| Tobias Grosser | 7580537 | 2011-04-29 06:27:02 +0000 | [diff] [blame] | 1 | ; ModuleID = 'parallel_loop_simple2.s' |
| 2 | ; RUN: opt %loadPolly %defaultOpts -polly-cloog -polly-codegen -enable-polly-openmp -analyze < %s | FileCheck %s |
| 3 | 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" |
| 4 | target triple = "x86_64-unknown-linux-gnu" |
| 5 | |
| 6 | @C = common global [1024 x float] zeroinitializer, align 16 |
| 7 | @X = common global [1024 x float] zeroinitializer, align 16 |
| 8 | |
| 9 | define float @parallel_loop_simple2() nounwind { |
| 10 | bb: |
| 11 | br label %bb5 |
| 12 | |
| 13 | bb5: ; preds = %bb7, %bb |
| 14 | %indvar1 = phi i64 [ %indvar.next2, %bb7 ], [ 0, %bb ] |
| 15 | %scevgep4 = getelementptr [1024 x float]* @C, i64 0, i64 %indvar1 |
| 16 | %j.0 = trunc i64 %indvar1 to i32 |
| 17 | %exitcond3 = icmp ne i64 %indvar1, 1024 |
| 18 | br i1 %exitcond3, label %bb6, label %bb8 |
| 19 | |
| 20 | bb6: ; preds = %bb5 |
| 21 | %tmp = sitofp i32 %j.0 to float |
| 22 | store float %tmp, float* %scevgep4, align 4 |
| 23 | br label %bb7 |
| 24 | |
| 25 | bb7: ; preds = %bb6 |
| 26 | %indvar.next2 = add i64 %indvar1, 1 |
| 27 | br label %bb5 |
| 28 | |
| 29 | bb8: ; preds = %bb5 |
| 30 | br label %bb9 |
| 31 | |
| 32 | bb9: ; preds = %bb14, %bb8 |
| 33 | %indvar = phi i64 [ %indvar.next, %bb14 ], [ 0, %bb8 ] |
| 34 | %scevgep = getelementptr [1024 x float]* @X, i64 0, i64 %indvar |
| 35 | %exitcond = icmp ne i64 %indvar, 1024 |
| 36 | br i1 %exitcond, label %bb10, label %bb15 |
| 37 | |
| 38 | bb10: ; preds = %bb9 |
| 39 | %tmp11 = load float* %scevgep, align 4 |
| 40 | %tmp12 = load float* %scevgep, align 4 |
| 41 | %tmp13 = fadd float %tmp12, %tmp11 |
| 42 | store float %tmp13, float* %scevgep, align 4 |
| 43 | br label %bb14 |
| 44 | |
| 45 | bb14: ; preds = %bb10 |
| 46 | %indvar.next = add i64 %indvar, 1 |
| 47 | br label %bb9 |
| 48 | |
| 49 | bb15: ; preds = %bb9 |
| 50 | %tmp16 = load float* getelementptr inbounds ([1024 x float]* @C, i64 0, i64 42), align 8 |
| 51 | %tmp17 = load float* getelementptr inbounds ([1024 x float]* @X, i64 0, i64 42), align 8 |
| 52 | %tmp18 = fadd float %tmp16, %tmp17 |
| 53 | ret float %tmp18 |
| 54 | } |
| 55 | |
| 56 | ; CHECK: for (c2=0;c2<=1023;c2++) { |
| 57 | ; CHECK: Stmt_bb6(c2); |
| 58 | ; CHECK: } |
| 59 | ; CHECK: for (c2=0;c2<=1023;c2++) { |
| 60 | ; CHECK: Stmt_bb10(c2); |
| 61 | ; CHECK: } |
| 62 | ; CHECK: Parallel loop with iterator 'c2' generated |
| 63 | ; CHECK: Parallel loop with iterator 'c2' generated |
| 64 | ; CHECK-NOT: Parallel loop |