Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / 572645cf84060c0fc25cb91d38cb9079918b3a88 / . / test / CodeGen / X86 / iv-users-in-other-loops.ll
blob: 408fb20b8d89189ef4fc136c534163c3f57fcb2e [file] [log] [blame]
Dan Gohman36a09472009-09-08 23:54:48 +0000[diff] [blame]1; RUN: llc < %s -march=x86-64 -o %t
Dan Gohman572645c2010-02-12 10:34:29 +0000[diff] [blame^]2; RUN: not grep inc %t
Dale Johannesenc1acc3f2009-05-11 17:15:42 +0000[diff] [blame]3; RUN: grep dec %t | count 2
Dan Gohman7979b722010-01-22 00:46:49 +0000[diff] [blame]4; RUN: grep addq %t | count 13
Dan Gohman4e8a9852009-06-18 16:54:06 +0000[diff] [blame]5; RUN: not grep addb %t
Dan Gohman572645c2010-02-12 10:34:29 +0000[diff] [blame^]6; RUN: not grep leaq %t
7; RUN: not grep leal %t
8; RUN: not grep movq %t
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]9
10; IV users in each of the loops from other loops shouldn't cause LSR
11; to insert new induction variables. Previously it would create a
12; flood of new induction variables.
Dale Johannesenc1acc3f2009-05-11 17:15:42 +0000[diff] [blame]13; Also, the loop reversal should kick in once.
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]14
15target 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"
16target triple = "x86_64-unknown-linux-gnu"
17
18define void @foo(float* %A, i32 %IA, float* %B, i32 %IB, float* nocapture %C, i32 %N) nounwind {
19entry:
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]20 %0 = xor i32 %IA, 1 ; <i32> [#uses=1]
21 %1 = xor i32 %IB, 1 ; <i32> [#uses=1]
22 %2 = or i32 %1, %0 ; <i32> [#uses=1]
23 %3 = icmp eq i32 %2, 0 ; <i1> [#uses=1]
24 br i1 %3, label %bb2, label %bb13
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]25
26bb: ; preds = %bb3
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]27 %4 = load float* %A_addr.0, align 4 ; <float> [#uses=1]
28 %5 = load float* %B_addr.0, align 4 ; <float> [#uses=1]
29 %6 = fmul float %4, %5 ; <float> [#uses=1]
30 %7 = fadd float %6, %Sum0.0 ; <float> [#uses=1]
31 %indvar.next154 = add i64 %B_addr.0.rec, 1 ; <i64> [#uses=1]
32 br label %bb2
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]33
34bb2: ; preds = %entry, %bb
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]35 %B_addr.0.rec = phi i64 [ %indvar.next154, %bb ], [ 0, %entry ] ; <i64> [#uses=14]
36 %Sum0.0 = phi float [ %7, %bb ], [ 0.000000e+00, %entry ] ; <float> [#uses=5]
37 %indvar146 = trunc i64 %B_addr.0.rec to i32 ; <i32> [#uses=1]
38 %N_addr.0 = sub i32 %N, %indvar146 ; <i32> [#uses=6]
39 %A_addr.0 = getelementptr float* %A, i64 %B_addr.0.rec ; <float*> [#uses=4]
40 %B_addr.0 = getelementptr float* %B, i64 %B_addr.0.rec ; <float*> [#uses=4]
41 %8 = icmp sgt i32 %N_addr.0, 0 ; <i1> [#uses=1]
42 br i1 %8, label %bb3, label %bb4
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]43
44bb3: ; preds = %bb2
