bf6e1c26cf
It's been quite some time the Dependence Analysis (DA) is broken,
as it uses the GEP representation to "identify" multi-dimensional arrays.
It even wrongly detects multi-dimensional arrays in single nested loops:
from test/Analysis/DependenceAnalysis/Coupled.ll, example @couple6
;; for (long int i = 0; i < 50; i++) {
;; A[i][3*i - 6] = i;
;; *B++ = A[i][i];
DA used to detect two subscripts, which makes no sense in the LLVM IR
or in C/C++ semantics, as there are no guarantees as in Fortran of
subscripts not overlapping into a next array dimension:
maximum nesting levels = 1
SrcPtrSCEV = %A
DstPtrSCEV = %A
using GEPs
subscript 0
src = {0,+,1}<nuw><nsw><%for.body>
dst = {0,+,1}<nuw><nsw><%for.body>
class = 1
loops = {1}
subscript 1
src = {-6,+,3}<nsw><%for.body>
dst = {0,+,1}<nuw><nsw><%for.body>
class = 1
loops = {1}
Separable = {}
Coupled = {1}
With the current patch, DA will correctly work on only one dimension:
maximum nesting levels = 1
SrcSCEV = {(-2424 + %A)<nsw>,+,1212}<%for.body>
DstSCEV = {%A,+,404}<%for.body>
subscript 0
src = {(-2424 + %A)<nsw>,+,1212}<%for.body>
dst = {%A,+,404}<%for.body>
class = 1
loops = {1}
Separable = {0}
Coupled = {}
This change removes all uses of GEP from DA, and we now only rely
on the SCEV representation.
The patch does not turn on -da-delinearize by default, and so the DA analysis
will be more conservative in the case of multi-dimensional memory accesses in
nested loops.
I disabled some interchange tests, as the DA is not able to disambiguate
the dependence anymore. To make DA stronger, we may need to
compute a bound on the number of iterations based on the access functions
and array dimensions.
The patch cleans up all the CHECKs in test/Transforms/LoopInterchange/*.ll to
avoid checking for snippets of LLVM IR: this form of checking is very hard to
maintain. Instead, we now check for output of the pass that are more meaningful
than dozens of lines of LLVM IR. Some tests now require -debug messages and thus
only enabled with asserts.
Patch written by Sebastian Pop and Aditya Kumar.
Differential Revision: https://reviews.llvm.org/D35430
llvm-svn: 326837
102 lines
4.1 KiB
LLVM
102 lines
4.1 KiB
LLVM
; RUN: opt < %s -analyze -basicaa -da -da-delinearize | FileCheck %s -check-prefix=DELIN
|
|
|
|
target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
|
|
target triple = "x86_64-apple-macosx10.6.0"
|
|
|
|
; for (int i = 0; i < 100; ++i) {
|
|
; int t0 = a[i][i];
|
|
; int t1 = t0 + 1;
|
|
; a[i][5] = t1;
|
|
; }
|
|
; The subscript 5 in a[i][5] is deliberately an i32, mismatching the types of
|
|
; other subscript. DependenceAnalysis before the fix crashed due to this
|
|
; mismatch.
|
|
define void @i32_subscript([100 x [100 x i32]]* %a, i32* %b) {
|
|
; DELIN-LABEL: 'Dependence Analysis' for function 'i32_subscript'
|
|
entry:
|
|
br label %for.body
|
|
|
|
for.body:
|
|
; DELIN: da analyze - none!
|
|
; DELIN: da analyze - anti [=|<]!
|
|
; DELIN: da analyze - none!
|
|
%i = phi i64 [ 0, %entry ], [ %i.inc, %for.body ]
|
|
%a.addr = getelementptr [100 x [100 x i32]], [100 x [100 x i32]]* %a, i64 0, i64 %i, i64 %i
|
|
%a.addr.2 = getelementptr [100 x [100 x i32]], [100 x [100 x i32]]* %a, i64 0, i64 %i, i32 5
|
|
%0 = load i32, i32* %a.addr, align 4
|
|
%1 = add i32 %0, 1
|
|
store i32 %1, i32* %a.addr.2, align 4
|
|
%i.inc = add nsw i64 %i, 1
|
|
%exitcond = icmp ne i64 %i.inc, 100
|
|
br i1 %exitcond, label %for.body, label %for.end
|
|
|
|
for.end:
|
|
ret void
|
|
}
|
|
|
|
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
|
|
target triple = "x86_64-unknown-linux-gnu"
|
|
|
|
; unsigned i, j;
|
|
; for (i = 1; i < SIZE; i++) {
|
|
; for (j = i; j < SIZE; j++) {
|
|
; a[i][j] = a[i+1][j-1] + 2;
|
|
; }
|
|
; }
|
|
; Extends the previous example to coupled MIV subscripts.
