llvm/test/CodeGen/X86/stack-align.ll
Dan Gohman 584fedf188 Teach two-address lowering how to unfold a load to open up commuting
opportunities. For example, this lets it emit this:

   movq (%rax), %rcx
   addq %rdx, %rcx

instead of this:

   movq %rdx, %rcx
   addq (%rax), %rcx

in the case where %rdx has subsequent uses. It's the same number
of instructions, and usually the same encoding size on x86, but
it appears faster, and in general, it may allow better scheduling
for the load.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@106493 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-21 22:17:20 +00:00

41 lines
1.4 KiB
LLVM

; RUN: llc < %s -relocation-model=static -realign-stack=1 -mcpu=yonah | FileCheck %s
; The double argument is at 4(esp) which is 16-byte aligned, allowing us to
; fold the load into the andpd.
target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128"
target triple = "i686-apple-darwin8"
@G = external global double
define void @test({ double, double }* byval %z, double* %P) {
entry:
%tmp3 = load double* @G, align 16 ; <double> [#uses=1]
%tmp4 = tail call double @fabs( double %tmp3 ) ; <double> [#uses=1]
volatile store double %tmp4, double* %P
%tmp = getelementptr { double, double }* %z, i32 0, i32 0 ; <double*> [#uses=1]
%tmp1 = volatile load double* %tmp, align 8 ; <double> [#uses=1]
%tmp2 = tail call double @fabs( double %tmp1 ) ; <double> [#uses=1]
; CHECK: andpd{{.*}}4(%esp), %xmm
%tmp6 = fadd double %tmp4, %tmp2 ; <double> [#uses=1]
volatile store double %tmp6, double* %P, align 8
ret void
}
define void @test2() alignstack(16) {
entry:
; CHECK: andl{{.*}}$-16, %esp
ret void
}
; Use a call to force a spill.
define <2 x double> @test3(<2 x double> %x, <2 x double> %y) alignstack(32) {
entry:
; CHECK: andl{{.*}}$-32, %esp
call void @test2()
%A = fmul <2 x double> %x, %y
ret <2 x double> %A
}
declare double @fabs(double)