1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-29 23:12:55 +01:00
llvm-mirror/test/CodeGen/X86/cmov.ll
Andrew Trick e3e67d4a0a Enable MI Sched for x86.
This changes the SelectionDAG scheduling preference to source
order. Soon, the SelectionDAG scheduler can be bypassed saving
a nice chunk of compile time.

Performance differences that result from this change are often a
consequence of register coalescing. The register coalescer is far from
perfect. Bugs can be filed for deficiencies.

On x86 SandyBridge/Haswell, the source order schedule is often
preserved, particularly for small blocks.

Register pressure is generally improved over the SD scheduler's ILP
mode. However, we are still able to handle large blocks that require
latency hiding, unlike the SD scheduler's BURR mode. MI scheduler also
attempts to discover the critical path in single-block loops and
adjust heuristics accordingly.

The MI scheduler relies on the new machine model. This is currently
unimplemented for AVX, so we may not be generating the best code yet.

Unit tests are updated so they don't depend on SD scheduling heuristics.

llvm-svn: 192750
2013-10-15 23:33:07 +00:00

158 lines
5.2 KiB
LLVM

; RUN: llc < %s -mtriple=x86_64-apple-darwin10 -disable-cgp-select2branch | 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"
define i32 @test1(i32 %x, i32 %n, i32 %w, i32* %vp) nounwind readnone {
entry:
; CHECK-LABEL: test1:
; CHECK: btl
; CHECK-NEXT: movl $12, %eax
; CHECK-NEXT: cmovael (%rcx), %eax
; CHECK-NEXT: ret
%0 = lshr i32 %x, %n ; <i32> [#uses=1]
%1 = and i32 %0, 1 ; <i32> [#uses=1]
%toBool = icmp eq i32 %1, 0 ; <i1> [#uses=1]
%v = load i32* %vp
%.0 = select i1 %toBool, i32 %v, i32 12 ; <i32> [#uses=1]
ret i32 %.0
}
define i32 @test2(i32 %x, i32 %n, i32 %w, i32* %vp) nounwind readnone {
entry:
; CHECK-LABEL: test2:
; CHECK: btl
; CHECK-NEXT: movl $12, %eax
; CHECK-NEXT: cmovbl (%rcx), %eax
; CHECK-NEXT: ret
%0 = lshr i32 %x, %n ; <i32> [#uses=1]
%1 = and i32 %0, 1 ; <i32> [#uses=1]
%toBool = icmp eq i32 %1, 0 ; <i1> [#uses=1]
%v = load i32* %vp
%.0 = select i1 %toBool, i32 12, i32 %v ; <i32> [#uses=1]
ret i32 %.0
}
; x86's 32-bit cmov doesn't clobber the high 32 bits of the destination
; if the condition is false. An explicit zero-extend (movl) is needed
; after the cmov.
declare void @bar(i64) nounwind
define void @test3(i64 %a, i64 %b, i1 %p) nounwind {
; CHECK-LABEL: test3:
; CHECK: cmovnel %edi, %esi
; CHECK-NEXT: movl %esi, %edi
%c = trunc i64 %a to i32
%d = trunc i64 %b to i32
%e = select i1 %p, i32 %c, i32 %d
%f = zext i32 %e to i64
call void @bar(i64 %f)
ret void
}
; CodeGen shouldn't try to do a setne after an expanded 8-bit conditional
; move without recomputing EFLAGS, because the expansion of the conditional
; move with control flow may clobber EFLAGS (e.g., with xor, to set the
; register to zero).
; The test is a little awkward; the important part is that there's a test before the
