maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/polly/test/ScopDetect/non-affine-loop.ll

65 lines
2.6 KiB
LLVM
Raw Normal View History

tests: Drop -polly-detect-unprofitable and -polly-no-early-exit These flags are now always passed to all tests and need to be disabled if not needed. Disabling these flags, rather than passing them to almost all tests, significantly simplfies our RUN: lines. llvm-svn: 249422
2015-10-06 23:36:44 +08:00
; RUN: opt %loadPolly -polly-detect -polly-allow-nonaffine-branches \
; RUN: -polly-allow-nonaffine-loops=false \
Do not detect Scops with only one loop. If a region does not have more than one loop, we do not identify it as a Scop in ScopDetection. The main optimizations Polly is currently performing (tiling, preparation for outer-loop vectorization and loop fusion) are unlikely to have a positive impact on individual loops. In some cases, Polly's run-time alias checks or conditional hoisting may still have a positive impact, but those are mostly enabling transformations which LLVM already performs for individual loops. As we do not focus on individual loops, we leave them untouched to not introduce compile time regressions and execution time noise. This results in good compile time reduction (oourafft: -73.99%, smg2000: -56.25%). Contributed-by: Pratik Bhatu <cs12b1010@iith.ac.in> Reviewers: grosser Differential Revision: http://reviews.llvm.org/D12268 llvm-svn: 246161
2015-08-28 00:55:18 +08:00
; RUN: -analyze < %s | FileCheck %s --check-prefix=REJECTNONAFFINELOOPS
tests: Drop -polly-detect-unprofitable and -polly-no-early-exit These flags are now always passed to all tests and need to be disabled if not needed. Disabling these flags, rather than passing them to almost all tests, significantly simplfies our RUN: lines. llvm-svn: 249422
2015-10-06 23:36:44 +08:00
; RUN: opt %loadPolly -polly-detect -polly-allow-nonaffine-branches \
; RUN: -polly-allow-nonaffine-loops=true \
Do not detect Scops with only one loop. If a region does not have more than one loop, we do not identify it as a Scop in ScopDetection. The main optimizations Polly is currently performing (tiling, preparation for outer-loop vectorization and loop fusion) are unlikely to have a positive impact on individual loops. In some cases, Polly's run-time alias checks or conditional hoisting may still have a positive impact, but those are mostly enabling transformations which LLVM already performs for individual loops. As we do not focus on individual loops, we leave them untouched to not introduce compile time regressions and execution time noise. This results in good compile time reduction (oourafft: -73.99%, smg2000: -56.25%). Contributed-by: Pratik Bhatu <cs12b1010@iith.ac.in> Reviewers: grosser Differential Revision: http://reviews.llvm.org/D12268 llvm-svn: 246161
2015-08-28 00:55:18 +08:00
; RUN: -analyze < %s | FileCheck %s --check-prefix=ALLOWNONAFFINELOOPS
tests: Drop -polly-detect-unprofitable and -polly-no-early-exit These flags are now always passed to all tests and need to be disabled if not needed. Disabling these flags, rather than passing them to almost all tests, significantly simplfies our RUN: lines. llvm-svn: 249422
2015-10-06 23:36:44 +08:00
; RUN: opt %loadPolly -polly-detect -polly-allow-nonaffine-branches \
[FIX] Count affine loops correctly The "unprofitable" heuristic was broken and counted boxed loops even though we do not represent and optimize them. llvm-svn: 249274
2015-10-04 22:56:08 +08:00
; RUN: -polly-allow-nonaffine-loops=false -polly-allow-nonaffine \
; RUN: -analyze < %s | FileCheck %s \
; RUN: --check-prefix=ALLOWNONAFFINEREGIONSANDACCESSES
tests: Drop -polly-detect-unprofitable and -polly-no-early-exit These flags are now always passed to all tests and need to be disabled if not needed. Disabling these flags, rather than passing them to almost all tests, significantly simplfies our RUN: lines. llvm-svn: 249422
2015-10-06 23:36:44 +08:00
; RUN: opt %loadPolly -polly-detect -polly-allow-nonaffine-branches \
Do not detect Scops with only one loop. If a region does not have more than one loop, we do not identify it as a Scop in ScopDetection. The main optimizations Polly is currently performing (tiling, preparation for outer-loop vectorization and loop fusion) are unlikely to have a positive impact on individual loops. In some cases, Polly's run-time alias checks or conditional hoisting may still have a positive impact, but those are mostly enabling transformations which LLVM already performs for individual loops. As we do not focus on individual loops, we leave them untouched to not introduce compile time regressions and execution time noise. This results in good compile time reduction (oourafft: -73.99%, smg2000: -56.25%). Contributed-by: Pratik Bhatu <cs12b1010@iith.ac.in> Reviewers: grosser Differential Revision: http://reviews.llvm.org/D12268 llvm-svn: 246161
2015-08-28 00:55:18 +08:00
; RUN: -polly-allow-nonaffine-loops=true -polly-allow-nonaffine \
