Subzero: Elide checks of known valid accesses of locals

tests_lit/asan_tests/instrumentstore.ll

Index: tests_lit/asan_tests/instrumentstore.ll
diff --git a/tests_lit/asan_tests/instrumentstore.ll b/tests_lit/asan_tests/instrumentstore.ll
index b9ac8e5a206b17215afb37000c42a96d47cd2801..44013158891433a0d024608862d114ecc41a069c 100644
--- a/tests_lit/asan_tests/instrumentstore.ll
+++ b/tests_lit/asan_tests/instrumentstore.ll
@@ -13,12 +13,13 @@
 @destGlobal128 = internal global [16 x i8] zeroinitializer
 
 ; A function with a local variable that does the stores
-define internal void @doStores(<4 x i32> %vecSrc) {
-  %destLocal8 = alloca i8, i32 1, align 4
-  %destLocal16 = alloca i8, i32 2, align 4
-  %destLocal32 = alloca i8, i32 4, align 4
-  %destLocal64 = alloca i8, i32 8, align 4
-  %destLocal128 = alloca i8, i32 16, align 4
+define internal void @doStores(<4 x i32> %vecSrc, i32 %arg8, i32 %arg16,
+                               i32 %arg32, i32 %arg64, i32 %arg128) {
+  %destLocal8 = inttoptr i32 %arg8 to i8*
+  %destLocal16 = inttoptr i32 %arg16 to i16*
+  %destLocal32 = inttoptr i32 %arg32 to i32*
+  %destLocal64 = inttoptr i32 %arg64 to i64*
+  %destLocal128 = inttoptr i32 %arg128 to <4 x i32>*
 
   %ptrGlobal8 = bitcast [1 x i8]* @destGlobal8 to i8*
   %ptrGlobal16 = bitcast [2 x i8]* @destGlobal16 to i16*
@@ -26,31 +27,25 @@ define internal void @doStores(<4 x i32> %vecSrc) {
   %ptrGlobal64 = bitcast [8 x i8]* @destGlobal64 to i64*
   %ptrGlobal128 = bitcast [16 x i8]* @destGlobal128 to <4 x i32>*
 
-  %ptrLocal8 = bitcast i8* %destLocal8 to i8*
-  %ptrLocal16 = bitcast i8* %destLocal16 to i16*
-  %ptrLocal32 = bitcast i8* %destLocal32 to i32*
-  %ptrLocal64 = bitcast i8* %destLocal64 to i64*
-  %ptrLocal128 = bitcast i8* %destLocal128 to <4 x i32>*
-
   store i8 42, i8* %ptrGlobal8, align 1
   store i16 42, i16* %ptrGlobal16, align 1
   store i32 42, i32* %ptrGlobal32, align 1
   store i64 42, i64* %ptrGlobal64, align 1
   store <4 x i32> %vecSrc, <4 x i32>* %ptrGlobal128, align 4
 
-  store i8 42, i8* %ptrLocal8, align 1
-  store i16 42, i16* %ptrLocal16, align 1
-  store i32 42, i32* %ptrLocal32, align 1
-  store i64 42, i64* %ptrLocal64, align 1
-  store <4 x i32> %vecSrc, <4 x i32>* %ptrLocal128, align 4
+  store i8 42, i8* %destLocal8, align 1
+  store i16 42, i16* %destLocal16, align 1
+  store i32 42, i32* %destLocal32, align 1
+  store i64 42, i64* %destLocal64, align 1
+  store <4 x i32> %vecSrc, <4 x i32>* %destLocal128, align 4
 
   ret void
 }
 
 ; DUMP-LABEL: ================ Instrumented CFG ================
-; DUMP-NEXT: define internal void @doStores(<4 x i32> %vecSrc) {
+; DUMP-NEXT: define internal void @doStores(
 ; DUMP-NEXT: __0:
-; DUMP:      call void @__asan_check_store(i32 @destGlobal8, i32 1)
+; DUMP-NEXT: call void @__asan_check_store(i32 @destGlobal8, i32 1)
 ; DUMP-NEXT: store i8 42, i8* @destGlobal8, align 1
 ; DUMP-NEXT: call void @__asan_check_store(i32 @destGlobal16, i32 2)
 ; DUMP-NEXT: store i16 42, i16* @destGlobal16, align 1
@@ -60,15 +55,15 @@ define internal void @doStores(<4 x i32> %vecSrc) {
 ; DUMP-NEXT: store i64 42, i64* @destGlobal64, align 1
 ; DUMP-NEXT: call void @__asan_check_store(i32 @destGlobal128, i32 16)
 ; DUMP-NEXT: store <4 x i32> %vecSrc, <4 x i32>* @destGlobal128, align 4
-; DUMP-NEXT: call void @__asan_check_store(i32 %destLocal8, i32 1)
-; DUMP-NEXT: store i8 42, i8* %destLocal8, align 1
-; DUMP-NEXT: call void @__asan_check_store(i32 %destLocal16, i32 2)
-; DUMP-NEXT: store i16 42, i16* %destLocal16, align 1
-; DUMP-NEXT: call void @__asan_check_store(i32 %destLocal32, i32 4)
-; DUMP-NEXT: store i32 42, i32* %destLocal32, align 1
-; DUMP-NEXT: call void @__asan_check_store(i32 %destLocal64, i32 8)
-; DUMP-NEXT: store i64 42, i64* %destLocal64, align 1
-; DUMP-NEXT: call void @__asan_check_store(i32 %destLocal128, i32 16)
-; DUMP-NEXT: store <4 x i32> %vecSrc, <4 x i32>* %destLocal128, align 4
+; DUMP-NEXT: call void @__asan_check_store(i32 %arg8, i32 1)
+; DUMP-NEXT: store i8 42, i8* %arg8, align 1
+; DUMP-NEXT: call void @__asan_check_store(i32 %arg16, i32 2)
+; DUMP-NEXT: store i16 42, i16* %arg16, align 1
+; DUMP-NEXT: call void @__asan_check_store(i32 %arg32, i32 4)
+; DUMP-NEXT: store i32 42, i32* %arg32, align 1
+; DUMP-NEXT: call void @__asan_check_store(i32 %arg64, i32 8)
+; DUMP-NEXT: store i64 42, i64* %arg64, align 1
+; DUMP-NEXT: call void @__asan_check_store(i32 %arg128, i32 16)
+; DUMP-NEXT: store <4 x i32> %vecSrc, <4 x i32>* %arg128, align 4
 ; DUMP:      ret void
 ; DUMP-NEXT: }