Skip to content

[LV][EVL] Pre-commit test case for fixed-order recurrence with EVL tail folding. (NFC) #122456

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 1 commit into from
Jan 13, 2025
Merged

[LV][EVL] Pre-commit test case for fixed-order recurrence with EVL tail folding. (NFC) #122456

merged 1 commit into from
Jan 13, 2025

Conversation

Mel-Chen
Copy link
Contributor

@Mel-Chen Mel-Chen commented Jan 10, 2025 •
edited
Loading

This test case is from SingleSource/UnitTests/Vectorizer/recurrences.test.
Pre-commit for #122458

@llvmbot
Copy link
Member

llvmbot commented Jan 10, 2025

@llvm/pr-subscribers-llvm-transforms

Author: Mel Chen (Mel-Chen)

Changes

This test case is from SingleSource/UnitTests/Vectorizer/recurrences.test.

Patch is 38.71 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/122456.diff

1 Files Affected:

  • (added) llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-fixed-order-recurrence.ll (+549)
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-fixed-order-recurrence.ll b/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-fixed-order-recurrence.ll
new file mode 100644
index 00000000000000..9f8cf169c05933
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-fixed-order-recurrence.ll
@@ -0,0 +1,549 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
+; RUN: opt -passes=loop-vectorize \
+; RUN: -prefer-inloop-reductions \
+; RUN: -force-tail-folding-style=data-with-evl \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -mtriple=riscv64 -mattr=+v,+f -S < %s| FileCheck %s --check-prefix=IF-EVL
+
+; RUN: opt -passes=loop-vectorize \
+; RUN: -prefer-inloop-reductions \
+; RUN: -force-tail-folding-style=none \
+; RUN: -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue \
+; RUN: -mtriple=riscv64 -mattr=+v,+f -S < %s| FileCheck %s --check-prefix=NO-VP
+
+define void @first_order_recurrence(ptr noalias %A, ptr noalias %B, i64 %TC) {
+; IF-EVL-LABEL: define void @first_order_recurrence(
+; IF-EVL-SAME: ptr noalias [[A:%.*]], ptr noalias [[B:%.*]], i64 [[TC:%.*]]) #[[ATTR0:[0-9]+]] {
+; IF-EVL-NEXT:  [[ENTRY:.*]]:
+; IF-EVL-NEXT:    [[TMP0:%.*]] = sub i64 -1, [[TC]]
+; IF-EVL-NEXT:    [[TMP1:%.*]] = call i64 @llvm.vscale.i64()
+; IF-EVL-NEXT:    [[TMP2:%.*]] = mul i64 [[TMP1]], 4
+; IF-EVL-NEXT:    [[TMP3:%.*]] = icmp ult i64 [[TMP0]], [[TMP2]]
+; IF-EVL-NEXT:    br i1 [[TMP3]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
+; IF-EVL:       [[VECTOR_PH]]:
+; IF-EVL-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
+; IF-EVL-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 4
+; IF-EVL-NEXT:    [[TMP6:%.*]] = sub i64 [[TMP5]], 1
+; IF-EVL-NEXT:    [[N_RND_UP:%.*]] = add i64 [[TC]], [[TMP6]]
+; IF-EVL-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP5]]
+; IF-EVL-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
+; IF-EVL-NEXT:    [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
+; IF-EVL-NEXT:    [[TMP8:%.*]] = mul i64 [[TMP7]], 4
