4 files changed, 453 insertions, 1 deletions
diff --git a/test/Analysis/CostModel/ARM/cast.ll b/test/Analysis/CostModel/ARM/cast.ll
index 464b6ec..96eb335 100644
--- a/test/Analysis/CostModel/ARM/cast.ll
+++ b/test/Analysis/CostModel/ARM/cast.ll
@@ -1,4 +1,4 @@
-; RUN: opt < %s  -cost-model -analyze -mtriple=thumbv7-apple-ios6.0.0 -mcpu=swift | FileCheck %s
+; RUN: opt < %s  -cost-model -analyze -mtriple=thumbv7-apple-ios6.0.0 -mcpu=cortex-a8 | FileCheck %s
 target datalayout = "e-p:32:32:32-i1:8:32-i8:8:32-i16:16:32-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:32:64-v128:32:128-a0:0:32-n32-S32"
 target triple = "thumbv7-apple-ios6.0.0"
 
@@ -152,6 +152,381 @@ define i32 @casts() {
   ; CHECK: cost of 10 {{.*}} uitofp
   %r69 = uitofp i64 undef to double
 
+  ; Vector cast cost of instructions lowering the cast to the stack.
+  ; CHECK: cost of 24 {{.*}} sext
+  %r70 = sext <8 x i8> undef to <8 x i32>
+  ; CHECK: cost of 48 {{.*}} sext
+  %r71 = sext <16 x i8> undef to <16 x i32>
+  ; CHECK: cost of 22 {{.*}} zext
+  %r72 = zext <8 x i8> undef to <8 x i32>
+  ; CHECK: cost of 44 {{.*}} zext
+  %r73 = zext <16 x i8> undef to <16 x i32>
+  ; CHECK: cost of 19 {{.*}} trunc
+  %r74 = trunc <8 x i32> undef to <8 x i8>
+  ; CHECK: cost of 38 {{.*}} trunc
+  %r75 = trunc <16 x i32> undef to <16 x i8>
+
+  ; Floating point truncation costs.
+  ; CHECK: cost of 1 {{.*}} fptrunc double
+  %r80 = fptrunc double undef to float
+  ; CHECK: cost of 2 {{.*}} fptrunc <2 x double
+  %r81 = fptrunc <2 x double> undef to <2 x float>
+  ; CHECK: cost of 4 {{.*}} fptrunc <4 x double
+  %r82 = fptrunc <4 x double> undef to <4 x float>
+  ; CHECK: cost of 8 {{.*}} fptrunc <8 x double
+  %r83 = fptrunc <8 x double> undef to <8 x float>
+  ; CHECK: cost of 16 {{.*}} fptrunc <16 x double
+  %r84 = fptrunc <16 x double> undef to <16 x float>
+
+  ; Floating point extension costs.
+  ; CHECK: cost of 1 {{.*}} fpext float
+  %r85 = fpext float undef to double
+  ; CHECK: cost of 2 {{.*}} fpext <2 x float
+  %r86 = fpext <2 x float> undef to <2 x double>
+  ; CHECK: cost of 4 {{.*}} fpext <4 x float
+  %r87 = fpext <4 x float> undef to <4 x double>
+  ; CHECK: cost of 8 {{.*}} fpext <8 x float
+  %r88 = fpext <8 x float> undef to <8 x double>
+  ; CHECK: cost of 16 {{.*}} fpext <16 x float
+  %r89 = fpext <16 x float> undef to <16 x double>
+
+  ;; Floating point to integer vector casts.
