[LFTR] Hoist extend expressions outside of loops w/o waiting for LICM

The motivation for this is two fold:
1) Make the output (and thus tests)  a bit more readable to a human trying to understand the result of the transform
2) Reduce spurious diffs in a potential future change to restructure all of this logic to use SCEVExpander (which hoists by default)

llvm-svn: 365066

diff --git a/llvm/lib/Transforms/Scalar/IndVarSimplify.cpp b/llvm/lib/Transforms/Scalar/IndVarSimplify.cpp
index 4d30160..4f939b2 100644 (file)
--- a/llvm/lib/Transforms/Scalar/IndVarSimplify.cpp
+++ b/llvm/lib/Transforms/Scalar/IndVarSimplify.cpp
@@ -2511,7 +2511,10 @@ linearFunctionTestReplace(Loop *L, BasicBlock *ExitingBB,
         }
       }
 
-      if (!Extended)
+      if (Extended) {
+        bool Discard;
+        L->makeLoopInvariant(ExitCnt, Discard);
+      } else 
         CmpIndVar = Builder.CreateTrunc(CmpIndVar, ExitCnt->getType(),
                                         "lftr.wideiv");
     }

diff --git a/llvm/test/Transforms/IndVarSimplify/elim-extend.ll b/llvm/test/Transforms/IndVarSimplify/elim-extend.ll
index 7fda91c..809ea9b 100644 (file)
--- a/llvm/test/Transforms/IndVarSimplify/elim-extend.ll
+++ b/llvm/test/Transforms/IndVarSimplify/elim-extend.ll
@@ -60,6 +60,7 @@ define void @postincVarIV(i8* %base, i32 %init, i32 %limit) nounwind {
 ; CHECK-NEXT:    br i1 [[PRECOND]], label [[LOOP_PREHEADER:%.*]], label [[RETURN:%.*]]
 ; CHECK:       loop.preheader:
 ; CHECK-NEXT:    [[TMP0:%.*]] = sext i32 [[INIT]] to i64
+; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = sext i32 [[LIMIT]] to i64
 ; CHECK-NEXT:    br label [[LOOP:%.*]]
 ; CHECK:       loop:
 ; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[TMP0]], [[LOOP_PREHEADER]] ], [ [[INDVARS_IV_NEXT:%.*]], [[LOOP]] ]
@@ -70,7 +71,6 @@ define void @postincVarIV(i8* %base, i32 %init, i32 %limit) nounwind {
 ; CHECK-NEXT:    store i8 0, i8* [[POSTADR]]
 ; CHECK-NEXT:    [[POSTADRNSW:%.*]] = getelementptr i8, i8* [[BASE]], i64 [[INDVARS_IV_NEXT]]
 ; CHECK-NEXT:    store i8 0, i8* [[POSTADRNSW]]
-; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = sext i32 [[LIMIT]] to i64
 ; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp ne i64 [[INDVARS_IV_NEXT]], [[WIDE_TRIP_COUNT]]
 ; CHECK-NEXT:    br i1 [[EXITCOND]], label [[LOOP]], label [[EXIT:%.*]]
 ; CHECK:       exit:

diff --git a/llvm/test/Transforms/IndVarSimplify/iv-widen-elim-ext.ll b/llvm/test/Transforms/IndVarSimplify/iv-widen-elim-ext.ll
index 08804ad..459d6e1 100644 (file)
--- a/llvm/test/Transforms/IndVarSimplify/iv-widen-elim-ext.ll
+++ b/llvm/test/Transforms/IndVarSimplify/iv-widen-elim-ext.ll
@@ -9,31 +9,31 @@ target datalayout = "e-m:e-i64:64-p:64:64:64-n8:16:32:64-S128"
 define void @foo(i32* %A, i32* %B, i32* %C, i32 %N) {
 ; CHECK-LABEL: @foo(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[CMP1:%.*]] = icmp slt i32 0, %N
-; CHECK-NEXT:    br i1 [[CMP1]], label %for.body.lr.ph, label %for.end
+; CHECK-NEXT:    [[CMP1:%.*]] = icmp slt i32 0, [[N:%.*]]
+; CHECK-NEXT:    br i1 [[CMP1]], label [[FOR_BODY_LR_PH:%.*]], label [[FOR_END:%.*]]
 ; CHECK:       for.body.lr.ph:
-; CHECK-NEXT:    br label %for.body
+; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 [[N]] to i64
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
-; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV:%.*]].next, %for.inc ], [ 0, %for.body.lr.ph ]
-; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, i32* %B, i64 [[INDVARS_IV]]
+; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[FOR_INC:%.*]] ], [ 0, [[FOR_BODY_LR_PH]] ]
+; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, i32* [[B:%.*]], i64 [[INDVARS_IV]]
 ; CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* [[ARRAYIDX]], align 4
 ; CHECK-NEXT:    [[TMP1:%.*]] = add nuw nsw i64 [[INDVARS_IV]], 2
-; CHECK-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds i32, i32* %C, i64 [[TMP1]]
+; CHECK-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds i32, i32* [[C:%.*]], i64 [[TMP1]]
 ; CHECK-NEXT:    [[TMP2:%.*]] = load i32, i32* [[ARRAYIDX2]], align 4
 ; CHECK-NEXT:    [[ADD3:%.*]] = add nsw i32 [[TMP0]], [[TMP2]]
-; CHECK-NEXT:    [[TRUNC0:%.*]] = trunc i64 [[TMP1]] to i32
-; CHECK-NEXT:    [[DIV0:%.*]] = udiv i32 5, [[TRUNC0]]
+; CHECK-NEXT:    [[TMP3:%.*]] = trunc i64 [[TMP1]] to i32
+; CHECK-NEXT:    [[DIV0:%.*]] = udiv i32 5, [[TMP3]]
 ; CHECK-NEXT:    [[ADD4:%.*]] = add nsw i32 [[ADD3]], [[DIV0]]
-; CHECK-NEXT:    [[ARRAYIDX5:%.*]] = getelementptr inbounds i32, i32* %A, i64 [[INDVARS_IV]]
+; CHECK-NEXT:    [[ARRAYIDX5:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i64 [[INDVARS_IV]]
 ; CHECK-NEXT:    store i32 [[ADD4]], i32* [[ARRAYIDX5]], align 4
-; CHECK-NEXT:    br label %for.inc
+; CHECK-NEXT:    br label [[FOR_INC]]
 ; CHECK:       for.inc:
-; CHECK-NEXT:    [[INDVARS_IV_NEXT:%.*]] = add nuw nsw i64 [[INDVARS_IV]], 1
-; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 %N to i64
+; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
 ; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp ne i64 [[INDVARS_IV_NEXT]], [[WIDE_TRIP_COUNT]]
-; CHECK-NEXT:    br i1 [[EXITCOND]], label %for.body, label %for.cond.for.end_crit_edge
+; CHECK-NEXT:    br i1 [[EXITCOND]], label [[FOR_BODY]], label [[FOR_COND_FOR_END_CRIT_EDGE:%.*]]
 ; CHECK:       for.cond.for.end_crit_edge:
-; CHECK-NEXT:    br label %for.end
+; CHECK-NEXT:    br label [[FOR_END]]
 ; CHECK:       for.end:
 ; CHECK-NEXT:    ret void
 ;
@@ -76,28 +76,28 @@ for.end:                                          ; preds = %for.cond.for.end_cr
 define void @foo1(i32* %A, i32* %B, i32* %C, i32 %N) {
 ; CHECK-LABEL: @foo1(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[CMP1:%.*]] = icmp slt i32 0, %N
-; CHECK-NEXT:    br i1 [[CMP1]], label %for.body.lr.ph, label %for.end
+; CHECK-NEXT:    [[CMP1:%.*]] = icmp slt i32 0, [[N:%.*]]
+; CHECK-NEXT:    br i1 [[CMP1]], label [[FOR_BODY_LR_PH:%.*]], label [[FOR_END:%.*]]
 ; CHECK:       for.body.lr.ph:
-; CHECK-NEXT:    br label %for.body
+; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 [[N]] to i64
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
-; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV:%.*]].next, %for.inc ], [ 0, %for.body.lr.ph ]
-; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, i32* %B, i64 [[INDVARS_IV]]
+; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[FOR_INC:%.*]] ], [ 0, [[FOR_BODY_LR_PH]] ]
+; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, i32* [[B:%.*]], i64 [[INDVARS_IV]]
 ; CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* [[ARRAYIDX]], align 4
 ; CHECK-NEXT:    [[TMP1:%.*]] = add nuw nsw i64 [[INDVARS_IV]], 2
-; CHECK-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds i32, i32* %C, i64 [[TMP1]]
+; CHECK-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds i32, i32* [[C:%.*]], i64 [[TMP1]]
 ; CHECK-NEXT:    [[TMP2:%.*]] = load i32, i32* [[ARRAYIDX2]], align 4
