Update aosp/master llvm for rebase to r233350

Change-Id: I07d935f8793ee8ec6b7da003f6483046594bca49

21 files changed, 284 insertions, 96 deletions

diff --git a/test/Transforms/LoopUnroll/2007-05-05-UnrollMiscomp.ll b/test/Transforms/LoopUnroll/2007-05-05-UnrollMiscomp.ll
index d4c8402..95e9dde 100644
--- a/test/Transforms/LoopUnroll/2007-05-05-UnrollMiscomp.ll
+++ b/test/Transforms/LoopUnroll/2007-05-05-UnrollMiscomp.ll
@@ -9,11 +9,11 @@ target triple = "i686-apple-darwin8"
 define void @Foo(%struct.__mpz_struct* %base) {
 entry:
         %want = alloca [1 x %struct.__mpz_struct], align 16             ; <[1 x %struct.__mpz_struct]*> [#uses=4]
-        %want1 = getelementptr [1 x %struct.__mpz_struct]* %want, i32 0, i32 0          ; <%struct.__mpz_struct*> [#uses=1]
+        %want1 = getelementptr [1 x %struct.__mpz_struct], [1 x %struct.__mpz_struct]* %want, i32 0, i32 0          ; <%struct.__mpz_struct*> [#uses=1]
         call void @__gmpz_init( %struct.__mpz_struct* %want1 )
-        %want27 = getelementptr [1 x %struct.__mpz_struct]* %want, i32 0, i32 0         ; <%struct.__mpz_struct*> [#uses=1]
-        %want3 = getelementptr [1 x %struct.__mpz_struct]* %want, i32 0, i32 0          ; <%struct.__mpz_struct*> [#uses=1]
-        %want2 = getelementptr [1 x %struct.__mpz_struct]* %want, i32 0, i32 0          ; <%struct.__mpz_struct*> [#uses=2]
+        %want27 = getelementptr [1 x %struct.__mpz_struct], [1 x %struct.__mpz_struct]* %want, i32 0, i32 0         ; <%struct.__mpz_struct*> [#uses=1]
+        %want3 = getelementptr [1 x %struct.__mpz_struct], [1 x %struct.__mpz_struct]* %want, i32 0, i32 0          ; <%struct.__mpz_struct*> [#uses=1]
+        %want2 = getelementptr [1 x %struct.__mpz_struct], [1 x %struct.__mpz_struct]* %want, i32 0, i32 0          ; <%struct.__mpz_struct*> [#uses=2]
         br label %bb
 
 bb:             ; preds = %bb, %entry
diff --git a/test/Transforms/LoopUnroll/2011-08-08-PhiUpdate.ll b/test/Transforms/LoopUnroll/2011-08-08-PhiUpdate.ll
index bf6d6d5..a87b16a 100644
--- a/test/Transforms/LoopUnroll/2011-08-08-PhiUpdate.ll
+++ b/test/Transforms/LoopUnroll/2011-08-08-PhiUpdate.ll
@@ -46,8 +46,8 @@ bb.nph:                                           ; preds = %entry
 bb:                                               ; preds = %bb.nph, %bb1
   %indvar = phi i64 [ 0, %bb.nph ], [ %indvar.next, %bb1 ] ; <i64> [#uses=2]
   %s.01 = phi i32 [ 0, %bb.nph ], [ %2, %bb1 ]    ; <i32> [#uses=1]
-  %scevgep = getelementptr i32* %p, i64 %indvar   ; <i32*> [#uses=1]
-  %1 = load i32* %scevgep, align 1                ; <i32> [#uses=1]
+  %scevgep = getelementptr i32, i32* %p, i64 %indvar   ; <i32*> [#uses=1]
+  %1 = load i32, i32* %scevgep, align 1                ; <i32> [#uses=1]
   %2 = add nsw i32 %1, %s.01                      ; <i32> [#uses=2]
   br label %bb1
 