+; IF-EVL-NEXT:    [[TMP9:%.*]] = call i32 @llvm.vscale.i32()
+; IF-EVL-NEXT:    [[TMP10:%.*]] = mul i32 [[TMP9]], 4
+; IF-EVL-NEXT:    [[TMP11:%.*]] = sub i32 [[TMP10]], 1
+; IF-EVL-NEXT:    [[VECTOR_RECUR_INIT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 33, i32 [[TMP11]]
+; IF-EVL-NEXT:    br label %[[VECTOR_BODY:.*]]
+; IF-EVL:       [[VECTOR_BODY]]:
+; IF-EVL-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
+; IF-EVL-NEXT:    [[EVL_BASED_IV:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_EVL_NEXT:%.*]], %[[VECTOR_BODY]] ]
+; IF-EVL-NEXT:    [[VECTOR_RECUR:%.*]] = phi <vscale x 4 x i32> [ [[VECTOR_RECUR_INIT]], %[[VECTOR_PH]] ], [ [[VP_OP_LOAD:%.*]], %[[VECTOR_BODY]] ]
+; IF-EVL-NEXT:    [[AVL:%.*]] = sub i64 [[TC]], [[EVL_BASED_IV]]
+; IF-EVL-NEXT:    [[TMP12:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 4, i1 true)
+; IF-EVL-NEXT:    [[TMP13:%.*]] = add i64 [[EVL_BASED_IV]], 0
+; IF-EVL-NEXT:    [[TMP14:%.*]] = getelementptr inbounds nuw i32, ptr [[A]], i64 [[TMP13]]
+; IF-EVL-NEXT:    [[TMP15:%.*]] = getelementptr inbounds nuw i32, ptr [[TMP14]], i32 0
+; IF-EVL-NEXT:    [[VP_OP_LOAD]] = call <vscale x 4 x i32> @llvm.vp.load.nxv4i32.p0(ptr align 4 [[TMP15]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP12]])
+; IF-EVL-NEXT:    [[TMP16:%.*]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> [[VECTOR_RECUR]], <vscale x 4 x i32> [[VP_OP_LOAD]], i32 -1)
+; IF-EVL-NEXT:    [[VP_OP:%.*]] = call <vscale x 4 x i32> @llvm.vp.add.nxv4i32(<vscale x 4 x i32> [[TMP16]], <vscale x 4 x i32> [[VP_OP_LOAD]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP12]])
+; IF-EVL-NEXT:    [[TMP17:%.*]] = getelementptr inbounds nuw i32, ptr [[B]], i64 [[TMP13]]
+; IF-EVL-NEXT:    [[TMP18:%.*]] = getelementptr inbounds nuw i32, ptr [[TMP17]], i32 0
+; IF-EVL-NEXT:    call void @llvm.vp.store.nxv4i32.p0(<vscale x 4 x i32> [[VP_OP]], ptr align 4 [[TMP18]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP12]])
+; IF-EVL-NEXT:    [[TMP19:%.*]] = zext i32 [[TMP12]] to i64
+; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add i64 [[TMP19]], [[EVL_BASED_IV]]
+; IF-EVL-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP8]]
+; IF-EVL-NEXT:    [[TMP20:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; IF-EVL-NEXT:    br i1 [[TMP20]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; IF-EVL:       [[MIDDLE_BLOCK]]:
+; IF-EVL-NEXT:    [[TMP21:%.*]] = call i32 @llvm.vscale.i32()
+; IF-EVL-NEXT:    [[TMP22:%.*]] = mul i32 [[TMP21]], 4
+; IF-EVL-NEXT:    [[TMP23:%.*]] = sub i32 [[TMP22]], 1
+; IF-EVL-NEXT:    [[VECTOR_RECUR_EXTRACT:%.*]] = extractelement <vscale x 4 x i32> [[VP_OP_LOAD]], i32 [[TMP23]]
+; IF-EVL-NEXT:    br i1 true, label %[[FOR_END:.*]], label %[[SCALAR_PH]]
+; IF-EVL:       [[SCALAR_PH]]:
+; IF-EVL-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ]
+; IF-EVL-NEXT:    [[SCALAR_RECUR_INIT:%.*]] = phi i32 [ [[VECTOR_RECUR_EXTRACT]], %[[MIDDLE_BLOCK]] ], [ 33, %[[ENTRY]] ]