 ; CHECK-NEXT:    [[ADD3:%.*]] = add nsw i32 [[TMP0]], [[TMP2]]
-; CHECK-NEXT:    [[ARRAYIDX5:%.*]] = getelementptr inbounds i32, i32* %A, i64 [[INDVARS_IV]]
+; CHECK-NEXT:    [[ARRAYIDX5:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i64 [[INDVARS_IV]]
 ; CHECK-NEXT:    store i32 [[ADD3]], i32* [[ARRAYIDX5]], align 4
-; CHECK-NEXT:    br label %for.inc
+; CHECK-NEXT:    br label [[FOR_INC]]
 ; CHECK:       for.inc:
-; CHECK-NEXT:    [[INDVARS_IV_NEXT:%.*]] = add nuw nsw i64 [[INDVARS_IV]], 1
-; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 %N to i64
+; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
 ; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp ne i64 [[INDVARS_IV_NEXT]], [[WIDE_TRIP_COUNT]]
-; CHECK-NEXT:    br i1 [[EXITCOND]], label %for.body, label %for.cond.for.end_crit_edge
+; CHECK-NEXT:    br i1 [[EXITCOND]], label [[FOR_BODY]], label [[FOR_COND_FOR_END_CRIT_EDGE:%.*]]
 ; CHECK:       for.cond.for.end_crit_edge:
-; CHECK-NEXT:    br label %for.end
+; CHECK-NEXT:    br label [[FOR_END]]
 ; CHECK:       for.end:
 ; CHECK-NEXT:    ret void
 ;
@@ -142,13 +142,14 @@ for.end:                                          ; preds = %for.cond.for.end_cr
 define i32 @foo2(i32 %M) {
 ; CHECK-LABEL: @foo2(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[CMP1:%.*]] = icmp slt i32 0, %M
-; CHECK-NEXT:    br i1 [[CMP1]], label %for.body.lr.ph, label %for.end
+; CHECK-NEXT:    [[CMP1:%.*]] = icmp slt i32 0, [[M:%.*]]
+; CHECK-NEXT:    br i1 [[CMP1]], label [[FOR_BODY_LR_PH:%.*]], label [[FOR_END:%.*]]
 ; CHECK:       for.body.lr.ph:
-; CHECK-NEXT:    [[TMP0:%.*]] = sext i32 %M to i64
-; CHECK-NEXT:    br label %for.body
+; CHECK-NEXT:    [[TMP0:%.*]] = sext i32 [[M]] to i64
+; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 [[M]] to i64
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
-; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV:%.*]].next, %for.inc ], [ 0, %for.body.lr.ph ]
+; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[FOR_INC:%.*]] ], [ 0, [[FOR_BODY_LR_PH]] ]
 ; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds [100 x i32], [100 x i32]* @a, i64 0, i64 [[INDVARS_IV]]
 ; CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* [[ARRAYIDX]], align 4
 ; CHECK-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds [100 x i32], [100 x i32]* @b, i64 0, i64 [[INDVARS_IV]]
@@ -157,14 +158,13 @@ define i32 @foo2(i32 %M) {
 ; CHECK-NEXT:    [[TMP3:%.*]] = add nsw i64 [[INDVARS_IV]], [[TMP0]]
 ; CHECK-NEXT:    [[ARRAYIDX5:%.*]] = getelementptr inbounds [100 x i32], [100 x i32]* @a, i64 0, i64 [[TMP3]]
 ; CHECK-NEXT:    store i32 [[ADD]], i32* [[ARRAYIDX5]], align 4
-; CHECK-NEXT:    br label %for.inc
+; CHECK-NEXT:    br label [[FOR_INC]]
 ; CHECK:       for.inc:
-; CHECK-NEXT:    [[INDVARS_IV_NEXT:%.*]] = add nuw nsw i64 [[INDVARS_IV]], 1
-; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 %M to i64
+; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
 ; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp ne i64 [[INDVARS_IV_NEXT]], [[WIDE_TRIP_COUNT]]
-; CHECK-NEXT:    br i1 [[EXITCOND]], label %for.body, label %for.cond.for.end_crit_edge
+; CHECK-NEXT:    br i1 [[EXITCOND]], label [[FOR_BODY]], label [[FOR_COND_FOR_END_CRIT_EDGE:%.*]]
 ; CHECK:       for.cond.for.end_crit_edge:
-; CHECK-NEXT:    br label %for.end
+; CHECK-NEXT:    br label [[FOR_END]]
 ; CHECK:       for.end:
