maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/polly/test/ScopInfo/multidim_single_and_multidi...

69 lines
2.6 KiB
LLVM
Raw Normal View History

ScopDetection: Only detect scops that have at least one read and one write Scops that only read seem generally uninteresting and scops that only write are most likely initializations where there is also little to optimize. To not waste compile time we bail early. Differential Revision: http://reviews.llvm.org/D7735 llvm-svn: 229820
2015-02-19 13:31:07 +08:00
; RUN: opt %loadPolly -polly-detect-unprofitable -polly-scops -analyze < %s | FileCheck %s
; RUN: opt %loadPolly -polly-detect-unprofitable -polly-scops -polly-allow-nonaffine -analyze < %s | FileCheck %s --check-prefix=NONAFFINE
; RUN: opt %loadPolly -polly-detect-unprofitable -polly-scops -polly-delinearize -analyze < %s | FileCheck %s --check-prefix=DELIN
; RUN: opt %loadPolly -polly-detect-unprofitable -polly-scops -polly-delinearize -polly-allow-nonaffine -analyze < %s | FileCheck %s --check-prefix=DELIN
Delinearize _all_ accesses to a multi-dimensional array Even though we previously correctly detected the multi-dimensional access pattern for accesses with a certain base address, we only delinearized non-affine accesses to this address. Affine accesses have not been touched and remained as single dimensional accesses. The result was an inconsistent description of accesses to the same array, with some being one dimensional and some being multi-dimensional. This patch ensures that all accesses are delinearized with the same dimensionality as soon as a single one of them has been detected as non-affine. While writing this patch, it became evident that the options -polly-allow-nonaffine and -polly-detect-keep-going have not been properly supported in case delinearization has been turned on. This patch adds relevant test coverage and addresses these issues as well. We also added some more documentation to the functions that are modified in this patch. This fixes llvm.org/PR20123 Differential Revision: http://reviews.llvm.org/D5329 llvm-svn: 217728
2014-09-13 22:47:55 +08:00
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
; void single-and-multi-dimensional-array(long n,float X[n][n]) {
; for (long i1 = 0; i1 < n; i1++)
; X[i1][0] = 1;
;
; for (long i2 = 0; i2 < n; i2++)
; X[n-1][i2] = 1;
; }
;
; In previous versions of Polly, the second access was detected as single
; dimensional access whereas the first one was detected as multi-dimensional.
; This test case checks that we now consistently delinearize the array accesses.
; CHECK-NOT: Stmt_for_i_1
; NONAFFINE: p0: %n
; NONAFFINE: p1: (4 * (-1 + %n) * %n)
; NONAFFINE: Statements {
; NONAFFINE: Stmt_for_i_1
; NONAFFINE: MayWriteAccess := [Reduction Type: NONE]
Add ScalarEvolution bounds to non-affine access functions llvm-svn: 230328
2015-02-24 19:58:30 +08:00
; NONAFFINE: [n, p_1] -> { Stmt_for_i_1[i0] -> MemRef_X[o0] : o0 >= -2305843009213693952 and o0 <= 2305843009213693949 };
Delinearize _all_ accesses to a multi-dimensional array Even though we previously correctly detected the multi-dimensional access pattern for accesses with a certain base address, we only delinearized non-affine accesses to this address. Affine accesses have not been touched and remained as single dimensional accesses. The result was an inconsistent description of accesses to the same array, with some being one dimensional and some being multi-dimensional. This patch ensures that all accesses are delinearized with the same dimensionality as soon as a single one of them has been detected as non-affine. While writing this patch, it became evident that the options -polly-allow-nonaffine and -polly-detect-keep-going have not been properly supported in case delinearization has been turned on. This patch adds relevant test coverage and addresses these issues as well. We also added some more documentation to the functions that are modified in this patch. This fixes llvm.org/PR20123 Differential Revision: http://reviews.llvm.org/D5329 llvm-svn: 217728
2014-09-13 22:47:55 +08:00
; NONAFFINE: Stmt_for_i_2
; NONAFFINE: MustWriteAccess := [Reduction Type: NONE]
; NONAFFINE: [n, p_1] -> { Stmt_for_i_2[i0] -> MemRef_X[o0] : 4o0 = p_1 + 4i0 };
; DELIN: Stmt_for_i_1
; DELIN: MustWriteAccess :=
; DELIN: [n] -> { Stmt_for_i_1[i0] -> MemRef_X[i0, 0] };
; DELIN: Stmt_for_i_2
; DELIN: MustWriteAccess :=
; DELIN: [n] -> { Stmt_for_i_2[i0] -> MemRef_X[-1 + n, i0] };
define void @single-and-multi-dimensional-array(i64 %n, float* %X) {
entry:
br label %for.i.1
for.i.1:
%indvar.1 = phi i64 [ 0, %entry ], [ %indvar.next.1, %for.i.1 ]
%offset.1 = mul i64 %n, %indvar.1
%arrayidx.1 = getelementptr float* %X, i64 %offset.1
store float 1.000000e+00, float* %arrayidx.1
%indvar.next.1 = add nsw i64 %indvar.1, 1
%exitcond.1 = icmp ne i64 %indvar.next.1, %n
br i1 %exitcond.1, label %for.i.1, label %next
next:
br label %for.i.2
for.i.2:
%indvar.2 = phi i64 [ 0, %next ], [ %indvar.next.2, %for.i.2 ]
%offset.2.a = add i64 %n, -1
%offset.2.b = mul i64 %n, %offset.2.a
%offset.2.c = add i64 %offset.2.b, %indvar.2
%arrayidx.2 = getelementptr float* %X, i64 %offset.2.c
store float 1.000000e+00, float* %arrayidx.2
%indvar.next.2 = add nsw i64 %indvar.2, 1
%exitcond.2 = icmp ne i64 %indvar.next.2, %n
br i1 %exitcond.2, label %for.i.2, label %exit
exit:
ret void
}