@@ -84,7 +84,7 @@ do.body:                                          ; preds = %do.cond, %if.end
   br i1 %cond2, label %exit, label %do.cond
 
 exit:                  ; preds = %do.body
-  %tmp7.i = load i32* undef, align 8
+  %tmp7.i = load i32, i32* undef, align 8
   br i1 undef, label %do.cond, label %land.lhs.true
 
 land.lhs.true:                                    ; preds = %exit
diff --git a/test/Transforms/LoopUnroll/2011-08-09-IVSimplify.ll b/test/Transforms/LoopUnroll/2011-08-09-IVSimplify.ll
index a43a4ff..0b48409 100644
--- a/test/Transforms/LoopUnroll/2011-08-09-IVSimplify.ll
+++ b/test/Transforms/LoopUnroll/2011-08-09-IVSimplify.ll
@@ -12,11 +12,11 @@ target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 ; CHECK: while.body:
 ; CHECK-NOT: while.body.1:
 ; CHECK: %shr.1 = lshr i32 %bit_addr.addr.01, 5
-; CHECK: %arrayidx.1 = getelementptr inbounds i32* %bitmap, i32 %shr.1
+; CHECK: %arrayidx.1 = getelementptr inbounds i32, i32* %bitmap, i32 %shr.1
 ; CHECK: %shr.2 = lshr i32 %bit_addr.addr.01, 5
-; CHECK: %arrayidx.2 = getelementptr inbounds i32* %bitmap, i32 %shr.2
+; CHECK: %arrayidx.2 = getelementptr inbounds i32, i32* %bitmap, i32 %shr.2
 ; CHECK: %shr.3 = lshr i32 %bit_addr.addr.01, 5
-; CHECK: %arrayidx.3 = getelementptr inbounds i32* %bitmap, i32 %shr.3
+; CHECK: %arrayidx.3 = getelementptr inbounds i32, i32* %bitmap, i32 %shr.3
 define void @FlipBit(i32* nocapture %bitmap, i32 %bit_addr, i32 %nbits) nounwind {
 entry:
   br label %while.body
@@ -28,8 +28,8 @@ while.body:
   %shr = lshr i32 %bit_addr.addr.01, 5
   %rem = and i32 %bit_addr.addr.01, 31
   %shl = shl i32 1, %rem
-  %arrayidx = getelementptr inbounds i32* %bitmap, i32 %shr
-  %tmp6 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %bitmap, i32 %shr
+  %tmp6 = load i32, i32* %arrayidx, align 4
   %xor = xor i32 %tmp6, %shl
   store i32 %xor, i32* %arrayidx, align 4
   %inc = add i32 %bit_addr.addr.01, 1
diff --git a/test/Transforms/LoopUnroll/2011-10-01-NoopTrunc.ll b/test/Transforms/LoopUnroll/2011-10-01-NoopTrunc.ll
index 617d4db..5f9eec7 100644
--- a/test/Transforms/LoopUnroll/2011-10-01-NoopTrunc.ll
+++ b/test/Transforms/LoopUnroll/2011-10-01-NoopTrunc.ll
@@ -21,8 +21,8 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %sum.02 = phi i32 [ 0, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds i8* %arr, i64 %indvars.iv
-  %0 = load i8* %arrayidx, align 1
+  %arrayidx = getelementptr inbounds i8, i8* %arr, i64 %indvars.iv
+  %0 = load i8, i8* %arrayidx, align 1
   %conv = sext i8 %0 to i32
   %add = add nsw i32 %conv, %sum.02
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopUnroll/AArch64/lit.local.cfg b/test/Transforms/LoopUnroll/AArch64/lit.local.cfg
new file mode 100644
index 0000000..cec29af
--- /dev/null
+++ b/test/Transforms/LoopUnroll/AArch64/lit.local.cfg
@@ -0,0 +1,3 @@
+if not 'AArch64' in config.root.targets:
+    config.unsupported = True
+
diff --git a/test/Transforms/LoopUnroll/AArch64/partial.ll b/test/Transforms/LoopUnroll/AArch64/partial.ll
new file mode 100644
index 0000000..8a1ea80
--- /dev/null
+++ b/test/Transforms/LoopUnroll/AArch64/partial.ll
@@ -0,0 +1,76 @@
+; RUN: opt < %s -S -loop-unroll -mtriple aarch64 -mcpu=cortex-a57 | FileCheck %s
+
+; Partial unroll 8 times for this loop.
+define void @unroll1() nounwind {
+entry:
+  br label %loop
+
+loop:
+  %iv = phi i32 [ 0, %entry ], [ %inc, %loop ]
+  %inc = add i32 %iv, 1
+  %exitcnd = icmp uge i32 %inc, 1024
+  br i1 %exitcnd, label %exit, label %loop
+
+exit:
+  ret void
+}
+
+; CHECK:      add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: icmp
+
+; Partial unroll 16 times for this loop.
+define void @unroll2() nounwind {
+entry:
+  br label %loop1
+
+loop1:
+  %iv1 = phi i32 [ 0, %entry ], [ %inc1, %loop1.latch ]
+  br label %loop2.header
+
+loop2.header:
+  br label %loop2
+
+loop2:
+  %iv2 = phi i32 [ 0, %loop2.header ], [ %inc2, %loop2 ]
+  %inc2 = add i32 %iv2, 1
+  %exitcnd2 = icmp uge i32 %inc2, 1024
+  br i1 %exitcnd2, label %exit2, label %loop2
+
+exit2:
+  br label %loop1.latch
+
+loop1.latch:
+  %inc1 = add i32 %iv1, 1
+  %exitcnd1 = icmp uge i32 %inc1, 1024
+  br i1 %exitcnd2, label %exit, label %loop1
+
+exit:
+  ret void
+}
+
+
+
+; CHECK:      add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: icmp
diff --git a/test/Transforms/LoopUnroll/AArch64/runtime-loop.ll b/test/Transforms/LoopUnroll/AArch64/runtime-loop.ll
new file mode 100644
index 0000000..d3dc081
--- /dev/null
+++ b/test/Transforms/LoopUnroll/AArch64/runtime-loop.ll
@@ -0,0 +1,33 @@
+; RUN: opt < %s -S -loop-unroll -mtriple aarch64 -mcpu=cortex-a57 | FileCheck %s
+
+; Tests for unrolling loops with run-time trip counts
+
+; CHECK:  %xtraiter = and i32 %n
+; CHECK:  %lcmp.mod = icmp ne i32 %xtraiter, 0
+; CHECK:  br i1 %lcmp.mod, label %for.body.prol, label %for.body.preheader.split
+
+; CHECK:  for.body.prol:
+; CHECK:  for.body:
+
+define i32 @test(i32* nocapture %a, i32 %n) nounwind uwtable readonly {
+entry:
+  %cmp1 = icmp eq i32 %n, 0
+  br i1 %cmp1, label %for.end, label %for.body
+
+for.body:                                         ; preds = %for.body, %entry
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %sum.02 = phi i32 [ %add, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %add = add nsw i32 %0, %sum.02
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  %sum.0.lcssa = phi i32 [ 0, %entry ], [ %add, %for.body ]
+  ret i32 %sum.0.lcssa
+}
+
+
diff --git a/test/Transforms/LoopUnroll/PowerPC/a2-unrolling.ll b/test/Transforms/LoopUnroll/PowerPC/a2-unrolling.ll
index aae79cb..e9aa1ac 100644
--- a/test/Transforms/LoopUnroll/PowerPC/a2-unrolling.ll
+++ b/test/Transforms/LoopUnroll/PowerPC/a2-unrolling.ll
@@ -27,8 +27,8 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %sum.02 = phi i32 [ %add, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %sum.02
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopUnroll/X86/partial.ll b/test/Transforms/LoopUnroll/X86/partial.ll
index a2b04c7..4566f79 100644
--- a/test/Transforms/LoopUnroll/X86/partial.ll
+++ b/test/Transforms/LoopUnroll/X86/partial.ll
@@ -9,20 +9,20 @@ entry:
 