Move more compile-time bailouts into -polly-detect-unprofitable Instead of having two separate options -polly-detect-scops-in-functions-without-loops and -polly-detect-scops-in-regions-without-loops we now just use -polly-detect-unprofitable to force the detection of scops ignoring any compile time saving bailout heuristics. llvm-svn: 247057
2015-09-09 03:46:41 +08:00
; RUN: -analyze < %s | FileCheck %s \
Do not detect Scops with only one loop. If a region does not have more than one loop, we do not identify it as a Scop in ScopDetection. The main optimizations Polly is currently performing (tiling, preparation for outer-loop vectorization and loop fusion) are unlikely to have a positive impact on individual loops. In some cases, Polly's run-time alias checks or conditional hoisting may still have a positive impact, but those are mostly enabling transformations which LLVM already performs for individual loops. As we do not focus on individual loops, we leave them untouched to not introduce compile time regressions and execution time noise. This results in good compile time reduction (oourafft: -73.99%, smg2000: -56.25%). Contributed-by: Pratik Bhatu <cs12b1010@iith.ac.in> Reviewers: grosser Differential Revision: http://reviews.llvm.org/D12268 llvm-svn: 246161
2015-08-28 00:55:18 +08:00
; RUN: --check-prefix=ALLOWNONAFFINELOOPSANDACCESSES
Introduce -polly-process-unprofitable This single option replaces -polly-detect-unprofitable and -polly-no-early-exit and is supposed to be the only option that disables compile-time heuristics that aim to bail out early on scops that are believed to not benefit from Polly optimizations. Suggested-by: Johannes Doerfert llvm-svn: 249426
2015-10-07 00:10:29 +08:00
; RUN: opt %loadPolly -polly-process-unprofitable=false \
tests: Explicitly state if profitability tests should be used Polly's profitability heuristic saves compile time by skipping trivial scops or scops were we know no good optimization can be applied. For almost all our tests this heuristic makes little sense as we aim for minimal test cases when testing functionality. Hence, in almost all cases this heuristic is better be disabled. In preparation of disabling Polly's compile time heuristic by default in the test suite we first explicitly enable it in the couple of test cases that really use it (or run with/without heuristic side-by-side). llvm-svn: 249418
2015-10-06 23:19:35 +08:00
; RUN: -polly-detect -polly-allow-nonaffine-branches \
[FIX] Count affine loops correctly The "unprofitable" heuristic was broken and counted boxed loops even though we do not represent and optimize them. llvm-svn: 249274
2015-10-04 22:56:08 +08:00
; RUN: -polly-allow-nonaffine-loops=true -polly-allow-nonaffine \
; RUN: -analyze < %s | FileCheck %s \
; RUN: --check-prefix=PROFIT
Allow loops in non-affine subregions -- SCoP Modeling This will allow the ScopInfo to build the polyhedral representation for non-affine regions that contain loops. Such loops are basically not visible in the SCoP representation. Accesses that are variant in such loops are therefor represented as non-affine accesses. Differential Revision: http://reviews.llvm.org/D8153 llvm-svn: 234713
2015-04-13 06:58:40 +08:00
;
; This function/region does contain a loop, however it is non-affine, hence the access
; A[i] is also. Furthermore, it is the only loop, thus when we over approximate
; non-affine loops __and__ accesses __and__ allow regins without a (affine) loop we will
; detect it, otherwise we won't.
;
; void f(int *A) {
; for (int i = 0; i < A[i]; i++)
; A[-1]++;
; }
;
[FIX] Count affine loops correctly The "unprofitable" heuristic was broken and counted boxed loops even though we do not represent and optimize them. llvm-svn: 249274
2015-10-04 22:56:08 +08:00
; REJECTNONAFFINELOOPS-NOT: Valid
; ALLOWNONAFFINELOOPS-NOT: Valid
; ALLOWNONAFFINEREGIONSANDACCESSES-NOT: Valid
; ALLOWNONAFFINELOOPSANDACCESSES: Valid
; PROFIT-NOT: Valid
Allow loops in non-affine subregions -- SCoP Modeling This will allow the ScopInfo to build the polyhedral representation for non-affine regions that contain loops. Such loops are basically not visible in the SCoP representation. Accesses that are variant in such loops are therefor represented as non-affine accesses. Differential Revision: http://reviews.llvm.org/D8153 llvm-svn: 234713
2015-04-13 06:58:40 +08:00
;
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
define void @f(i32* %A) {
bb:
br label %bb1
bb1: ; preds = %bb9, %bb
%indvars.iv = phi i64 [ %indvars.iv.next, %bb9 ], [ 0, %bb ]
%tmp = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
%tmp2 = load i32, i32* %tmp, align 4
%tmp3 = sext i32 %tmp2 to i64
%tmp4 = icmp slt i64 %indvars.iv, %tmp3
br i1 %tmp4, label %bb5, label %bb10
bb5: ; preds = %bb1
%tmp6 = getelementptr inbounds i32, i32* %A, i64 -1
%tmp7 = load i32, i32* %tmp6, align 4
%tmp8 = add nsw i32 %tmp7, 1
store i32 %tmp8, i32* %tmp6, align 4
br label %bb9
bb9: ; preds = %bb5
%indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
br label %bb1
bb10: ; preds = %bb1
ret void
}