mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-23 19:23:23 +01:00
db3c93e416
This will remove suboptimal branching from the generated ll/sc loops. The extra simplification pass affects a lot of testcases, which have been modified to accommodate this change: either by modifying the test to become immune to the CFG simplification, or (less preferablt) by adding option -hexagon-initial-cfg-clenaup=0. llvm-svn: 338774
63 lines
3.3 KiB
LLVM
63 lines
3.3 KiB
LLVM
; RUN: llc -march=hexagon < %s | FileCheck %s
|
|
|
|
; Test that we generate a .cur
|
|
|
|
; CHECK: v{{[0-9]*}}.cur
|
|
|
|
; Function Attrs: nounwind
|
|
define void @f0(i8* noalias nocapture readonly %a0, i32 %a1, i32 %a2, <16 x i32>* %a3, <16 x i32>* %a4) #0 {
|
|
b0:
|
|
br i1 undef, label %b1, label %b3
|
|
|
|
b1: ; preds = %b0
|
|
br label %b2
|
|
|
|
b2: ; preds = %b2, %b1
|
|
%v0 = phi i8* [ %a0, %b1 ], [ %v4, %b2 ]
|
|
%v1 = phi i32 [ 0, %b1 ], [ %v23, %b2 ]
|
|
%v2 = phi <16 x i32> [ zeroinitializer, %b1 ], [ %v6, %b2 ]
|
|
%v3 = phi <16 x i32> [ zeroinitializer, %b1 ], [ zeroinitializer, %b2 ]
|
|
%v4 = getelementptr inbounds i8, i8* %v0, i32 64
|
|
%v5 = bitcast i8* %v4 to <16 x i32>*
|
|
%v6 = load <16 x i32>, <16 x i32>* %v5, align 64, !tbaa !0
|
|
%v7 = load <16 x i32>, <16 x i32>* %a3, align 64, !tbaa !0
|
|
%v8 = tail call <16 x i32> @llvm.hexagon.V6.valignbi(<16 x i32> %v6, <16 x i32> %v2, i32 4)
|
|
%v9 = tail call <16 x i32> @llvm.hexagon.V6.valignbi(<16 x i32> zeroinitializer, <16 x i32> %v3, i32 4)
|
|
%v10 = tail call <16 x i32> @llvm.hexagon.V6.valignbi(<16 x i32> %v7, <16 x i32> zeroinitializer, i32 4)
|
|
%v11 = tail call <32 x i32> @llvm.hexagon.V6.vcombine(<16 x i32> %v8, <16 x i32> %v2)
|
|
%v12 = tail call <32 x i32> @llvm.hexagon.V6.vcombine(<16 x i32> %v10, <16 x i32> zeroinitializer)
|
|
%v13 = tail call <32 x i32> @llvm.hexagon.V6.vrmpybusi(<32 x i32> %v11, i32 0, i32 0)
|
|
%v14 = tail call <32 x i32> @llvm.hexagon.V6.vrmpybusi.acc(<32 x i32> %v13, <32 x i32> zeroinitializer, i32 undef, i32 0)
|
|
%v15 = tail call <32 x i32> @llvm.hexagon.V6.vrmpybusi.acc(<32 x i32> %v14, <32 x i32> undef, i32 undef, i32 0)
|
|
%v16 = tail call <16 x i32> @llvm.hexagon.V6.hi(<32 x i32> %v15)
|
|
%v17 = tail call <16 x i32> @llvm.hexagon.V6.vasrwh(<16 x i32> %v16, <16 x i32> undef, i32 %a1)
|
|
%v18 = tail call <16 x i32> @llvm.hexagon.V6.vsathub(<16 x i32> undef, <16 x i32> %v17)
|
|
store <16 x i32> %v18, <16 x i32>* %a3, align 64, !tbaa !0
|
|
%v19 = tail call <32 x i32> @llvm.hexagon.V6.vrmpybusi.acc(<32 x i32> zeroinitializer, <32 x i32> %v12, i32 undef, i32 1)
|
|
%v20 = tail call <16 x i32> @llvm.hexagon.V6.hi(<32 x i32> %v19)
|
|
%v21 = tail call <16 x i32> @llvm.hexagon.V6.vasrwh(<16 x i32> %v20, <16 x i32> undef, i32 %a1)
|
|
%v22 = tail call <16 x i32> @llvm.hexagon.V6.vsathub(<16 x i32> %v21, <16 x i32> undef)
|
|
store <16 x i32> %v22, <16 x i32>* %a4, align 64, !tbaa !0
|
|
%v23 = add nsw i32 %v1, 64
|
|
%v24 = icmp slt i32 %v23, %a2
|
|
br i1 %v24, label %b2, label %b3
|
|
|
|
b3: ; preds = %b2, %b0
|
|
ret void
|
|
}
|
|
|
|
declare <16 x i32> @llvm.hexagon.V6.valignbi(<16 x i32>, <16 x i32>, i32) #1
|
|
declare <32 x i32> @llvm.hexagon.V6.vcombine(<16 x i32>, <16 x i32>) #1
|
|
declare <32 x i32> @llvm.hexagon.V6.vrmpybusi(<32 x i32>, i32, i32) #1
|
|
declare <32 x i32> @llvm.hexagon.V6.vrmpybusi.acc(<32 x i32>, <32 x i32>, i32, i32) #1
|
|
declare <16 x i32> @llvm.hexagon.V6.vasrwh(<16 x i32>, <16 x i32>, i32) #1
|
|
declare <16 x i32> @llvm.hexagon.V6.hi(<32 x i32>) #1
|
|
declare <16 x i32> @llvm.hexagon.V6.vsathub(<16 x i32>, <16 x i32>) #1
|
|
|
|
attributes #0 = { nounwind "target-cpu"="hexagonv60" "target-features"="+hvxv60,+hvx-length64b" }
|
|
attributes #1 = { nounwind readnone }
|
|
|
|
!0 = !{!1, !1, i64 0}
|
|
!1 = !{!"omnipotent char", !2, i64 0}
|
|
!2 = !{!"Simple C/C++ TBAA"}
|