[Polly] DeLICM/DePRE (WIP)

[Meinersbur]

Summary

Implement -polly-delicm pass. The pass intends to undo the effects of LoadInvariantCodeMotion and GVN's Partial Redundancy Elimination (Load PTR) which adds additional scalar dependencies into scops.

DeLICM/DePRE will try to map those scalars back to the array elements they were promoted from, as long as the array element is unused.

This is work in progress. The patch is not as tidy as it could be, not all functions have been commended yet, most test cases fail and some refactoring is still planned.

[grosser]

include/polly/DeLICM.h

61

InclRedef=true RETURNS a

67

ARE ignored

70

"as where" -- Not sure what this means. Do you want to say:

"Whether to include the definition's timepoint to the set of well-defined elements"

You can probably also drop the "Whether to" at the beginning of the sentence.

72

enabling THIS OPTION adds

97

RESULTS in:

102

IS overwritten THE FIRST TIME by Write

"when before timepoint" : when what is before timepoint? Can you add the object for clarity? Is it "the write at timepoint [i] is before timepoint 0"?

108

ARE ignored

111

"an overwrite timepoints": the numerus seems to not match

146

The result IS

159

ARE ignored

164

load READS

166

loads use

167

THAT it had

175

written IS used

176

IS considered

lib/Transform/DeLICM.cpp

23

i.e.,

24

isl rational

40

i.e., strictly speaking

59

isl reference (isl is written lowercase)

follow isl syntax

230

I would place the "." after the ")"

256

"." after ")"

296

IN the result

298

IN the result

303

isl max

330

i.e.,

353

does not

355

i.e.,

500

are ignored.

504

comment flow seems wrong (below as well)

552

AND

does not need

611

comment flow seems wrong

650

compute_divs can sometimes make a map a lot more complex. Did you had a case where this is necessary.

We also should ask sven at some point if he can provide a simplify function for isl that does all simplifications in the right order.

678

"." at the end?

696

"st."?

708

of the contents <sth missing here?> any array elements in any zone

711

one OF three

712

so A transformation

715

Leftover documentation from the 'Known' part?

731

problems problem?

732

known ? Is this a leftover?

733

Last sentence is incomplete

838

Comment flow broken.

849

ARE the new lifetimes required for proposed UNUSED in existing?

895

? at the end

917

isl

942

a reference a counter????

949

I doubt this has any performance impact. I would prefer to keep the state local and not pollute global space with such caching things.

1076

IS as

1115

Get

1117

IS as

1209

A MemoryKind::Value ... A MemoryKind

1214

A ... A

1226

the transformation IS applied to the

1276

WITH each other

1339

of A

1399

HAS.

1520

comment flow broken

1532

USED?

1657

Map A

unittests/DeLICM/DeLICMTest.cpp

23

indef -> undef?

[Meinersbur]

lib/Transform/DeLICM.cpp

650

I am not aware that this can make sets more complex, it just provides an additional explicit representation of a div that should make follow-up computations faster and was sometimes necessary for coalescing.

According to the source comments the issue is rather that compute_divs can be an expensive operation, but so is coalescing. If possible I'd not call simplify() at all but that results in other operations taking much longer, e.g. lexmin.

unittests/DeLICM/DeLICMTest.cpp

23

This refers to the definition of indef below

[Meinersbur]

• Address Tobias' comments

[grosser]

Hi Michael,

some of the comments from our discussion.

Best,
Tobias

include/polly/DeLICM.h

36

Add comment?

lib/Transform/DeLICM.cpp

40

Comma after i.e., missing.

83

Maybe give an overview comment describing the high-level structure of this approach.

355

Comma after i.e., missing.

496

Several of these functions above are not really tied to your DeLICM approach, but are generally useful. I would suggest to add these separately in an ISLTools.cpp file, in the optimal case with separate unit tests.

You can probably already upstream these today.

650

OK. We probably need to evaluate this in practice. Let's start with what you have.

708

Pull in getAccessRelation.

942

Drop this one as well?

1076

IS BE?? Just 'is as'?

1111

The above all really seem to be belong into ScopInfo. Let's leave them here for know, as this makes updating this patch easier, but maybe we can keep in mind that we want to unify this at some point.

1791

OK. I think this functionality should at some point be merged with ScopInfo, but let's leave it here for know to reduce the dependences to ScopInfo.

1864

I believe we should not use auto so much. As you correctly pointed out, auto is suggested by LLVM to be only used in rare cases. We had in Barcelone a discussion how much we should use auto and it seems we choose a rather liberal direction. I now believe that following LLVM here closer is better.

Now, currently auto makes sense as it saves the long IslPTr<.....> type so you probably do not need to change anything here immediately. We should just keep this in mind, when potentially switching to the C++ bindings.

unittests/DeLICM/DeLICMTest.cpp

424

To fix formatting, maybe use:

struct {
  type *Known, type *undef, type *written, type *unkown
} input;

void foo() {
  EXPECT_FALSE(conflicts(
      {aasdfsadf, basdfsadf, casdfasf, dsadfsadfsadfdsafdsafsafdfasdfsadf},
      {a, basdfsafd, casdfdsaf, dsadfsadf}));
}

427

It also seems as if these tests still cover the Known part. It would be nice if we could have a set of simpler tests that just take the current simpler isConflicting implementation and do not go through the complexity of checkIsConflicting.

[Meinersbur]

• Rebase to r293429
• Remove MapReport
• Review isConflicting unittests.
• Address remaining comments

[grosser]

Hi Michael,

thanks a lot for the very nice cleanup. I am very glad this patch proceeds this way. I may now seem to be very picky, but I really try to understand this one in detail. I added a bunch of smaller comments. I am not yet fully through, but believe I identified a new set of changes we can commit. All of checkConvertZoneToTimepoint, computeReachingDefinition, computeReachingOverwrite, and checkComputeArrayUnused are clearly self-contained and thanks to your nice unit tests are now easy to test independently. I suggest you commit them one-by-one. I did not go through all corner cases, but as they are well tested that seems to be fine.

