mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-24 03:33:20 +01:00
7d64810efd
The benchmarking summarized in http://lists.llvm.org/pipermail/llvm-dev/2017-May/113525.html showed this is beneficial for a wide range of cores. As is to be expected, quite a few small adaptations are needed to the regressions tests, as the difference in scheduling results in: - Quite a few small instruction schedule differences. - A few changes in register allocation decisions caused by different instruction schedules. - A few changes in IfConversion decisions, due to a difference in instruction schedule and/or the estimated cost of a branch mispredict. llvm-svn: 306514
94 lines
2.8 KiB
LLVM
94 lines
2.8 KiB
LLVM
; RUN: llc < %s -mtriple=thumbv7-apple-ios -arm-atomic-cfg-tidy=0 | FileCheck %s
|
|
; RUN: llc < %s -mtriple=thumbv7-apple-ios -arm-atomic-cfg-tidy=0 -arm-default-it | FileCheck %s
|
|
; RUN: llc < %s -mtriple=thumbv8-apple-ios -arm-atomic-cfg-tidy=0 -arm-no-restrict-it | FileCheck %s
|
|
|
|
define void @foo(i32 %X, i32 %Y) {
|
|
entry:
|
|
; CHECK-LABEL: foo:
|
|
; CHECK: it ne
|
|
; CHECK: cmpne
|
|
; CHECK: it hi
|
|
; CHECK: bxhi lr
|
|
%tmp1 = icmp ult i32 %X, 4 ; <i1> [#uses=1]
|
|
%tmp4 = icmp eq i32 %Y, 0 ; <i1> [#uses=1]
|
|
%tmp7 = or i1 %tmp4, %tmp1 ; <i1> [#uses=1]
|
|
br i1 %tmp7, label %cond_true, label %UnifiedReturnBlock
|
|
|
|
cond_true: ; preds = %entry
|
|
%tmp10 = call i32 (...) @bar( ) ; <i32> [#uses=0]
|
|
ret void
|
|
|
|
UnifiedReturnBlock: ; preds = %entry
|
|
ret void
|
|
}
|
|
|
|
declare i32 @bar(...)
|
|
|
|
; FIXME: Need post-ifcvt branch folding to get rid of the extra br at end of BB1.
|
|
|
|
%struct.quad_struct = type { i32, i32, %struct.quad_struct*, %struct.quad_struct*, %struct.quad_struct*, %struct.quad_struct*, %struct.quad_struct* }
|
|
|
|
define fastcc i32 @CountTree(%struct.quad_struct* %tree) {
|
|
entry:
|
|
; CHECK-LABEL: CountTree:
|
|
; CHECK: bne
|
|
; CHECK: cmp
|
|
; CHECK: it eq
|
|
; CHECK: cmpeq
|
|
br label %tailrecurse
|
|
|
|
tailrecurse: ; preds = %bb, %entry
|
|
%tmp6 = load %struct.quad_struct*, %struct.quad_struct** null ; <%struct.quad_struct*> [#uses=1]
|
|
%tmp9 = load %struct.quad_struct*, %struct.quad_struct** null ; <%struct.quad_struct*> [#uses=2]
|
|
%tmp12 = load %struct.quad_struct*, %struct.quad_struct** null ; <%struct.quad_struct*> [#uses=1]
|
|
%tmp14 = icmp eq %struct.quad_struct* null, null ; <i1> [#uses=1]
|
|
%tmp17 = icmp eq %struct.quad_struct* %tmp6, null ; <i1> [#uses=1]
|
|
%tmp23 = icmp eq %struct.quad_struct* %tmp9, null ; <i1> [#uses=1]
|
|
%tmp29 = icmp eq %struct.quad_struct* %tmp12, null ; <i1> [#uses=1]
|
|
%bothcond = and i1 %tmp17, %tmp14 ; <i1> [#uses=1]
|
|
%bothcond1 = and i1 %bothcond, %tmp23 ; <i1> [#uses=1]
|
|
%bothcond2 = and i1 %bothcond1, %tmp29 ; <i1> [#uses=1]
|
|
br i1 %bothcond2, label %return, label %bb
|
|
|
|
bb: ; preds = %tailrecurse
|
|
%tmp41 = tail call fastcc i32 @CountTree( %struct.quad_struct* %tmp9 ) ; <i32> [#uses=0]
|
|
br label %tailrecurse
|
|
|
|
return: ; preds = %tailrecurse
|
|
ret i32 0
|
|
}
|
|
|
|
%struct.SString = type { i8*, i32, i32 }
|
|
|
|
declare void @abort()
|
|
|
|
define fastcc void @t1(%struct.SString* %word, i8 signext %c) {
|
|
entry:
|
|
; CHECK-LABEL: t1:
|
|
; CHECK: it ne
|
|
; CHECK: bxne lr
|
|
%tmp1 = icmp eq %struct.SString* %word, null ; <i1> [#uses=1]
|
|
br i1 %tmp1, label %cond_true, label %cond_false
|
|
|
|
cond_true: ; preds = %entry
|
|
tail call void @abort( )
|
|
unreachable
|
|
|
|
cond_false: ; preds = %entry
|
|
ret void
|
|
}
|
|
|
|
define fastcc void @t2() nounwind {
|
|
entry:
|
|
; CHECK-LABEL: t2:
|
|
; CHECK: cmp r0, #0
|
|
; CHECK: %growMapping.exit
|
|
br i1 undef, label %bb.i.i3, label %growMapping.exit
|
|
|
|
bb.i.i3: ; preds = %entry
|
|
unreachable
|
|
|
|
growMapping.exit: ; preds = %entry
|
|
unreachable
|
|
}
|