+  ; CHECK: cost of 1 {{.*}} fptoui
+  %r90 = fptoui <2 x float> undef to <2 x i1>
+  ; CHECK: cost of 1 {{.*}} fptosi
+  %r91 = fptosi <2 x float> undef to <2 x i1>
+  ; CHECK: cost of 1 {{.*}} fptoui
+  %r92 = fptoui <2 x float> undef to <2 x i8>
+  ; CHECK: cost of 1 {{.*}} fptosi
+  %r93 = fptosi <2 x float> undef to <2 x i8>
+  ; CHECK: cost of 1 {{.*}} fptoui
+  %r94 = fptoui <2 x float> undef to <2 x i16>
+  ; CHECK: cost of 1 {{.*}} fptosi
+  %r95 = fptosi <2 x float> undef to <2 x i16>
+  ; CHECK: cost of 1 {{.*}} fptoui
+  %r96 = fptoui <2 x float> undef to <2 x i32>
+  ; CHECK: cost of 1 {{.*}} fptosi
+  %r97 = fptosi <2 x float> undef to <2 x i32>
+  ; CHECK: cost of 24 {{.*}} fptoui
+  %r98 = fptoui <2 x float> undef to <2 x i64>
+  ; CHECK: cost of 24 {{.*}} fptosi
+  %r99 = fptosi <2 x float> undef to <2 x i64>
+
+  ; CHECK: cost of 8 {{.*}} fptoui
+  %r100 = fptoui <2 x double> undef to <2 x i1>
+  ; CHECK: cost of 8 {{.*}} fptosi
+  %r101 = fptosi <2 x double> undef to <2 x i1>
+  ; CHECK: cost of 8 {{.*}} fptoui
+  %r102 = fptoui <2 x double> undef to <2 x i8>
+  ; CHECK: cost of 8 {{.*}} fptosi
+  %r103 = fptosi <2 x double> undef to <2 x i8>
+  ; CHECK: cost of 8 {{.*}} fptoui
+  %r104 = fptoui <2 x double> undef to <2 x i16>
+  ; CHECK: cost of 8 {{.*}} fptosi
+  %r105 = fptosi <2 x double> undef to <2 x i16>
+  ; CHECK: cost of 2 {{.*}} fptoui
+  %r106 = fptoui <2 x double> undef to <2 x i32>
+  ; CHECK: cost of 2 {{.*}} fptosi
+  %r107 = fptosi <2 x double> undef to <2 x i32>
+  ; CHECK: cost of 24 {{.*}} fptoui
+  %r108 = fptoui <2 x double> undef to <2 x i64>
+  ; CHECK: cost of 24 {{.*}} fptosi
+  %r109 = fptosi <2 x double> undef to <2 x i64>
+
+  ; CHECK: cost of 16 {{.*}} fptoui
+  %r110 = fptoui <4 x float> undef to <4 x i1>
+  ; CHECK: cost of 16 {{.*}} fptosi
+  %r111 = fptosi <4 x float> undef to <4 x i1>
+  ; CHECK: cost of 3 {{.*}} fptoui
+  %r112 = fptoui <4 x float> undef to <4 x i8>
+  ; CHECK: cost of 3 {{.*}} fptosi
+  %r113 = fptosi <4 x float> undef to <4 x i8>
+  ; CHECK: cost of 2 {{.*}} fptoui
+  %r114 = fptoui <4 x float> undef to <4 x i16>
+  ; CHECK: cost of 2 {{.*}} fptosi
+  %r115 = fptosi <4 x float> undef to <4 x i16>
+  ; CHECK: cost of 1 {{.*}} fptoui
+  %r116 = fptoui <4 x float> undef to <4 x i32>
+  ; CHECK: cost of 1 {{.*}} fptosi
+  %r117 = fptosi <4 x float> undef to <4 x i32>
+  ; CHECK: cost of 48 {{.*}} fptoui
+  %r118 = fptoui <4 x float> undef to <4 x i64>
+  ; CHECK: cost of 48 {{.*}} fptosi
+  %r119 = fptosi <4 x float> undef to <4 x i64>
+
+  ; CHECK: cost of 16 {{.*}} fptoui
+  %r120 = fptoui <4 x double> undef to <4 x i1>
+  ; CHECK: cost of 16 {{.*}} fptosi