The only concern I have is that there is some -- possibly unnecessary -- code duplication between computeReachingDefinition and computeReachingOverwrite, which I would prefer to avoid.

I also noted that you provide implementations for all corner cases always including or not including bounds. On the other side, most of the actual code is currently not using any of these (there is commonly only one configuration used). As this all looks very well tested, I don't want to ask you to drop all the special cases, especially as I have the impression that these may be useful for some of the later code or later discussions on how to model certain things best. Hence, I am fine leaving them in. However, if you have any cases in mind that are certainly not needed any more, don't hesitate to drop them.

As a next step, I will go through Knowledge and isConflicting.

Best,
Tobias

include/polly/DeLICM.h

45

The set OF timepoints

51

IS overwritten

79

This is only ever called with InclStart = true and InclEnd=false. Is this intentional and do you keep the other options just for completness?

146

(dover)write -> (over)write

180

NOT read in between

lib/Transform/DeLICM.cpp

135

one definition PER SCoP

143

The definition (write MemoryAccess) of a MemoryKind::Value scalar.

definitions -> definion
/ -> ()
an ->a

148

uses (read MemoryAccesses) of a MemoryKind::Value

/ -> ()
an -> a

152

The PHI instruction (read Memory Access) of a MemoryKind::PHI or MemoryKind::ExitPHI.

/ -> ()
an -> a
/ -> or (makes clear if this is an 'or' or an 'and')

156

List of all incoming values (write MemoryAccesses) for a MemoryKind::PHI or MemoryKind::ExitPHI scalar.

175

This fits into two lines

245

This is only ever called with

computeScalarReachingOverwrite(Schedule, TargetDom, false, true);

295

This is only ever called with "computeScalarReachingDefinition(Schedule, Domain, false, true)".

1489

The previous four lines can be dropped, but changing !MA->isWrite() to !MA->isMustWrite() a couple of lines earlier.

1709

This is only called with:

auto ArrayUnused = computeArrayUnused(Schedule, AllMustWrites, AllReads,

unittests/DeLICM/DeLICMTest.cpp

23

Right, I still don't understand why you call this 'indef'. This seems to be an abbreviation of "undefined", but then I would expect this word to be called "undef". What does "indef" stand for?

[grosser]

lib/Transform/DeLICM.cpp

187

This fits into two lines.

235

What is an "overwrite timepoints"?

284

What does not need to be specified?

and therefore THIS ELEMENT does not ...

356

of TWO states

370

the theses?

KnowledgeS need to know

468

This -> Existing
That -> Proposed

?

560

must be the declared .. Grammar?

619

before other accessES AND

621

Is this true? What about region statements that write to a PHI node?

638

Maybe add a test case?

641

If you add a continue above, the second MA->isWrite is not needed. This reduces indentation and also makes clear that this is an if-else construct.

647

What would be such a case? Is there an easy opportunity for a test case?

653

Should we have a test case for this? (including a comment in the test case that states that this is the reason it is not DeLICMed)

655

Should we have a test case for this? (including a comment in the test case that states that this is the reason it is not DeLICMed.

685

No {}.

689

No {}.

712

is as narrow

Drop 'be'

820

Are the last two definitions dead?

1617

Should this be ScatterRead instead of ScatterUse? You do not seem to modify the range. In computeReachingOverwrite you seem to call it ScatterRead as well.

1651

This assert is missing in computeReachingDefinition

1668

This should be called AfterMap, right? Maybe an oversight when renaming and adopting computeReachingDefinition?

1692

In computeReachingDefinition SelfUse has been hoisted outside the condition. Not a semantic difference, but a minor change that lets the code look different.

1702

Besides the comparisions isl_map_lex_lt/isl_map_lex_le this function seems to be indentical with computeReachingDefinition. It feels to me as if this is unnecessary complication of rather complicated code. Would it make sense to have one common implementation that just takes a comparision function to then deliver the two different implementations?

[grosser]

Hi Michael,

I just looked closer into Knowledge and isConflicting.

To me it is not well documented what "implicit" means. As far as I understand implicit means that only one of unknown and unused is actually stored and the other one is "implicitly" the opposite. It would be good to document this and also give a reason why you implemented it this way. I assume because the complement of one of these sets can possibly be very large.

I am not sure if Knowledges must be implicit or if also both sets can be explicit. In isConflicting you state "current implementation requires it to be implicit", but when testing you seem to compute two explicit sets using completeLifetime. This seems inconsistent.

Do you have an example that makes clear why you need zones, and cannot just track the values / life ranges? That seems like an interesting test for isConflicting.

Assuming both implicit and explicit representation are supported. I wonder if one of them would be enough? Or is there a reason to keep both? I have the feeling that having both currently makes the code (unnecessarily) hard to read?

That's all for now. I will later look into the remaining stuff.

lib/Transform/DeLICM.cpp

403

a nullptr

435

What does it mean to be implicitly unused. Can you somehow print this, e.g. as "universe - Occupied".

What happens if the first set does not contain elements from each space? Assuming we never write to a given array? This would mean it does not appear in Occupied and is completely unused. Now, the universe we compute won't contain an element from this space, after the (implicit) subtraction it won't appear, right?

You seem to be working around this in the tests, but it is not clear how this is resolved in general or which preconditions need to be satisfied to make this work.

450

The previous line does not have any test coverage? Should we just assert that such a condition is not expected to arise? Or is there a test case that would cover this situation? (Same for the condition below).

454

The previous line does not have any test coverage.

465

WITH each other

477

What do you mean by "use case X" here?

477

What do you mean with 'use case X'?

I am surprised that the current implementation only works with the first parameter being "implicit" unknown. It seems the tests provide an explicit representation of "unknown", no?

I tried to assert(!Existing.Occupied), but this fails (due to the tests using explicit constraints.

1761

Why do we move this stuff here? This does not seem to be performance sensitive.

unittests/DeLICM/DeLICMTest.cpp

379

Why do you compute the universe here? I was under the impression that Knowledges can also be created with implicit Unkown, Unused sets and should work well with them. In fact, there is a comment in isConflicting that claims that it _only_ works with implicit sets?