+; IF-EVL-NEXT:    br label %[[FOR_BODY:.*]]
+; IF-EVL:       [[FOR_BODY]]:
+; IF-EVL-NEXT:    [[INDVARS:%.*]] = phi i64 [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ], [ [[INDVARS_NEXT:%.*]], %[[FOR_BODY]] ]
+; IF-EVL-NEXT:    [[FOR1:%.*]] = phi i32 [ [[SCALAR_RECUR_INIT]], %[[SCALAR_PH]] ], [ [[TMP24:%.*]], %[[FOR_BODY]] ]
+; IF-EVL-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds nuw i32, ptr [[A]], i64 [[INDVARS]]
+; IF-EVL-NEXT:    [[TMP24]] = load i32, ptr [[ARRAYIDX]], align 4
+; IF-EVL-NEXT:    [[ADD:%.*]] = add nsw i32 [[FOR1]], [[TMP24]]
+; IF-EVL-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds nuw i32, ptr [[B]], i64 [[INDVARS]]
+; IF-EVL-NEXT:    store i32 [[ADD]], ptr [[ARRAYIDX2]], align 4
+; IF-EVL-NEXT:    [[INDVARS_NEXT]] = add nuw nsw i64 [[INDVARS]], 1
+; IF-EVL-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i64 [[INDVARS_NEXT]], [[TC]]
+; IF-EVL-NEXT:    br i1 [[EXITCOND_NOT]], label %[[FOR_END]], label %[[FOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
+; IF-EVL:       [[FOR_END]]:
+; IF-EVL-NEXT:    ret void
+;
+; NO-VP-LABEL: define void @first_order_recurrence(
+; NO-VP-SAME: ptr noalias [[A:%.*]], ptr noalias [[B:%.*]], i64 [[TC:%.*]]) #[[ATTR0:[0-9]+]] {
+; NO-VP-NEXT:  [[ENTRY:.*]]:
+; NO-VP-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 4
+; NO-VP-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TC]], [[TMP1]]
+; NO-VP-NEXT:    br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
+; NO-VP:       [[VECTOR_PH]]:
+; NO-VP-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 4
+; NO-VP-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TC]], [[TMP3]]
+; NO-VP-NEXT:    [[N_VEC:%.*]] = sub i64 [[TC]], [[N_MOD_VF]]
+; NO-VP-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 4
+; NO-VP-NEXT:    [[TMP6:%.*]] = call i32 @llvm.vscale.i32()
+; NO-VP-NEXT:    [[TMP7:%.*]] = mul i32 [[TMP6]], 4
+; NO-VP-NEXT:    [[TMP8:%.*]] = sub i32 [[TMP7]], 1
+; NO-VP-NEXT:    [[VECTOR_RECUR_INIT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 33, i32 [[TMP8]]
+; NO-VP-NEXT:    br label %[[VECTOR_BODY:.*]]
+; NO-VP:       [[VECTOR_BODY]]:
+; NO-VP-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
+; NO-VP-NEXT:    [[VECTOR_RECUR:%.*]] = phi <vscale x 4 x i32> [ [[VECTOR_RECUR_INIT]], %[[VECTOR_PH]] ], [ [[WIDE_LOAD:%.*]], %[[VECTOR_BODY]] ]
+; NO-VP-NEXT:    [[TMP9:%.*]] = add i64 [[INDEX]], 0
+; NO-VP-NEXT:    [[TMP10:%.*]] = getelementptr inbounds nuw i32, ptr [[A]], i64 [[TMP9]]
+; NO-VP-NEXT:    [[TMP11:%.*]] = getelementptr inbounds nuw i32, ptr [[TMP10]], i32 0
+; NO-VP-NEXT:    [[WIDE_LOAD]] = load <vscale x 4 x i32>, ptr [[TMP11]], align 4
+; NO-VP-NEXT:    [[TMP12:%.*]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> [[VECTOR_RECUR]], <vscale x 4 x i32> [[WIDE_LOAD]], i32 -1)
+; NO-VP-NEXT:    [[TMP13:%.*]] = add nsw <vscale x 4 x i32> [[TMP12]], [[WIDE_LOAD]]
+; NO-VP-NEXT:    [[TMP14:%.*]] = getelementptr inbounds nuw i32, ptr [[B]], i64 [[TMP9]]
