1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-25 20:23:11 +01:00
llvm-mirror/test/CodeGen/ARM/fpcmp-opt.ll
Eli Friedman 925690f23e [ARM] Prefer BIC over BFC in ARM mode.
BIC is generally faster, and it can put the output in a different
register from the input.

We already do this in Thumb2 mode; not sure why the equivalent fix
never got applied to ARM mode.

Differential Revision: https://reviews.llvm.org/D31797

llvm-svn: 299803
2017-04-07 22:01:23 +00:00

82 lines
1.7 KiB
LLVM

; RUN: llc -mtriple=arm-eabi -mcpu=cortex-a8 -mattr=+vfp2 -enable-unsafe-fp-math %s -o - \
; RUN: | FileCheck %s
; rdar://7461510
; rdar://10964603
; Disable this optimization unless we know one of them is zero.
define arm_apcscc i32 @t1(float* %a, float* %b) nounwind {
entry:
; CHECK-LABEL: t1:
; CHECK: vldr [[S0:s[0-9]+]],
; CHECK: vldr [[S1:s[0-9]+]],
; CHECK: vcmp.f32 [[S1]], [[S0]]
; CHECK: vmrs APSR_nzcv, fpscr
; CHECK: beq
%0 = load float, float* %a
%1 = load float, float* %b
%2 = fcmp une float %0, %1
br i1 %2, label %bb1, label %bb2
bb1:
%3 = call i32 @bar()
ret i32 %3
bb2:
%4 = call i32 @foo()
ret i32 %4
}
; If one side is zero, the other size sign bit is masked off to allow
; +0.0 == -0.0
define arm_apcscc i32 @t2(double* %a, double* %b) nounwind {
entry:
; CHECK-LABEL: t2:
; CHECK-NOT: vldr
; CHECK: ldrd [[REG1:(r[0-9]+)]], [[REG2:(r[0-9]+)]], [r0]
; CHECK-NOT: b LBB
; CHECK: bic [[REG2]], [[REG2]], #-2147483648
; CHECK: cmp [[REG1]], #0
; CHECK: cmpeq [[REG2]], #0
; CHECK-NOT: vcmp.f32
; CHECK-NOT: vmrs
; CHECK: bne
%0 = load double, double* %a
%1 = fcmp oeq double %0, 0.000000e+00
br i1 %1, label %bb1, label %bb2
bb1:
%2 = call i32 @bar()
ret i32 %2
bb2:
%3 = call i32 @foo()
ret i32 %3
}
define arm_apcscc i32 @t3(float* %a, float* %b) nounwind {
entry:
; CHECK-LABEL: t3:
; CHECK-NOT: vldr
; CHECK: ldr [[REG3:(r[0-9]+)]], [r0]
; CHECK: mvn [[REG4:(r[0-9]+)]], #-2147483648
; CHECK: tst [[REG3]], [[REG4]]
; CHECK-NOT: vcmp.f32
; CHECK-NOT: vmrs
; CHECK: bne
%0 = load float, float* %a
%1 = fcmp oeq float %0, 0.000000e+00
br i1 %1, label %bb1, label %bb2
bb1:
%2 = call i32 @bar()
ret i32 %2
bb2:
%3 = call i32 @foo()
ret i32 %3
}
declare i32 @bar()
declare i32 @foo()