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]45 %9 = ptrtoint float* %A_addr.0 to i64 ; <i64> [#uses=1]
46 %10 = and i64 %9, 15 ; <i64> [#uses=1]
47 %11 = icmp eq i64 %10, 0 ; <i1> [#uses=1]
48 br i1 %11, label %bb4, label %bb
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]49
50bb4: ; preds = %bb3, %bb2
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]51 %12 = ptrtoint float* %B_addr.0 to i64 ; <i64> [#uses=1]
52 %13 = and i64 %12, 15 ; <i64> [#uses=1]
53 %14 = icmp eq i64 %13, 0 ; <i1> [#uses=1]
54 %15 = icmp sgt i32 %N_addr.0, 15 ; <i1> [#uses=2]
55 br i1 %14, label %bb6.preheader, label %bb10.preheader
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]56
57bb10.preheader: ; preds = %bb4
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]58 br i1 %15, label %bb9, label %bb12.loopexit
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]59
60bb6.preheader: ; preds = %bb4
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]61 br i1 %15, label %bb5, label %bb8.loopexit
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]62
63bb5: ; preds = %bb5, %bb6.preheader
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]64 %indvar143 = phi i64 [ 0, %bb6.preheader ], [ %indvar.next144, %bb5 ] ; <i64> [#uses=3]
65 %vSum0.072 = phi <4 x float> [ zeroinitializer, %bb6.preheader ], [ %21, %bb5 ] ; <<4 x float>> [#uses=1]
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]66 %vSum1.070 = phi <4 x float> [ zeroinitializer, %bb6.preheader ], [ %29, %bb5 ] ; <<4 x float>> [#uses=1]
67 %vSum2.069 = phi <4 x float> [ zeroinitializer, %bb6.preheader ], [ %37, %bb5 ] ; <<4 x float>> [#uses=1]
68 %vSum3.067 = phi <4 x float> [ zeroinitializer, %bb6.preheader ], [ %45, %bb5 ] ; <<4 x float>> [#uses=1]
69 %indvar145 = trunc i64 %indvar143 to i32 ; <i32> [#uses=1]
70 %tmp150 = mul i32 %indvar145, -16 ; <i32> [#uses=1]
71 %N_addr.268 = add i32 %tmp150, %N_addr.0 ; <i32> [#uses=1]
72 %A_addr.273.rec = shl i64 %indvar143, 4 ; <i64> [#uses=5]
73 %B_addr.0.sum180 = add i64 %B_addr.0.rec, %A_addr.273.rec ; <i64> [#uses=2]
74 %B_addr.271 = getelementptr float* %B, i64 %B_addr.0.sum180 ; <float*> [#uses=1]
75 %A_addr.273 = getelementptr float* %A, i64 %B_addr.0.sum180 ; <float*> [#uses=1]
76 tail call void asm sideeffect ";# foo", "~{dirflag},~{fpsr},~{flags}"() nounwind
77 %16 = bitcast float* %A_addr.273 to <4 x float>* ; <<4 x float>*> [#uses=1]
78 %17 = load <4 x float>* %16, align 16 ; <<4 x float>> [#uses=1]
79 %18 = bitcast float* %B_addr.271 to <4 x float>* ; <<4 x float>*> [#uses=1]
80 %19 = load <4 x float>* %18, align 16 ; <<4 x float>> [#uses=1]
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]81 %20 = fmul <4 x float> %17, %19 ; <<4 x float>> [#uses=1]
82 %21 = fadd <4 x float> %20, %vSum0.072 ; <<4 x float>> [#uses=2]
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]83 %A_addr.273.sum163 = or i64 %A_addr.273.rec, 4 ; <i64> [#uses=1]
