From dc7bdc1e7121693df112f2fdb11cc6b88580ba4b Mon Sep 17 00:00:00 2001
From: Sander de Smalen <sander.desmalen@arm.com>
Date: Thu, 15 Jul 2021 12:22:24 +0100
Subject: [PATCH] [LV] Fix determinism for failing scalable-call.ll test.

The sort function for emitting an OptRemark was not deterministic,
which caused scalable-call.ll to fail on some buildbots. This patch
fixes that.

This patch also fixes an issue where `Instruction::comesBefore()`
is called when two Instructions are in different basic blocks,
which would otherwise cause an assertion failure.
---
 .../Transforms/Vectorize/LoopVectorize.cpp    |  9 ++++-
 .../LoopVectorize/AArch64/scalable-call.ll    | 37 +++++++++++++++++++
 2 files changed, 44 insertions(+), 2 deletions(-)

diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index e70c7f1e7195..e263a5d2038a 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -6111,10 +6111,15 @@ VectorizationFactor LoopVectorizationCostModel::selectVectorizationFactor(
 
   // Emit a report of VFs with invalid costs in the loop.
   if (!InvalidCosts.empty()) {
-    // Sort/group per instruction
+    // Sort/group per instruction (lexicographically within basic blocks).
     llvm::sort(InvalidCosts, [](InstructionVFPair &A, InstructionVFPair &B) {
+      const Instruction *AI = A.first, *BI = B.first;
+      if (AI->getParent() != BI->getParent())
+        return AI->getParent() < BI->getParent();
       ElementCountComparator ECC;
-      return A.first->comesBefore(B.first) || ECC(A.second, B.second);
+      if (AI != BI)
+        return AI->comesBefore(BI);
+      return ECC(A.second, B.second);
     });
 
     // For a list of ordered instruction-vf pairs:
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/scalable-call.ll b/llvm/test/Transforms/LoopVectorize/AArch64/scalable-call.ll
index e118e98d44af..28270e89193a 100644
--- a/llvm/test/Transforms/LoopVectorize/AArch64/scalable-call.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/scalable-call.ll
@@ -123,6 +123,42 @@ for.cond.cleanup:                                 ; preds = %for.body
   ret void
 }
 
+; CHECK-REMARKS: UserVF ignored because of invalid costs.
+; CHECK-REMARKS-NEXT: t.c:3:10: Instruction with invalid costs prevented vectorization at VF=(vscale x 1): load
+; CHECK-REMARKS-NEXT: t.c:3:40: Instruction with invalid costs prevented vectorization at VF=(vscale x 1): store
+; CHECK-REMARKS-NEXT: t.c:3:20: Instruction with invalid costs prevented vectorization at VF=(vscale x 1, vscale x 2): call to llvm.sin.f32
+; CHECK-REMARKS-NEXT: t.c:3:30: Instruction with invalid costs prevented vectorization at VF=(vscale x 1, vscale x 2): call to llvm.sin.f32
+define void @vec_sin_no_mapping_ite(float* noalias nocapture %dst, float* noalias nocapture readonly %src, i64 %n) {
+; CHECK: @vec_sin_no_mapping_ite
+; CHECK-NOT: <vscale x
+; CHECK: ret
+entry:
+  br label %for.body
+
+for.body:                                         ; preds = %entry, %if.end
+  %i.07 = phi i64 [ %inc, %if.end ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %src, i64 %i.07
+  %0 = load float, float* %arrayidx, align 4, !dbg !11
+  %cmp = fcmp ugt float %0, 0.0000
+  br i1 %cmp, label %if.then, label %if.else
+if.then:
+  %1 = tail call fast float @llvm.sin.f32(float %0), !dbg !12
+  br label %if.end
+if.else:
+  %2 = tail call fast float @llvm.sin.f32(float 0.0), !dbg !13
+  br label %if.end
+if.end:
+  %3 = phi float [%1, %if.then], [%2, %if.else]
+  %arrayidx1 = getelementptr inbounds float, float* %dst, i64 %i.07
+  store float %3, float* %arrayidx1, align 4, !dbg !14
+  %inc = add nuw nsw i64 %i.07, 1
+  %exitcond.not = icmp eq i64 %inc, %n
+  br i1 %exitcond.not, label %for.cond.cleanup, label %for.body, !llvm.loop !1
+
+for.cond.cleanup:                                 ; preds = %for.body
+  ret void
+}
+
 ; CHECK-REMARKS: UserVF ignored because of invalid costs.
 ; CHECK-REMARKS-NEXT: t.c:3:10: Instruction with invalid costs prevented vectorization at VF=(vscale x 1): load
 ; CHECK-REMARKS-NEXT: t.c:3:20: Instruction with invalid costs prevented vectorization at VF=(vscale x 1, vscale x 2): call to llvm.sin.f32
@@ -208,3 +244,4 @@ attributes #3 = { "vector-function-abi-variant"="_ZGV_LLVM_N2v_llvm.sin.f64(sin_
 !11 = !DILocation(line: 3, column: 10, scope: !9)
 !12 = !DILocation(line: 3, column: 20, scope: !9)
 !13 = !DILocation(line: 3, column: 30, scope: !9)
+!14 = !DILocation(line: 3, column: 40, scope: !9)