; setne.
; PR4814
@g_3 = external global i8 ; <i8*> [#uses=1]
@g_96 = external global i8 ; <i8*> [#uses=2]
@g_100 = external global i8 ; <i8*> [#uses=2]
@_2E_str = external constant [15 x i8], align 1 ; <[15 x i8]*> [#uses=1]
define i32 @test4() nounwind {
entry:
%0 = load i8* @g_3, align 1 ; <i8> [#uses=2]
%1 = sext i8 %0 to i32 ; <i32> [#uses=1]
%.lobit.i = lshr i8 %0, 7 ; <i8> [#uses=1]
%tmp.i = zext i8 %.lobit.i to i32 ; <i32> [#uses=1]
%tmp.not.i = xor i32 %tmp.i, 1 ; <i32> [#uses=1]
%iftmp.17.0.i.i = ashr i32 %1, %tmp.not.i ; <i32> [#uses=1]
%retval56.i.i = trunc i32 %iftmp.17.0.i.i to i8 ; <i8> [#uses=1]
%2 = icmp eq i8 %retval56.i.i, 0 ; <i1> [#uses=2]
%g_96.promoted.i = load i8* @g_96 ; <i8> [#uses=3]
%3 = icmp eq i8 %g_96.promoted.i, 0 ; <i1> [#uses=2]
br i1 %3, label %func_4.exit.i, label %bb.i.i.i
bb.i.i.i: ; preds = %entry
%4 = load volatile i8* @g_100, align 1 ; <i8> [#uses=0]
br label %func_4.exit.i
; CHECK-LABEL: test4:
; CHECK: g_100
; CHECK: testb
; CHECK-NOT: xor
; CHECK: setne
; CHECK: testb
func_4.exit.i: ; preds = %bb.i.i.i, %entry
%.not.i = xor i1 %2, true ; <i1> [#uses=1]
%brmerge.i = or i1 %3, %.not.i ; <i1> [#uses=1]
%.mux.i = select i1 %2, i8 %g_96.promoted.i, i8 0 ; <i8> [#uses=1]
br i1 %brmerge.i, label %func_1.exit, label %bb.i.i
bb.i.i: ; preds = %func_4.exit.i
%5 = load volatile i8* @g_100, align 1 ; <i8> [#uses=0]
br label %func_1.exit
func_1.exit: ; preds = %bb.i.i, %func_4.exit.i
%g_96.tmp.0.i = phi i8 [ %g_96.promoted.i, %bb.i.i ], [ %.mux.i, %func_4.exit.i ] ; <i8> [#uses=2]
store i8 %g_96.tmp.0.i, i8* @g_96
%6 = zext i8 %g_96.tmp.0.i to i32 ; <i32> [#uses=1]
%7 = tail call i32 (i8*, ...)* @printf(i8* noalias getelementptr ([15 x i8]* @_2E_str, i64 0, i64 0), i32 %6) nounwind ; <i32> [#uses=0]
ret i32 0
}
declare i32 @printf(i8* nocapture, ...) nounwind
; Should compile to setcc | -2.
; rdar://6668608
define i32 @test5(i32* nocapture %P) nounwind readonly {
entry:
; CHECK-LABEL: test5:
; CHECK: setg %al
; CHECK: movzbl %al, %eax
; CHECK: orl $-2, %eax
; CHECK: ret
%0 = load i32* %P, align 4 ; <i32> [#uses=1]
%1 = icmp sgt i32 %0, 41 ; <i1> [#uses=1]
%iftmp.0.0 = select i1 %1, i32 -1, i32 -2 ; <i32> [#uses=1]
ret i32 %iftmp.0.0
}
define i32 @test6(i32* nocapture %P) nounwind readonly {
entry:
; CHECK-LABEL: test6:
; CHECK: setl %al
; CHECK: movzbl %al, %eax
; CHECK: leal 4(%rax,%rax,8), %eax
; CHECK: ret
%0 = load i32* %P, align 4 ; <i32> [#uses=1]
%1 = icmp sgt i32 %0, 41 ; <i1> [#uses=1]
%iftmp.0.0 = select i1 %1, i32 4, i32 13 ; <i32> [#uses=1]
ret i32 %iftmp.0.0
}
; Don't try to use a 16-bit conditional move to do an 8-bit select,
; because it isn't worth it. Just use a branch instead.
define i8 @test7(i1 inreg %c, i8 inreg %a, i8 inreg %b) nounwind {
; CHECK-LABEL: test7:
; CHECK: testb $1, %dil
; CHECK-NEXT: jne LBB
%d = select i1 %c, i8 %a, i8 %b
ret i8 %d
}