Would the tests still work if you do not create explicit representations of Unknown and Unused?

397

vs.

Also end this and the sentences below with "."

419

Maybe add a comment that states why these two conflict. Dom[1] is a zone that covers [0,1]. This took me a little while to understand.

[Meinersbur]

Thanks for the review

In D24716#665836, @grosser wrote:

I also noted that you provide implementations for all corner cases always including or not including bounds. On the other side, most of the actual code is currently not using any of these (there is commonly only one configuration used). As this all looks very well tested, I don't want to ask you to drop all the special cases, especially as I have the impression that these may be useful for some of the later code or later discussions on how to model certain things best. Hence, I am fine leaving them in. However, if you have any cases in mind that are certainly not needed any more, don't hesitate to drop them.

I feel relative strongly about keeping them. They were certainly helpful in development when I did not yet have a singular definition of what a "zone" is. Currently a zone 0 < 1 < 1 is consistently represented by the integer set { [1] }. There is no strict necessity to fo this everywhere. E.g. it might be more intuitive to use the start of the range instead. These computations are currently agnostic to what a zone is (could be moved to ISLTools.cpp as well maybe?) and I would perceive it as a loss when they suddenly would.

In D24716#666155, @grosser wrote:

I am not sure if Knowledges must be implicit or if also both sets can be explicit. In isConflicting you state "current implementation requires it to be implicit", but when testing you seem to compute two explicit sets using completeLifetime. This seems inconsistent.

That is comment was outdated. In previous revisions Unused, Known and Unknown was stored in the same map (instead in one Unsued and one Occupied). Two of them had to be defined, the third was implicit, hence "required" to be implicit.

Do you have an example that makes clear why you need zones, and cannot just track the values / life ranges? That seems like an interesting test for isConflicting.

See file comment in DeLICM.cpp

Assuming both implicit and explicit representation are supported. I wonder if one of them would be enough? Or is there a reason to keep both? I have the feeling that having both currently makes the code (unnecessarily) hard to read?

It's just the opposite, it makes it clearer that Unused/Occupied can be nullptr and therefore easier to understand. Previous versions of this patch used a flag which one is implicit, hence also requiring that one of the is implicit. I invite you to look into these versions to see if whether they are easier to understand.

The sets can be nullptr only because we want avoid computing them, not because of something conceptional. If they are still given, they are ignored. This is fine for the unit tests which are small, but the complement operation blew up easily in my experiments with even small programs.

Michael

lib/Transform/DeLICM.cpp

435

What does it mean to be implicitly unused.

The unused set has not been computed explicitly, but is assumed to be the complement of Occupied

Can you somehow print this, e.g. as "universe - Occupied".

I don't understand this suggestion.

What happens if the first set does not contain elements from each space?

Does not happen. In ZoneAlgorithm one of the two sets will always be implicit. In DeLICMTests unionSpace() is used.

Assuming we never write to a given array? This would mean it does not appear in Occupied and is completely unused. Now, the universe we compute won't contain an element from this space, after the (implicit) subtraction it won't appear, right?

As there is no write to the array, greedyCollapse() will not try to map anything to it.

621

True since commit r258809: Unique phi write accesses

638

You previously asked me to remove the non-"After Accesses" parts from the test cases. This currently cannot be tested effectively without more diagnostics.

There would be the possibility to copy reduction_preheader.ll a few times and add variations. In most cases the variations would already cause it not to be mapped (longer lifetimes), not because the SCoP is incompatible (Test would succeed even if this check was removed)

So I could add loads/store combinations to other array elements than the ones necessary for at least one mapping. But in the mid term I'd only want to mark the elements as incompatible, not the entire SCoP. That would then require to rewrite all tests.

Originally I intended to more systematically test all combinations of loads and stores orders in one ScopStmt when more diagnostics is available. I'd add that separately sometime later.

647

memcpy, memmove, memset

About testing, see comment above

653

About testing, see comment above

655

About testing, see comment above

1489

That wouldn't print a debug message.

1761

This is to avoid to exposing Knowledge only for testing isConflicting.

unittests/DeLICM/DeLICMTest.cpp

379

Why do you compute the universe here?

To compute an argument that got only a nullptr.

I was under the impression that Knowledges can also be created with implicit Unkown, Unused sets and should work well with them. In fact, there is a comment in isConflicting that claims that it _only_ works with implicit sets?

See main comment.

Would the tests still work if you do not create explicit representations of Unknown and Unused?

Yes, but the explicit representation is ignored.

[Meinersbur]

• Rebase to r293890
• Address Tobias' comments
• Merge reachingDefinition and reachingOverwrite to reachingWrite
• Revise unittests

[grosser]

In D24716#666384, @Meinersbur wrote:

Thanks for the review

In D24716#665836, @grosser wrote:

I also noted that you provide implementations for all corner cases always including or not including bounds. On the other side, most of the actual code is currently not using any of these (there is commonly only one configuration used). As this all looks very well tested, I don't want to ask you to drop all the special cases, especially as I have the impression that these may be useful for some of the later code or later discussions on how to model certain things best. Hence, I am fine leaving them in. However, if you have any cases in mind that are certainly not needed any more, don't hesitate to drop them.

I feel relative strongly about keeping them. They were certainly helpful in development when I did not yet have a singular definition of what a "zone" is. Currently a zone 0 < 1 < 1 is consistently represented by the integer set { [1] }. There is no strict necessity to fo this everywhere. E.g. it might be more intuitive to use the start of the range instead. These computations are currently agnostic to what a zone is (could be moved to ISLTools.cpp as well maybe?) and I would perceive it as a loss when they suddenly would.

Sure, that's why I said I am fine with these to be kept in the current generality, as long as they are well

In D24716#666155, @grosser wrote:

I am not sure if Knowledges must be implicit or if also both sets can be explicit. In isConflicting you state "current implementation requires it to be implicit", but when testing you seem to compute two explicit sets using completeLifetime. This seems inconsistent.

That is comment was outdated. In previous revisions Unused, Known and Unknown was stored in the same map (instead in one Unsued and one Occupied). Two of them had to be defined, the third was implicit, hence "required" to be implicit.