 vector.body:                                      ; preds = %vector.body, %entry
   %index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
-  %0 = getelementptr inbounds double* %b, i64 %index
+  %0 = getelementptr inbounds double, double* %b, i64 %index
   %1 = bitcast double* %0 to <2 x double>*
-  %wide.load = load <2 x double>* %1, align 8
+  %wide.load = load <2 x double>, <2 x double>* %1, align 8
   %.sum9 = or i64 %index, 2
-  %2 = getelementptr double* %b, i64 %.sum9
+  %2 = getelementptr double, double* %b, i64 %.sum9
   %3 = bitcast double* %2 to <2 x double>*
-  %wide.load8 = load <2 x double>* %3, align 8
+  %wide.load8 = load <2 x double>, <2 x double>* %3, align 8
   %4 = fadd <2 x double> %wide.load, <double 1.000000e+00, double 1.000000e+00>
   %5 = fadd <2 x double> %wide.load8, <double 1.000000e+00, double 1.000000e+00>
-  %6 = getelementptr inbounds double* %a, i64 %index
+  %6 = getelementptr inbounds double, double* %a, i64 %index
   %7 = bitcast double* %6 to <2 x double>*
   store <2 x double> %4, <2 x double>* %7, align 8
   %.sum10 = or i64 %index, 2
-  %8 = getelementptr double* %a, i64 %.sum10
+  %8 = getelementptr double, double* %a, i64 %.sum10
   %9 = bitcast double* %8 to <2 x double>*
   store <2 x double> %5, <2 x double>* %9, align 8
   %index.next = add i64 %index, 4
@@ -45,12 +45,12 @@ entry:
 