 ; CHECK-NEXT:    [[CALL:%.*]] = call i32 @dummy(i32* getelementptr inbounds ([100 x i32], [100 x i32]* @a, i32 0, i32 0), i32* getelementptr inbounds ([100 x i32], [100 x i32]* @b, i32 0, i32 0))
 ; CHECK-NEXT:    ret i32 0
@@ -210,13 +210,14 @@ declare i32 @dummy(i32*, i32*)
 define i32 @foo3(i32 %M) {
 ; CHECK-LABEL: @foo3(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[CMP1:%.*]] = icmp slt i32 0, %M
-; CHECK-NEXT:    br i1 [[CMP1]], label %for.body.lr.ph, label %for.end
+; CHECK-NEXT:    [[CMP1:%.*]] = icmp slt i32 0, [[M:%.*]]
+; CHECK-NEXT:    br i1 [[CMP1]], label [[FOR_BODY_LR_PH:%.*]], label [[FOR_END:%.*]]
 ; CHECK:       for.body.lr.ph:
-; CHECK-NEXT:    [[TMP0:%.*]] = sext i32 %M to i64
-; CHECK-NEXT:    br label %for.body
+; CHECK-NEXT:    [[TMP0:%.*]] = sext i32 [[M]] to i64
+; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 [[M]] to i64
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
-; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV:%.*]].next, %for.inc ], [ 0, %for.body.lr.ph ]
+; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[FOR_INC:%.*]] ], [ 0, [[FOR_BODY_LR_PH]] ]
 ; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds [100 x i32], [100 x i32]* @a, i64 0, i64 [[INDVARS_IV]]
 ; CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* [[ARRAYIDX]], align 4
 ; CHECK-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds [100 x i32], [100 x i32]* @b, i64 0, i64 [[INDVARS_IV]]
@@ -227,14 +228,13 @@ define i32 @foo3(i32 %M) {
 ; CHECK-NEXT:    [[IDXPROM4:%.*]] = zext i32 [[TMP4]] to i64
 ; CHECK-NEXT:    [[ARRAYIDX5:%.*]] = getelementptr inbounds [100 x i32], [100 x i32]* @a, i64 0, i64 [[IDXPROM4]]
 ; CHECK-NEXT:    store i32 [[ADD]], i32* [[ARRAYIDX5]], align 4
-; CHECK-NEXT:    br label %for.inc
+; CHECK-NEXT:    br label [[FOR_INC]]
 ; CHECK:       for.inc:
-; CHECK-NEXT:    [[INDVARS_IV_NEXT:%.*]] = add nuw nsw i64 [[INDVARS_IV]], 1
-; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 %M to i64
+; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
 ; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp ne i64 [[INDVARS_IV_NEXT]], [[WIDE_TRIP_COUNT]]
-; CHECK-NEXT:    br i1 [[EXITCOND]], label %for.body, label %for.cond.for.end_crit_edge
+; CHECK-NEXT:    br i1 [[EXITCOND]], label [[FOR_BODY]], label [[FOR_COND_FOR_END_CRIT_EDGE:%.*]]
 ; CHECK:       for.cond.for.end_crit_edge:
-; CHECK-NEXT:    br label %for.end
+; CHECK-NEXT:    br label [[FOR_END]]
 ; CHECK:       for.end:
 ; CHECK-NEXT:    [[CALL:%.*]] = call i32 @dummy(i32* getelementptr inbounds ([100 x i32], [100 x i32]* @a, i32 0, i32 0), i32* getelementptr inbounds ([100 x i32], [100 x i32]* @b, i32 0, i32 0))
 ; CHECK-NEXT:    ret i32 0
@@ -276,6 +276,38 @@ for.end:                                          ; preds = %for.cond.for.end_cr
 
 %struct.image = type {i32, i32}
 define i32 @foo4(%struct.image* %input, i32 %length, i32* %in) {
+; CHECK-LABEL: @foo4(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[STRIDE:%.*]] = getelementptr inbounds [[STRUCT_IMAGE:%.*]], %struct.image* [[INPUT:%.*]], i64 0, i32 1
+; CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* [[STRIDE]], align 4
+; CHECK-NEXT:    [[CMP17:%.*]] = icmp sgt i32 [[LENGTH:%.*]], 1
+; CHECK-NEXT:    br i1 [[CMP17]], label [[FOR_BODY_LR_PH:%.*]], label [[FOR_COND_CLEANUP:%.*]]
+; CHECK:       for.body.lr.ph:
+; CHECK-NEXT:    [[CHANNEL:%.*]] = getelementptr inbounds [[STRUCT_IMAGE]], %struct.image* [[INPUT]], i64 0, i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = sext i32 [[TMP0]] to i64
+; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 [[LENGTH]] to i64
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
+; CHECK:       for.cond.cleanup.loopexit:
+; CHECK-NEXT:    [[TMP2:%.*]] = phi i32 [ [[TMP10:%.*]], [[FOR_BODY]] ]
+; CHECK-NEXT:    br label [[FOR_COND_CLEANUP]]
+; CHECK:       for.cond.cleanup:
+; CHECK-NEXT:    [[TMP3:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[TMP2]], [[FOR_COND_CLEANUP_LOOPEXIT:%.*]] ]
+; CHECK-NEXT:    ret i32 [[TMP3]]
+; CHECK:       for.body:
+; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ], [ 1, [[FOR_BODY_LR_PH]] ]
+; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
+; CHECK-NEXT:    [[TMP4:%.*]] = load i32, i32* [[CHANNEL]], align 8
+; CHECK-NEXT:    [[TMP5:%.*]] = sext i32 [[TMP4]] to i64
+; CHECK-NEXT:    [[TMP6:%.*]] = mul nsw i64 [[TMP5]], [[INDVARS_IV_NEXT]]
+; CHECK-NEXT:    [[ADD_PTR:%.*]] = getelementptr inbounds i32, i32* [[IN:%.*]], i64 [[TMP6]]
+; CHECK-NEXT:    [[TMP7:%.*]] = load i32, i32* [[ADD_PTR]], align 4
+; CHECK-NEXT:    [[TMP8:%.*]] = mul nsw i64 [[TMP1]], [[INDVARS_IV_NEXT]]
+; CHECK-NEXT:    [[ADD_PTR1:%.*]] = getelementptr inbounds i32, i32* [[IN]], i64 [[TMP8]]
+; CHECK-NEXT:    [[TMP9:%.*]] = load i32, i32* [[ADD_PTR1]], align 4
+; CHECK-NEXT:    [[TMP10]] = add i32 [[TMP7]], [[TMP9]]
+; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp ne i64 [[INDVARS_IV_NEXT]], [[WIDE_TRIP_COUNT]]
+; CHECK-NEXT:    br i1 [[EXITCOND]], label [[FOR_BODY]], label [[FOR_COND_CLEANUP_LOOPEXIT]]
+;
 entry:
   %stride = getelementptr inbounds %struct.image, %struct.image* %input, i64 0, i32 1
   %0 = load i32, i32* %stride, align 4
@@ -296,8 +328,6 @@ for.cond.cleanup:                                 ; preds = %for.cond.cleanup.lo
 
 ; mul instruction below is widened instead of generating a truncate instruction for it
 ; regardless if Load operand of mul is inside or outside the loop (we have both cases).
-; CHECK: for.body:
-; CHECK-NOT: trunc
 for.body:                                         ; preds = %for.body.lr.ph, %for.body
   %x.018 = phi i32 [ 1, %for.body.lr.ph ], [ %add, %for.body ]
   %add = add nuw nsw i32 %x.018, 1
@@ -317,6 +347,40 @@ for.body:                                         ; preds = %for.body.lr.ph, %fo
 
 
 define i32 @foo5(%struct.image* %input, i32 %length, i32* %in) {
+; CHECK-LABEL: @foo5(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[STRIDE:%.*]] = getelementptr inbounds [[STRUCT_IMAGE:%.*]], %struct.image* [[INPUT:%.*]], i64 0, i32 1
+; CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* [[STRIDE]], align 4
+; CHECK-NEXT:    [[CMP17:%.*]] = icmp sgt i32 [[LENGTH:%.*]], 1
+; CHECK-NEXT:    br i1 [[CMP17]], label [[FOR_BODY_LR_PH:%.*]], label [[FOR_COND_CLEANUP:%.*]]
+; CHECK:       for.body.lr.ph:
+; CHECK-NEXT:    [[CHANNEL:%.*]] = getelementptr inbounds [[STRUCT_IMAGE]], %struct.image* [[INPUT]], i64 0, i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = sext i32 [[TMP0]] to i64
+; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 [[LENGTH]] to i64
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
+; CHECK:       for.cond.cleanup.loopexit:
+; CHECK-NEXT:    [[TMP2:%.*]] = phi i32 [ [[TMP10:%.*]], [[FOR_BODY]] ]
+; CHECK-NEXT:    br label [[FOR_COND_CLEANUP]]
+; CHECK:       for.cond.cleanup:
+; CHECK-NEXT:    [[TMP3:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[TMP2]], [[FOR_COND_CLEANUP_LOOPEXIT:%.*]] ]
+; CHECK-NEXT:    ret i32 [[TMP3]]