Do you have an example that makes clear why you need zones, and cannot just track the values / life ranges? That seems like an interesting test for isConflicting.

See file comment in DeLICM.cpp

Assuming both implicit and explicit representation are supported. I wonder if one of them would be enough? Or is there a reason to keep both? I have the feeling that having both currently makes the code (unnecessarily) hard to read?

It's just the opposite, it makes it clearer that Unused/Occupied can be nullptr and therefore easier to understand. Previous versions of this patch used a flag which one is implicit, hence also requiring that one of the is implicit. I invite you to look into these versions to see if whether they are easier to understand.

The sets can be nullptr only because we want avoid computing them, not because of something conceptional. If they are still given, they are ignored. This is fine for the unit tests which are small, but the complement operation blew up easily in my experiments with even small programs.

Michael

In D24716#666384, @Meinersbur wrote:

Thanks for the review

In D24716#665836, @grosser wrote:

I also noted that you provide implementations for all corner cases always including or not including bounds. On the other side, most of the actual code is currently not using any of these (there is commonly only one configuration used). As this all looks very well tested, I don't want to ask you to drop all the special cases, especially as I have the impression that these may be useful for some of the later code or later discussions on how to model certain things best. Hence, I am fine leaving them in. However, if you have any cases in mind that are certainly not needed any more, don't hesitate to drop them.

I feel relative strongly about keeping them. They were certainly helpful in development when I did not yet have a singular definition of what a "zone" is. Currently a zone 0 < 1 < 1 is consistently represented by the integer set { [1] }. There is no strict necessity to fo this everywhere. E.g. it might be more intuitive to use the start of the range instead. These computations are currently agnostic to what a zone is (could be moved to ISLTools.cpp as well maybe?) and I would perceive it as a loss when they suddenly would.

In D24716#666155, @grosser wrote:

I am not sure if Knowledges must be implicit or if also both sets can be explicit. In isConflicting you state "current implementation requires it to be implicit", but when testing you seem to compute two explicit sets using completeLifetime. This seems inconsistent.

That is comment was outdated. In previous revisions Unused, Known and Unknown was stored in the same map (instead in one Unsued and one Occupied). Two of them had to be defined, the third was implicit, hence "required" to be implicit.

Do you have an example that makes clear why you need zones, and cannot just track the values / life ranges? That seems like an interesting test for isConflicting.

See file comment in DeLICM.cpp

Assuming both implicit and explicit representation are supported. I wonder if one of them would be enough? Or is there a reason to keep both? I have the feeling that having both currently makes the code (unnecessarily) hard to read?

It's just the opposite, it makes it clearer that Unused/Occupied can be nullptr and therefore easier to understand. Previous versions of this patch used a flag which one is implicit, hence also requiring that one of the is implicit. I invite you to look into these versions to see if whether they are easier to understand.

The sets can be nullptr only because we want avoid computing them, not because of something conceptional. If they are still given, they are ignored. This is fine for the unit tests which are small, but the complement operation blew up easily in my experiments with even small programs.

Michael

In D24716#666384, @Meinersbur wrote:

Thanks for the review

In D24716#665836, @grosser wrote:

I also noted that you provide implementations for all corner cases always including or not including bounds. On the other side, most of the actual code is currently not using any of these (there is commonly only one configuration used). As this all looks very well tested, I don't want to ask you to drop all the special cases, especially as I have the impression that these may be useful for some of the later code or later discussions on how to model certain things best. Hence, I am fine leaving them in. However, if you have any cases in mind that are certainly not needed any more, don't hesitate to drop them.

I feel relative strongly about keeping them. They were certainly helpful in development when I did not yet have a singular definition of what a "zone" is. Currently a zone 0 < 1 < 1 is consistently represented by the integer set { [1] }. There is no strict necessity to fo this everywhere. E.g. it might be more intuitive to use the start of the range instead. These computations are currently agnostic to what a zone is (could be moved to ISLTools.cpp as well maybe?) and I would perceive it as a loss when they suddenly would.

In D24716#666155, @grosser wrote:

I am not sure if Knowledges must be implicit or if also both sets can be explicit. In isConflicting you state "current implementation requires it to be implicit", but when testing you seem to compute two explicit sets using completeLifetime. This seems inconsistent.

That is comment was outdated. In previous revisions Unused, Known and Unknown was stored in the same map (instead in one Unsued and one Occupied). Two of them had to be defined, the third was implicit, hence "required" to be implicit.

Do you have an example that makes clear why you need zones, and cannot just track the values / life ranges? That seems like an interesting test for isConflicting.

See file comment in DeLICM.cpp

Assuming both implicit and explicit representation are supported. I wonder if one of them would be enough? Or is there a reason to keep both? I have the feeling that having both currently makes the code (unnecessarily) hard to read?

It's just the opposite, it makes it clearer that Unused/Occupied can be nullptr and therefore easier to understand. Previous versions of this patch used a flag which one is implicit, hence also requiring that one of the is implicit. I invite you to look into these versions to see if whether they are easier to understand.

The sets can be nullptr only because we want avoid computing them, not because of something conceptional. If they are still given, they are ignored. This is fine for the unit tests which are small, but the complement operation blew up easily in my experiments with even small programs.

Michael

In D24716#666384, @Meinersbur wrote:

Thanks for the review

In D24716#665836, @grosser wrote:

I also noted that you provide implementations for all corner cases always including or not including bounds. On the other side, most of the actual code is currently not using any of these (there is commonly only one configuration used). As this all looks very well tested, I don't want to ask you to drop all the special cases, especially as I have the impression that these may be useful for some of the later code or later discussions on how to model certain things best. Hence, I am fine leaving them in. However, if you have any cases in mind that are certainly not needed any more, don't hesitate to drop them.

I feel relative strongly about keeping them. They were certainly helpful in development when I did not yet have a singular definition of what a "zone" is. Currently a zone 0 < 1 < 1 is consistently represented by the integer set { [1] }. There is no strict necessity to fo this everywhere. E.g. it might be more intuitive to use the start of the range instead. These computations are currently agnostic to what a zone is (could be moved to ISLTools.cpp as well maybe?) and I would perceive it as a loss when they suddenly would.