 vector.body:                                      ; preds = %vector.body, %entry
   %index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
-  %v0 = getelementptr inbounds double* %b, i64 %index
+  %v0 = getelementptr inbounds double, double* %b, i64 %index
   %v1 = bitcast double* %v0 to <2 x double>*
-  %wide.load = load <2 x double>* %v1, align 8
+  %wide.load = load <2 x double>, <2 x double>* %v1, align 8
   %v4 = fadd <2 x double> %wide.load, <double 1.000000e+00, double 1.000000e+00>
   %v5 = fmul <2 x double> %v4, <double 8.000000e+00, double 8.000000e+00>
-  %v6 = getelementptr inbounds double* %a, i64 %index
+  %v6 = getelementptr inbounds double, double* %a, i64 %index
   %v7 = bitcast double* %v6 to <2 x double>*
   store <2 x double> %v5, <2 x double>* %v7, align 8
   %index.next = add i64 %index, 2
@@ -84,18 +84,18 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %reduction.026 = phi i16 [ %add14, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i16* %arr, i64 %indvars.iv
-  %0 = load i16* %arrayidx, align 2
+  %arrayidx = getelementptr inbounds i16, i16* %arr, i64 %indvars.iv
+  %0 = load i16, i16* %arrayidx, align 2
   %add = add i16 %0, %reduction.026
   %sext = mul i64 %indvars.iv, 12884901888
   %idxprom3 = ashr exact i64 %sext, 32
-  %arrayidx4 = getelementptr inbounds i16* %arr, i64 %idxprom3
-  %1 = load i16* %arrayidx4, align 2
+  %arrayidx4 = getelementptr inbounds i16, i16* %arr, i64 %idxprom3
+  %1 = load i16, i16* %arrayidx4, align 2
   %add7 = add i16 %add, %1
   %sext28 = mul i64 %indvars.iv, 21474836480
   %idxprom10 = ashr exact i64 %sext28, 32
-  %arrayidx11 = getelementptr inbounds i16* %arr, i64 %idxprom10
-  %2 = load i16* %arrayidx11, align 2
+  %arrayidx11 = getelementptr inbounds i16, i16* %arr, i64 %idxprom10
+  %2 = load i16, i16* %arrayidx11, align 2
   %add14 = add i16 %add7, %2
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopUnroll/ephemeral.ll b/test/Transforms/LoopUnroll/ephemeral.ll
index 9d40613..d16eba7 100644
--- a/test/Transforms/LoopUnroll/ephemeral.ll
+++ b/test/Transforms/LoopUnroll/ephemeral.ll
@@ -12,8 +12,8 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %sum.01 = phi i32 [ 0, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
 
   ; This loop will be completely unrolled, even with these extra instructions,
   ; but only because they're ephemeral (and, thus, free).
diff --git a/test/Transforms/LoopUnroll/full-unroll-heuristics.ll b/test/Transforms/LoopUnroll/full-unroll-heuristics.ll
index a1bb4c5..a9104ad 100644
--- a/test/Transforms/LoopUnroll/full-unroll-heuristics.ll
+++ b/test/Transforms/LoopUnroll/full-unroll-heuristics.ll
@@ -24,14 +24,14 @@
 
 ; If the absolute threshold is too low, or if we can't optimize away requested
 ; percent of instructions, we shouldn't unroll:
-; TEST1: %array_const_idx = getelementptr inbounds [9 x i32]* @known_constant, i64 0, i64 %iv
-; TEST3: %array_const_idx = getelementptr inbounds [9 x i32]* @known_constant, i64 0, i64 %iv
+; TEST1: %array_const_idx = getelementptr inbounds [9 x i32], [9 x i32]* @known_constant, i64 0, i64 %iv
+; TEST3: %array_const_idx = getelementptr inbounds [9 x i32], [9 x i32]* @known_constant, i64 0, i64 %iv
 
