX10 and GFX11, 2) the AssemblerPredicate = isGFX10Plus on those real instruction definitions in VOP1Instructions.td and 3) Me accidentally putting in extra tests :)
It's easiest to change the commit message at this point however, so I will plan to just do that unless you have an objection.

> Also I noticed that there are instructions defined by this change that have no tests, for example: v_minmax*, v_maxmin*, v_div_scale*, v_sub_nc_i16, v_ashrrev_i64.

True, I don't have any test created for these at the moment.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D126475/new/

https://reviews.llvm.org/D126475