In D24716#666155, @grosser wrote:

I am not sure if Knowledges must be implicit or if also both sets can be explicit. In isConflicting you state "current implementation requires it to be implicit", but when testing you seem to compute two explicit sets using completeLifetime. This seems inconsistent.

That is comment was outdated. In previous revisions Unused, Known and Unknown was stored in the same map (instead in one Unsued and one Occupied). Two of them had to be defined, the third was implicit, hence "required" to be implicit.

Do you have an example that makes clear why you need zones, and cannot just track the values / life ranges? That seems like an interesting test for isConflicting.

See file comment in DeLICM.cpp

Assuming both implicit and explicit representation are supported. I wonder if one of them would be enough? Or is there a reason to keep both? I have the feeling that having both currently makes the code (unnecessarily) hard to read?

It's just the opposite, it makes it clearer that Unused/Occupied can be nullptr and therefore easier to understand. Previous versions of this patch used a flag which one is implicit, hence also requiring that one of the is implicit. I invite you to look into these versions to see if whether they are%2

lib/Transform/DeLICM.cpp

1657

Nice, this nicely removes the code duplication.

unittests/DeLICM/DeLICMTest.cpp

211

reduction_embedded.ll

[grosser]

Hi Michael,

please ignore the previous email. It got sent incomplete.

First, I like the changes to computeReachingWrite. The code duplication is nicely removed. Very cool!

I feel relative strongly about keeping them. They were certainly helpful in development when I did not yet have a singular definition of what a "zone" is. Currently a zone 0 < 1 < 1 is consistently represented by the integer set { [1] }. There is no strict necessity to fo this everywhere. E.g. it might be more intuitive to use the start of the range instead. These computations are currently agnostic to what a zone is (could be moved to ISLTools.cpp as well maybe?) and I would perceive it as a loss when they suddenly would.

I also agree with you that keeping the additional cases for now makes sense -- especially as we have good test coverage. In fact, I like the idea of adding this functionality to ISLTools.cpp. It clearly is self-contained, well documented, and well tested. Please feel free to push these out already.

In D24716#666155, @grosser wrote:

I am not sure if Knowledges must be implicit or if also both sets can be explicit. In isConflicting you state "current implementation requires it to be implicit", but when testing you seem to compute two explicit sets using completeLifetime. This seems inconsistent.

That is comment was outdated. In previous revisions Unused, Known and Unknown was stored in the same map (instead in one Unsued and one Occupied). Two of them had to be defined, the third was implicit, hence "required" to be implicit.

Do you have an example that makes clear why you need zones, and cannot just track the values / life ranges? That seems like an interesting test for isConflicting.

See file comment in DeLICM.cpp

Could such an example be added to the tests of isConflicting that illustrates why such a merge would result in an incorrect / unprecise answer.

Assuming both implicit and explicit representation are supported. I wonder if one of them would be enough? Or is there a reason to keep both? I have the feeling that having both currently makes the code (unnecessarily) hard to read?

It's just the opposite, it makes it clearer that Unused/Occupied can be nullptr and therefore easier to understand. Previous versions of this patch used a flag which one is implicit, hence also requiring that one of the is implicit. I invite you to look into these versions to see if whether they are%2

I did not mean that we should merge Unused/Occupied. I agree that having both makes a lot of sense. What I was suggesting above was if we should prohibit that both can be set simultaniously. The reason for this is that we now need to always support an explicit and an implicit representation of these sets. As such I always need to convince myself that both code paths are correct (or that ignoring one is correct). In the actual code that we run, we always seem to use implicit sets. So I wonder why we do not just limit our implementation to implicit sets and also test it with implicit sets.

[Meinersbur]

In D24716#666573, @grosser wrote:

Do you have an example that makes clear why you need zones, and cannot just track the values / life ranges? That seems like an interesting test for isConflicting.

See file comment in DeLICM.cpp

Could such an example be added to the tests of isConflicting that illustrates why such a merge would result in an incorrect / unprecise answer.

Could you suggest an EXPECT(...) line? I have no clue what kind of example you have in mind.

[Meinersbur]

• Rebase to r294094
• Use implicit sets in unittests

[grosser]

Hi Michael,

here a first update. I would like to think a little bit more about this to get some illustrative examples (am busy on CC at the moment), but here already some comments on what I am thinking. (No need to change anything yet. I currently try to complete my understanding of this code).

In D24716#666792, @Meinersbur wrote:

In D24716#666573, @grosser wrote:

Do you have an example that makes clear why you need zones, and cannot just track the values / life ranges? That seems like an interesting test for isConflicting.

See file comment in DeLICM.cpp

Could such an example be added to the tests of isConflicting that illustrates why such a merge would result in an incorrect / unprecise answer.

Could you suggest an EXPECT(...) line? I have no clue what kind of example you have in mind.

I am not sure how such test case can be constructed. You use "zones" to model occupied and undef sets in a knowledge. In isConflicting you always convert the zones to a set of timepoints (when comparing to writes). To me it looks as if we could as well store the zones as set of timepoints. I am looking for an argument that explains why we need to store occupied and undef indeed as zones.

lib/Transform/DeLICM.cpp

412

In your documentation and the code, you use "Unused" and "Undef" for the same thing. The set here is e.g., called Unused, but in the class documentation and the tests you talk about "Undef". Would it not be better to just use one of the two?

435

What does it mean to be implicitly unused.

The unused set has not been computed explicitly, but is assumed to be the complement of Occupied

OK. The comments now make this very clear. Thank you!

Can you somehow print this, e.g. as "universe - Occupied".

I don't understand this suggestion.

In "print(llvm::raw_ostream &OS, unsigned Indent = 0)" you just print "<implicit>" in case the corresponding set is a nullptr. Without reading the source code comments, I do not know what <implicit> means. Hence, this is difficult to understand. I wonder if we could print this in a way that one can easily understand what this means. E.g. by printing a string that shows how "implicit" is computed.

What happens if the first set does not contain elements from each space?