 ; Otherwise, we should:
-; TEST2-NOT: %array_const_idx = getelementptr inbounds [9 x i32]* @known_constant, i64 0, i64 %iv
+; TEST2-NOT: %array_const_idx = getelementptr inbounds [9 x i32], [9 x i32]* @known_constant, i64 0, i64 %iv
 
 ; Also, we should unroll if the 'unroll-threshold' is big enough:
-; TEST4-NOT: %array_const_idx = getelementptr inbounds [9 x i32]* @known_constant, i64 0, i64 %iv
+; TEST4-NOT: %array_const_idx = getelementptr inbounds [9 x i32], [9 x i32]* @known_constant, i64 0, i64 %iv
 
 ; And check that we don't crash when we're not allowed to do any analysis.
 ; RUN: opt < %s -loop-unroll -unroll-max-iteration-count-to-analyze=0 -disable-output
@@ -46,10 +46,10 @@ entry:
 loop:                                                ; preds = %loop, %entry
   %iv = phi i64 [ 0, %entry ], [ %inc, %loop ]
   %r  = phi i32 [ 0, %entry ], [ %add, %loop ]
-  %arrayidx = getelementptr inbounds i32* %src, i64 %iv
-  %src_element = load i32* %arrayidx, align 4
-  %array_const_idx = getelementptr inbounds [9 x i32]* @known_constant, i64 0, i64 %iv
-  %const_array_element = load i32* %array_const_idx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %src, i64 %iv
+  %src_element = load i32, i32* %arrayidx, align 4
+  %array_const_idx = getelementptr inbounds [9 x i32], [9 x i32]* @known_constant, i64 0, i64 %iv
+  %const_array_element = load i32, i32* %array_const_idx, align 4
   %mul = mul nsw i32 %src_element, %const_array_element
   %add = add nsw i32 %mul, %r
   %inc = add nuw nsw i64 %iv, 1
diff --git a/test/Transforms/LoopUnroll/ignore-annotation-intrinsic-cost.ll b/test/Transforms/LoopUnroll/ignore-annotation-intrinsic-cost.ll
index dcb5d1c..2101b63 100644
--- a/test/Transforms/LoopUnroll/ignore-annotation-intrinsic-cost.ll
+++ b/test/Transforms/LoopUnroll/ignore-annotation-intrinsic-cost.ll
@@ -18,7 +18,7 @@ for.body:                                         ; preds = %entry, %for.body
   %i.01 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
   ; The real loop.
   %mul = mul nsw i32 %B, %C
-  %arrayidx = getelementptr inbounds i32* %A, i32 %i.01
+  %arrayidx = getelementptr inbounds i32, i32* %A, i32 %i.01
   store i32 %mul, i32* %arrayidx, align 4
   %inc = add nsw i32 %i.01, 1
   %exitcond = icmp ne i32 %inc, 4
diff --git a/test/Transforms/LoopUnroll/runtime-loop.ll b/test/Transforms/LoopUnroll/runtime-loop.ll
index 80571ec..9e78edf 100644
--- a/test/Transforms/LoopUnroll/runtime-loop.ll
+++ b/test/Transforms/LoopUnroll/runtime-loop.ll
@@ -20,8 +20,8 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %sum.02 = phi i32 [ %add, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %sum.02
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -47,8 +47,8 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %sum.01 = phi i32 [ 0, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %sum.01
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -96,8 +96,8 @@ for.body:                                         ; preds = %for.body, %entry
   %p.addr.05 = phi i16* [ %incdec.ptr, %for.body ], [ %p, %entry ]
   %len.addr.04 = phi i32 [ %sub, %for.body ], [ %len, %entry ]
   %res.03 = phi i32 [ %add, %for.body ], [ 0, %entry ]
-  %incdec.ptr = getelementptr inbounds i16* %p.addr.05, i64 1
-  %0 = load i16* %p.addr.05, align 2
+  %incdec.ptr = getelementptr inbounds i16, i16* %p.addr.05, i64 1
+  %0 = load i16, i16* %p.addr.05, align 2
   %conv = zext i16 %0 to i32
   %add = add i32 %conv, %res.03
   %sub = add nsw i32 %len.addr.04, -2
@@ -113,6 +113,39 @@ for.end:                                          ; preds = %for.cond.for.end_cr
   ret i16 %res.0.lcssa
 }
 