+; CHECK:       for.body:
+; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ], [ 1, [[FOR_BODY_LR_PH]] ]
+; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
+; CHECK-NEXT:    [[TMP4:%.*]] = load i32, i32* [[CHANNEL]], align 8
+; CHECK-NEXT:    [[TMP5:%.*]] = trunc i64 [[INDVARS_IV_NEXT]] to i32
+; CHECK-NEXT:    [[MUL:%.*]] = mul nsw i32 [[TMP4]], [[TMP5]]
+; CHECK-NEXT:    [[IDX_EXT:%.*]] = sext i32 [[MUL]] to i64
+; CHECK-NEXT:    [[ADD_PTR:%.*]] = getelementptr inbounds i32, i32* [[IN:%.*]], i64 [[IDX_EXT]]
+; CHECK-NEXT:    [[TMP6:%.*]] = load i32, i32* [[ADD_PTR]], align 4
+; CHECK-NEXT:    [[TMP7:%.*]] = mul nsw i64 [[TMP1]], [[INDVARS_IV_NEXT]]
+; CHECK-NEXT:    [[ADD_PTR1:%.*]] = getelementptr inbounds i32, i32* [[IN]], i64 [[TMP7]]
+; CHECK-NEXT:    [[TMP8:%.*]] = load i32, i32* [[ADD_PTR1]], align 4
+; CHECK-NEXT:    [[TMP9:%.*]] = add i32 [[TMP6]], [[TMP8]]
+; CHECK-NEXT:    [[TMP10]] = add i32 [[TMP9]], [[MUL]]
+; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp ne i64 [[INDVARS_IV_NEXT]], [[WIDE_TRIP_COUNT]]
+; CHECK-NEXT:    br i1 [[EXITCOND]], label [[FOR_BODY]], label [[FOR_COND_CLEANUP_LOOPEXIT]]
+;
 entry:
   %stride = getelementptr inbounds %struct.image, %struct.image* %input, i64 0, i32 1
   %0 = load i32, i32* %stride, align 4
@@ -338,8 +402,6 @@ for.cond.cleanup:                                 ; preds = %for.cond.cleanup.lo
 ; This example is the same as above except that the first mul is used in two places
 ; and this may result in having two versions of the multiply: an i32 and i64 version.
 ; In this case, keep the trucate instructions to avoid this redundancy.
-; CHECK: for.body:
-; CHECK: trunc
 for.body:                                         ; preds = %for.body.lr.ph, %for.body
   %x.018 = phi i32 [ 1, %for.body.lr.ph ], [ %add, %for.body ]
   %add = add nuw nsw i32 %x.018, 1

diff --git a/llvm/test/Transforms/IndVarSimplify/iv-widen.ll b/llvm/test/Transforms/IndVarSimplify/iv-widen.ll
index a8d89b1..9370bcd 100644 (file)
--- a/llvm/test/Transforms/IndVarSimplify/iv-widen.ll
+++ b/llvm/test/Transforms/IndVarSimplify/iv-widen.ll
@@ -173,6 +173,7 @@ define void @loop_2(i32 %size, i32 %nsteps, i32 %hsize, i32* %lined, i8 %tmp1) {
 ; CHECK-NEXT:    [[TMP4:%.*]] = add nsw i64 [[TMP3]], [[TMP1]]
 ; CHECK-NEXT:    br i1 [[CMP215]], label [[FOR_BODY2_PREHEADER:%.*]], label [[FOR_INC]]
 ; CHECK:       for.body2.preheader:
+; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 [[SIZE]] to i64
 ; CHECK-NEXT:    br label [[FOR_BODY2:%.*]]
 ; CHECK:       for.body2:
 ; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ 1, [[FOR_BODY2_PREHEADER]] ], [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY2]] ]
@@ -180,12 +181,12 @@ define void @loop_2(i32 %size, i32 %nsteps, i32 %hsize, i32* %lined, i8 %tmp1) {
 ; CHECK-NEXT:    [[ADD_PTR:%.*]] = getelementptr inbounds i8, i8* [[BC0]], i64 [[TMP5]]
 ; CHECK-NEXT:    store i8 [[TMP1:%.*]], i8* [[ADD_PTR]], align 1
 ; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
-; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 [[SIZE]] to i64
 ; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp ne i64 [[INDVARS_IV_NEXT]], [[WIDE_TRIP_COUNT]]
 ; CHECK-NEXT:    br i1 [[EXITCOND]], label [[FOR_BODY2]], label [[FOR_BODY3_PREHEADER:%.*]]
 ; CHECK:       for.body3.preheader:
 ; CHECK-NEXT:    [[TMP6:%.*]] = trunc i64 [[TMP4]] to i32
 ; CHECK-NEXT:    [[TMP7:%.*]] = zext i32 [[TMP6]] to i64
