6502/llvm-6502
1
0
Fork 0
mirror of https://github.com/c64scene-ar/llvm-6502.git synced 2024-12-28 04:33:05 +00:00
Code Issues Projects Releases Wiki Activity
db89809037
llvm-6502/test/CodeGen/ARM/fpcmp-opt.ll

83 lines
1.6 KiB
LLVM
Raw Normal View History

Extend the r107852 optimization which turns some fp compare to code sequence using only i32 operations. It now optimize some f64 compares when fp compare is exceptionally slow (e.g. cortex-a8). It also catches comparison against 0.0. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@108258 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-13 19:27:42 +00:00
; RUN: llc < %s -march=arm -mcpu=cortex-a8 -mattr=+vfp2 -enable-unsafe-fp-math -enable-finite-only-fp-math | FileCheck -check-prefix=FINITE %s
; RUN: llc < %s -march=arm -mcpu=cortex-a8 -mattr=+vfp2 -enable-unsafe-fp-math | FileCheck -check-prefix=NAN %s
Optimize some vfp comparisons to integer ones. This patch implements the simplest case when the following conditions are met: 1. The arguments are f32. 2. The arguments are loads and they have no uses other than the comparison. 3. The comparison code is EQ or NE. e.g. vldr.32 s0, [r1] vldr.32 s1, [r0] vcmpe.f32 s1, s0 vmrs apsr_nzcv, fpscr beq LBB0_2 => ldr r1, [r1] ldr r0, [r0] cmp r0, r1 beq LBB0_2 More complicated cases will be implemented in subsequent patches. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@107852 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-08 02:08:50 +00:00
; rdar://7461510
define arm_apcscc i32 @t1(float* %a, float* %b) nounwind {
entry:
Extend the r107852 optimization which turns some fp compare to code sequence using only i32 operations. It now optimize some f64 compares when fp compare is exceptionally slow (e.g. cortex-a8). It also catches comparison against 0.0. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@108258 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-13 19:27:42 +00:00
; FINITE: t1:
; FINITE-NOT: vldr
; FINITE: ldr
; FINITE: ldr
; FINITE: cmp r0, r1
; FINITE-NOT: vcmpe.f32
; FINITE-NOT: vmrs
; FINITE: beq
; NAN: t1:
; NAN: vldr.32 s0,
; NAN: vldr.32 s1,
; NAN: vcmpe.f32 s1, s0
; NAN: vmrs apsr_nzcv, fpscr
; NAN: beq
Optimize some vfp comparisons to integer ones. This patch implements the simplest case when the following conditions are met: 1. The arguments are f32. 2. The arguments are loads and they have no uses other than the comparison. 3. The comparison code is EQ or NE. e.g. vldr.32 s0, [r1] vldr.32 s1, [r0] vcmpe.f32 s1, s0 vmrs apsr_nzcv, fpscr beq LBB0_2 => ldr r1, [r1] ldr r0, [r0] cmp r0, r1 beq LBB0_2 More complicated cases will be implemented in subsequent patches. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@107852 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-08 02:08:50 +00:00
%0 = load float* %a
%1 = load 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
}
Extend the r107852 optimization which turns some fp compare to code sequence using only i32 operations. It now optimize some f64 compares when fp compare is exceptionally slow (e.g. cortex-a8). It also catches comparison against 0.0. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@108258 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-13 19:27:42 +00:00
define arm_apcscc i32 @t2(double* %a, double* %b) nounwind {
entry:
; FINITE: t2:
; FINITE-NOT: vldr
; FINITE: ldrd r0, [r0]
; FINITE: cmp r0, #0
; FINITE: cmpeq r1, #0
; FINITE-NOT: vcmpe.f32
; FINITE-NOT: vmrs
; FINITE: bne
%0 = load 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:
; FINITE: t3:
; FINITE-NOT: vldr
; FINITE: ldr r0, [r0]
; FINITE: cmp r0, #0
; FINITE-NOT: vcmpe.f32
; FINITE-NOT: vmrs
; FINITE: bne
%0 = load 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
}
Optimize some vfp comparisons to integer ones. This patch implements the simplest case when the following conditions are met: 1. The arguments are f32. 2. The arguments are loads and they have no uses other than the comparison. 3. The comparison code is EQ or NE. e.g. vldr.32 s0, [r1] vldr.32 s1, [r0] vcmpe.f32 s1, s0 vmrs apsr_nzcv, fpscr beq LBB0_2 => ldr r1, [r1] ldr r0, [r0] cmp r0, r1 beq LBB0_2 More complicated cases will be implemented in subsequent patches. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@107852 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-08 02:08:50 +00:00
declare i32 @bar()
declare i32 @foo()
Reference in New Issue Copy Permalink