+; NO-VP-NEXT:    [[TMP15:%.*]] = getelementptr inbounds nuw i32, ptr [[TMP14]], i32 0
+; NO-VP-NEXT:    store <vscale x 4 x i32> [[TMP13]], ptr [[TMP15]], align 4
+; NO-VP-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
+; NO-VP-NEXT:    [[TMP16:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; NO-VP-NEXT:    br i1 [[TMP16]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; NO-VP:       [[MIDDLE_BLOCK]]:
+; NO-VP-NEXT:    [[TMP17:%.*]] = call i32 @llvm.vscale.i32()
+; NO-VP-NEXT:    [[TMP18:%.*]] = mul i32 [[TMP17]], 4
+; NO-VP-NEXT:    [[TMP19:%.*]] = sub i32 [[TMP18]], 1
+; NO-VP-NEXT:    [[VECTOR_RECUR_EXTRACT:%.*]] = extractelement <vscale x 4 x i32> [[WIDE_LOAD]], i32 [[TMP19]]
+; NO-VP-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TC]], [[N_VEC]]
+; NO-VP-NEXT:    br i1 [[CMP_N]], label %[[FOR_END:.*]], label %[[SCALAR_PH]]
+; NO-VP:       [[SCALAR_PH]]:
+; NO-VP-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ]
+; NO-VP-NEXT:    [[SCALAR_RECUR_INIT:%.*]] = phi i32 [ [[VECTOR_RECUR_EXTRACT]], %[[MIDDLE_BLOCK]] ], [ 33, %[[ENTRY]] ]
+; NO-VP-NEXT:    br label %[[FOR_BODY:.*]]
+; NO-VP:       [[FOR_BODY]]:
+; NO-VP-NEXT:    [[INDVARS:%.*]] = phi i64 [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ], [ [[INDVARS_NEXT:%.*]], %[[FOR_BODY]] ]
+; NO-VP-NEXT:    [[FOR1:%.*]] = phi i32 [ [[SCALAR_RECUR_INIT]], %[[SCALAR_PH]] ], [ [[TMP20:%.*]], %[[FOR_BODY]] ]
+; NO-VP-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds nuw i32, ptr [[A]], i64 [[INDVARS]]
+; NO-VP-NEXT:    [[TMP20]] = load i32, ptr [[ARRAYIDX]], align 4
+; NO-VP-NEXT:    [[ADD:%.*]] = add nsw i32 [[FOR1]], [[TMP20]]
+; NO-VP-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds nuw i32, ptr [[B]], i64 [[INDVARS]]
+; NO-VP-NEXT:    store i32 [[ADD]], ptr [[ARRAYIDX2]], align 4
+; NO-VP-NEXT:    [[INDVARS_NEXT]] = add nuw nsw i64 [[INDVARS]], 1
+; NO-VP-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i64 [[INDVARS_NEXT]], [[TC]]
+; NO-VP-NEXT:    br i1 [[EXITCOND_NOT]], label %[[FOR_END]], label %[[FOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
+; NO-VP:       [[FOR_END]]:
+; NO-VP-NEXT:    ret void
+;
+entry:
+  br label %for.body
+
+for.body:
+  %indvars = phi i64 [ 0, %entry ], [ %indvars.next, %for.body ]
+  %for1 = phi i32 [ 33, %entry ], [ %0, %for.body ]
+  %arrayidx = getelementptr inbounds nuw i32, ptr %A, i64 %indvars
+  %0 = load i32, ptr %arrayidx, align 4
+  %add = add nsw i32 %for1, %0
+  %arrayidx2 = getelementptr inbounds nuw i32, ptr %B, i64 %indvars
+  store i32 %add, ptr %arrayidx2, align 4
+  %indvars.next = add nuw nsw i64 %indvars, 1
+  %exitcond.not = icmp eq i64 %indvars.next, %TC
+  br i1 %exitcond.not, label %for.end, label %for.body, !llvm.loop !0
+
+for.end:
+  ret void
+}
+
+define void @second_order_recurrence(ptr noalias %A, ptr noalias %B, i64 %TC) {
+; IF-EVL-LABEL: define void @second_order_recurrence(
+; IF-EVL-SAME: ptr noalias [[A:%.*]], ptr noalias [[B:%.*]], i64 [[TC:%.*]]) #[[ATTR0]] {
