AMDGPU: Fix counting debug instructions in execz skip threshold

[arsenm]

b953155b492ddd2080e01d939da13f5320b5bb1b

llvm/lib/Target/AMDGPU/SIPreEmitPeephole.cpp

for (MachineFunction::const_iterator MBBI(&From), ToI(&To), End = MF->end();
 
     for (const MachineInstr &MI : MBB) {
       // When a uniform loop is inside non-uniform control flow, the branch
       // leaving the loop might never be taken when EXEC = 0.
       // Hence we should retain cbranch out of the loop lest it become infinite.
       if (MI.isConditionalBranch())
         return true;
 
+      if (MI.isMetaInstruction())
+        continue;
+
       if (TII->hasUnwantedEffectsWhenEXECEmpty(MI))
         return true;
 
       // These instructions are potentially expensive even if EXEC = 0.
       if (TII->isSMRD(MI) || TII->isVMEM(MI) || TII->isFLAT(MI) ||
           TII->isDS(MI) || MI.getOpcode() == AMDGPU::S_WAITCNT)
         return true;
 

llvm/test/CodeGen/AMDGPU/insert-skips-ignored-insts.mir

 # RUN: llc -mtriple=amdgcn-amd-amdhsa -run-pass si-pre-emit-peephole -amdgpu-skip-threshold=3 %s -o - | FileCheck %s
 
 ---
 
-# FIXME: RemoveShortExecBranches should not count dbg_value instructions.
 # CHECK-LABEL: name: no_count_dbg_value
 # CHECK: $vgpr1 = V_MOV_B32_e32 7, implicit $exec
 # CHECK-NOT: S_CBRANCH_EXECZ
 name: no_count_dbg_value
 body: |
   bb.0:
     successors: %bb.1, %bb.2
 
     $vgpr1 = V_MOV_B32_e32 7, implicit $exec
     S_CBRANCH_EXECZ %bb.2, implicit $exec
 
   bb.1:
     successors: %bb.2
     $vgpr0 = V_MOV_B32_e32 0, implicit $exec
     DBG_VALUE
+    DBG_VALUE
+    DBG_VALUE
+    DBG_VALUE
+    DBG_VALUE
+    DBG_VALUE
 
   bb.2:
     $vgpr0 = V_MOV_B32_e32 1, implicit $exec
 
   bb.3:
     S_ENDPGM 0
 ...