[llvm] VPlan: use worklist in simplifyRecipes (PR #93998)

[Ramkumar Ramachandra via llvm-commits]

================
@@ -7,135 +7,109 @@ define void @test(ptr %p, i40 %a) {
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
 ; CHECK:       vector.ph:
-; CHECK-NEXT:    [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <16 x i40> poison, i40 [[A]], i64 0
-; CHECK-NEXT:    [[BROADCAST_SPLAT2:%.*]] = shufflevector <16 x i40> [[BROADCAST_SPLATINSERT1]], <16 x i40> poison, <16 x i32> zeroinitializer
 ; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; CHECK:       vector.body:
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE32:%.*]] ]
 ; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <16 x i32> poison, i32 [[INDEX]], i64 0
 ; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <16 x i32> [[BROADCAST_SPLATINSERT]], <16 x i32> poison, <16 x i32> zeroinitializer
 ; CHECK-NEXT:    [[VEC_IV:%.*]] = add <16 x i32> [[BROADCAST_SPLAT]], <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
 ; CHECK-NEXT:    [[TMP0:%.*]] = icmp ule <16 x i32> [[VEC_IV]], <i32 9, i32 9, i32 9, i32 9, i32 9, i32 9, i32 9, i32 9, i32 9, i32 9, i32 9, i32 9, i32 9, i32 9, i32 9, i32 9>
-; CHECK-NEXT:    [[TMP1:%.*]] = shl <16 x i40> [[BROADCAST_SPLAT2]], <i40 24, i40 24, i40 24, i40 24, i40 24, i40 24, i40 24, i40 24, i40 24, i40 24, i40 24, i40 24, i40 24, i40 24, i40 24, i40 24>
-; CHECK-NEXT:    [[TMP2:%.*]] = ashr <16 x i40> [[TMP1]], <i40 28, i40 28, i40 28, i40 28, i40 28, i40 28, i40 28, i40 28, i40 28, i40 28, i40 28, i40 28, i40 28, i40 28, i40 28, i40 28>
-; CHECK-NEXT:    [[TMP3:%.*]] = trunc <16 x i40> [[TMP2]] to <16 x i32>
-; CHECK-NEXT:    [[TMP4:%.*]] = trunc <16 x i32> [[TMP3]] to <16 x i1>
-; CHECK-NEXT:    [[TMP5:%.*]] = icmp eq <16 x i1> [[TMP4]], zeroinitializer
-; CHECK-NEXT:    [[TMP6:%.*]] = icmp ult <16 x i1> zeroinitializer, [[TMP5]]
-; CHECK-NEXT:    [[TMP7:%.*]] = or <16 x i1> [[TMP6]], <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>
-; CHECK-NEXT:    [[TMP8:%.*]] = icmp sgt <16 x i1> [[TMP7]], zeroinitializer
----------------
artagnon wrote:

So, I stepped through the code and didn't see anything surprising happening. The final VPlan, emitted as:

```
VPlan 'Final VPlan for VF={4},UF>=1' {
Live-in vp<%0> = VF * UF
Live-in vp<%1> = vector-trip-count
Live-in vp<%2> = backedge-taken count
Live-in ir<10> = original trip-count

vector.ph:
Successor(s): vector loop

<x1> vector loop: {
  vector.body:
    EMIT vp<%3> = CANONICAL-INDUCTION ir<0>, vp<%6>
    EMIT vp<%4> = WIDEN-CANONICAL-INDUCTION vp<%3>
    EMIT vp<%5> = icmp ule vp<%4>, vp<%2>
    WIDEN ir<%icmp.sgt> = icmp sgt ir<true>, ir<false>
  Successor(s): pred.store

  <xVFxUF> pred.store: {
    pred.store.entry:
      BRANCH-ON-MASK vp<%5>
    Successor(s): pred.store.if, pred.store.continue

    pred.store.if:
      REPLICATE store ir<%icmp.sgt>, ir<%p>
    Successor(s): pred.store.continue

    pred.store.continue:
    No successors
  }
  Successor(s): for.body.0

  for.body.0:
    EMIT vp<%6> = add vp<%3>, vp<%0>
    EMIT branch-on-count vp<%6>, vp<%1>
  No successors
}
Successor(s): middle.block

middle.block:
  EMIT branch-on-cond ir<true>
Successor(s): ir-bb<exit>, scalar.ph

ir-bb<exit>:
No successors

scalar.ph:
No successors
}
```

seems to be correct too. I'm confused about why `icmp sgt ir<true>, ir<false>` is lowered to `false`: is there an underlying bug somewhere (independent of this patch)?

https://github.com/llvm/llvm-project/pull/93998