+; IF-EVL-NEXT:  [[ENTRY:.*]]:
+; IF-EVL-NEXT:    [[TMP0:%.*]] = sub i64 -1, [[TC]]
+; IF-EVL-NEXT:    [[TMP1:%.*]] = call i64 @llvm.vscale.i64()
+; IF-EVL-NEXT:    [[TMP2:%.*]] = mul i64 [[TMP1]], 4
+; IF-EVL-NEXT:    [[TMP3:%.*]] = icmp ult i64 [[TMP0]], [[TMP2]]
+; IF-EVL-NEXT:    br i1 [[TMP3]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
+; IF-EVL:       [[VECTOR_PH]]:
+; IF-EVL-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
+; IF-EVL-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 4
+; IF-EVL-NEXT:    [[TMP6:%.*]] = sub i64 [[TMP5]], 1
+; IF-EVL-NEXT:    [[N_RND_UP:%.*]] = add i64 [[TC]], [[TMP6]]
+; IF-EVL-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP5]]
+; IF-EVL-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
+; IF-EVL-NEXT:    [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
+; IF-EVL-NEXT:    [[TMP8:%.*]] = mul i64 [[TMP7]], 4
+; IF-EVL-NEXT:    [[TMP9:%.*]] = call i32 @llvm.vscale.i32()
+; IF-EVL-NEXT:    [[TMP10:%.*]] = mul i32 [[TMP9]], 4
+; IF-EVL-NEXT:    [[TMP11:%.*]] = sub i32 [[TMP10]], 1
+; IF-EVL-NEXT:    [[VECTOR_RECUR_INIT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 33, i32 [[TMP11]]
+; IF-EVL-NEXT:    [[TMP12:%.*]] = call i32 @llvm.vscale.i32()
+; IF-EVL-NEXT:    [[TMP13:%.*]] = mul i32 [[TMP12]], 4
+; IF-EVL-NEXT:    [[TMP14:%.*]] = sub i32 [[TMP13]], 1
+; IF-EVL-NEXT:    [[VECTOR_RECUR_INIT1:%.*]] = insertelement <vscale x 4 x i32> poison, i32 22, i32 [[TMP14]]
+; IF-EVL-NEXT:    br label %[[VECTOR_BODY:.*]]
+; IF-EVL:       [[VECTOR_BODY]]:
+; IF-EVL-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
+; IF-EVL-NEXT:    [[EVL_BASED_IV:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_EVL_NEXT:%.*]], %[[VECTOR_BODY]] ]
+; IF-EVL-NEXT:    [[VECTOR_RECUR:%.*]] = phi <vscale x 4 x i32> [ [[VECTOR_RECUR_INIT]], %[[VECTOR_PH]] ], [ [[VP_OP_LOAD:%.*]], %[[VECTOR_BODY]] ]
+; IF-EVL-NEXT:    [[VECTOR_RECUR2:%.*]] = phi <vscale x 4 x i32> [ [[VECTOR_RECUR_INIT1]], %[[VECTOR_PH]] ], [ [[TMP19:%.*]], %[[VECTOR_BODY]] ]
+; IF-EVL-NEXT:    [[AVL:%.*]] = sub i64 [[TC]], [[EVL_BASED_IV]]
+; IF-EVL-NEXT:    [[TMP15:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 4, i1 true)
+; IF-EVL-NEXT:    [[TMP16:%.*]] = add i64 [[EVL_BASED_IV]], 0
+; IF-EVL-NEXT:    [[TMP17:%.*]] = getelementptr inbounds nuw i32, ptr [[A]], i64 [[TMP16]]
+; IF-EVL-NEXT:    [[TMP18:%.*]] = getelementptr inbounds nuw i32, ptr [[TMP17]], i32 0
+; IF-EVL-NEXT:    [[VP_OP_LOAD]] = call <vscale x 4 x i32> @llvm.vp.load.nxv4i32.p0(ptr align 4 [[TMP18]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP15]])
+; IF-EVL-NEXT:    [[TMP19]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> [[VECTOR_RECUR]], <vscale x 4 x i32> [[VP_OP_LOAD]], i32 -1)
+; IF-EVL-NEXT:    [[TMP20:%.*]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> [[VECTOR_RECUR2]], <vscale x 4 x i32> [[TMP19]], i32 -1)
+; IF-EVL-NEXT:    [[VP_OP:%.*]] = call <vscale x 4 x i32> @llvm.vp.add.nxv4i32(<vscale x 4 x i32> [[TMP19]], <vscale x 4 x i32> [[TMP20]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP15]])
+; IF-EVL-NEXT:    [[TMP21:%.*]] = getelementptr inbounds nuw i32, ptr [[B]], i64 [[TMP16]]
+; IF-EVL-NEXT:    [[TMP22:%.*]] = getelementptr inbounds nuw i32, ptr [[TMP21]], i32 0
+; IF-EVL-NEXT:    call void @llvm.vp.store.nxv4i32.p0(<vscale x 4 x i32> [[VP_OP]], ptr align 4 [[TMP22]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP15]])
+; IF-EVL-NEXT:    [[TMP23:%.*]] = zext i32 [[TMP15]] to i64
+; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add i64 [[TMP23]], [[EVL_BASED_IV]]
+; IF-EVL-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP8]]
+; IF-EVL-NEXT:    [[TMP24:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; IF-EVL-NEXT:    br i1 [[TMP24]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
+; IF-EVL:       [[MIDDLE_BLOCK]]:
+; IF-EVL-NEXT:    [[TMP25:%.*]] = call i32 @llvm.vscale.i32()
+; IF-EVL-NEXT:    [[TMP26:%.*]] = mul i32 [[TMP25]], 4
+; IF-EVL-NEXT:    [[TMP27:%.*]] = sub i32 [[TMP26]], 1
+; IF-EVL-NEXT:    [[VECTOR_RECUR_EXTRACT:%.*]] = extractelement <vscale x 4 x i32> [[VP_OP_LOAD]], i32 [[TMP27]]
+; IF-EVL-NEXT:    [[TMP28:%.*]] = call i32 @llvm.vscale.i32()
+; IF-EVL-NEXT:    [[TMP29:%.*]] = mul i32 [[TMP28]], 4
+; IF-EVL-NEXT:    [[TMP30:%.*]] = sub i32 [[TMP29]], 1
+; IF-EVL-NEXT:    [[VECTOR_RECUR_EXTRACT3:%.*]] = extractelement <vscale x 4 x i32> [[TMP19]], i32 [[TMP30]]
+; IF-EVL-NEXT:    br i1 true, label %[[FOR_END:.*]], label %[[SCALAR_PH]]
+; IF-EVL:       [[SCALAR_PH]]:
+; IF-EVL-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ]
+; IF-EVL-NEXT:    [[SCALAR_RECUR_INIT:%.*]] = phi i32 [ [[VECTOR_RECUR_EXTRACT]], %[[MIDDLE_BLOCK]] ], [ 33, %[[ENTRY]] ]
+; IF-EVL-NEXT:    [[SCALAR_RECUR_INIT4:%.*]] = phi i32 [ [[VECTOR_RECUR_EXTRACT3]], %[[MIDDLE_BLOCK]] ], [ 22, %[[ENTRY]] ]
+; IF-EVL-NEXT:    br label %[[FOR_BODY:.*]]
+; IF-EVL:       [[FOR_BODY]]:
+; IF-EVL-NEXT:    [[INDVARS:%.*]] = phi i64 [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ], [ [[INDVARS_NEXT:%.*]], %[[FOR_BODY]] ]
+; IF-EVL-NEXT:    [[FOR1:%.*]] = phi i32 [ [[SCALAR_RECUR_INIT]], %[[SCALAR_PH]] ], [ [[TMP31:%.*]], %[[FOR_BODY]] ]
+; IF-EVL-NEXT:    [[FOR2:%.*]] = phi i32 [ [[SCALAR_RECUR_INIT4]], %[[SCALAR_PH]] ], [ [[FOR1]], %[[FOR_BODY]] ]
+; IF-EVL-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds nuw i32, ptr [[A]], i64 [[INDVARS]]
+; IF-EVL-NEXT:    [[TMP31]] = load i32, ptr [[ARRAYIDX]], align 4
+; IF-EVL-NEXT:    [[ADD:%.*]] = add nsw i32 [[FOR1]], [[FOR2]]
+; IF-EVL-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds nuw i32, ptr [[B]], i64 [[INDVARS]]
+; IF-EVL-NEXT:    store i32 [[ADD]], ptr [[ARRAYIDX2]], align 4
+; IF-EVL-NEXT:    [[INDVARS_NEXT]] = add nuw nsw i64 [[INDVARS]], 1
+; IF-EVL-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i64 [[INDVARS_NEXT]], [[TC]]