+; Test run-time unrolling disable metadata.
+; CHECK: for.body:
+; CHECK-NOT: for.body.prol:
+
+define zeroext i16 @test2(i16* nocapture %p, i32 %len) nounwind uwtable readonly {
+entry:
+  %cmp2 = icmp eq i32 %len, 0
+  br i1 %cmp2, label %for.end, label %for.body
+
+for.body:                                         ; preds = %for.body, %entry
+  %p.addr.05 = phi i16* [ %incdec.ptr, %for.body ], [ %p, %entry ]
+  %len.addr.04 = phi i32 [ %sub, %for.body ], [ %len, %entry ]
+  %res.03 = phi i32 [ %add, %for.body ], [ 0, %entry ]
+  %incdec.ptr = getelementptr inbounds i16, i16* %p.addr.05, i64 1
+  %0 = load i16, i16* %p.addr.05, align 2
+  %conv = zext i16 %0 to i32
+  %add = add i32 %conv, %res.03
+  %sub = add nsw i32 %len.addr.04, -2
+  %cmp = icmp eq i32 %sub, 0
+  br i1 %cmp, label %for.cond.for.end_crit_edge, label %for.body, !llvm.loop !0
+
+for.cond.for.end_crit_edge:                       ; preds = %for.body
+  %phitmp = trunc i32 %add to i16
+  br label %for.end
+
+for.end:                                          ; preds = %for.cond.for.end_crit_edge, %entry
+  %res.0.lcssa = phi i16 [ %phitmp, %for.cond.for.end_crit_edge ], [ 0, %entry ]
+  ret i16 %res.0.lcssa
+}
+
+!0 = distinct !{!0, !1}
+!1 = !{!"llvm.loop.unroll.runtime.disable"}
+
 ; CHECK: !0 = distinct !{!0, !1}
 ; CHECK: !1 = !{!"llvm.loop.unroll.disable"}
 