84 %A_addr.0.sum175 = add i64 %B_addr.0.rec, %A_addr.273.sum163 ; <i64> [#uses=2]
85 %22 = getelementptr float* %A, i64 %A_addr.0.sum175 ; <float*> [#uses=1]
86 %23 = bitcast float* %22 to <4 x float>* ; <<4 x float>*> [#uses=1]
87 %24 = load <4 x float>* %23, align 16 ; <<4 x float>> [#uses=1]
88 %25 = getelementptr float* %B, i64 %A_addr.0.sum175 ; <float*> [#uses=1]
89 %26 = bitcast float* %25 to <4 x float>* ; <<4 x float>*> [#uses=1]
90 %27 = load <4 x float>* %26, align 16 ; <<4 x float>> [#uses=1]
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]91 %28 = fmul <4 x float> %24, %27 ; <<4 x float>> [#uses=1]
92 %29 = fadd <4 x float> %28, %vSum1.070 ; <<4 x float>> [#uses=2]
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]93 %A_addr.273.sum161 = or i64 %A_addr.273.rec, 8 ; <i64> [#uses=1]
94 %A_addr.0.sum174 = add i64 %B_addr.0.rec, %A_addr.273.sum161 ; <i64> [#uses=2]
95 %30 = getelementptr float* %A, i64 %A_addr.0.sum174 ; <float*> [#uses=1]
96 %31 = bitcast float* %30 to <4 x float>* ; <<4 x float>*> [#uses=1]
97 %32 = load <4 x float>* %31, align 16 ; <<4 x float>> [#uses=1]
98 %33 = getelementptr float* %B, i64 %A_addr.0.sum174 ; <float*> [#uses=1]
99 %34 = bitcast float* %33 to <4 x float>* ; <<4 x float>*> [#uses=1]
100 %35 = load <4 x float>* %34, align 16 ; <<4 x float>> [#uses=1]
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]101 %36 = fmul <4 x float> %32, %35 ; <<4 x float>> [#uses=1]
102 %37 = fadd <4 x float> %36, %vSum2.069 ; <<4 x float>> [#uses=2]
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]103 %A_addr.273.sum159 = or i64 %A_addr.273.rec, 12 ; <i64> [#uses=1]
104 %A_addr.0.sum173 = add i64 %B_addr.0.rec, %A_addr.273.sum159 ; <i64> [#uses=2]
105 %38 = getelementptr float* %A, i64 %A_addr.0.sum173 ; <float*> [#uses=1]
106 %39 = bitcast float* %38 to <4 x float>* ; <<4 x float>*> [#uses=1]
107 %40 = load <4 x float>* %39, align 16 ; <<4 x float>> [#uses=1]
108 %41 = getelementptr float* %B, i64 %A_addr.0.sum173 ; <float*> [#uses=1]
109 %42 = bitcast float* %41 to <4 x float>* ; <<4 x float>*> [#uses=1]
110 %43 = load <4 x float>* %42, align 16 ; <<4 x float>> [#uses=1]
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]111 %44 = fmul <4 x float> %40, %43 ; <<4 x float>> [#uses=1]
112 %45 = fadd <4 x float> %44, %vSum3.067 ; <<4 x float>> [#uses=2]
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]113 %.rec83 = add i64 %A_addr.273.rec, 16 ; <i64> [#uses=1]
114 %A_addr.0.sum172 = add i64 %B_addr.0.rec, %.rec83 ; <i64> [#uses=2]
115 %46 = getelementptr float* %A, i64 %A_addr.0.sum172 ; <float*> [#uses=1]
116 %47 = getelementptr float* %B, i64 %A_addr.0.sum172 ; <float*> [#uses=1]
117 %48 = add i32 %N_addr.268, -16 ; <i32> [#uses=2]
118 %49 = icmp sgt i32 %48, 15 ; <i1> [#uses=1]
119 %indvar.next144 = add i64 %indvar143, 1 ; <i64> [#uses=1]
120 br i1 %49, label %bb5, label %bb8.loopexit
121
122bb7: ; preds = %bb7, %bb8.loopexit
123 %indvar130 = phi i64 [ 0, %bb8.loopexit ], [ %indvar.next131, %bb7 ] ; <i64> [#uses=3]
124 %vSum0.260 = phi <4 x float> [ %vSum0.0.lcssa, %bb8.loopexit ], [ %55, %bb7 ] ; <<4 x float>> [#uses=1]
125 %indvar132 = trunc i64 %indvar130 to i32 ; <i32> [#uses=1]
126 %tmp133 = mul i32 %indvar132, -4 ; <i32> [#uses=1]
127 %N_addr.358 = add i32 %tmp133, %N_addr.2.lcssa ; <i32> [#uses=1]
128 %A_addr.361.rec = shl i64 %indvar130, 2 ; <i64> [#uses=3]
129 %B_addr.359 = getelementptr float* %B_addr.2.lcssa, i64 %A_addr.361.rec ; <float*> [#uses=1]
130 %A_addr.361 = getelementptr float* %A_addr.2.lcssa, i64 %A_addr.361.rec ; <float*> [#uses=1]
131 %50 = bitcast float* %A_addr.361 to <4 x float>* ; <<4 x float>*> [#uses=1]
132 %51 = load <4 x float>* %50, align 16 ; <<4 x float>> [#uses=1]
133 %52 = bitcast float* %B_addr.359 to <4 x float>* ; <<4 x float>*> [#uses=1]
134 %53 = load <4 x float>* %52, align 16 ; <<4 x float>> [#uses=1]
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]135 %54 = fmul <4 x float> %51, %53 ; <<4 x float>> [#uses=1]
136 %55 = fadd <4 x float> %54, %vSum0.260 ; <<4 x float>> [#uses=2]
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]137 %.rec85 = add i64 %A_addr.361.rec, 4 ; <i64> [#uses=2]
138 %56 = getelementptr float* %A_addr.2.lcssa, i64 %.rec85 ; <float*> [#uses=1]
139 %57 = getelementptr float* %B_addr.2.lcssa, i64 %.rec85 ; <float*> [#uses=1]
140 %58 = add i32 %N_addr.358, -4 ; <i32> [#uses=2]
141 %59 = icmp sgt i32 %58, 3 ; <i1> [#uses=1]
142 %indvar.next131 = add i64 %indvar130, 1 ; <i64> [#uses=1]
143 br i1 %59, label %bb7, label %bb13
144
145bb8.loopexit: ; preds = %bb5, %bb6.preheader
146 %A_addr.2.lcssa = phi float* [ %A_addr.0, %bb6.preheader ], [ %46, %bb5 ] ; <float*> [#uses=3]
147 %vSum0.0.lcssa = phi <4 x float> [ zeroinitializer, %bb6.preheader ], [ %21, %bb5 ] ; <<4 x float>> [#uses=2]
148 %B_addr.2.lcssa = phi float* [ %B_addr.0, %bb6.preheader ], [ %47, %bb5 ] ; <float*> [#uses=3]
149 %vSum1.0.lcssa = phi <4 x float> [ zeroinitializer, %bb6.preheader ], [ %29, %bb5 ] ; <<4 x float>> [#uses=2]
150 %vSum2.0.lcssa = phi <4 x float> [ zeroinitializer, %bb6.preheader ], [ %37, %bb5 ] ; <<4 x float>> [#uses=2]
151 %N_addr.2.lcssa = phi i32 [ %N_addr.0, %bb6.preheader ], [ %48, %bb5 ] ; <i32> [#uses=3]
152 %vSum3.0.lcssa = phi <4 x float> [ zeroinitializer, %bb6.preheader ], [ %45, %bb5 ] ; <<4 x float>> [#uses=2]
153 %60 = icmp sgt i32 %N_addr.2.lcssa, 3 ; <i1> [#uses=1]
154 br i1 %60, label %bb7, label %bb13
155
156bb9: ; preds = %bb9, %bb10.preheader
157 %indvar106 = phi i64 [ 0, %bb10.preheader ], [ %indvar.next107, %bb9 ] ; <i64> [#uses=3]
158 %vSum0.339 = phi <4 x float> [ zeroinitializer, %bb10.preheader ], [ %75, %bb9 ] ; <<4 x float>> [#uses=1]
159 %vSum1.237 = phi <4 x float> [ zeroinitializer, %bb10.preheader ], [ %80, %bb9 ] ; <<4 x float>> [#uses=1]
160 %vSum2.236 = phi <4 x float> [ zeroinitializer, %bb10.preheader ], [ %85, %bb9 ] ; <<4 x float>> [#uses=1]
161 %vSum3.234 = phi <4 x float> [ zeroinitializer, %bb10.preheader ], [ %90, %bb9 ] ; <<4 x float>> [#uses=1]
162 %indvar108 = trunc i64 %indvar106 to i32 ; <i32> [#uses=1]
163 %tmp113 = mul i32 %indvar108, -16 ; <i32> [#uses=1]
164 %N_addr.435 = add i32 %tmp113, %N_addr.0 ; <i32> [#uses=1]
165 %A_addr.440.rec = shl i64 %indvar106, 4 ; <i64> [#uses=5]
166 %B_addr.0.sum = add i64 %B_addr.0.rec, %A_addr.440.rec ; <i64> [#uses=2]
167 %B_addr.438 = getelementptr float* %B, i64 %B_addr.0.sum ; <float*> [#uses=1]
168 %A_addr.440 = getelementptr float* %A, i64 %B_addr.0.sum ; <float*> [#uses=1]
Dan Gohman926b0a22009-02-16 00:44:23 +0000[diff] [blame]169 %61 = bitcast float* %B_addr.438 to <4 x float>* ; <i8*> [#uses=1]
170 %62 = load <4 x float>* %61, align 1
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]171 %B_addr.438.sum169 = or i64 %A_addr.440.rec, 4 ; <i64> [#uses=1]
172 %B_addr.0.sum187 = add i64 %B_addr.0.rec, %B_addr.438.sum169 ; <i64> [#uses=2]
173 %63 = getelementptr float* %B, i64 %B_addr.0.sum187 ; <float*> [#uses=1]
Dan Gohman926b0a22009-02-16 00:44:23 +0000[diff] [blame]174 %64 = bitcast float* %63 to <4 x float>* ; <i8*> [#uses=1]
175 %65 = load <4 x float>* %64, align 1
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]176 %B_addr.438.sum168 = or i64 %A_addr.440.rec, 8 ; <i64> [#uses=1]
177 %B_addr.0.sum186 = add i64 %B_addr.0.rec, %B_addr.438.sum168 ; <i64> [#uses=2]
178 %66 = getelementptr float* %B, i64 %B_addr.0.sum186 ; <float*> [#uses=1]
Dan Gohman926b0a22009-02-16 00:44:23 +0000[diff] [blame]179 %67 = bitcast float* %66 to <4 x float>* ; <i8*> [#uses=1]
180 %68 = load <4 x float>* %67, align 1
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]181 %B_addr.438.sum167 = or i64 %A_addr.440.rec, 12 ; <i64> [#uses=1]
182 %B_addr.0.sum185 = add i64 %B_addr.0.rec, %B_addr.438.sum167 ; <i64> [#uses=2]
183 %69 = getelementptr float* %B, i64 %B_addr.0.sum185 ; <float*> [#uses=1]
Dan Gohman926b0a22009-02-16 00:44:23 +0000[diff] [blame]184 %70 = bitcast float* %69 to <4 x float>* ; <i8*> [#uses=1]
185 %71 = load <4 x float>* %70, align 1
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]186 %72 = bitcast float* %A_addr.440 to <4 x float>* ; <<4 x float>*> [#uses=1]
187 %73 = load <4 x float>* %72, align 16 ; <<4 x float>> [#uses=1]
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]188 %74 = fmul <4 x float> %73, %62 ; <<4 x float>> [#uses=1]
189 %75 = fadd <4 x float> %74, %vSum0.339 ; <<4 x float>> [#uses=2]
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]190 %76 = getelementptr float* %A, i64 %B_addr.0.sum187 ; <float*> [#uses=1]
191 %77 = bitcast float* %76 to <4 x float>* ; <<4 x float>*> [#uses=1]
192 %78 = load <4 x float>* %77, align 16 ; <<4 x float>> [#uses=1]
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]193 %79 = fmul <4 x float> %78, %65 ; <<4 x float>> [#uses=1]
194 %80 = fadd <4 x float> %79, %vSum1.237 ; <<4 x float>> [#uses=2]
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]195 %81 = getelementptr float* %A, i64 %B_addr.0.sum186 ; <float*> [#uses=1]
196 %82 = bitcast float* %81 to <4 x float>* ; <<4 x float>*> [#uses=1]
197 %83 = load <4 x float>* %82, align 16 ; <<4 x float>> [#uses=1]
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]198 %84 = fmul <4 x float> %83, %68 ; <<4 x float>> [#uses=1]
199 %85 = fadd <4 x float> %84, %vSum2.236 ; <<4 x float>> [#uses=2]
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]200 %86 = getelementptr float* %A, i64 %B_addr.0.sum185 ; <float*> [#uses=1]
201 %87 = bitcast float* %86 to <4 x float>* ; <<4 x float>*> [#uses=1]
202 %88 = load <4 x float>* %87, align 16 ; <<4 x float>> [#uses=1]
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]203 %89 = fmul <4 x float> %88, %71 ; <<4 x float>> [#uses=1]
204 %90 = fadd <4 x float> %89, %vSum3.234 ; <<4 x float>> [#uses=2]
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]205 %.rec89 = add i64 %A_addr.440.rec, 16 ; <i64> [#uses=1]
206 %A_addr.0.sum170 = add i64 %B_addr.0.rec, %.rec89 ; <i64> [#uses=2]
207 %91 = getelementptr float* %A, i64 %A_addr.0.sum170 ; <float*> [#uses=1]
208 %92 = getelementptr float* %B, i64 %A_addr.0.sum170 ; <float*> [#uses=1]
209 %93 = add i32 %N_addr.435, -16 ; <i32> [#uses=2]
210 %94 = icmp sgt i32 %93, 15 ; <i1> [#uses=1]
211 %indvar.next107 = add i64 %indvar106, 1 ; <i64> [#uses=1]
212 br i1 %94, label %bb9, label %bb12.loopexit
213
214bb11: ; preds = %bb11, %bb12.loopexit
215 %indvar = phi i64 [ 0, %bb12.loopexit ], [ %indvar.next, %bb11 ] ; <i64> [#uses=3]
216 %vSum0.428 = phi <4 x float> [ %vSum0.3.lcssa, %bb12.loopexit ], [ %100, %bb11 ] ; <<4 x float>> [#uses=1]
217 %indvar96 = trunc i64 %indvar to i32 ; <i32> [#uses=1]
218 %tmp = mul i32 %indvar96, -4 ; <i32> [#uses=1]
219 %N_addr.526 = add i32 %tmp, %N_addr.4.lcssa ; <i32> [#uses=1]
220 %A_addr.529.rec = shl i64 %indvar, 2 ; <i64> [#uses=3]
221 %B_addr.527 = getelementptr float* %B_addr.4.lcssa, i64 %A_addr.529.rec ; <float*> [#uses=1]
222 %A_addr.529 = getelementptr float* %A_addr.4.lcssa, i64 %A_addr.529.rec ; <float*> [#uses=1]
Dan Gohman926b0a22009-02-16 00:44:23 +0000[diff] [blame]223 %95 = bitcast float* %B_addr.527 to <4 x float>* ; <i8*> [#uses=1]
224 %96 = load <4 x float>* %95, align 1
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]225 %97 = bitcast float* %A_addr.529 to <4 x float>* ; <<4 x float>*> [#uses=1]
226 %98 = load <4 x float>* %97, align 16 ; <<4 x float>> [#uses=1]
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]227 %99 = fmul <4 x float> %98, %96 ; <<4 x float>> [#uses=1]
228 %100 = fadd <4 x float> %99, %vSum0.428 ; <<4 x float>> [#uses=2]
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]229 %.rec91 = add i64 %A_addr.529.rec, 4 ; <i64> [#uses=2]
230 %101 = getelementptr float* %A_addr.4.lcssa, i64 %.rec91 ; <float*> [#uses=1]
231 %102 = getelementptr float* %B_addr.4.lcssa, i64 %.rec91 ; <float*> [#uses=1]
232 %103 = add i32 %N_addr.526, -4 ; <i32> [#uses=2]
233 %104 = icmp sgt i32 %103, 3 ; <i1> [#uses=1]
234 %indvar.next = add i64 %indvar, 1 ; <i64> [#uses=1]
235 br i1 %104, label %bb11, label %bb13
236
237bb12.loopexit: ; preds = %bb9, %bb10.preheader
238 %A_addr.4.lcssa = phi float* [ %A_addr.0, %bb10.preheader ], [ %91, %bb9 ] ; <float*> [#uses=3]
239 %vSum0.3.lcssa = phi <4 x float> [ zeroinitializer, %bb10.preheader ], [ %75, %bb9 ] ; <<4 x float>> [#uses=2]
240 %B_addr.4.lcssa = phi float* [ %B_addr.0, %bb10.preheader ], [ %92, %bb9 ] ; <float*> [#uses=3]
241 %vSum1.2.lcssa = phi <4 x float> [ zeroinitializer, %bb10.preheader ], [ %80, %bb9 ] ; <<4 x float>> [#uses=2]
242 %vSum2.2.lcssa = phi <4 x float> [ zeroinitializer, %bb10.preheader ], [ %85, %bb9 ] ; <<4 x float>> [#uses=2]
243 %N_addr.4.lcssa = phi i32 [ %N_addr.0, %bb10.preheader ], [ %93, %bb9 ] ; <i32> [#uses=3]
244 %vSum3.2.lcssa = phi <4 x float> [ zeroinitializer, %bb10.preheader ], [ %90, %bb9 ] ; <<4 x float>> [#uses=2]
245 %105 = icmp sgt i32 %N_addr.4.lcssa, 3 ; <i1> [#uses=1]
246 br i1 %105, label %bb11, label %bb13
247
248bb13: ; preds = %bb12.loopexit, %bb11, %bb8.loopexit, %bb7, %entry
249 %Sum0.1 = phi float [ 0.000000e+00, %entry ], [ %Sum0.0, %bb7 ], [ %Sum0.0, %bb8.loopexit ], [ %Sum0.0, %bb11 ], [ %Sum0.0, %bb12.loopexit ] ; <float> [#uses=1]
250 %vSum3.1 = phi <4 x float> [ zeroinitializer, %entry ], [ %vSum3.0.lcssa, %bb7 ], [ %vSum3.0.lcssa, %bb8.loopexit ], [ %vSum3.2.lcssa, %bb11 ], [ %vSum3.2.lcssa, %bb12.loopexit ] ; <<4 x float>> [#uses=1]
251 %N_addr.1 = phi i32 [ %N, %entry ], [ %N_addr.2.lcssa, %bb8.loopexit ], [ %58, %bb7 ], [ %N_addr.4.lcssa, %bb12.loopexit ], [ %103, %bb11 ] ; <i32> [#uses=2]
252 %vSum2.1 = phi <4 x float> [ zeroinitializer, %entry ], [ %vSum2.0.lcssa, %bb7 ], [ %vSum2.0.lcssa, %bb8.loopexit ], [ %vSum2.2.lcssa, %bb11 ], [ %vSum2.2.lcssa, %bb12.loopexit ] ; <<4 x float>> [#uses=1]
253 %vSum1.1 = phi <4 x float> [ zeroinitializer, %entry ], [ %vSum1.0.lcssa, %bb7 ], [ %vSum1.0.lcssa, %bb8.loopexit ], [ %vSum1.2.lcssa, %bb11 ], [ %vSum1.2.lcssa, %bb12.loopexit ] ; <<4 x float>> [#uses=1]
254 %B_addr.1 = phi float* [ %B, %entry ], [ %B_addr.2.lcssa, %bb8.loopexit ], [ %57, %bb7 ], [ %B_addr.4.lcssa, %bb12.loopexit ], [ %102, %bb11 ] ; <float*> [#uses=1]
255 %vSum0.1 = phi <4 x float> [ zeroinitializer, %entry ], [ %vSum0.0.lcssa, %bb8.loopexit ], [ %55, %bb7 ], [ %vSum0.3.lcssa, %bb12.loopexit ], [ %100, %bb11 ] ; <<4 x float>> [#uses=1]
256 %A_addr.1 = phi float* [ %A, %entry ], [ %A_addr.2.lcssa, %bb8.loopexit ], [ %56, %bb7 ], [ %A_addr.4.lcssa, %bb12.loopexit ], [ %101, %bb11 ] ; <float*> [#uses=1]
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]257 %106 = fadd <4 x float> %vSum0.1, %vSum2.1 ; <<4 x float>> [#uses=1]
258 %107 = fadd <4 x float> %vSum1.1, %vSum3.1 ; <<4 x float>> [#uses=1]
259 %108 = fadd <4 x float> %106, %107 ; <<4 x float>> [#uses=4]
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]260 %tmp23 = extractelement <4 x float> %108, i32 0 ; <float> [#uses=1]
261 %tmp21 = extractelement <4 x float> %108, i32 1 ; <float> [#uses=1]
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]262 %109 = fadd float %tmp23, %tmp21 ; <float> [#uses=1]
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]263 %tmp19 = extractelement <4 x float> %108, i32 2 ; <float> [#uses=1]
264 %tmp17 = extractelement <4 x float> %108, i32 3 ; <float> [#uses=1]
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]265 %110 = fadd float %tmp19, %tmp17 ; <float> [#uses=1]
266 %111 = fadd float %109, %110 ; <float> [#uses=1]
267 %Sum0.254 = fadd float %111, %Sum0.1 ; <float> [#uses=2]
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]268 %112 = icmp sgt i32 %N_addr.1, 0 ; <i1> [#uses=1]
269 br i1 %112, label %bb.nph56, label %bb16
270
271bb.nph56: ; preds = %bb13
272 %tmp. = zext i32 %N_addr.1 to i64 ; <i64> [#uses=1]
273 br label %bb14
274
275bb14: ; preds = %bb14, %bb.nph56
276 %indvar117 = phi i64 [ 0, %bb.nph56 ], [ %indvar.next118, %bb14 ] ; <i64> [#uses=3]
277 %Sum0.255 = phi float [ %Sum0.254, %bb.nph56 ], [ %Sum0.2, %bb14 ] ; <float> [#uses=1]
278 %tmp.122 = sext i32 %IB to i64 ; <i64> [#uses=1]
279 %B_addr.652.rec = mul i64 %indvar117, %tmp.122 ; <i64> [#uses=1]
280 %tmp.124 = sext i32 %IA to i64 ; <i64> [#uses=1]
281 %A_addr.653.rec = mul i64 %indvar117, %tmp.124 ; <i64> [#uses=1]
282 %B_addr.652 = getelementptr float* %B_addr.1, i64 %B_addr.652.rec ; <float*> [#uses=1]
283 %A_addr.653 = getelementptr float* %A_addr.1, i64 %A_addr.653.rec ; <float*> [#uses=1]
284 %113 = load float* %A_addr.653, align 4 ; <float> [#uses=1]
285 %114 = load float* %B_addr.652, align 4 ; <float> [#uses=1]
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]286 %115 = fmul float %113, %114 ; <float> [#uses=1]
287 %Sum0.2 = fadd float %115, %Sum0.255 ; <float> [#uses=2]
Dan Gohman9194e8b2009-02-13 03:58:31 +0000[diff] [blame]288 %indvar.next118 = add i64 %indvar117, 1 ; <i64> [#uses=2]
289 %exitcond = icmp eq i64 %indvar.next118, %tmp. ; <i1> [#uses=1]
290 br i1 %exitcond, label %bb16, label %bb14
291
292bb16: ; preds = %bb14, %bb13
293 %Sum0.2.lcssa = phi float [ %Sum0.254, %bb13 ], [ %Sum0.2, %bb14 ] ; <float> [#uses=1]
294 store float %Sum0.2.lcssa, float* %C, align 4
295 ret void
296}