Files
clang-p2996/llvm/test/Transforms/SLPVectorizer/X86/opaque-ptr.ll
Alex Richardson e39f6c1844 [opt] Infer DataLayout from triple if not specified
There are many tests that specify a target triple/CPU flags but no
DataLayout which can lead to IR being generated that has unusual
behaviour. This commit attempts to use the default DataLayout based
on the relevant flags if there is no explicit override on the command
line or in the IR file.

One thing that is not currently possible to differentiate from a missing
datalayout `target datalayout = ""` in the IR file since the current
APIs don't allow detecting this case. If it is considered useful to
support this case (instead of passing "-data-layout=" on the command
line), I can change IR parsers to track whether they have seen such a
directive and change the callback type.

Differential Revision: https://reviews.llvm.org/D141060
2023-10-26 12:07:37 -07:00

81 lines
3.4 KiB
LLVM

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt -S -passes=slp-vectorizer -mtriple=x86_64-apple-macosx -mcpu=haswell < %s | FileCheck %s
define void @test(ptr %r, ptr %p, ptr %q) #0 {
; CHECK-LABEL: @test(
; CHECK-NEXT: [[P0:%.*]] = getelementptr inbounds i64, ptr [[P:%.*]], i64 0
; CHECK-NEXT: [[Q0:%.*]] = getelementptr inbounds i64, ptr [[Q:%.*]], i64 0
; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, ptr [[P0]], align 2
; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, ptr [[Q0]], align 2
; CHECK-NEXT: [[TMP3:%.*]] = sub nsw <4 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i64> [[TMP3]], i32 0
; CHECK-NEXT: [[G0:%.*]] = getelementptr inbounds i32, ptr [[R:%.*]], i64 [[TMP4]]
; CHECK-NEXT: [[TMP5:%.*]] = extractelement <4 x i64> [[TMP3]], i32 1
; CHECK-NEXT: [[G1:%.*]] = getelementptr inbounds i32, ptr [[R]], i64 [[TMP5]]
; CHECK-NEXT: [[TMP6:%.*]] = extractelement <4 x i64> [[TMP3]], i32 2
; CHECK-NEXT: [[G2:%.*]] = getelementptr inbounds i32, ptr [[R]], i64 [[TMP6]]
; CHECK-NEXT: [[TMP7:%.*]] = extractelement <4 x i64> [[TMP3]], i32 3
; CHECK-NEXT: [[G3:%.*]] = getelementptr inbounds i32, ptr [[R]], i64 [[TMP7]]
; CHECK-NEXT: ret void
;
%p0 = getelementptr inbounds i64, ptr %p, i64 0
%p1 = getelementptr inbounds i64, ptr %p, i64 1
%p2 = getelementptr inbounds i64, ptr %p, i64 2
%p3 = getelementptr inbounds i64, ptr %p, i64 3
%q0 = getelementptr inbounds i64, ptr %q, i64 0
%q1 = getelementptr inbounds i64, ptr %q, i64 1
%q2 = getelementptr inbounds i64, ptr %q, i64 2
%q3 = getelementptr inbounds i64, ptr %q, i64 3
%x0 = load i64, ptr %p0, align 2
%x1 = load i64, ptr %p1, align 2
%x2 = load i64, ptr %p2, align 2
%x3 = load i64, ptr %p3, align 2
%y0 = load i64, ptr %q0, align 2
%y1 = load i64, ptr %q1, align 2
%y2 = load i64, ptr %q2, align 2
%y3 = load i64, ptr %q3, align 2
%sub0 = sub nsw i64 %x0, %y0
%sub1 = sub nsw i64 %x1, %y1
%sub2 = sub nsw i64 %x2, %y2
%sub3 = sub nsw i64 %x3, %y3
%g0 = getelementptr inbounds i32, ptr %r, i64 %sub0
%g1 = getelementptr inbounds i32, ptr %r, i64 %sub1
%g2 = getelementptr inbounds i32, ptr %r, i64 %sub2
%g3 = getelementptr inbounds i32, ptr %r, i64 %sub3
ret void
}
define void @test2(i64* %a, i64* %b) {
; CHECK-LABEL: @test2(
; CHECK-NEXT: [[A1:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 1
; CHECK-NEXT: [[A2:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 2
; CHECK-NEXT: [[I1:%.*]] = ptrtoint ptr [[A1]] to i64
; CHECK-NEXT: [[B3:%.*]] = getelementptr inbounds i64, ptr [[B:%.*]], i64 3
; CHECK-NEXT: [[I2:%.*]] = ptrtoint ptr [[B3]] to i64
; CHECK-NEXT: [[V1:%.*]] = load i64, ptr [[A1]], align 8
; CHECK-NEXT: [[V2:%.*]] = load i64, ptr [[A2]], align 8
; CHECK-NEXT: [[ADD1:%.*]] = add i64 [[I1]], [[V1]]
; CHECK-NEXT: [[ADD2:%.*]] = add i64 [[I2]], [[V2]]
; CHECK-NEXT: store i64 [[ADD1]], ptr [[A1]], align 8
; CHECK-NEXT: store i64 [[ADD2]], ptr [[A2]], align 8
; CHECK-NEXT: ret void
;
%a1 = getelementptr inbounds i64, i64* %a, i64 1
%a2 = getelementptr inbounds i64, i64* %a, i64 2
%i1 = ptrtoint i64* %a1 to i64
%b3 = getelementptr inbounds i64, i64* %b, i64 3
%i2 = ptrtoint i64* %b3 to i64
%v1 = load i64, i64* %a1, align 8
%v2 = load i64, i64* %a2, align 8
%add1 = add i64 %i1, %v1
%add2 = add i64 %i2, %v2
store i64 %add1, i64* %a1, align 8
store i64 %add2, i64* %a2, align 8
ret void
}