+  %r121 = fptosi <4 x double> undef to <4 x i1>
+  ; CHECK: cost of 16 {{.*}} fptoui
+  %r122 = fptoui <4 x double> undef to <4 x i8>
+  ; CHECK: cost of 16 {{.*}} fptosi
+  %r123 = fptosi <4 x double> undef to <4 x i8>
+  ; CHECK: cost of 16 {{.*}} fptoui
+  %r124 = fptoui <4 x double> undef to <4 x i16>
+  ; CHECK: cost of 16 {{.*}} fptosi
+  %r125 = fptosi <4 x double> undef to <4 x i16>
+  ; CHECK: cost of 16 {{.*}} fptoui
+  %r126 = fptoui <4 x double> undef to <4 x i32>
+  ; CHECK: cost of 16 {{.*}} fptosi
+  %r127 = fptosi <4 x double> undef to <4 x i32>
+  ; CHECK: cost of 48 {{.*}} fptoui
+  %r128 = fptoui <4 x double> undef to <4 x i64>
+  ; CHECK: cost of 48 {{.*}} fptosi
+  %r129 = fptosi <4 x double> undef to <4 x i64>
+
+  ; CHECK: cost of 32 {{.*}} fptoui
+  %r130 = fptoui <8 x float> undef to <8 x i1>
+  ; CHECK: cost of 32 {{.*}} fptosi
+  %r131 = fptosi <8 x float> undef to <8 x i1>
+  ; CHECK: cost of 32 {{.*}} fptoui
+  %r132 = fptoui <8 x float> undef to <8 x i8>
+  ; CHECK: cost of 32 {{.*}} fptosi
+  %r133 = fptosi <8 x float> undef to <8 x i8>
+  ; CHECK: cost of 4 {{.*}} fptoui
+  %r134 = fptoui <8 x float> undef to <8 x i16>
+  ; CHECK: cost of 4 {{.*}} fptosi
+  %r135 = fptosi <8 x float> undef to <8 x i16>
+  ; CHECK: cost of 2 {{.*}} fptoui
+  %r136 = fptoui <8 x float> undef to <8 x i32>
+  ; CHECK: cost of 2 {{.*}} fptosi
+  %r137 = fptosi <8 x float> undef to <8 x i32>
+  ; CHECK: cost of 96 {{.*}} fptoui
+  %r138 = fptoui <8 x float> undef to <8 x i64>
+  ; CHECK: cost of 96 {{.*}} fptosi
+  %r139 = fptosi <8 x float> undef to <8 x i64>
+
+  ; CHECK: cost of 32 {{.*}} fptoui
+  %r140 = fptoui <8 x double> undef to <8 x i1>
+  ; CHECK: cost of 32 {{.*}} fptosi
+  %r141 = fptosi <8 x double> undef to <8 x i1>
+  ; CHECK: cost of 32 {{.*}} fptoui
+  %r142 = fptoui <8 x double> undef to <8 x i8>
+  ; CHECK: cost of 32 {{.*}} fptosi
+  %r143 = fptosi <8 x double> undef to <8 x i8>
+  ; CHECK: cost of 32 {{.*}} fptoui
+  %r144 = fptoui <8 x double> undef to <8 x i16>
+  ; CHECK: cost of 32 {{.*}} fptosi
+  %r145 = fptosi <8 x double> undef to <8 x i16>
+  ; CHECK: cost of 32 {{.*}} fptoui
+  %r146 = fptoui <8 x double> undef to <8 x i32>
+  ; CHECK: cost of 32 {{.*}} fptosi
+  %r147 = fptosi <8 x double> undef to <8 x i32>
+  ; CHECK: cost of 96 {{.*}} fptoui
+  %r148 = fptoui <8 x double> undef to <8 x i64>
+  ; CHECK: cost of 96 {{.*}} fptosi
+  %r149 = fptosi <8 x double> undef to <8 x i64>
+
+  ; CHECK: cost of 64 {{.*}} fptoui
+  %r150 = fptoui <16 x float> undef to <16 x i1>
+  ; CHECK: cost of 64 {{.*}} fptosi
+  %r151 = fptosi <16 x float> undef to <16 x i1>
+  ; CHECK: cost of 64 {{.*}} fptoui