+; CHECK-NEXT:    [[WIDE_TRIP_COUNT5:%.*]] = zext i32 [[SIZE]] to i64
 ; CHECK-NEXT:    br label [[FOR_BODY3:%.*]]
 ; CHECK:       for.body3:
 ; CHECK-NEXT:    [[INDVARS_IV2:%.*]] = phi i64 [ 1, [[FOR_BODY3_PREHEADER]] ], [ [[INDVARS_IV_NEXT3:%.*]], [[FOR_BODY3]] ]
@@ -193,7 +194,6 @@ define void @loop_2(i32 %size, i32 %nsteps, i32 %hsize, i32* %lined, i8 %tmp1) {
 ; CHECK-NEXT:    [[ADD_PTR2:%.*]] = getelementptr inbounds i8, i8* [[BC0]], i64 [[TMP8]]
 ; CHECK-NEXT:    store i8 [[TMP1]], i8* [[ADD_PTR2]], align 1
 ; CHECK-NEXT:    [[INDVARS_IV_NEXT3]] = add nuw nsw i64 [[INDVARS_IV2]], 1
-; CHECK-NEXT:    [[WIDE_TRIP_COUNT5:%.*]] = zext i32 [[SIZE]] to i64
 ; CHECK-NEXT:    [[EXITCOND6:%.*]] = icmp ne i64 [[INDVARS_IV_NEXT3]], [[WIDE_TRIP_COUNT5]]
 ; CHECK-NEXT:    br i1 [[EXITCOND6]], label [[FOR_BODY3]], label [[FOR_INC_LOOPEXIT:%.*]]
 ; CHECK:       for.inc.loopexit:

diff --git a/llvm/test/Transforms/IndVarSimplify/lftr-reuse.ll b/llvm/test/Transforms/IndVarSimplify/lftr-reuse.ll
index c380f41..c14b452 100644 (file)
--- a/llvm/test/Transforms/IndVarSimplify/lftr-reuse.ll
+++ b/llvm/test/Transforms/IndVarSimplify/lftr-reuse.ll
@@ -135,6 +135,7 @@ define void @guardedloop([0 x double]* %matrix, [0 x double]* %vector,
 ; CHECK-NEXT:    br i1 [[CMP]], label [[LOOP_PREHEADER:%.*]], label [[RETURN:%.*]]
 ; CHECK:       loop.preheader:
 ; CHECK-NEXT:    [[TMP0:%.*]] = sext i32 [[ILEAD:%.*]] to i64
+; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 [[IROW]] to i64
 ; CHECK-NEXT:    br label [[LOOP:%.*]]
 ; CHECK:       loop:
 ; CHECK-NEXT:    [[INDVARS_IV2:%.*]] = phi i64 [ 0, [[LOOP_PREHEADER]] ], [ [[INDVARS_IV_NEXT3:%.*]], [[LOOP]] ]
@@ -148,7 +149,6 @@ define void @guardedloop([0 x double]* %matrix, [0 x double]* %vector,
 ; CHECK-NEXT:    call void @use(double [[V2]])
 ; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add nsw i64 [[INDVARS_IV]], [[TMP0]]
 ; CHECK-NEXT:    [[INDVARS_IV_NEXT3]] = add nuw nsw i64 [[INDVARS_IV2]], 1
-; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 [[IROW]] to i64
 ; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp ne i64 [[INDVARS_IV_NEXT3]], [[WIDE_TRIP_COUNT]]
 ; CHECK-NEXT:    br i1 [[EXITCOND]], label [[LOOP]], label [[RETURN_LOOPEXIT:%.*]]
 ; CHECK:       return.loopexit:

diff --git a/llvm/test/Transforms/IndVarSimplify/lftr.ll b/llvm/test/Transforms/IndVarSimplify/lftr.ll
index 9f0172f..b72778b 100644 (file)
--- a/llvm/test/Transforms/IndVarSimplify/lftr.ll
+++ b/llvm/test/Transforms/IndVarSimplify/lftr.ll
@@ -429,6 +429,7 @@ define float @wide_trip_count_test2(float* %a,
 ; CHECK-NEXT:    [[CMP5:%.*]] = icmp ugt i32 [[M:%.*]], 500
 ; CHECK-NEXT:    br i1 [[CMP5]], label [[FOR_BODY_PREHEADER:%.*]], label [[FOR_END:%.*]]
 ; CHECK:       for.body.preheader:
+; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 [[M]] to i64
 ; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
 ; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ], [ 500, [[FOR_BODY_PREHEADER]] ]
@@ -440,7 +441,6 @@ define float @wide_trip_count_test2(float* %a,
 ; CHECK-NEXT:    [[MUL:%.*]] = fmul float [[TEMP]], [[TEMP1]]
 ; CHECK-NEXT:    [[ADD]] = fadd float [[SUM_07]], [[MUL]]
 ; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
-; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 [[M]] to i64
 ; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp ne i64 [[INDVARS_IV_NEXT]], [[WIDE_TRIP_COUNT]]
 ; CHECK-NEXT:    br i1 [[EXITCOND]], label [[FOR_BODY]], label [[FOR_END_LOOPEXIT:%.*]]
