From 62848c9c2459cd724a884f008044807a54dece30 Mon Sep 17 00:00:00 2001 From: Pekka Jaaskelainen Date: Fri, 15 Feb 2013 21:50:19 +0000 Subject: [PATCH] Forgot to 'svn add' the LoopVectorizer tests for the new parallel loop metadata, sorry. llvm-svn: 175311 --- .../X86/parallel-loops-after-reg2mem.ll | 52 ++++++++++ .../Transforms/LoopVectorize/X86/parallel-loops.ll | 114 +++++++++++++++++++++ 2 files changed, 166 insertions(+) create mode 100644 llvm/test/Transforms/LoopVectorize/X86/parallel-loops-after-reg2mem.ll create mode 100644 llvm/test/Transforms/LoopVectorize/X86/parallel-loops.ll diff --git a/llvm/test/Transforms/LoopVectorize/X86/parallel-loops-after-reg2mem.ll b/llvm/test/Transforms/LoopVectorize/X86/parallel-loops-after-reg2mem.ll new file mode 100644 index 0000000..452d0df --- /dev/null +++ b/llvm/test/Transforms/LoopVectorize/X86/parallel-loops-after-reg2mem.ll @@ -0,0 +1,52 @@ +; RUN: opt < %s -loop-vectorize -force-vector-unroll=1 -force-vector-width=4 -dce -instcombine -S | 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-unknown-linux-gnu" + +; The parallel loop has been invalidated by the new memory accesses introduced +; by reg2mem (Loop::isParallel() starts to return false). Ensure the loop is +; now non-vectorizable. + +;CHECK-NOT: <4 x i32> +define void @parallel_loop(i32* nocapture %a, i32* nocapture %b) nounwind uwtable { +entry: + %indvars.iv.next.reg2mem = alloca i64 + %indvars.iv.reg2mem = alloca i64 + %"reg2mem alloca point" = bitcast i32 0 to i32 + store i64 0, i64* %indvars.iv.reg2mem + br label %for.body + +for.body: ; preds = %for.body.for.body_crit_edge, %entry + %indvars.iv.reload = load i64* %indvars.iv.reg2mem + %arrayidx = getelementptr inbounds i32* %b, i64 %indvars.iv.reload + %0 = load i32* %arrayidx, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !3 + %arrayidx2 = getelementptr inbounds i32* %a, i64 %indvars.iv.reload + %1 = load i32* %arrayidx2, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !3 + %idxprom3 = sext i32 %1 to i64 + %arrayidx4 = getelementptr inbounds i32* %a, i64 %idxprom3 + store i32 %0, i32* %arrayidx4, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !3 + %indvars.iv.next = add i64 %indvars.iv.reload, 1 + ; A new store without the parallel metadata here: + store i64 %indvars.iv.next, i64* %indvars.iv.next.reg2mem + %indvars.iv.next.reload1 = load i64* %indvars.iv.next.reg2mem + %arrayidx6 = getelementptr inbounds i32* %b, i64 %indvars.iv.next.reload1 + %2 = load i32* %arrayidx6, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !3 + store i32 %2, i32* %arrayidx2, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !3 + %indvars.iv.next.reload = load i64* %indvars.iv.next.reg2mem + %lftr.wideiv = trunc i64 %indvars.iv.next.reload to i32 + %exitcond = icmp eq i32 %lftr.wideiv, 512 + br i1 %exitcond, label %for.end, label %for.body.for.body_crit_edge, !llvm.loop.parallel !3 + +for.body.for.body_crit_edge: ; preds = %for.body + %indvars.iv.next.reload2 = load i64* %indvars.iv.next.reg2mem + store i64 %indvars.iv.next.reload2, i64* %indvars.iv.reg2mem + br label %for.body + +for.end: ; preds = %for.body + ret void +} + +!0 = metadata !{metadata !"int", metadata !1} +!1 = metadata !{metadata !"omnipotent char", metadata !2} +!2 = metadata !{metadata !"Simple C/C++ TBAA"} +!3 = metadata !{metadata !3} diff --git a/llvm/test/Transforms/LoopVectorize/X86/parallel-loops.ll b/llvm/test/Transforms/LoopVectorize/X86/parallel-loops.ll new file mode 100644 index 0000000..f648722 --- /dev/null +++ b/llvm/test/Transforms/LoopVectorize/X86/parallel-loops.ll @@ -0,0 +1,114 @@ +; RUN: opt < %s -loop-vectorize -force-vector-unroll=1 -force-vector-width=4 -dce -instcombine -S | 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-unknown-linux-gnu" + +; A tricky loop: +; +; void loop(int *a, int *b) { +; for (int i = 0; i < 512; ++i) { +; a[a[i]] = b[i]; +; a[i] = b[i+1]; +; } +;} + +;CHECK: @loop +;CHECK-NOT: <4 x i32> +define void @loop(i32* nocapture %a, i32* nocapture %b) nounwind uwtable { +entry: + br label %for.body + +for.body: ; preds = %for.body, %entry + %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ] + %arrayidx = getelementptr inbounds i32* %b, i64 %indvars.iv + %0 = load i32* %arrayidx, align 4, !tbaa !0 + %arrayidx2 = getelementptr inbounds i32* %a, i64 %indvars.iv + %1 = load i32* %arrayidx2, align 4, !tbaa !0 + %idxprom3 = sext i32 %1 to i64 + %arrayidx4 = getelementptr inbounds i32* %a, i64 %idxprom3 + store i32 %0, i32* %arrayidx4, align 4, !tbaa !0 + %indvars.iv.next = add i64 %indvars.iv, 1 + %arrayidx6 = getelementptr inbounds i32* %b, i64 %indvars.iv.next + %2 = load i32* %arrayidx6, align 4, !tbaa !0 + store i32 %2, i32* %arrayidx2, align 4, !tbaa !0 + %lftr.wideiv = trunc i64 %indvars.iv.next to i32 + %exitcond = icmp eq i32 %lftr.wideiv, 512 + br i1 %exitcond, label %for.end, label %for.body + +for.end: ; preds = %for.body + ret void +} + +; The same loop with parallel loop metadata added to the loop branch +; and the memory instructions. + +;CHECK: @parallel_loop +;CHECK: <4 x i32> +define void @parallel_loop(i32* nocapture %a, i32* nocapture %b) nounwind uwtable { +entry: + br label %for.body + +for.body: ; preds = %for.body, %entry + %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ] + %arrayidx = getelementptr inbounds i32* %b, i64 %indvars.iv + %0 = load i32* %arrayidx, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !3 + %arrayidx2 = getelementptr inbounds i32* %a, i64 %indvars.iv + %1 = load i32* %arrayidx2, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !3 + %idxprom3 = sext i32 %1 to i64 + %arrayidx4 = getelementptr inbounds i32* %a, i64 %idxprom3 + ; This store might have originated from inlining a function with a parallel + ; loop. Refers to a list with the "original loop reference" (!4) also included. + store i32 %0, i32* %arrayidx4, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !5 + %indvars.iv.next = add i64 %indvars.iv, 1 + %arrayidx6 = getelementptr inbounds i32* %b, i64 %indvars.iv.next + %2 = load i32* %arrayidx6, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !3 + store i32 %2, i32* %arrayidx2, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !3 + %lftr.wideiv = trunc i64 %indvars.iv.next to i32 + %exitcond = icmp eq i32 %lftr.wideiv, 512 + br i1 %exitcond, label %for.end, label %for.body, !llvm.loop.parallel !3 + +for.end: ; preds = %for.body + ret void +} + +; The same loop with an illegal parallel loop metadata: the memory +; accesses refer to a different loop's identifier. + +;CHECK: @mixed_metadata +;CHECK-NOT: <4 x i32> + +define void @mixed_metadata(i32* nocapture %a, i32* nocapture %b) nounwind uwtable { +entry: + br label %for.body + +for.body: ; preds = %for.body, %entry + %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ] + %arrayidx = getelementptr inbounds i32* %b, i64 %indvars.iv + %0 = load i32* %arrayidx, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !6 + %arrayidx2 = getelementptr inbounds i32* %a, i64 %indvars.iv + %1 = load i32* %arrayidx2, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !6 + %idxprom3 = sext i32 %1 to i64 + %arrayidx4 = getelementptr inbounds i32* %a, i64 %idxprom3 + ; This refers to the loop marked with !7 which we are not in at the moment. + ; It should prevent detecting as a parallel loop. + store i32 %0, i32* %arrayidx4, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !7 + %indvars.iv.next = add i64 %indvars.iv, 1 + %arrayidx6 = getelementptr inbounds i32* %b, i64 %indvars.iv.next + %2 = load i32* %arrayidx6, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !6 + store i32 %2, i32* %arrayidx2, align 4, !tbaa !0, !llvm.mem.parallel_loop_access !6 + %lftr.wideiv = trunc i64 %indvars.iv.next to i32 + %exitcond = icmp eq i32 %lftr.wideiv, 512 + br i1 %exitcond, label %for.end, label %for.body, !llvm.loop.parallel !6 + +for.end: ; preds = %for.body + ret void +} + +!0 = metadata !{metadata !"int", metadata !1} +!1 = metadata !{metadata !"omnipotent char", metadata !2} +!2 = metadata !{metadata !"Simple C/C++ TBAA"} +!3 = metadata !{metadata !3} +!4 = metadata !{metadata !4} +!5 = metadata !{metadata !3, metadata !4} +!6 = metadata !{metadata !6} +!7 = metadata !{metadata !7} -- 2.7.4