+; IF-EVL-NEXT:    br i1 [[EXITCOND_NOT]], label %[[FOR_END]], label %[[FOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
+; IF-EVL:       [[FOR_END]]:
+; IF-EVL-NEXT:    ret void
+;
+; NO-VP-LABEL: define void @second_order_recurrence(
+; NO-VP-SAME: ptr noalias [[A:%.*]], ptr noalias [[B:%.*]], i64 [[TC:%.*]]) #[[ATTR0]] {
+; NO-VP-NEXT:  [[ENTRY:.*]]:
+; NO-VP-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 4
+; NO-VP-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TC]], [[TMP1]]
+; NO-VP-NEXT:    br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
+; NO-VP:       [[VECTOR_PH]]:
+; NO-VP-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 4
+; NO-VP-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TC]], [[TMP3]]
+; NO-VP-NEXT:    [[N_VEC:%.*]] = sub i64 [[TC]], [[N_MOD_VF]]
+; NO-VP-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 4
+; NO-VP-NEXT:    [[TMP6:%.*]] = call i32 @llvm.vscale.i32()
+; NO-VP-NEXT:    [[TMP7:%.*]] = mul i32 [[TMP6]], 4
+; NO-VP-NEXT:    [[TMP8:%.*]] = sub i32 [[TMP7]], 1
+; NO-VP-NEXT:    [[VECTOR_RECUR_INIT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 33, i32 [[TMP8]]
+; NO-VP-NEXT:    [[TMP9:%.*]] = call i32 @llvm.vscale.i32()
+; NO-VP-NEXT:    [[TMP10:%.*]] = mul i32 [[TMP9]], 4
+; NO-VP-NEXT:    [[TMP11:%.*]] = sub i32 [[TMP10]], 1
+; NO-VP-NEXT:    [[VECTOR_RECUR_INIT1:%.*]] = insertelement <vscale x 4 x i32> poison, i32 22, i32 [[TMP11]]
+; NO-VP-NEXT:    br label %[[VECTOR_BODY:.*]]
+; NO-VP:       [[VECTOR_BODY]]:
+; NO-VP-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
+; NO-VP-NEXT:    [[VECTOR_RECUR:%.*]] = phi <vscale x 4 x i32> [ [[VECTOR_RECUR_INIT]], %[[VECTOR_PH]] ], [ [[WIDE_LOAD:%.*]], %[[VECTOR_BODY]] ]
+; NO-VP-NEXT:    [[VECTOR_RECUR2:%.*]] = phi <vscale x 4 x i32> [ [[VECTOR_RECUR_INIT1]], %[[VECTOR_PH]] ], [ [[TMP15:%.*]], %[[VECTOR_BODY]] ]
+; NO-VP-NEXT:    [[TMP12:%.*]] = add i64 [[INDEX]], 0
+; NO-VP-NEXT:    [[TMP13:%.*]] = getelementptr inbounds nuw i32, ptr [[A]], i64 [[TMP12]]
+; NO-VP-NEXT:    [[TMP14:%.*]] = getelementptr inbounds nuw i32, ptr [[TMP13]], i32 0
+; NO-VP-NEXT:    [[WIDE_LOAD]] = load <vscale x 4 x i32>, ptr [[TMP14]], align 4
+; NO-VP-NEXT:    [[TMP15]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> [[VECTOR_RECUR]], <vscale x 4 x i32> [[WIDE_LOAD]], i32 -1)
+; NO-VP-NEXT:    [[TMP16:%.*]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> [[VECTOR_RECUR2]], <vscale x 4 x i32> [[TMP15]], i32 -1)
+; NO-VP-NEXT:    [[TMP17:%.*]] = add nsw <vscale x 4 x i32> [[TMP15]], [[TMP16]]
+; NO-VP-NEXT:    [[TMP18:%.*]] = getelementptr inbounds nuw i32, ptr [[B]], i64 [[TMP12]]
+; NO-VP-NEXT:    [[TMP19:%.*]] = getelementptr inbounds nuw i32, ptr [[TMP18]], i32 0
+; NO-VP-NEXT:    store <vscale x 4 x i32> [[TMP17]], ptr [[TMP19]], align 4
+; NO-VP-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
+; NO-VP-NEXT:    [[TMP20:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; NO-VP-NEXT:    br i1 [[TMP20]], label %[[MIDDLE_BLOCK:.*]], la...
[truncated]