 ; CHECK:       for.end.loopexit:
@@ -488,6 +488,7 @@ define float @wide_trip_count_test3(float* %b,
 ; CHECK-NEXT:    [[CMP5:%.*]] = icmp sgt i32 [[M:%.*]], -10
 ; CHECK-NEXT:    br i1 [[CMP5]], label [[FOR_BODY_PREHEADER:%.*]], label [[FOR_END:%.*]]
 ; CHECK:       for.body.preheader:
+; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = sext i32 [[M]] to i64
 ; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
 ; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ], [ -10, [[FOR_BODY_PREHEADER]] ]
@@ -500,7 +501,6 @@ define float @wide_trip_count_test3(float* %b,
 ; CHECK-NEXT:    [[MUL:%.*]] = fmul float [[CONV]], [[TEMP]]
 ; CHECK-NEXT:    [[ADD1]] = fadd float [[SUM_07]], [[MUL]]
 ; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add nsw i64 [[INDVARS_IV]], 1
-; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = sext i32 [[M]] to i64
 ; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp ne i64 [[INDVARS_IV_NEXT]], [[WIDE_TRIP_COUNT]]
 ; CHECK-NEXT:    br i1 [[EXITCOND]], label [[FOR_BODY]], label [[FOR_END_LOOPEXIT:%.*]]
 ; CHECK:       for.end.loopexit:
@@ -547,6 +547,7 @@ define float @wide_trip_count_test4(float* %b,
 ; CHECK-NEXT:    [[CMP5:%.*]] = icmp sgt i32 [[M:%.*]], 10
 ; CHECK-NEXT:    br i1 [[CMP5]], label [[FOR_BODY_PREHEADER:%.*]], label [[FOR_END:%.*]]
 ; CHECK:       for.body.preheader:
+; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 [[M]] to i64
 ; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
 ; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ], [ 10, [[FOR_BODY_PREHEADER]] ]
@@ -559,7 +560,6 @@ define float @wide_trip_count_test4(float* %b,
 ; CHECK-NEXT:    [[MUL:%.*]] = fmul float [[CONV]], [[TEMP]]
 ; CHECK-NEXT:    [[ADD1]] = fadd float [[SUM_07]], [[MUL]]
 ; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
-; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 [[M]] to i64
 ; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp ne i64 [[INDVARS_IV_NEXT]], [[WIDE_TRIP_COUNT]]
 ; CHECK-NEXT:    br i1 [[EXITCOND]], label [[FOR_BODY]], label [[FOR_END_LOOPEXIT:%.*]]
 ; CHECK:       for.end.loopexit:

diff --git a/llvm/test/Transforms/IndVarSimplify/ult-sub-to-eq.ll b/llvm/test/Transforms/IndVarSimplify/ult-sub-to-eq.ll
index 0bd5064..3650877 100644 (file)
--- a/llvm/test/Transforms/IndVarSimplify/ult-sub-to-eq.ll
+++ b/llvm/test/Transforms/IndVarSimplify/ult-sub-to-eq.ll
@@ -12,6 +12,7 @@ define void @test1(float* nocapture %autoc, float* nocapture %data, float %d, i3
 ; CHECK-NEXT:    [[CMP4:%.*]] = icmp eq i32 [[DATA_LEN]], [[SAMPLE]]
 ; CHECK-NEXT:    br i1 [[CMP4]], label [[FOR_END:%.*]], label [[FOR_BODY_PREHEADER:%.*]]
 ; CHECK:       for.body.preheader:
+; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 [[SUB]] to i64
 ; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
 ; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ], [ 0, [[FOR_BODY_PREHEADER]] ]
@@ -26,7 +27,6 @@ define void @test1(float* nocapture %autoc, float* nocapture %data, float %d, i3
 ; CHECK-NEXT:    [[ADD3:%.*]] = fadd float [[TMP2]], [[MUL]]
 ; CHECK-NEXT:    store float [[ADD3]], float* [[ARRAYIDX2]], align 4
 ; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
-; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext i32 [[SUB]] to i64
 ; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp ne i64 [[INDVARS_IV_NEXT]], [[WIDE_TRIP_COUNT]]
 ; CHECK-NEXT:    br i1 [[EXITCOND]], label [[FOR_BODY]], label [[FOR_END_LOOPEXIT:%.*]]
 ; CHECK:       for.end.loopexit: