10 files changed, 103 insertions, 27 deletions
diff --git a/test/Analysis/CostModel/X86/cmp.ll b/test/Analysis/CostModel/X86/cmp.ll
index 90b09c1154..713b3742e9 100644
--- a/test/Analysis/CostModel/X86/cmp.ll
+++ b/test/Analysis/CostModel/X86/cmp.ll
@@ -1,38 +1,52 @@
-; RUN: opt < %s  -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mcpu=corei7-avx | FileCheck %s
+; RUN: opt < %s  -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mcpu=corei7-avx | FileCheck --check-prefix=AVX1 %s
+; RUN: opt < %s  -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mcpu=core-avx2 | FileCheck --check-prefix=AVX2 %s
 
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
 target triple = "x86_64-apple-macosx10.8.0"
 
 define i32 @cmp(i32 %arg) {
   ;  -- floats --
-  ;CHECK: cost of 1 {{.*}} fcmp
+  ;AVX1: cost of 1 {{.*}} fcmp
+  ;AVX2: cost of 1 {{.*}} fcmp
   %A = fcmp olt <2 x float> undef, undef
-  ;CHECK: cost of 1 {{.*}} fcmp
+  ;AVX1: cost of 1 {{.*}} fcmp
+  ;AVX2: cost of 1 {{.*}} fcmp
   %B = fcmp olt <4 x float> undef, undef
-  ;CHECK: cost of 1 {{.*}} fcmp
+  ;AVX1: cost of 1 {{.*}} fcmp
+  ;AVX2: cost of 1 {{.*}} fcmp
   %C = fcmp olt <8 x float> undef, undef
-  ;CHECK: cost of 1 {{.*}} fcmp
+  ;AVX1: cost of 1 {{.*}} fcmp
+  ;AVX2: cost of 1 {{.*}} fcmp
   %D = fcmp olt <2 x double> undef, undef
-  ;CHECK: cost of 1 {{.*}} fcmp
+  ;AVX1: cost of 1 {{.*}} fcmp
+  ;AVX2: cost of 1 {{.*}} fcmp
   %E = fcmp olt <4 x double> undef, undef
 
   ;  -- integers --
 
-  ;CHECK: cost of 1 {{.*}} icmp
+  ;AVX1: cost of 1 {{.*}} icmp
+  ;AVX2: cost of 1 {{.*}} icmp
   %F = icmp eq <16 x i8> undef, undef
-  ;CHECK: cost of 1 {{.*}} icmp
+  ;AVX1: cost of 1 {{.*}} icmp
+  ;AVX2: cost of 1 {{.*}} icmp
   %G = icmp eq <8 x i16> undef, undef
-  ;CHECK: cost of 1 {{.*}} icmp
+  ;AVX1: cost of 1 {{.*}} icmp
+  ;AVX2: cost of 1 {{.*}} icmp
   %H = icmp eq <4 x i32> undef, undef
-  ;CHECK: cost of 1 {{.*}} icmp
+  ;AVX1: cost of 1 {{.*}} icmp
+  ;AVX2: cost of 1 {{.*}} icmp
   %I = icmp eq <2 x i64> undef, undef
-  ;CHECK: cost of 4 {{.*}} icmp
+  ;AVX1: cost of 4 {{.*}} icmp
+  ;AVX2: cost of 1 {{.*}} icmp
   %J = icmp eq <4 x i64> undef, undef
-  ;CHECK: cost of 4 {{.*}} icmp
+  ;AVX1: cost of 4 {{.*}} icmp
+  ;AVX2: cost of 1 {{.*}} icmp
   %K = icmp eq <8 x i32> undef, undef
-  ;CHECK: cost of 4 {{.*}} icmp
+  ;AVX1: cost of 4 {{.*}} icmp
+  ;AVX2: cost of 1 {{.*}} icmp
   %L = icmp eq <16 x i16> undef, undef
-  ;CHECK: cost of 4 {{.*}} icmp
+  ;AVX1: cost of 4 {{.*}} icmp
+  ;AVX2: cost of 1 {{.*}} icmp
   %M = icmp eq <32 x i8> undef, undef
 
   ;CHECK: cost of 0 {{.*}} ret
diff --git a/test/Analysis/CostModel/X86/i32.ll b/test/Analysis/CostModel/X86/i32.ll
index 52c295934c..c2dce762a0 100644
--- a/test/Analysis/CostModel/X86/i32.ll
+++ b/test/Analysis/CostModel/X86/i32.ll
@@ -1,7 +1,5 @@
 ; RUN: opt < %s  -cost-model -analyze -mtriple=i386 -mcpu=corei7-avx | FileCheck %s
 
