[llvm] r336454 - [InstCombine] add more tests for potentially poisonous shifts; NFC
Sanjay Patel via llvm-commits
llvm-commits at lists.llvm.org
Fri Jul 6 10:44:57 PDT 2018
Author: spatel
Date: Fri Jul 6 10:44:57 2018
New Revision: 336454
URL: http://llvm.org/viewvc/llvm-project?rev=336454&view=rev
Log:
[InstCombine] add more tests for potentially poisonous shifts; NFC
Modified:
llvm/trunk/test/Transforms/InstCombine/shuffle_select.ll
Modified: llvm/trunk/test/Transforms/InstCombine/shuffle_select.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Transforms/InstCombine/shuffle_select.ll?rev=336454&r1=336453&r2=336454&view=diff
==============================================================================
--- llvm/trunk/test/Transforms/InstCombine/shuffle_select.ll (original)
+++ llvm/trunk/test/Transforms/InstCombine/shuffle_select.ll Fri Jul 6 10:44:57 2018
@@ -998,6 +998,49 @@ define <4 x i32> @lshr_2_vars(<4 x i32>
ret <4 x i32> %t3
}
+define <4 x i32> @lshr_2_vars_exact(<4 x i32> %v0, <4 x i32> %v1) {
+; CHECK-LABEL: @lshr_2_vars_exact(
+; CHECK-NEXT: [[TMP1:%.*]] = shufflevector <4 x i32> [[V0:%.*]], <4 x i32> [[V1:%.*]], <4 x i32> <i32 4, i32 5, i32 2, i32 7>
+; CHECK-NEXT: [[T3:%.*]] = lshr exact <4 x i32> <i32 5, i32 6, i32 3, i32 8>, [[TMP1]]
+; CHECK-NEXT: ret <4 x i32> [[T3]]
+;
+ %t1 = lshr exact <4 x i32> <i32 1, i32 2, i32 3, i32 4>, %v0
+ %t2 = lshr exact <4 x i32> <i32 5, i32 6, i32 7, i32 8>, %v1
+ %t3 = shufflevector <4 x i32> %t1, <4 x i32> %t2, <4 x i32> <i32 4, i32 5, i32 2, i32 7>
+ ret <4 x i32> %t3
+}
+
+; FIXME: This is not safe for the same reason that div/rem with undef mask is not safe.
+; Analysis could determine that the shift has an undef shift amount creating poison.
+
+define <4 x i32> @lshr_2_vars_undef_mask_elt(<4 x i32> %v0, <4 x i32> %v1) {
+; CHECK-LABEL: @lshr_2_vars_undef_mask_elt(
+; CHECK-NEXT: [[TMP1:%.*]] = shufflevector <4 x i32> [[V0:%.*]], <4 x i32> [[V1:%.*]], <4 x i32> <i32 undef, i32 5, i32 2, i32 7>
+; CHECK-NEXT: [[T3:%.*]] = lshr <4 x i32> <i32 undef, i32 6, i32 3, i32 8>, [[TMP1]]
+; CHECK-NEXT: ret <4 x i32> [[T3]]
+;
+ %t1 = lshr <4 x i32> <i32 1, i32 2, i32 3, i32 4>, %v0
+ %t2 = lshr <4 x i32> <i32 5, i32 6, i32 7, i32 8>, %v1
+ %t3 = shufflevector <4 x i32> %t1, <4 x i32> %t2, <4 x i32> <i32 undef, i32 5, i32 2, i32 7>
+ ret <4 x i32> %t3
+}
+
+; FIXME: This is not safe for the same reason that div/rem with undef mask is not safe.
+; Analysis could determine that the shift has an undef shift amount creating poison.
+; The 'exact' could also result in poison by choosing a -1 shift value.
+
+define <4 x i32> @lshr_2_vars_exact_undef_mask_elt(<4 x i32> %v0, <4 x i32> %v1) {
+; CHECK-LABEL: @lshr_2_vars_exact_undef_mask_elt(
+; CHECK-NEXT: [[TMP1:%.*]] = shufflevector <4 x i32> [[V0:%.*]], <4 x i32> [[V1:%.*]], <4 x i32> <i32 undef, i32 5, i32 2, i32 7>
+; CHECK-NEXT: [[T3:%.*]] = lshr exact <4 x i32> <i32 undef, i32 6, i32 3, i32 8>, [[TMP1]]
+; CHECK-NEXT: ret <4 x i32> [[T3]]
+;
+ %t1 = lshr exact <4 x i32> <i32 1, i32 2, i32 3, i32 4>, %v0
+ %t2 = lshr exact <4 x i32> <i32 5, i32 6, i32 7, i32 8>, %v1
+ %t3 = shufflevector <4 x i32> %t1, <4 x i32> %t2, <4 x i32> <i32 undef, i32 5, i32 2, i32 7>
+ ret <4 x i32> %t3
+}
+
; Try weird types.
define <3 x i32> @ashr_2_vars(<3 x i32> %v0, <3 x i32> %v1) {
More information about the llvm-commits
mailing list