Changes in multidim_nested_start_integer.ll

5c028081 - [polly] Introduce -polly-print-* passes to replace -analyze.

Mon, 14 Mar 2022 13:39:25 +0000

[polly] Introduce -polly-print-* passes to replace -analyze.The `opt -analyze` option only works with the legacy pass manager and might be removed in the future, as explained in llvm.org/PR53733. This patch introduced -polly-print-* passes that print what the pass would print with the `-analyze` option and replaces all uses of `-analyze` in the regression tests.There are two exceptions: `CodeGen\single_loop_param_less_equal.ll` and `CodeGen\loop_with_condition_nested.ll` use `-analyze on the `-loops` pass which is not part of Polly.Reviewed By: aeubanksDifferential Revision: https://reviews.llvm.org/D120782 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll

4ba65a56 - [GSoC 2016]New function pass ScopInfoWrapperPass

Mon, 27 Jun 2016 09:32:30 +0000

[GSoC 2016]New function pass ScopInfoWrapperPassThis patch adds a new function pass ScopInfoWrapperPass so that thepolyhedral description of a region, the SCoP, can be constructed andused in a function pass.Patch by Utpal Bora Differential Revision: http://reviews.llvm.org/D20962llvm-svn: 273856 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll

25e8ebe2 - Drop explicit -polly-delinearize parameter

Wed, 23 Mar 2016 13:21:02 +0000

Drop explicit -polly-delinearize parameterDelinearization is now enabled by default and does not need to explicitly needto be enabled in our tests.llvm-svn: 264154 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll

959a8dc3 - Update to ISL 0.16.1

Fri, 15 Jan 2016 15:54:45 +0000

Update to ISL 0.16.1llvm-svn: 257898 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll

5a9a65e4 - Prepare unit tests for update to ISL 0.16

Fri, 15 Jan 2016 00:48:42 +0000

Prepare unit tests for update to ISL 0.16ISL 0.16 will change how sets are printed which breaks 117 unit teststhat text-compare printed sets. This patch re-formats most of these unittests using a script and small manual editing on top of that. Whenactually updating ISL, most work is done by just re-running the scriptto adapt to the changed output.Some tests that compare IR and tests with single CHECK-lines that can beeasily updated manually are not included here.The re-format script will also be committed afterwards. The per-testformatter invocation command lines options will not be added in the nearfuture because it is ad hoc and would overwrite the manual edits.Ideally it also shouldn't be required anymore because ISL's set printinghas become more stable in 0.16.Differential Revision: http://reviews.llvm.org/D16095llvm-svn: 257851 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll

f4ee371e - tests: Drop -polly-detect-unprofitable and -polly-no-early-exit

Tue, 06 Oct 2015 15:36:44 +0000

tests: Drop -polly-detect-unprofitable and -polly-no-early-exitThese flags are now always passed to all tests and need to be disabled ifnot needed. Disabling these flags, rather than passing them to almost alltests, significantly simplfies our RUN: lines.llvm-svn: 249422 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll

96425c25 - Traverse the SCoP to compute non-loop-carried domain conditions

Sun, 30 Aug 2015 21:13:53 +0000

Traverse the SCoP to compute non-loop-carried domain conditions In order to compute domain conditions for conditionals we will now traverse the region in the ScopInfo once and build the domains for each block in the region. The SCoP statements can then use these constraints when they build their domain. The reason behind this change is twofold: 1) This removes a big chunk of preprocessing logic from the TempScopInfo, namely the Conditionals we used to build there. Additionally to moving this logic it is also simplified. Instead of walking the dominance tree up for each basic block in the region (as we did before), we now traverse the region only once in order to collect the domain conditions. 2) This is the first step towards the isl based domain creation. The second step will traverse the region similar to this step, however it will propagate back edge conditions. Once both are in place this conditional handling will allow multiple exit loops additional logic.Reviewers: grosserDifferential Revision: http://reviews.llvm.org/D12428llvm-svn: 246398 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll

173ecab7 - Remove target triples from test cases

Tue, 21 Apr 2015 14:28:02 +0000

Remove target triples from test casesI just learned that target triples prevent test cases to be run on otherarchitectures. Polly test cases are until now sufficiently target independentto not require any target triples. Hence, we drop them.llvm-svn: 235384 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll

54839311 - Rename 'scattering' to 'schedule'

Tue, 21 Apr 2015 11:37:25 +0000

Rename 'scattering' to 'schedule'In Polly we used both the term 'scattering' and the term 'schedule' to describethe execution order of a statement without actually distinguishing between them.We now uniformly use the term 'schedule' for the execution order. Thiscorresponds to the terminology of isl.History: CLooG introduced the term scattering as the generated code can be usedas a sequential execution order (schedule) or as a parallel dimensionenumerating different threads of execution (placement). In Polly and/or isl theterm placement was never used, but we uniformly refer to an execution order as aschedule and only later introduce parallelism. When doing so we do not talkabout about specific placement dimensions.llvm-svn: 235380 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll

bad3ff20 - Update Polly tests to handle explicitly typed gep changes in LLVM

Fri, 27 Feb 2015 19:20:19 +0000

Update Polly tests to handle explicitly typed gep changes in LLVMllvm-svn: 230784 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll

d1e33e70 - ScopDetection: Only detect scops that have at least one read and one write