diff --git a/test/Transforms/LoopUnroll/runtime-loop1.ll b/test/Transforms/LoopUnroll/runtime-loop1.ll
index 5ff75e3..7684e39 100644
--- a/test/Transforms/LoopUnroll/runtime-loop1.ll
+++ b/test/Transforms/LoopUnroll/runtime-loop1.ll
@@ -16,8 +16,8 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %sum.02 = phi i32 [ %add, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %sum.02
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopUnroll/runtime-loop2.ll b/test/Transforms/LoopUnroll/runtime-loop2.ll
index 176362a..7c6bb96 100644
--- a/test/Transforms/LoopUnroll/runtime-loop2.ll
+++ b/test/Transforms/LoopUnroll/runtime-loop2.ll
@@ -16,8 +16,8 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %sum.02 = phi i32 [ %add, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %sum.02
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopUnroll/runtime-loop3.ll b/test/Transforms/LoopUnroll/runtime-loop3.ll
index aa928cc..fd13ebf 100644
--- a/test/Transforms/LoopUnroll/runtime-loop3.ll
+++ b/test/Transforms/LoopUnroll/runtime-loop3.ll
@@ -23,8 +23,8 @@ for.body3:                                        ; preds = %for.cond1.preheader
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3 ], [ 0, %for.cond1.preheader ]
   %sum.19 = phi i32 [ %add4, %for.body3 ], [ %sum.012, %for.cond1.preheader ]
   %0 = add nsw i64 %indvars.iv, %indvars.iv16
-  %arrayidx = getelementptr inbounds i32* %a, i64 %0
-  %1 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %0
+  %1 = load i32, i32* %arrayidx, align 4
   %add4 = add nsw i32 %1, %sum.19
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopUnroll/runtime-loop4.ll b/test/Transforms/LoopUnroll/runtime-loop4.ll
new file mode 100644
index 0000000..9be0ffd
--- /dev/null
+++ b/test/Transforms/LoopUnroll/runtime-loop4.ll
@@ -0,0 +1,43 @@
+; RUN: opt < %s -S -O2 -unroll-runtime=true | FileCheck %s
+
+; Check runtime unrolling prologue can be promoted by LICM pass.
+
+; CHECK: entry:
+; CHECK: %xtraiter
+; CHECK: %lcmp.mod
+; CHECK: loop1:
+; CHECK: br i1 %lcmp.mod
+; CHECK: loop2.prol:
+
+define void @unroll(i32 %iter, i32* %addr1, i32* %addr2) nounwind {
+entry:
+  br label %loop1
+
+loop1:
+  %iv1 = phi i32 [ 0, %entry ], [ %inc1, %loop1.latch ]
+  %offset1 = getelementptr i32, i32* %addr1, i32 %iv1
+  store i32 %iv1, i32* %offset1, align 4
+  br label %loop2.header
+
+loop2.header:
+  br label %loop2
+
+loop2:
+  %iv2 = phi i32 [ 0, %loop2.header ], [ %inc2, %loop2 ]
+  %offset2 = getelementptr i32, i32* %addr2, i32 %iv2
+  store i32 %iv2, i32* %offset2, align 4
+  %inc2 = add i32 %iv2, 1
+  %exitcnd2 = icmp uge i32 %inc2, %iter
+  br i1 %exitcnd2, label %exit2, label %loop2
+
+exit2:
+  br label %loop1.latch
+
+loop1.latch:
+  %inc1 = add i32 %iv1, 1
+  %exitcnd1 = icmp uge i32 %inc1, 1024
+  br i1 %exitcnd1, label %exit, label %loop1
+
+exit:
+  ret void
+}
diff --git a/test/Transforms/LoopUnroll/scevunroll.ll b/test/Transforms/LoopUnroll/scevunroll.ll
index 20161d7..a5c9a6e 100644
--- a/test/Transforms/LoopUnroll/scevunroll.ll
+++ b/test/Transforms/LoopUnroll/scevunroll.ll
@@ -19,8 +19,8 @@ while.body:
   %iv = phi i64 [ 10, %entry ], [ %iv.next, %while.body ]
   %sum = phi i32 [ 0, %entry ], [ %sum.next, %while.body ]
   %iv.next = add i64 %iv, -1
-  %adr = getelementptr inbounds i32* %base, i64 %iv.next
-  %tmp = load i32* %adr, align 8
+  %adr = getelementptr inbounds i32, i32* %base, i64 %iv.next
+  %tmp = load i32, i32* %adr, align 8
   %sum.next = add i32 %sum, %tmp
   %iv.narrow = trunc i64 %iv.next to i32
   %cmp.i65 = icmp sgt i32 %iv.narrow, 0
@@ -46,8 +46,8 @@ entry:
 loop:
   %iv = phi i64 [ 0, %entry ], [ %inc, %tail ]
   %s = phi i64 [ 0, %entry ], [ %s.next, %tail ]
-  %adr = getelementptr i64* %base, i64 %iv
-  %val = load i64* %adr
+  %adr = getelementptr i64, i64* %base, i64 %iv
+  %val = load i64, i64* %adr
   %s.next = add i64 %s, %val
   %inc = add i64 %iv, 1
   %cmp = icmp ne i64 %inc, 4
@@ -67,8 +67,8 @@ exit2:
 ; SCEV properly unrolls multi-exit loops.
 ;
 ; CHECK-LABEL: @multiExit(
-; CHECK: getelementptr i32* %base, i32 10
-; CHECK-NEXT: load i32*
+; CHECK: getelementptr i32, i32* %base, i32 10
+; CHECK-NEXT: load i32, i32*
 ; CHECK: br i1 false, label %l2.10, label %exit1
 ; CHECK: l2.10:
 ; CHECK-NOT: br
@@ -81,8 +81,8 @@ l1:
   %iv2 = phi i32 [ 0, %entry ], [ %inc2, %l2 ]
   %inc1 = add i32 %iv1, 1
   %inc2 = add i32 %iv2, 1
-  %adr = getelementptr i32* %base, i32 %iv1
-  %val = load i32* %adr
+  %adr = getelementptr i32, i32* %base, i32 %iv1
+  %val = load i32, i32* %adr
   %cmp1 = icmp slt i32 %iv1, 5
   br i1 %cmp1, label %l2, label %exit1
 l2:
@@ -112,8 +112,8 @@ l1:
   %iv2 = phi i32 [ 0, %entry ], [ %inc2, %l3 ]
   %inc1 = add i32 %iv1, 1
   %inc2 = add i32 %iv2, 1
-  %adr = getelementptr i32* %base, i32 %iv1
-  %val = load i32* %adr
+  %adr = getelementptr i32, i32* %base, i32 %iv1
+  %val = load i32, i32* %adr
   %cmp1 = icmp slt i32 %iv1, 5
   br i1 %cmp1, label %l2, label %exit1
 l2:
diff --git a/test/Transforms/LoopUnroll/shifted-tripcount.ll b/test/Transforms/LoopUnroll/shifted-tripcount.ll
index a118a46..4c21698 100644
--- a/test/Transforms/LoopUnroll/shifted-tripcount.ll
+++ b/test/Transforms/LoopUnroll/shifted-tripcount.ll
@@ -13,11 +13,11 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %i.013 = phi i64 [ %tmp16, %for.body ], [ 0, %entry ] ; <i64> [#uses=2]
-  %arrayidx7 = getelementptr double* %p, i64 %i.013 ; <double*> [#uses=2]
+  %arrayidx7 = getelementptr double, double* %p, i64 %i.013 ; <double*> [#uses=2]
   %tmp16 = add i64 %i.013, 1                      ; <i64> [#uses=3]
-  %arrayidx = getelementptr double* %p, i64 %tmp16 ; <double*> [#uses=1]
-  %tmp4 = load double* %arrayidx                  ; <double> [#uses=1]
-  %tmp8 = load double* %arrayidx7                 ; <double> [#uses=1]
+  %arrayidx = getelementptr double, double* %p, i64 %tmp16 ; <double*> [#uses=1]
+  %tmp4 = load double, double* %arrayidx                  ; <double> [#uses=1]
+  %tmp8 = load double, double* %arrayidx7                 ; <double> [#uses=1]
   %mul9 = fmul double %tmp8, %tmp4                ; <double> [#uses=1]
   store double %mul9, double* %arrayidx7
   %exitcond = icmp eq i64 %tmp16, %mul10          ; <i1> [#uses=1]
diff --git a/test/Transforms/LoopUnroll/unroll-pragmas-disabled.ll b/test/Transforms/LoopUnroll/unroll-pragmas-disabled.ll
index 4f934a6..dc812fb 100644
--- a/test/Transforms/LoopUnroll/unroll-pragmas-disabled.ll
+++ b/test/Transforms/LoopUnroll/unroll-pragmas-disabled.ll
@@ -19,8 +19,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -51,8 +51,8 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -78,8 +78,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -110,8 +110,8 @@ entry:
 
