[llvm] [LoopVectorize] Add test case for minloc reduction (PR #141556)

[Mel Chen via llvm-commits]

================
@@ -0,0 +1,86 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
+; RUN: opt -passes=loop-vectorize -force-vector-width=4 -S %s | FileCheck %s --check-prefix=CHECK-REV-MIN
+
+; This test case is extracted from rnflow (fortran) benchmark in polyhedron benchmark suite.
+; The function minlst primarily takes two indices (i.e. range), scans backwards in the range
+; and returns the firstIV of the minimum value.
+
+define fastcc i32 @minlst(i32 %first_index, i32 %last_index, ptr %array) {
+; CHECK-REV-MIN-LABEL: define fastcc i32 @minlst(
+; CHECK-REV-MIN-SAME: i32 [[FIRST_INDEX:%.*]], i32 [[LAST_INDEX:%.*]], ptr [[ARRAY:%.*]]) {
+; CHECK-REV-MIN-NEXT:  [[ENTRY:.*]]:
+; CHECK-REV-MIN-NEXT:    [[FIRST_INDEX_SEXT:%.*]] = sext i32 [[FIRST_INDEX]] to i64
+; CHECK-REV-MIN-NEXT:    [[LAST_INDEX_NEG:%.*]] = sub i32 0, [[LAST_INDEX]]
+; CHECK-REV-MIN-NEXT:    [[LAST_INDEX_NEG_SEXT:%.*]] = sext i32 [[LAST_INDEX_NEG]] to i64
+; CHECK-REV-MIN-NEXT:    [[ADD:%.*]] = add nsw i64 [[FIRST_INDEX_SEXT]], [[LAST_INDEX_NEG_SEXT]]
+; CHECK-REV-MIN-NEXT:    [[SUB:%.*]] = sub nsw i64 0, [[ADD]]
+; CHECK-REV-MIN-NEXT:    [[INVARIANT_GEP:%.*]] = getelementptr i8, ptr [[ARRAY]], i64 -8
+; CHECK-REV-MIN-NEXT:    [[INVARIANT_GEP5:%.*]] = getelementptr i8, ptr [[ARRAY]], i64 -4
+; CHECK-REV-MIN-NEXT:    [[EARLY_EXIT_COND:%.*]] = icmp slt i64 [[ADD]], 0
+; CHECK-REV-MIN-NEXT:    br i1 [[EARLY_EXIT_COND]], label %[[LOOP_PREHEADER:.*]], [[DOT_CRIT_EDGE:label %.*]]
+; CHECK-REV-MIN:       [[LOOP_PREHEADER]]:
+; CHECK-REV-MIN-NEXT:    [[LAST_INDEX_SEXT:%.*]] = sext i32 [[LAST_INDEX]] to i64
+; CHECK-REV-MIN-NEXT:    br label %[[LOOP:.*]]
+; CHECK-REV-MIN:       [[LOOP]]:
+; CHECK-REV-MIN-NEXT:    [[IV:%.*]] = phi i64 [ [[LAST_INDEX_SEXT]], %[[LOOP_PREHEADER]] ], [ [[IV_NEXT:%.*]], %[[LOOP]] ]
+; CHECK-REV-MIN-NEXT:    [[TMP0:%.*]] = phi i64 [ [[TMP1:%.*]], %[[LOOP]] ], [ [[SUB]], %[[LOOP_PREHEADER]] ]
+; CHECK-REV-MIN-NEXT:    [[INDEX:%.*]] = phi i32 [ [[SELECT:%.*]], %[[LOOP]] ], [ [[LAST_INDEX]], %[[LOOP_PREHEADER]] ]
+; CHECK-REV-MIN-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], -1
+; CHECK-REV-MIN-NEXT:    [[GEP:%.*]] = getelementptr float, ptr [[INVARIANT_GEP]], i64 [[IV]]
+; CHECK-REV-MIN-NEXT:    [[LOAD1:%.*]] = load float, ptr [[GEP]], align 4
+; CHECK-REV-MIN-NEXT:    [[INDEX_SEXT:%.*]] = sext i32 [[INDEX]] to i64
+; CHECK-REV-MIN-NEXT:    [[GEP6:%.*]] = getelementptr float, ptr [[INVARIANT_GEP5]], i64 [[INDEX_SEXT]]
+; CHECK-REV-MIN-NEXT:    [[LOAD2:%.*]] = load float, ptr [[GEP6]], align 4
+; CHECK-REV-MIN-NEXT:    [[CMP:%.*]] = fcmp contract olt float [[LOAD1]], [[LOAD2]]
+; CHECK-REV-MIN-NEXT:    [[IV_NEXT_TRUNC:%.*]] = trunc nsw i64 [[IV_NEXT]] to i32