Thu, 19 Feb 2015 05:31:07 +0000

ScopDetection: Only detect scops that have at least one read and one writeScops that only read seem generally uninteresting and scops that only write aremost likely initializations where there is also little to optimize. To notwaste compile time we bail early.Differential Revision: http://reviews.llvm.org/D7735llvm-svn: 229820 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll

9282076e - [NFC] Drop the "scattering" tuple name

Mon, 02 Feb 2015 13:45:54 +0000

[NFC] Drop the "scattering" tuple namellvm-svn: 227801 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll

3f296196 - Drop all constant scheduling dimensions

Thu, 01 Jan 2015 23:01:11 +0000

Drop all constant scheduling dimensionsSchedule dimensions that have the same constant value accross all statements donot carry any information, but due to the increased dimensionality of theschedule cost compile time. To not pay this cost, we remove constant dimensionsif possible.llvm-svn: 225067 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll

683b8e44 - Remove -polly-codegen-scev option and related code

Sun, 30 Nov 2014 14:33:31 +0000

Remove -polly-codegen-scev option and related codeSCEV based code generation has been the default for two weeks after havingbeen tested for a long time. We now drop the support the non-scev-based codegeneration.llvm-svn: 222978 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll

5e6813d1 - Derive run-time conditions for delinearization

Wed, 02 Jul 2014 17:47:48 +0000

Derive run-time conditions for delinearizationAs our delinearization works optimistically, we need in some cases run-timechecks that verify our optimistic assumptions. A simple example is thefollowing code:void foo(long n, long m, long o, double A[n][m][o]) { for (long i = 0; i < 100; i++) for (long j = 0; j < 150; j++) for (long k = 0; k < 200; k++) A[i][j][k] = 1.0;}After clang linearized the access to A and we delinearized it again toA[i][j][k] we need to ensure that we do not access the delinearized arrayout of bounds (this information is not available in LLVM-IR). Hence, weneed to verify the following constraints at run-time:CHECK: Assumed Context:CHECK: [o, m] -> { : m >= 150 and o >= 200 }llvm-svn: 212198 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll

5416a039 - Adjust multidim test cases to not access out-of-bound memory

Wed, 04 Jun 2014 11:47:54 +0000

Adjust multidim test cases to not access out-of-bound memoryWe do this currently only for test cases where we have integer offsets thatclearly access array dimensions out-of-bound.-; for (long i = 0; i < n; i++)-; for (long j = 0; j < m; j++)-; for (long k = 0; k < o; k++)+; for (long i = 0; i < n - 3; i++)+; for (long j = 4; j < m; j++)+; for (long k = 0; k < o - 7; k++) ; A[i+3][j-4][k+7] = 1.0;This will be helpful if we later want to simplify the access functions under theassumption that they do not access memory out of bounds.llvm-svn: 210179 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll

1801668a - delinearize memory access functions

Tue, 08 Apr 2014 21:20:44 +0000

delinearize memory access functionsllvm-svn: 205799 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll

e42ddb9a - ScopInfo: Split start value from SCEVAddRecExpr to enable parameter sharing.

Mon, 05 Aug 2013 15:14:15 +0000

ScopInfo: Split start value from SCEVAddRecExpr to enable parameter sharing.SCoP invariant parameters with the different start value would deter parametersharing. For example, when compiling the following C code: void foo(float *input) { for (long j = 0; j < 8; j++) { // SCoP begin for (long i = 0; i < 8; i++) { float x = input[j * 64 + i + 1]; input[j * 64 + i] = x * x; } } }Polly would creat two parameters for these memory accesses: p_0: {0,+,256} p_2: {4,+,256} [j * 64 + i + 1] => MemRef_input[o0] : 4o0 = p_1 + 4i0 [j * 64 + i] => MemRef_input[o0] : 4o0 = p_0 + 4i0These parameters only differ from start value. To enable parameter sharing,we split the start value from SCEVAddRecExpr, so they would share a singleparameter that always has zero start value: p0: {0,+,256}<%for.cond1.preheader> [j * 64 + i + 1] => MemRef_input[o0] : 4o0 = 4 + p_1 + 4i0 [j * 64 + i] => MemRef_input[o0] : 4o0 = p_0 + 4i0Such translation can make the polly-dependence much faster.Contributed-by: Star Tan llvm-svn: 187728 List of files: /llvm-project-15.0.7/polly/test/ScopInfo/multidim_nested_start_integer.ll