+  %r152 = fptoui <16 x float> undef to <16 x i8>
+  ; CHECK: cost of 64 {{.*}} fptosi
+  %r153 = fptosi <16 x float> undef to <16 x i8>
+  ; CHECK: cost of 8 {{.*}} fptoui
+  %r154 = fptoui <16 x float> undef to <16 x i16>
+  ; CHECK: cost of 8 {{.*}} fptosi
+  %r155 = fptosi <16 x float> undef to <16 x i16>
+  ; CHECK: cost of 4 {{.*}} fptoui
+  %r156 = fptoui <16 x float> undef to <16 x i32>
+  ; CHECK: cost of 4 {{.*}} fptosi
+  %r157 = fptosi <16 x float> undef to <16 x i32>
+  ; CHECK: cost of 192 {{.*}} fptoui
+  %r158 = fptoui <16 x float> undef to <16 x i64>
+  ; CHECK: cost of 192 {{.*}} fptosi
+  %r159 = fptosi <16 x float> undef to <16 x i64>
+
+  ; CHECK: cost of 64 {{.*}} fptoui
+  %r160 = fptoui <16 x double> undef to <16 x i1>
+  ; CHECK: cost of 64 {{.*}} fptosi
+  %r161 = fptosi <16 x double> undef to <16 x i1>
+  ; CHECK: cost of 64 {{.*}} fptoui
+  %r162 = fptoui <16 x double> undef to <16 x i8>
+  ; CHECK: cost of 64 {{.*}} fptosi
+  %r163 = fptosi <16 x double> undef to <16 x i8>
+  ; CHECK: cost of 64 {{.*}} fptoui
+  %r164 = fptoui <16 x double> undef to <16 x i16>
+  ; CHECK: cost of 64 {{.*}} fptosi
+  %r165 = fptosi <16 x double> undef to <16 x i16>
+  ; CHECK: cost of 64 {{.*}} fptoui
+  %r166 = fptoui <16 x double> undef to <16 x i32>
+  ; CHECK: cost of 64 {{.*}} fptosi
+  %r167 = fptosi <16 x double> undef to <16 x i32>
+  ; CHECK: cost of 192 {{.*}} fptoui
+  %r168 = fptoui <16 x double> undef to <16 x i64>
+  ; CHECK: cost of 192 {{.*}} fptosi
+  %r169 = fptosi <16 x double> undef to <16 x i64>
+
+  ; CHECK: cost of 8 {{.*}} uitofp
+  %r170 = uitofp <2 x i1> undef to <2 x float>
+  ; CHECK: cost of 8 {{.*}} sitofp
+  %r171 = sitofp <2 x i1> undef to <2 x float>
+  ; CHECK: cost of 3 {{.*}} uitofp
+  %r172 = uitofp <2 x i8> undef to <2 x float>
+  ; CHECK: cost of 3 {{.*}} sitofp
+  %r173 = sitofp <2 x i8> undef to <2 x float>
+  ; CHECK: cost of 2 {{.*}} uitofp
+  %r174 = uitofp <2 x i16> undef to <2 x float>
+  ; CHECK: cost of 2 {{.*}} sitofp
+  %r175 = sitofp <2 x i16> undef to <2 x float>
+  ; CHECK: cost of 1 {{.*}} uitofp
+  %r176 = uitofp <2 x i32> undef to <2 x float>
+  ; CHECK: cost of 1 {{.*}} sitofp
+  %r177 = sitofp <2 x i32> undef to <2 x float>
+  ; CHECK: cost of 24 {{.*}} uitofp
+  %r178 = uitofp <2 x i64> undef to <2 x float>
+  ; CHECK: cost of 24 {{.*}} sitofp
+  %r179 = sitofp <2 x i64> undef to <2 x float>
+
+  ; CHECK: cost of 8 {{.*}} uitofp
+  %r180 = uitofp <2 x i1> undef to <2 x double>
+  ; CHECK: cost of 8 {{.*}} sitofp
+  %r181 = sitofp <2 x i1> undef to <2 x double>
+  ; CHECK: cost of 4 {{.*}} uitofp
+  %r182 = uitofp <2 x i8> undef to <2 x double>