Does not happen. In ZoneAlgorithm one of the two sets will always be implicit. In DeLICMTests unionSpace() is used.

Assuming we never write to a given array? This would mean it does not appear in Occupied and is completely unused. Now, the universe we compute won't contain an element from this space, after the (implicit) subtraction it won't appear, right?

As there is no write to the array, greedyCollapse() will not try to map anything to it.

I wonder if there is an implicit assumption hidden. I need to think a little bit more about this to see if I can construct a test case that would validate this assumption and consequently result in incorrect answers being given by isConflicting. Overall, the correctness and the behavior of Knowledge should not depend on how Knowledge is used, but knowledge should have defined behavior for any input.

(In some way, we should make very clear which spaces a complement is actually formed. And how isConflicting works in case the sets are inconsistent -- if this can happen at all).

443

I wonder if it makes sense to document the implicit Occupied and Unused sets at this position and to explain precisely that the complement here means the complement for all spaces that are mentioned in the explicit of the two sets. In terms of interface, I could use see us enforcing and documenting that one of Occupied and Unused must be nullptr.

478

This "if" is not needed. It is always implied by the asserts.

510

I tried to replace these lines by:

+ assert(Existing.Unused && !Existing.Occupied);
+ assert(!Proposed.Unused && Proposed.Occupied);

This seems to work. Would this be correct and could we make Knowledge consistently limited to this pattern? Would this make sense at all? Are there cases where isConflicting would return wrong results for certain values of Existing.Occupied or Proposed.Unused (which are not useful).

[grosser]

Hi Michael,

thanks for explaining me this patch again on the phone in great detail. I think I now got the information I was lacking and added a couple of comments which I think might help others to get this information as well.

Would be great if you could cross-check this information and integrate it where appropriate. After this, we should be able to commit the Knowledge part and move on to the next pieces.

Best,
Tobias

Best,
Tobias

lib/Transform/DeLICM.cpp

405

Could we possibly explain why zones are used to represent lifetimes? Something like:

The set of alive array elements is represented as zone, as the set of live values can differ depending on how the elements are interpreted.

Assuming a value X is written at timestep [0] and read at timestep [1] without being used at any later point, then the value is alive in the interval ]0,1[. This interval cannot be represented by an integer set, as it does not contain any integer point. Zones allow us to represent this interval and can be converted to sets of timepoints when needed (e.g., in isConflicting when comparing to the write sets).

@see convertZoneToTimepoints for more details.

This might be a little verbose, but might help the reader.

461

What about "everything not in 'Occupied'"?

What about "everything not in 'Unused'"?

504

Can you add a comment that ensures that the universe of both Existing and Proposed need to be identical.

Also, if both are fully defined, can we add asserts to validate this.

527

Maybe an additional comment:

"
We convert here the set of lifetimes to actual timesteps. A lifetime is in conflict with a set of write timepoints, if either a lite timepoint is clearly within the lifetime or if a write happens at the beginning of the lifetime (where it would conflict with the value that actually writes the value alive). There is no conflict at the end of a lifetime, as the alive value will always be read, before it is overwritten again. The last property holds in Polly for all scalar values and we expect all users of Knowledge to check this property also for accesses to MemoryKind::Array.
"

unittests/DeLICM/DeLICMTest.cpp

109

Maybe test four cases?

[Meinersbur]

• Refine documentation about zones
• Testing and assertions for Knowledges with Occupied and Unused defined
• Rebase to r294894
• Address Tobias' other comments

[grosser]

Hi Michael,

thanks for the very fast update. This was exactly what I was looking for. I have a couple of minor typos and final questions on the Knowledge and Zone description. Otherwise, I think the knowledge stuff is good to go.

(I believe this was the most difficult part of the patch. The remaining part is still a lot larger, but I believe most of it is well explained and also not technically too complicated).

lib/Transform/DeLICM.cpp

47

AT the

at THE end

50

e.g., a LOAD

51

WE exclude ... starting the zone FROM THE LIVE-RANGE.

57

It is unclear to me why a write may overwrite a variable's value. Is it because it is a may write, due to the order of the writes in the statement, or because it may write an identical value (and consequently the write has no effect)?

Also, I am not sure why you mention "undefined behavior" here. In the context of C this term has a very strong meaning. Is this really what you mean here? Can you explain what exactly is undefined and why?

60

Hence, WE include

62

What is a contradiction of a live-range. You can contradict statements that state something, but what does contradicting a live-range mean?

64

starts

65

IN the live-range

68

e.g.,

69

e.g.,

498

I don't get the grammar of the sentence in the assert. Could you look at it again?

[Meinersbur]

• Clarify considerations in comment about zones.

[grosser]

Hi Michael,

thank you for the quick update. The knowledge stuff is now really well polished. Thanks! Feel free to commit it.

I will try to have a look at the remaining parts later tonight!

lib/Transform/DeLICM.cpp

50

IN the following

[Meinersbur]

• Rebase to r295204

[grosser]

Hi Michael,

this patch looks good to me. Just two minor typos. Also, I would really appreciate if you could add some negative tests.

Best,
Tobias

lib/Transform/DeLICM.cpp

855

NOT yet processed

1028

processed ELEMENT

[grosser]

Also mark officially as accepted.

include/polly/DeLICM.h

 //===------ DeLICM.h --------------------------------------------*- C++ -*-===//
 //
 //                     The LLVM Compiler Infrastructure
 //
 // This file is distributed under the University of Illinois Open Source
 // License. See LICENSE.TXT for details.
 //
 //===----------------------------------------------------------------------===//
 //
 // Undo the effect of Loop Invariant Code Motion (LICM) and
 // GVN Partial Redundancy Elimination (PRE) on SCoP-level.
 //
 // Namely, remove register/scalar dependencies by mapping them back to array
 // elements.
 //
 //===----------------------------------------------------------------------===//
 
