[PATCH] D130579: AMDGPU: Use tablegen patterns for buffer global and flat atomic fadd

[Matt Arsenault via Phabricator via llvm-commits]

arsenm accepted this revision.
arsenm added a comment.
This revision is now accepted and ready to land.

LGTM



================
Comment at: llvm/lib/Target/AMDGPU/FLATInstructions.td:1422
 defm : GlobalFLATAtomicPats <"GLOBAL_ATOMIC_XOR", "atomic_load_xor_global", i32>;
-defm : GlobalFLATAtomicPatsRtn <"GLOBAL_ATOMIC_CSUB", int_amdgcn_global_atomic_csub, i32>;
+defm : GlobalFLATAtomicIntrPatsRet <"GLOBAL_ATOMIC_CSUB", "int_amdgcn_global_atomic_csub", i32>;
 
----------------
Can drop the Intr from the multiclass name for consistency


================
Comment at: llvm/test/CodeGen/AMDGPU/GlobalISel/flat-atomic-fadd-f32.ll:5
-
-; GFX11: LLVM ERROR: cannot select: %4:vgpr(s32) = G_INTRINSIC_W_SIDE_EFFECTS intrinsic(@llvm.amdgcn.flat.atomic.fadd), %0:vgpr(p0), %1:vgpr(s32) :: (volatile dereferenceable load store (s32) on %ir.ptr) (in function: flat_atomic_fadd_f32_no_rtn_intrinsic)
 
----------------
I'm assuming this really does exist on gfx11


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D130579/new/

https://reviews.llvm.org/D130579