|
|
|
|
|
|
@a = global [10004 x [10004 x i32]] zeroinitializer, align 16
|
|
|
|
; Function Attrs: nounwind uwtable
|
|
define void @coupled_miv_type_mismatch(i32 %n) #0 {
|
|
; DELIN-LABEL: 'Dependence Analysis' for function 'coupled_miv_type_mismatch'
|
|
entry:
|
|
br label %for.cond
|
|
|
|
; DELIN: da analyze - input [* *]!
|
|
; DELIN: da analyze - anti [* *|<]!
|
|
; DELIN: da analyze - none!
|
|
for.cond: ; preds = %for.inc11, %entry
|
|
%indvars.iv11 = phi i64 [ %indvars.iv.next12, %for.inc11 ], [ 1, %entry ]
|
|
%exitcond14 = icmp ne i64 %indvars.iv11, 10000
|
|
br i1 %exitcond14, label %for.cond1.preheader, label %for.end13
|
|
|
|
for.cond1.preheader: ; preds = %for.cond
|
|
%0 = trunc i64 %indvars.iv11 to i32
|
|
br label %for.cond1
|
|
|
|
for.cond1: ; preds = %for.cond1.preheader, %for.body3
|
|
%indvars.iv8 = phi i64 [ %indvars.iv11, %for.cond1.preheader ], [ %indvars.iv.next9, %for.body3 ]
|
|
%j.0 = phi i32 [ %inc, %for.body3 ], [ %0, %for.cond1.preheader ]
|
|
%lftr.wideiv = trunc i64 %indvars.iv8 to i32
|
|
%exitcond = icmp ne i32 %lftr.wideiv, 10000
|
|
br i1 %exitcond, label %for.body3, label %for.inc11
|
|
|
|
for.body3: ; preds = %for.cond1
|
|
%sub = add nsw i32 %j.0, -1
|
|
%idxprom = zext i32 %sub to i64
|
|
%1 = add nuw nsw i64 %indvars.iv11, 1
|
|
%arrayidx5 = getelementptr inbounds [10004 x [10004 x i32]], [10004 x [10004 x i32]]* @a, i64 0, i64 %1, i64 %idxprom
|
|
%2 = load i32, i32* %arrayidx5, align 4
|
|
%add6 = add nsw i32 %2, 2
|
|
%arrayidx10 = getelementptr inbounds [10004 x [10004 x i32]], [10004 x [10004 x i32]]* @a, i64 0, i64 %indvars.iv11, i64 %indvars.iv8
|
|
store i32 %add6, i32* %arrayidx10, align 4
|
|
%indvars.iv.next9 = add nuw nsw i64 %indvars.iv8, 1
|
|
%inc = add nuw nsw i32 %j.0, 1
|
|
br label %for.cond1
|
|
|
|
for.inc11: ; preds = %for.cond1
|
|
%indvars.iv.next12 = add nuw nsw i64 %indvars.iv11, 1
|
|
br label %for.cond
|
|
|
|
for.end13: ; preds = %for.cond
|
|
ret void
|
|
}
|
|
|
|
attributes #0 = { nounwind uwtable "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+sse,+sse2" "unsafe-fp-math"="false" "use-soft-float"="false" }
|
|
|
|
!llvm.ident = !{!0}
|
|
|
|
!0 = !{!"clang version 3.7.0 (https://vaivaswatha@bitbucket.org/compilertree/amd_clang.git 93a05fb75ee3411d24e8b2b184fc766a5318403e) (https://vaivaswatha@bitbucket.org/compilertree/amd_llvm.git 166d93d26efc912b517739f64d054a435e8e95cd)"}
|