 #ifndef POLLY_DELICM_H
 #define POLLY_DELICM_H
 
+#include "polly/Support/GICHelper.h"
+
 namespace llvm {
 class PassRegistry;
 class Pass;
 } // anonymous namespace
 
 namespace polly {
+class Scop;
+
 /// Create a new DeLICM pass instance.
 llvm::Pass *createDeLICMPass();
+
+/// Convert a zone (range between timepoints) to timepoints.
+///
+/// A zone represents the time between (integer) timepoints, but not the
+/// timepoints themselves. This function can be used to determine whether a
+/// timepoint lies within a zone.
+///
+/// For instance, the range (1,3), representing the time between 1 and 3, is
+/// represented by the zone
+///
+/// { [i] : 1 < i <= 3 }
+///
+/// The set timepoints that lie completely within this range is
+///
+/// { [i] : 1 < i < 3 }
+///
+/// A typical use-case is the range in which a value written by a store is
+/// available until it is overwritten by another value. If the write is at
+/// timepoint 1 and its value overwritten by another value at timepoint 3, the
+/// value is available between those timepoints: timepoint 2 in this example.
+///
+///
+/// When InclStart is true, the range is interpreted left-inclusive, i.e. adds
+/// the timepoint 1 to the result:
+///
+/// { [i] : 1 <= i < 3 }
+///
+/// In the use-case mentioned above that means that the value written at
+/// timepoint 1 is already available in timepoint 1 (write takes place before
+/// any read of it even if executed at the same timepoint)
+///
+/// When InclEnd is true, the range is interpreted right-inclusive, i.e. adds
+/// the timepoint 3 to the result:
+///
+/// { [i] : 1 < i <= 3 }
+///
+/// In the use-case mentioned above that means that although the value is
+/// overwritten in timepoint 3, the old value is still available at timepoint 3
+/// (write takes place after any read even if executed at the same timepoint)
+///
+/// @param Zone      { Zone[] }
+/// @param InclStart Include timepoints adjacent to the beginning of a zone.
+/// @param InclEnd   Include timepoints adjacent to the ending of a zone.
+///
+/// @return { Scatter[] }
+IslPtr<isl_union_set> convertZoneToTimepoints(IslPtr<isl_union_set> Zone,
+                                              bool InclStart, bool InclEnd);
+
+/// Compute the reaching definition statement for each definition of an array
+/// element.
+///
+/// The reaching definition of an array element at a specific timepoint is the
+/// statement instance that has written the current element's content. This
+/// function computes all reaching definitions for all array elements and
+/// timepoints. For example:
+///
+/// Schedule := { Write[] -> [0]; Overwrite[] -> [10] }
+/// Defs := { Write[] -> A[5]; Overwrite[] -> A[5] }
+///
+/// If index 5 of array A is written at timepoint 0 and 10, the resulting
+/// reaching definitions are:
+///
+/// { [A[5] -> [i]] -> Write[] : 0 < i < 10;
+///   [A[5] -> [i]] -> Overwrite[] : 10 < i }
+///
+/// Between timepoint 0 (Write[]) and timepoint 10 (Overwrite[]), the
+/// content of A[5] is written by statement instance Write[] and after
+/// timepoint 10 by Overwrite[]. Values not defined in the map have no known
+/// definition. This includes the statement instance timepoints themselves,
+/// because reads at those timepoints could either read the old or the new
+/// value, defined only by the statement itself. But this can be changed by @p
+/// InclDef and @p InclRedef. InclDef=false and InclRedef=true returns a
+/// zone. Unless @p InclDef and @p InclRedef are both true, there is only one
+/// unique definition per element and timepoint.
+///
+/// @param Schedule  { DomainWrite[] -> Scatter[] }
+///                  Schedule of (at least) all array writes. Instances not in
+///                  @p Writes are ignored.
+/// @param Writes    { DomainWrite[] -> Element[] }
+///                  Elements written to by the statement instances.
+/// @param InclDef   Include the definition's timepoint to the set of
+///                  well-defined elements (any load at that timepoint happen at
+///                  the writes). In the example, enabling this option adds
+///                  { [A[5] -> [0]] -> Write[]; [A[5] -> [10]] -> Overwrite[] }
+///                  to the result.
+/// @param InclRedef Whether to assume that at the timepoint where an element
+///                  is overwritten, it still contains the old value (any load
+///                  at that timepoint would happen before the overwrite). In
+///                  this example, enabling this adds
+///                  { [A[] -> [10]] -> Write[] } to the result.
+///
+/// @return { [Element[] -> Scatter[]] -> DomainWrite[] }
+///         The reaching definitions as described above, or nullptr if either @p
+///         Schedule or @p Writes is nullptr, or the ISL max operations count
+///         has exceeded.
+IslPtr<isl_union_map> computeReachingDefinition(IslPtr<isl_union_map> Schedule,
+                                                IslPtr<isl_union_map> Writes,
+                                                bool InclDef, bool InclRedef);
+
+/// Compute the next overwrite for each array element.
+///
+/// This is computeReachingDefinition() "in reverse"; instead of looking for the
+/// most recent write to an element, look for the next (over)write. For example:
+///
+/// Schedule := { Write[] -> [0]; Overwrite[] -> [10] }
+/// Writes := { Write[] -> A[5]; Overwrite[] -> A[5] }
+///
+/// results in:
+///
+/// { [A[5] -> [i]] -> Write[] : i < 0;
+///   [A[5] -> [i]] -> Overwrite[] : 0 < i < 10 }
+///
+/// That is, Write[] is the next write to A[5] when before timepoint 0, and when
+/// between timepoints 0 and 10, Overwrite[] is the following (dover)write. Use
+/// InclPrevWrite=false and InclOverwrite=true to interpret the result as a
+/// Zone.
+///
+/// @param Schedule      { DomainWrite[] -> Scatter[] }
+///                      Schedule of (at least) all array writes. Instances not
+///                      in @p Writes are ignored.
+/// @param Writes        { DomainWrite[] -> Element[] }
+///                      Elements written to by the statement instances.
+/// @param InclPrevWrite Whether to extend any overwrite timepoints to include
+///                      the timepoint where the previous write happens (the
+///                      previous write would happen at the beginning of its
+///                      timepoint). In this example, this adds
+///                      { [A[5] -> [0]] -> Overwrite[] } to the result.
+/// @param InclOverwrite Whether the reaching overwrite includes the timepoint
+///                      of the overwrite itself (so the overwrite would happen
+///                      at then end of its timepoint). In the example, this
+///                      adds
+///                      { [A[5] -> [0]] -> Write[];
+///                        [A[5] -> [10]] -> Overwrite[] }
+///                      to the result.
+///
+/// @return { [Element[] -> Scatter[]] -> DomainWrite[] }
+///         The reaching overwrite as defined above, or nullptr if either @p
+///         Schedule or @p Writes is nullptr, or the ISL max operations count
+///         has exceeded.
+IslPtr<isl_union_map> computeReachingOverwrite(IslPtr<isl_union_map> Schedule,
+                                               IslPtr<isl_union_map> Writes,
+                                               bool InclPrevWrite,
+                                               bool InclOverwrite);
+
+/// Compute the timepoints where the contents of an array element are not used.
+///
+/// An element is unused at a timepoint when the element is overwritten in
+/// the future, but it is no read in between. Another way to express this: the
+/// time from when the element is written, to the most recent read before it, or
+/// infinitely into the past if there is no read before. Such unused elements
+/// can be overwritten by any value without changing the scop's semantics. An
+/// example:
+///
+/// Schedule := { Read[] -> [0]; Write[] -> [10]; Def[] -> [20] }
+/// Writes := { Write[] -> A[5]; Def[] -> A[6] }
+/// Reads := { Read[] -> A[5] }
+///
+/// The result is:
+///
+/// { A[5] -> [i] : 0 < i < 10;
+///   A[6] -> [i] : i < 20 }
+///
+/// That is, A[5] is unused between timepoint 0 (the read) and timepoint 10 (the
+/// write). A[6] is unused before timepoint 20, but might be used after the
+/// scop's execution (A[5] and any other A[i] as well). Use InclLastRead=false
+/// and InclWrite=true to interpret the result as zone.
+///
+/// @param Schedule          { Domain[] -> Scatter[] }
+///                          The schedule of (at least) all statement instances
+///                          occurring in @p Writes or @p Reads. All other
+///                          instances are ignored.
+/// @param Writes            { DomainWrite[] -> Element[] }
+///                          Elements written to by the statement instances.
+/// @param Reads             { DomainRead[] -> Element[] }
+///                          Elements read from by the statement instances.
+/// @param ReadEltInSameInst Whether a load reads the value from a write
+///                          that is scheduled at the same timepoint (Writes
+///                          happen before reads). Otherwise, loads use the
+///                          value of an element that it had before the
+///                          timepoint (Reads before writes). For example:
+///                          { Read[] -> [0]; Write[] -> [0] }
+///                          With ReadEltInSameInst=false it is assumed that the
+///                          read happens before the write, such that the
+///                          element is never unused, or just at timepoint 0,
+///                          depending on InclLastRead/InclWrite.
+///                          With ReadEltInSameInst=false it assumes that the
+///                          value just written is used. Anything before
+///                          timepoint 0 is considered unused.
+/// @param InclLastRead      Whether a timepoint where an element is last read
+///                          counts as unused (the read happens at the beginning
+///                          of its timepoint, and nothing (else) can use it
+///                          during the timepoint). In the example, this option
+///                          adds { A[5] -> [0] } to the result.
+/// @param InclWrite         Whether the timepoint where an element is written
+///                          itself counts as unused (the write happens at the
+///                          end of its timepoint; no (other) operations uses
+///                          the element during the timepoint). In this example,
+///                          this adds
+///                          { A[5] -> [10]; A[6] -> [20] } to the result.
+///
+/// @return { Element[] -> Scatter[] }
+///         The unused timepoints as defined above, or nullptr if either @p
+///         Schedule, @p Writes are @p Reads is nullptr, or the ISL max
+///         operations count is exceeded.
+IslPtr<isl_union_map> computeArrayUnused(IslPtr<isl_union_map> Schedule,
+                                         IslPtr<isl_union_map> Writes,
+                                         IslPtr<isl_union_map> Reads,
+                                         bool ReadEltInSameInst,
+                                         bool InclLastRead, bool InclWrite);
+
+/// Determine whether two lifetimes are conflicting.
+///
+/// Used by unittesting.
+bool isConflicting(IslPtr<isl_union_set> ExistingOccupied,
+                   IslPtr<isl_union_set> ExistingUnused,
+                   IslPtr<isl_union_set> ExistingWrites,
+                   IslPtr<isl_union_set> ProposedOccupied,
+                   IslPtr<isl_union_set> ProposedUnused,
+                   IslPtr<isl_union_set> ProposedWrites,
+                   llvm::raw_ostream *OS = nullptr, unsigned Indent = 0);
 } // namespace polly
 