+; CHECK-REV-MIN-NEXT:    [[SELECT]] = select i1 [[CMP]], i32 [[IV_NEXT_TRUNC]], i32 [[INDEX]]
+; CHECK-REV-MIN-NEXT:    [[TMP1]] = add nsw i64 [[TMP0]], -1
+; CHECK-REV-MIN-NEXT:    [[LOOP_COND:%.*]] = icmp sgt i64 [[TMP0]], 1
+; CHECK-REV-MIN-NEXT:    br i1 [[LOOP_COND]], label %[[LOOP]], label %[[DOT_CRIT_EDGE_LOOPEXIT:.*]]
+; CHECK-REV-MIN:       [[__CRIT_EDGE_LOOPEXIT:.*:]]
+; CHECK-REV-MIN-NEXT:    [[SELECT_LCSSA:%.*]] = phi i32 [ [[SELECT]], %[[LOOP]] ]
+; CHECK-REV-MIN-NEXT:    br [[DOT_CRIT_EDGE]]
+; CHECK-REV-MIN:       [[__CRIT_EDGE:.*:]]
+; CHECK-REV-MIN-NEXT:    [[LAST_INDEX_RET:%.*]] = phi i32 [ [[LAST_INDEX]], %[[ENTRY]] ], [ [[SELECT_LCSSA]], %[[DOT_CRIT_EDGE_LOOPEXIT]] ]
+; CHECK-REV-MIN-NEXT:    ret i32 [[LAST_INDEX_RET]]
+;
+entry:
+  %first_index_sext = sext i32 %first_index to i64
+  %last_index_neg = sub i32 0, %last_index
+  %last_index_neg_sext = sext i32 %last_index_neg to i64
+  %add = add nsw i64 %first_index_sext, %last_index_neg_sext
+  %diff = sub nsw i64 0, %add
+  %first_ptr = getelementptr i8, ptr %array, i64 -8
+  %second_ptr = getelementptr i8, ptr %array, i64 -4
+  %early_exit_cond = icmp slt i64 %add, 0
+  br i1 %early_exit_cond, label %loop.preheader, label %._crit_edge
+
+loop.preheader:                                 ; preds = %entry
+  %last_index_sext = sext i32 %last_index to i64
+  br label %loop
+
+loop:                                           ; preds = %loop.preheader, %loop
+  %iv = phi i64 [ %last_index_sext, %loop.preheader ], [ %iv.next, %loop ]
+  %dec_iv = phi i64 [ %dec, %loop ], [ %diff, %loop.preheader ]
+  %index = phi i32 [ %select, %loop ], [ %last_index, %loop.preheader ]
+  %iv.next = add nsw i64 %iv, -1
+  %load1_ptr = getelementptr float, ptr %first_ptr, i64 %iv
+  %load1 = load float, ptr %load1_ptr, align 4
+  %index_sext = sext i32 %index to i64
+  %load2_ptr = getelementptr float, ptr %second_ptr, i64 %index_sext
+  %load2 = load float, ptr %load2_ptr, align 4
+  %cmp = fcmp contract olt float %load1, %load2
+  %iv.next.trunc = trunc nsw i64 %iv.next to i32
+  %select = select i1 %cmp, i32 %iv.next.trunc, i32 %index
+  %dec = add nsw i64 %dec_iv, -1
+  %loop_cond = icmp sgt i64 %dec_iv, 1
+  br i1 %loop_cond, label %loop, label %._crit_edge.loopexit
+
+._crit_edge.loopexit:                             ; preds = %loop
+  %select.lcssa = phi i32 [ %select, %loop ]
+  br label %._crit_edge
+
+._crit_edge:                                      ; preds = %._crit_edge.loopexit, %0
+  %last_index_ret = phi i32 [ %last_index, %entry ], [ %select.lcssa, %._crit_edge.loopexit ]
+  ret i32 %last_index_ret
+}
----------------
Mel-Chen wrote:

Merge ._crit_edge.loopexit and ._crit_edge into one basic block.

https://github.com/llvm/llvm-project/pull/141556