alexey-bataev
alexey-bataev approved these changes Jan 10, 2025
View reviewed changes
Copy link
Member

@alexey-bataev alexey-bataev left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LG

@Mel-Chen Mel-Chen merged commit 418f5cd into llvm:main Jan 13, 2025
10 checks passed
@Mel-Chen Mel-Chen deleted the disable-fixed-order-recurrence-pre-commit branch January 13, 2025 13:15
kazutakahirata pushed a commit to kazutakahirata/llvm-project that referenced this pull request Jan 13, 2025
@Mel-Chen @kazutakahirata
[LV][EVL] Pre-commit test case for fixed-order recurrence with EVL ta…
ef85aa9 
…il folding. (NFC) (llvm#122456)

This test case is from
SingleSource/UnitTests/Vectorizer/recurrences.test.
Pre-commit for llvm#122458
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@alexey-bataev alexey-bataev alexey-bataev approved these changes

@fhahn fhahn Awaiting requested review from fhahn

@lukel97 lukel97 Awaiting requested review from lukel97

@arcbbb arcbbb Awaiting requested review from arcbbb

@LiqinWeng LiqinWeng Awaiting requested review from LiqinWeng

@ayalz ayalz Awaiting requested review from ayalz

Assignees
No one assigned
Labels
llvm:transforms
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@Mel-Chen @llvmbot @alexey-bataev