test/Transforms/PhaseOrdering/inlining-alignment-assumptions.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt -S -O2 -preserve-alignment-assumptions-during-inlining=0 < %s | FileCheck %s --check-prefixes=CHECK,ASSUMPTIONS-OFF
; RUN: opt -S -O2 -preserve-alignment-assumptions-during-inlining=1 < %s | FileCheck %s --check-prefixes=CHECK,ASSUMPTIONS-ON
; RUN: opt -S -O2 < %s | FileCheck %s --check-prefixes=CHECK,ASSUMPTIONS-OFF

target datalayout = "e-p:64:64-p5:32:32-A5"

; This illustrates an optimization difference caused by instruction counting
; heuristics, which are affected by the additional instructions of the
; alignment assumption.

define internal i1 @callee1(i1 %c, ptr align 8 %ptr) {
  store volatile i64 0, ptr %ptr
  ret i1 %c
}

define void @caller1(i1 %c, ptr align 1 %ptr) {
; ASSUMPTIONS-OFF-LABEL: @caller1(
; ASSUMPTIONS-OFF-NEXT:    br i1 [[C:%.*]], label [[COMMON_RET:%.*]], label [[FALSE2:%.*]]
; ASSUMPTIONS-OFF:       common.ret:
; ASSUMPTIONS-OFF-NEXT:    [[DOTSINK:%.*]] = phi i64 [ 3, [[FALSE2]] ], [ 2, [[TMP0:%.*]] ]
; ASSUMPTIONS-OFF-NEXT:    store volatile i64 0, ptr [[PTR:%.*]], align 4
; ASSUMPTIONS-OFF-NEXT:    store volatile i64 -1, ptr [[PTR]], align 4
; ASSUMPTIONS-OFF-NEXT:    store volatile i64 -1, ptr [[PTR]], align 4
; ASSUMPTIONS-OFF-NEXT:    store volatile i64 -1, ptr [[PTR]], align 4
; ASSUMPTIONS-OFF-NEXT:    store volatile i64 -1, ptr [[PTR]], align 4
; ASSUMPTIONS-OFF-NEXT:    store volatile i64 -1, ptr [[PTR]], align 4
; ASSUMPTIONS-OFF-NEXT:    store volatile i64 [[DOTSINK]], ptr [[PTR]], align 4
; ASSUMPTIONS-OFF-NEXT:    ret void
; ASSUMPTIONS-OFF:       false2:
; ASSUMPTIONS-OFF-NEXT:    store volatile i64 1, ptr [[PTR]], align 4
; ASSUMPTIONS-OFF-NEXT:    br label [[COMMON_RET]]
;
; ASSUMPTIONS-ON-LABEL: @caller1(
; ASSUMPTIONS-ON-NEXT:    br i1 [[C:%.*]], label [[COMMON_RET:%.*]], label [[FALSE2:%.*]]
; ASSUMPTIONS-ON:       common.ret:
; ASSUMPTIONS-ON-NEXT:    [[DOTSINK:%.*]] = phi i64 [ 3, [[FALSE2]] ], [ 2, [[TMP0:%.*]] ]
; ASSUMPTIONS-ON-NEXT:    call void @llvm.assume(i1 true) [ "align"(ptr [[PTR:%.*]], i64 8) ]
; ASSUMPTIONS-ON-NEXT:    store volatile i64 0, ptr [[PTR]], align 8
; ASSUMPTIONS-ON-NEXT:    store volatile i64 -1, ptr [[PTR]], align 8
; ASSUMPTIONS-ON-NEXT:    store volatile i64 -1, ptr [[PTR]], align 8
; ASSUMPTIONS-ON-NEXT:    store volatile i64 -1, ptr [[PTR]], align 8
; ASSUMPTIONS-ON-NEXT:    store volatile i64 -1, ptr [[PTR]], align 8
; ASSUMPTIONS-ON-NEXT:    store volatile i64 -1, ptr [[PTR]], align 8
; ASSUMPTIONS-ON-NEXT:    store volatile i64 [[DOTSINK]], ptr [[PTR]], align 8
; ASSUMPTIONS-ON-NEXT:    ret void
; ASSUMPTIONS-ON:       false2:
; ASSUMPTIONS-ON-NEXT:    store volatile i64 1, ptr [[PTR]], align 4
; ASSUMPTIONS-ON-NEXT:    br label [[COMMON_RET]]
;
  br i1 %c, label %true1, label %false1

true1:
  %c2 = call i1 @callee1(i1 %c, ptr %ptr)
  store volatile i64 -1, ptr %ptr
  store volatile i64 -1, ptr %ptr
  store volatile i64 -1, ptr %ptr
  store volatile i64 -1, ptr %ptr
  store volatile i64 -1, ptr %ptr
  br i1 %c2, label %true2, label %false2

false1:
  store volatile i64 1, ptr %ptr
  br label %true1

true2:
  store volatile i64 2, ptr %ptr
  ret void

false2:
  store volatile i64 3, ptr %ptr
  ret void
}

; This test checks that alignment assumptions do not prevent SROA.
; See PR45763.

define internal void @callee2(ptr noalias sret(i64) align 32 %arg) {
  store i64 0, ptr %arg, align 8
  ret void
}

define amdgpu_kernel void @caller2() {
; CHECK-LABEL: @caller2(
; CHECK-NEXT:    ret void
;
  %alloca = alloca i64, align 8, addrspace(5)
  %cast = addrspacecast ptr addrspace(5) %alloca to ptr
  call void @callee2(ptr sret(i64) align 32 %cast)
  ret void
}