-
-;CHECK: cost of 2 {{.*}} add
 ;CHECK: cost of 0 {{.*}} ret
 define i32 @no_info(i32 %arg) {
   %e = add i64 undef, undef
diff --git a/test/Analysis/CostModel/X86/load_store.ll b/test/Analysis/CostModel/X86/load_store.ll
new file mode 100644
index 0000000000..4195b1d879
--- /dev/null
+++ b/test/Analysis/CostModel/X86/load_store.ll
@@ -0,0 +1,64 @@
+; RUN: opt < %s  -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mcpu=corei7-avx | FileCheck %s
+
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
+target triple = "x86_64-apple-macosx10.8.0"
+
+define i32 @stores(i32 %arg) {
+
+  ;CHECK: cost of 1 {{.*}} store
+  store i8 undef, i8* undef, align 4
+  ;CHECK: cost of 1 {{.*}} store
+  store i16 undef, i16* undef, align 4
+  ;CHECK: cost of 1 {{.*}} store
+  store i32 undef, i32* undef, align 4
+  ;CHECK: cost of 1 {{.*}} store
+  store i64 undef, i64* undef, align 4
+  ;CHECK: cost of 2 {{.*}} store
+  store i128 undef, i128* undef, align 4
+
+  ;CHECK: cost of 1 {{.*}} store
+  store <4 x i16> undef, <4 x i16>* undef, align 4
+  ;CHECK: cost of 1 {{.*}} store
+  store <4 x i32> undef, <4 x i32>* undef, align 4
+  ;CHECK: cost of 2 {{.*}} store
+  store <4 x i64> undef, <4 x i64>* undef, align 4
+
+  ;CHECK: cost of 1 {{.*}} store
+  store <8 x i16> undef, <8 x i16>* undef, align 4
+  ;CHECK: cost of 2 {{.*}} store
+  store <8 x i32> undef, <8 x i32>* undef, align 4
+  ;CHECK: cost of 4 {{.*}} store
+  store <8 x i64> undef, <8 x i64>* undef, align 4
+
+  ret i32 undef
+}
+define i32 @loads(i32 %arg) {
+  ;CHECK: cost of 1 {{.*}} load
+  load i8* undef, align 4
+  ;CHECK: cost of 1 {{.*}} load
+  load i16* undef, align 4
+  ;CHECK: cost of 1 {{.*}} load
+  load i32* undef, align 4
+  ;CHECK: cost of 1 {{.*}} load
+  load i64* undef, align 4
+  ;CHECK: cost of 2 {{.*}} load
+  load i128* undef, align 4
+
+  ;CHECK: cost of 1 {{.*}} load
+  load <2 x i32>* undef, align 4
+  ;CHECK: cost of 1 {{.*}} load
+  load <4 x i32>* undef, align 4
+  ;CHECK: cost of 2 {{.*}} load
+  load <8 x i32>* undef, align 4
+
+
+  ;CHECK: cost of 1 {{.*}} load
+  load <2 x i64>* undef, align 4
+  ;CHECK: cost of 2 {{.*}} load
+  load <4 x i64>* undef, align 4
+  ;CHECK: cost of 4 {{.*}} load
+  load <8 x i64>* undef, align 4
+
+  ret i32 undef
+}
+
diff --git a/test/Analysis/CostModel/X86/vectorized-loop.ll b/test/Analysis/CostModel/X86/vectorized-loop.ll
index 6c9e111bb1..25b11145c6 100644
--- a/test/Analysis/CostModel/X86/vectorized-loop.ll
+++ b/test/Analysis/CostModel/X86/vectorized-loop.ll
@@ -28,16 +28,17 @@ vector.body:                                      ; preds = %for.body.lr.ph, %ve
   %4 = getelementptr inbounds i32* %B, i64 %3
   ;CHECK: cost of 0 {{.*}} bitcast
   %5 = bitcast i32* %4 to <8 x i32>*
-  ;CHECK: cost of 1 {{.*}} load
+  ;CHECK: cost of 2 {{.*}} load
   %6 = load <8 x i32>* %5, align 4
   ;CHECK: cost of 4 {{.*}} mul
   %7 = mul nsw <8 x i32> %6, <i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5>
   %8 = getelementptr inbounds i32* %A, i64 %index
   %9 = bitcast i32* %8 to <8 x i32>*
+  ;CHECK: cost of 2 {{.*}} load
   %10 = load <8 x i32>* %9, align 4
   ;CHECK: cost of 4 {{.*}} add
   %11 = add nsw <8 x i32> %10, %7
-  ;CHECK: cost of 1 {{.*}} store
+  ;CHECK: cost of 2 {{.*}} store
   store <8 x i32> %11, <8 x i32>* %9, align 4
   %index.next = add i64 %index, 8
   %12 = icmp eq i64 %index.next, %end.idx.rnd.down
diff --git a/test/Analysis/CostModel/no_info.ll b/test/Analysis/CostModel/no_info.ll
index d20d56b79a..f3f165b1b5 100644
--- a/test/Analysis/CostModel/no_info.ll
+++ b/test/Analysis/CostModel/no_info.ll
@@ -1,11 +1,8 @@
 ; RUN: opt < %s -cost-model -analyze | FileCheck %s
 
