# Basic machine sched model test for Thumb2 int instructions
# RUN: llc -o /dev/null %s -mtriple=thumbv7-eabi -mcpu=swift -run-pass machine-scheduler -enable-misched -verify-misched \
# RUN: -debug-only=machine-scheduler 2>&1 | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK_SWIFT
# RUN: llc -o /dev/null %s -mtriple=thumbv7--eabi -mcpu=cortex-a9 -run-pass machine-scheduler -enable-misched -verify-misched \
# RUN: -debug-only=machine-scheduler 2>&1 | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK_A9
# RUN: llc -o /dev/null %s -mtriple=thumbv8r-eabi -mcpu=cortex-r52 -run-pass machine-scheduler -enable-misched -verify-misched \
# RUN: -debug-only=machine-scheduler 2>&1 | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK_R52
# RUN: llc -o /dev/null %s -mtriple=thumbv8r-eabi -mcpu=cortex-r52plus -run-pass machine-scheduler -enable-misched -verify-misched \
# RUN: -debug-only=machine-scheduler 2>&1 | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK_R52
# REQUIRES: asserts
--- |
; ModuleID = 'foo.ll'
source_filename = "foo.ll"
target datalayout = "e-m:e-p:32:32-i64:64-v128:64:128-a:0:32-n32-S64"
target triple = "thumbv7---eabi"
@g1 = common global i32 0, align 4
@g2 = common global i32 0, align 4
define i64 @foo(i16 signext %a, i16 signext %b) {
entry:
%0 = load i32, ptr @g1, align 4
%1 = load i32, ptr @g2, align 4
%2 = add nuw nsw i32 %0, %0
%3 = sdiv i32 %2, %1
store i32 %3, ptr @g1, align 4
%d = mul nsw i16 %a, %a
%e = mul nsw i16 %b, %b
%f = add nuw nsw i16 %e, %d
%c = zext i16 %f to i32
%mul8 = mul nsw i32 %c, %3
%mul9 = mul nsw i32 %mul8, %mul8
%add10 = add nuw nsw i32 %mul9, %mul8
%conv1130 = zext i32 %add10 to i64
%mul12 = mul nuw nsw i64 %conv1130, %conv1130
%mul13 = mul nsw i64 %mul12, %mul12
%add14 = add nuw nsw i64 %mul13, %mul12
ret i64 %add14
}
#
# CHECK: ********** MI Scheduling **********
# CHECK: SU(2): %2:rgpr = t2MOVi32imm @g1
# CHECK_A9: Latency : 2
# CHECK_SWIFT: Latency : 2
# CHECK_R52: Latency : 2
#
# CHECK: SU(3): %3:rgpr = t2LDRi12 %2:rgpr, 0, 14, $noreg :: (dereferenceable load (s32) from @g1)
# CHECK_A9: Latency : 1
# CHECK_SWIFT: Latency : 3
# CHECK_R52: Latency : 4
#
# CHECK: SU(6): %6:rgpr = t2ADDrr %3:rgpr, %3:rgpr, 14, $noreg, $noreg
# CHECK_A9: Latency : 1
# CHECK_SWIFT: Latency : 1
# CHECK_R52: Latency : 3
# CHECK: SU(7): %7:rgpr = t2SDIV %6:rgpr, %5:rgpr, 14, $noreg
# CHECK_A9: Latency : 0
# CHECK_SWIFT: Latency : 14
# CHECK_R52: Latency : 8
# CHECK: SU(8): t2STRi12 %7:rgpr, %2:rgpr, 0, 14, $noreg :: (store (s32) into @g1)
# CHECK_A9: Latency : 1
# CHECK_SWIFT: Latency : 0
# CHECK_R52: Latency : 4
#
# CHECK: SU(9): %8:rgpr = t2SMULBB %1:rgpr, %1:rgpr, 14, $noreg
# CHECK_A9: Latency : 2
# CHECK_SWIFT: Latency : 4
# CHECK_R52: Latency : 4
#
# CHECK: SU(10): %9:rgpr = t2SMLABB %0:rgpr, %0:rgpr, %8:rgpr, 14, $noreg
# CHECK_A9: Latency : 2
# CHECK_SWIFT: Latency : 4
# CHECK_R52: Latency : 4
#
# CHECK: SU(11): %10:rgpr = t2UXTH %9:rgpr, 0, 14, $noreg
# CHECK_A9: Latency : 1
# CHECK_SWIFT: Latency : 1
# CHECK_R52: Latency : 3
#
# CHECK: SU(12): %11:rgpr = t2MUL %10:rgpr, %7:rgpr, 14, $noreg
# CHECK_A9: Latency : 2
# CHECK_SWIFT: Latency : 4
# CHECK_R52: Latency : 4
#
# CHECK: SU(13): %12:rgpr = t2MLA %11:rgpr, %11:rgpr, %11:rgpr, 14, $noreg
# CHECK_A9: Latency : 2
# CHECK_SWIFT: Latency : 4
# CHECK_R52: Latency : 4
#
# CHECK: SU(14): %13:rgpr, %14:rgpr = t2UMULL %12:rgpr, %12:rgpr, 14, $noreg
# CHECK_A9: Latency : 3
# CHECK_SWIFT: Latency : 5
# CHECK_R52: Latency : 4
#
# CHECK: SU(18): %19:rgpr, %20:rgpr = t2UMLAL %12:rgpr, %12:rgpr, %19:rgpr(tied-def 0), %20:rgpr(tied-def 1), 14, $noreg
# CHECK_A9: Latency : 3
# CHECK_SWIFT: Latency : 7
# CHECK_R52: Latency : 4
# CHECK: ** ScheduleDAGMILive::schedule picking next node
...
---
name: foo
alignment: 2
exposesReturnsTwice: false
legalized: false
regBankSelected: false
selected: false
tracksRegLiveness: true
registers:
- { id: 0, class: rgpr }
- { id: 1, class: rgpr }
- { id: 2, class: rgpr }
- { id: 3, class: rgpr }
- { id: 4, class: rgpr }
- { id: 5, class: rgpr }
- { id: 6, class: rgpr }
- { id: 7, class: rgpr }
- { id: 8, class: rgpr }
- { id: 9, class: rgpr }
- { id: 10, class: rgpr }
- { id: 11, class: rgpr }
- { id: 12, class: rgpr }
- { id: 13, class: rgpr }
- { id: 14, class: rgpr }
- { id: 15, class: rgpr }
- { id: 16, class: rgpr }
- { id: 17, class: rgpr }
- { id: 18, class: rgpr }
- { id: 19, class: rgpr }
- { id: 20, class: rgpr }
liveins:
- { reg: '$r0', virtual-reg: '%0' }
- { reg: '$r1', virtual-reg: '%1' }
frameInfo:
isFrameAddressTaken: false
isReturnAddressTaken: false
hasStackMap: false
hasPatchPoint: false
stackSize: 0
offsetAdjustment: 0
maxAlignment: 0
adjustsStack: false
hasCalls: false
maxCallFrameSize: 0
hasOpaqueSPAdjustment: false
hasVAStart: false
hasMustTailInVarArgFunc: false
body: |
bb.0.entry:
liveins: $r0, $r1
%1 = COPY $r1
%0 = COPY $r0
%2 = t2MOVi32imm @g1
%3 = t2LDRi12 %2, 0, 14, $noreg :: (dereferenceable load (s32) from @g1)
%4 = t2MOVi32imm @g2
%5 = t2LDRi12 %4, 0, 14, $noreg :: (dereferenceable load (s32) from @g2)
%6 = t2ADDrr %3, %3, 14, $noreg, $noreg
%7 = t2SDIV %6, %5, 14, $noreg
t2STRi12 %7, %2, 0, 14, $noreg :: (store (s32) into @g1)
%8 = t2SMULBB %1, %1, 14, $noreg
%9 = t2SMLABB %0, %0, %8, 14, $noreg
%10 = t2UXTH %9, 0, 14, $noreg
%11 = t2MUL %10, %7, 14, $noreg
%12 = t2MLA %11, %11, %11, 14, $noreg
%13, %14 = t2UMULL %12, %12, 14, $noreg
%19, %16 = t2UMULL %13, %13, 14, $noreg
%17 = t2MLA %13, %14, %16, 14, $noreg
%20 = t2MLA %13, %14, %17, 14, $noreg
%19, %20 = t2UMLAL %12, %12, %19, %20, 14, $noreg
$r0 = COPY %19
$r1 = COPY %20
tBX_RET 14, $noreg, implicit $r0, implicit $r1
...
Codebase Browser
llvm