[AMDGPU] Shrinking V_SUBBREV_U32

[rampitec]

Summary

V_SUBBREV_U32 is a commute opcode for V_SUBB_U32. However, when
we try to commute V_SUBB_U32 in order to shrink it we do not then
process V_SUBBREV_U32 and it stay VOP3. This is fixed.

[msearles]

LGTM

lib/Target/AMDGPU/SIShrinkInstructions.cpp

static bool canShrink(MachineInstr &MI, const SIInstrInfo *TII,
   // a register allocation hint pre-regalloc and then do the shrinking
   // post-regalloc.
   if (Src2) {
     switch (MI.getOpcode()) {
       default: return false;
 
       case AMDGPU::V_ADDC_U32_e64:
       case AMDGPU::V_SUBB_U32_e64:
-        if (TII->getNamedOperand(MI, AMDGPU::OpName::src1)->isImm())
+      case AMDGPU::V_SUBBREV_U32_e64:
+        if (!isVGPR(TII->getNamedOperand(MI, AMDGPU::OpName::src1), TRI, MRI))
           return false;
         // Additional verification is needed for sdst/src2.
         return true;
 
       case AMDGPU::V_MAC_F32_e64:
       case AMDGPU::V_MAC_F16_e64:
         if (!isVGPR(Src2, TRI, MRI) ||
             TII->hasModifiersSet(MI, AMDGPU::OpName::src2_modifiers))

test/CodeGen/AMDGPU/combine-cond-add-sub.ll

bb:
   %cmp = icmp ugt i32 %x, %y
   %ext = zext i1 %cmp to i32
   %add = add i32 %v, %ext
   store i32 %add, i32 addrspace(1)* %gep, align 4
   ret void
 }
 
 ; GCN-LABEL: {{^}}sub1:
-; GCN: v_cmp_gt_u32_e{{32|64}} [[CC:[^,]+]], v{{[0-9]+}}, v{{[0-9]+}}
-; GCN: v_subb_u32_e{{32|64}} v{{[0-9]+}}, {{[^,]+}}, v{{[0-9]+}}, 0, [[CC]]
+; GCN: v_cmp_gt_u32_e32 vcc, v{{[0-9]+}}, v{{[0-9]+}}
+; GCN: v_subbrev_u32_e32 v{{[0-9]+}}, vcc, 0, v{{[0-9]+}}, vcc
 ; GCN-NOT: v_cndmask
 
 define amdgpu_kernel void @sub1(i32 addrspace(1)* nocapture %arg) {
 bb:
   %x = tail call i32 @llvm.amdgcn.workitem.id.x()
   %y = tail call i32 @llvm.amdgcn.workitem.id.y()
   %gep = getelementptr inbounds i32, i32 addrspace(1)* %arg, i32 %x
   %v = load i32, i32 addrspace(1)* %gep, align 4
bb:
   %cmp = tail call zeroext i1 @llvm.amdgcn.class.f32(float %x, i32 608)
   %ext = zext i1 %cmp to i32
   %add = add i32 %v, %ext
   store i32 %add, i32 addrspace(1)* %gep, align 4
   ret void
 }
 
 ; GCN-LABEL: {{^}}sext_flclass:
-; GCN: v_cmp_class_f32_e{{32|64}} [[CC:[^,]+]],
-; GCN: v_subb_u32_e{{32|64}} v{{[0-9]+}}, {{[^,]+}}, v{{[0-9]+}}, 0, [[CC]]
+; GCN: v_cmp_class_f32_e32 vcc,
+; GCN: v_subbrev_u32_e32 v{{[0-9]+}}, vcc, 0, v{{[0-9]+}}, vcc
 ; GCN-NOT: v_cndmask
 
 define amdgpu_kernel void @sext_flclass(i32 addrspace(1)* nocapture %arg, float %x) {
 bb:
   %id = tail call i32 @llvm.amdgcn.workitem.id.x()
   %gep = getelementptr inbounds i32, i32 addrspace(1)* %arg, i32 %id
   %v = load i32, i32 addrspace(1)* %gep, align 4
   %cmp = tail call zeroext i1 @llvm.amdgcn.class.f32(float %x, i32 608)

test/CodeGen/AMDGPU/shrink-carry.mir

 # RUN: llc -march=amdgcn -verify-machineinstrs -start-before si-shrink-instructions -stop-before si-insert-skips -o - %s | FileCheck -check-prefix=GCN %s
 
 # GCN-LABEL: name: subbrev{{$}}
-# GCN:       V_SUBBREV_U32_e64 0, undef $vgpr0, killed $vcc, implicit $exec
+# GCN:       V_SUBBREV_U32_e32 0, undef $vgpr0, implicit-def $vcc, implicit killed $vcc, implicit $exec
 
 ---
 name:            subbrev
 tracksRegLiveness: true
 registers:
   - { id: 0, class: vgpr_32 }
   - { id: 1, class: vgpr_32 }
   - { id: 2, class: vgpr_32 }
   - { id: 3, class: sreg_64_xexec }
   - { id: 4, class: vgpr_32 }
   - { id: 5, class: sreg_64_xexec }
 body:             |
   bb.0:
 
     %0 = IMPLICIT_DEF
     %1 = IMPLICIT_DEF
     %2 = IMPLICIT_DEF
     %3 = V_CMP_GT_U32_e64 %0, %1, implicit $exec
     %4, %5 = V_SUBBREV_U32_e64 0, %0, %3, implicit $exec
 
 ...
 
 # GCN-LABEL: name: subb{{$}}
-# GCN:       V_SUBB_U32_e64 undef $vgpr0, 0, killed $vcc, implicit $exec
+# GCN:       V_SUBBREV_U32_e32 0, undef $vgpr0, implicit-def $vcc, implicit killed $vcc, implicit $exec
 
 ---
 name:            subb
 tracksRegLiveness: true
 registers:
   - { id: 0, class: vgpr_32 }
   - { id: 1, class: vgpr_32 }
   - { id: 2, class: vgpr_32 }