 ; The cost model does not have any target information so it can't make a decision.
-; Notice that OPT does not read the triple information from the module itself, only through the command line.
 
-; This info ignored:
-target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
-target triple = "x86_64-apple-macosx10.8.0"
+; -- No triple in this module --
 
 ;CHECK: Unknown cost {{.*}} add
 ;CHECK: Unknown cost {{.*}} ret
diff --git a/test/Analysis/Dominators/invoke.ll b/test/Analysis/Dominators/invoke.ll
index f935750c98..da0b246165 100644
--- a/test/Analysis/Dominators/invoke.ll
+++ b/test/Analysis/Dominators/invoke.ll
@@ -1,4 +1,4 @@
-; RUN: opt -verify -disable-output %s
+; RUN: opt -verify -disable-output < %s
 ; This tests that we handle unreachable blocks correctly
 
 define void @f() {
diff --git a/test/Analysis/RegionInfo/20100809_bb_not_in_domtree.ll b/test/Analysis/RegionInfo/20100809_bb_not_in_domtree.ll
index 218b4375f7..0dfa0bf9cd 100644
--- a/test/Analysis/RegionInfo/20100809_bb_not_in_domtree.ll
+++ b/test/Analysis/RegionInfo/20100809_bb_not_in_domtree.ll
@@ -1,4 +1,4 @@
-; RUN: opt -regions %s
+; RUN: opt -regions < %s
 define i32 @main() nounwind {
 entry:
   br label %for.cond
diff --git a/test/Analysis/ScalarEvolution/2010-09-03-RequiredTransitive.ll b/test/Analysis/ScalarEvolution/2010-09-03-RequiredTransitive.ll
index aba0ce7467..5a02398104 100644
--- a/test/Analysis/ScalarEvolution/2010-09-03-RequiredTransitive.ll
+++ b/test/Analysis/ScalarEvolution/2010-09-03-RequiredTransitive.ll
@@ -1,8 +1,10 @@
-; RUN: opt -indvars -scalar-evolution -analyze %s
+; RUN: opt -indvars -scalar-evolution -analyze < %s | FileCheck %s
 ; This test checks if the SCEV analysis is printed out at all.
 ; It failed once as the RequiredTransitive option was not implemented
 ; correctly.
 
+; CHECK: Classifying expressions for: @main
+
 define i32 @main() nounwind {
 entry:
   br label %for.cond
diff --git a/test/Analysis/ScalarEvolution/2011-03-09-ExactNoMaxBECount.ll b/test/Analysis/ScalarEvolution/2011-03-09-ExactNoMaxBECount.ll
index 9f17e27577..49e944dcd2 100644
--- a/test/Analysis/ScalarEvolution/2011-03-09-ExactNoMaxBECount.ll
+++ b/test/Analysis/ScalarEvolution/2011-03-09-ExactNoMaxBECount.ll
@@ -1,4 +1,4 @@
-; RUN: opt -indvars  %s
+; RUN: opt -indvars < %s
 ; PR9424: Attempt to use a SCEVCouldNotCompute object!
 ; The inner loop computes the Step and Start of the outer loop.
 ; Call that Vexit. The outer End value is max(2,Vexit), because
diff --git a/test/Analysis/ScalarEvolution/fold.ll b/test/Analysis/ScalarEvolution/fold.ll
index 4e2adf187e..57006dd9bb 100644
--- a/test/Analysis/ScalarEvolution/fold.ll
+++ b/test/Analysis/ScalarEvolution/fold.ll
@@ -1,4 +1,4 @@
-; RUN: opt -analyze -scalar-evolution %s -S | FileCheck %s
+; RUN: opt -analyze -scalar-evolution -S < %s | FileCheck %s
 
 define i16 @test1(i8 %x) {
   %A = zext i8 %x to i12