+  ; CHECK: cost of 4 {{.*}} sitofp
+  %r183 = sitofp <2 x i8> undef to <2 x double>
+  ; CHECK: cost of 3 {{.*}} uitofp
+  %r184 = uitofp <2 x i16> undef to <2 x double>
+  ; CHECK: cost of 3 {{.*}} sitofp
+  %r185 = sitofp <2 x i16> undef to <2 x double>
+  ; CHECK: cost of 2 {{.*}} uitofp
+  %r186 = uitofp <2 x i32> undef to <2 x double>
+  ; CHECK: cost of 2 {{.*}} sitofp
+  %r187 = sitofp <2 x i32> undef to <2 x double>
+  ; CHECK: cost of 24 {{.*}} uitofp
+  %r188 = uitofp <2 x i64> undef to <2 x double>
+  ; CHECK: cost of 24 {{.*}} sitofp
+  %r189 = sitofp <2 x i64> undef to <2 x double>
+
+  ; CHECK: cost of 3 {{.*}} uitofp
+  %r190 = uitofp <4 x i1> undef to <4 x float>
+  ; CHECK: cost of 3 {{.*}} sitofp
+  %r191 = sitofp <4 x i1> undef to <4 x float>
+  ; CHECK: cost of 3 {{.*}} uitofp
+  %r192 = uitofp <4 x i8> undef to <4 x float>
+  ; CHECK: cost of 3 {{.*}} sitofp
+  %r193 = sitofp <4 x i8> undef to <4 x float>
+  ; CHECK: cost of 2 {{.*}} uitofp
+  %r194 = uitofp <4 x i16> undef to <4 x float>
+  ; CHECK: cost of 2 {{.*}} sitofp
+  %r195 = sitofp <4 x i16> undef to <4 x float>
+  ; CHECK: cost of 1 {{.*}} uitofp
+  %r196 = uitofp <4 x i32> undef to <4 x float>
+  ; CHECK: cost of 1 {{.*}} sitofp
+  %r197 = sitofp <4 x i32> undef to <4 x float>
+  ; CHECK: cost of 48 {{.*}} uitofp
+  %r198 = uitofp <4 x i64> undef to <4 x float>
+  ; CHECK: cost of 48 {{.*}} sitofp
+  %r199 = sitofp <4 x i64> undef to <4 x float>
+
+  ; CHECK: cost of 16 {{.*}} uitofp
+  %r200 = uitofp <4 x i1> undef to <4 x double>
+  ; CHECK: cost of 16 {{.*}} sitofp
+  %r201 = sitofp <4 x i1> undef to <4 x double>
+  ; CHECK: cost of 16 {{.*}} uitofp
+  %r202 = uitofp <4 x i8> undef to <4 x double>
+  ; CHECK: cost of 16 {{.*}} sitofp
+  %r203 = sitofp <4 x i8> undef to <4 x double>
+  ; CHECK: cost of 16 {{.*}} uitofp
+  %r204 = uitofp <4 x i16> undef to <4 x double>
+  ; CHECK: cost of 16 {{.*}} sitofp
+  %r205 = sitofp <4 x i16> undef to <4 x double>
+  ; CHECK: cost of 16 {{.*}} uitofp
+  %r206 = uitofp <4 x i32> undef to <4 x double>
+  ; CHECK: cost of 16 {{.*}} sitofp
+  %r207 = sitofp <4 x i32> undef to <4 x double>
+  ; CHECK: cost of 48 {{.*}} uitofp
+  %r208 = uitofp <4 x i64> undef to <4 x double>
+  ; CHECK: cost of 48 {{.*}} sitofp
+  %r209 = sitofp <4 x i64> undef to <4 x double>
+
+  ; CHECK: cost of 32 {{.*}} uitofp
+  %r210 = uitofp <8 x i1> undef to <8 x float>
+  ; CHECK: cost of 32 {{.*}} sitofp
+  %r211 = sitofp <8 x i1> undef to <8 x float>
+  ; CHECK: cost of 32 {{.*}} uitofp
+  %r212 = uitofp <8 x i8> undef to <8 x float>
+  ; CHECK: cost of 32 {{.*}} sitofp
+  %r213 = sitofp <8 x i8> undef to <8 x float>
+  ; CHECK: cost of 4 {{.*}} uitofp
+  %r214 = uitofp <8 x i16> undef to <8 x float>
+  ; CHECK: cost of 4 {{.*}} sitofp
+  %r215 = sitofp <8 x i16> undef to <8 x float>
+  ; CHECK: cost of 2 {{.*}} uitofp
+  %r216 = uitofp <8 x i32> undef to <8 x float>
+  ; CHECK: cost of 2 {{.*}} sitofp
+  %r217 = sitofp <8 x i32> undef to <8 x float>
+  ; CHECK: cost of 96 {{.*}} uitofp
+  %r218 = uitofp <8 x i64> undef to <8 x float>
+  ; CHECK: cost of 96 {{.*}} sitofp
+  %r219 = sitofp <8 x i64> undef to <8 x float>
+
+  ; CHECK: cost of 32 {{.*}} uitofp
+  %r220 = uitofp <8 x i1> undef to <8 x double>
+  ; CHECK: cost of 32 {{.*}} sitofp
+  %r221 = sitofp <8 x i1> undef to <8 x double>
+  ; CHECK: cost of 32 {{.*}} uitofp
+  %r222 = uitofp <8 x i8> undef to <8 x double>
+  ; CHECK: cost of 32 {{.*}} sitofp
+  %r223 = sitofp <8 x i8> undef to <8 x double>
+  ; CHECK: cost of 32 {{.*}} uitofp
+  %r224 = uitofp <8 x i16> undef to <8 x double>
+  ; CHECK: cost of 32 {{.*}} sitofp
+  %r225 = sitofp <8 x i16> undef to <8 x double>
+  ; CHECK: cost of 32 {{.*}} uitofp
+  %r226 = uitofp <8 x i16> undef to <8 x double>
+  ; CHECK: cost of 32 {{.*}} sitofp
+  %r227 = sitofp <8 x i16> undef to <8 x double>
+  ; CHECK: cost of 96 {{.*}} uitofp
+  %r228 = uitofp <8 x i64> undef to <8 x double>
+  ; CHECK: cost of 96 {{.*}} sitofp
+  %r229 = sitofp <8 x i64> undef to <8 x double>
+
+  ; CHECK: cost of 64 {{.*}} uitofp
+  %r230 = uitofp <16 x i1> undef to <16 x float>
+  ; CHECK: cost of 64 {{.*}} sitofp
+  %r231 = sitofp <16 x i1> undef to <16 x float>
+  ; CHECK: cost of 64 {{.*}} uitofp
+  %r232 = uitofp <16 x i8> undef to <16 x float>
+  ; CHECK: cost of 64 {{.*}} sitofp
+  %r233 = sitofp <16 x i8> undef to <16 x float>
+  ; CHECK: cost of 8 {{.*}} uitofp
+  %r234 = uitofp <16 x i16> undef to <16 x float>
+  ; CHECK: cost of 8 {{.*}} sitofp
+  %r235 = sitofp <16 x i16> undef to <16 x float>
+  ; CHECK: cost of 4 {{.*}} uitofp
+  %r236 = uitofp <16 x i32> undef to <16 x float>
+  ; CHECK: cost of 4 {{.*}} sitofp
+  %r237 = sitofp <16 x i32> undef to <16 x float>
+  ; CHECK: cost of 192 {{.*}} uitofp
+  %r238 = uitofp <16 x i64> undef to <16 x float>
+  ; CHECK: cost of 192 {{.*}} sitofp
+  %r239 = sitofp <16 x i64> undef to <16 x float>
+
+  ; CHECK: cost of 64 {{.*}} uitofp
+  %r240 = uitofp <16 x i1> undef to <16 x double>
+  ; CHECK: cost of 64 {{.*}} sitofp
+  %r241 = sitofp <16 x i1> undef to <16 x double>
+  ; CHECK: cost of 64 {{.*}} uitofp
+  %r242 = uitofp <16 x i8> undef to <16 x double>
+  ; CHECK: cost of 64 {{.*}} sitofp
+  %r243 = sitofp <16 x i8> undef to <16 x double>
+  ; C4ECK: cost of 64 {{.*}} uitofp
+  %r244 = uitofp <16 x i16> undef to <16 x double>
+  ; CHECK: cost of 64 {{.*}} sitofp
+  %r245 = sitofp <16 x i16> undef to <16 x double>
+  ; CHECK: cost of 64 {{.*}} uitofp
+  %r246 = uitofp <16 x i16> undef to <16 x double>
+  ; CHECK: cost of 64 {{.*}} sitofp
+  %r247 = sitofp <16 x i16> undef to <16 x double>
+  ; CHECK: cost of 192 {{.*}} uitofp
+  %r248 = uitofp <16 x i64> undef to <16 x double>
+  ; CHECK: cost of 192 {{.*}} sitofp
+  %r249 = sitofp <16 x i64> undef to <16 x double>
+
   ;CHECK: cost of 0 {{.*}} ret
   ret i32 undef
 }
diff --git a/test/Analysis/CostModel/ARM/select.ll b/test/Analysis/CostModel/ARM/select.ll
index 96afccf..34ed1ee 100644
--- a/test/Analysis/CostModel/ARM/select.ll
+++ b/test/Analysis/CostModel/ARM/select.ll
@@ -34,13 +34,26 @@ define void @casts() {
   %v12 = select <4 x i1>  undef, <4 x i16> undef, <4 x i16> undef
   ; CHECK: cost of 1 {{.*}} select
   %v13 = select <8 x i1>  undef, <8 x i16> undef, <8 x i16> undef
+  ; CHECK: cost of 40 {{.*}} select
+  %v13b = select <16 x i1>  undef, <16 x i16> undef, <16 x i16> undef
 
   ; CHECK: cost of 1 {{.*}} select
   %v14 = select <2 x i1> undef, <2 x i32> undef, <2 x i32> undef
   ; CHECK: cost of 1 {{.*}} select
   %v15 = select <4 x i1>  undef, <4 x i32> undef, <4 x i32> undef
+  ; CHECK: cost of 41 {{.*}} select
+  %v15b = select <8 x i1>  undef, <8 x i32> undef, <8 x i32> undef
+  ; CHECK: cost of 82 {{.*}} select
+  %v15c = select <16 x i1>  undef, <16 x i32> undef, <16 x i32> undef
+
   ; CHECK: cost of 1 {{.*}} select
   %v16 = select <2 x i1> undef, <2 x i64> undef, <2 x i64> undef
+  ; CHECK: cost of 19 {{.*}} select
+  %v16a = select <4 x i1> undef, <4 x i64> undef, <4 x i64> undef
+  ; CHECK: cost of 50 {{.*}} select
+  %v16b = select <8 x i1> undef, <8 x i64> undef, <8 x i64> undef
+  ; CHECK: cost of 100 {{.*}} select
+  %v16c = select <16 x i1> undef, <16 x i64> undef, <16 x i64> undef
 
   ; CHECK: cost of 1 {{.*}} select
   %v17 = select <2 x i1> undef, <2 x float> undef, <2 x float> undef
diff --git a/test/Analysis/CostModel/X86/arith.ll b/test/Analysis/CostModel/X86/arith.ll
index ae78d44..f0521ba 100644
--- a/test/Analysis/CostModel/X86/arith.ll
+++ b/test/Analysis/CostModel/X86/arith.ll
@@ -1,4 +1,6 @@
 ; 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=core2 | FileCheck %s --check-prefix=SSE3
+; RUN: opt < %s  -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mcpu=core-avx2 | FileCheck %s --check-prefix=AVX2
 
 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"
@@ -32,7 +34,37 @@ define i32 @xor(i32 %arg) {
   ret i32 undef
 }
 
+; CHECK: mul
+define void @mul() {
+  ; A <2 x i32> gets expanded to a <2 x i64> vector.
+  ; A <2 x i64> vector multiply is implemented using
+  ; 3 PMULUDQ and 2 PADDS and 4 shifts.
+  ;CHECK: cost of 9 {{.*}} mul
+  %A0 = mul <2 x i32> undef, undef
+  ;CHECK: cost of 9 {{.*}} mul
+  %A1 = mul <2 x i64> undef, undef
+  ;CHECK: cost of 18 {{.*}} mul
+  %A2 = mul <4 x i64> undef, undef
+  ret void
+}
+
+; SSE3: sse3mull
+define void @sse3mull() {
+  ; SSE3: cost of 6 {{.*}} mul
+  %A0 = mul <4 x i32> undef, undef
+  ret void
+  ; SSE3: avx2mull
+}
+
+; AVX2: avx2mull
+define void @avx2mull() {
+  ; AVX2: cost of 9 {{.*}} mul
+  %A0 = mul <4 x i64> undef, undef
+  ret void
+  ; AVX2: fmul
+}
 
+; CHECK: fmul
 define i32 @fmul(i32 %arg) {
   ;CHECK: cost of 1 {{.*}} fmul
   %A = fmul <4 x float> undef, undef
diff --git a/test/Analysis/CostModel/X86/intrinsic-cost.ll b/test/Analysis/CostModel/X86/intrinsic-cost.ll
new file mode 100644
index 0000000..e235a36
--- /dev/null
+++ b/test/Analysis/CostModel/X86/intrinsic-cost.ll
@@ -0,0 +1,32 @@
+; RUN: opt -S -mtriple=x86_64-apple-darwin -mcpu=core2 -cost-model -analyze < %s | FileCheck %s -check-prefix=CORE2
+; RUN: opt -S -mtriple=x86_64-apple-darwin -mcpu=corei7 -cost-model -analyze < %s | FileCheck %s -check-prefix=COREI7
+
+; If SSE4.1 roundps instruction is available it is cheap to lower, otherwise
+; it'll be scalarized into calls which are expensive.
+define void @test1(float* nocapture %f) nounwind {
+vector.ph:
+  br label %vector.body
+
+vector.body:                                      ; preds = %vector.body, %vector.ph
+  %index = phi i64 [ 0, %vector.ph ], [ %index.next, %vector.body ]
+  %0 = getelementptr inbounds float* %f, i64 %index
+  %1 = bitcast float* %0 to <4 x float>*
+  %wide.load = load <4 x float>* %1, align 4
+  %2 = call <4 x float> @llvm.ceil.v4f32(<4 x float> %wide.load)
+  store <4 x float> %2, <4 x float>* %1, align 4
+  %index.next = add i64 %index, 4
+  %3 = icmp eq i64 %index.next, 1024
+  br i1 %3, label %for.end, label %vector.body
+
+for.end:                                          ; preds = %vector.body
+  ret void
+
+; CORE2: Printing analysis 'Cost Model Analysis' for function 'test1':
+; CORE2: Cost Model: Found an estimated cost of 400 for instruction:   %2 = call <4 x float> @llvm.ceil.v4f32(<4 x float> %wide.load)
+
+; COREI7: Printing analysis 'Cost Model Analysis' for function 'test1':
+; COREI7: Cost Model: Found an estimated cost of 1 for instruction:   %2 = call <4 x float> @llvm.ceil.v4f32(<4 x float> %wide.load)
+
+}
+
+declare <4 x float> @llvm.ceil.v4f32(<4 x float>)  nounwind readnone