 for.body3:                                        ; preds = %for.body3, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body3 ]
-  %arrayidx = getelementptr inbounds i32* %List, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %List, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add4 = add nsw i32 %0, 10
   store i32 %add4, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -124,8 +124,8 @@ for.body3.1.preheader:                            ; preds = %for.body3
 for.body3.1:                                      ; preds = %for.body3.1.preheader, %for.body3.1
   %indvars.iv.1 = phi i64 [ %1, %for.body3.1 ], [ 0, %for.body3.1.preheader ]
   %1 = add nsw i64 %indvars.iv.1, 1
-  %arrayidx.1 = getelementptr inbounds i32* %List, i64 %1
-  %2 = load i32* %arrayidx.1, align 4
+  %arrayidx.1 = getelementptr inbounds i32, i32* %List, i64 %1
+  %2 = load i32, i32* %arrayidx.1, align 4
   %add4.1 = add nsw i32 %2, 10
   store i32 %add4.1, i32* %arrayidx.1, align 4
   %exitcond.1 = icmp eq i64 %1, 4
diff --git a/test/Transforms/LoopUnroll/unroll-pragmas.ll b/test/Transforms/LoopUnroll/unroll-pragmas.ll
index 5831557..1354181 100644
--- a/test/Transforms/LoopUnroll/unroll-pragmas.ll
+++ b/test/Transforms/LoopUnroll/unroll-pragmas.ll
@@ -19,8 +19,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -43,8 +43,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -71,8 +71,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -94,8 +94,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -124,8 +124,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -153,8 +153,8 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -190,8 +190,8 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -217,8 +217,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -245,8 +245,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1