 namespace llvm {
 void initializeDeLICMPass(llvm::PassRegistry &);
 } // namespace llvm
 
 #endif /* POLLY_DELICM_H */

lib/Transform/DeLICM.cpp

 //===------ DeLICM.cpp -----------------------------------------*- C++ -*-===//
 //
 //                     The LLVM Compiler Infrastructure
 //
 // This file is distributed under the University of Illinois Open Source
 // License. See LICENSE.TXT for details.
 //
 //===----------------------------------------------------------------------===//
 //
 // Undo the effect of Loop Invariant Code Motion (LICM) and
 // GVN Partial Redundancy Elimination (PRE) on SCoP-level.
 //
 // Namely, remove register/scalar dependencies by mapping them back to array
 // elements.
 //
+// The algorithms here work on the scatter space - the image space of the
+// schedule returned by Scop::getSchedule(). We call an element in that space a
+// "timepoint". Timepoints are lexicographically ordered such that we can
+// defined ranges in the scatter space. We use two flavors of such ranges:
+// Timepoint sets and zones. A timepoint set is simply a subset of the scatter
+// space and is directly stored as isl_set.
+//
+// Zones are used to describe the space between timepoints as open sets, i.e.
+// they do not contain the extrema. Using isl rational sets to express these
+// would be overkill. We also cannot store them as the integer timepoints they
+// contain; the (nonempty) zone between 1 and 2 would be empty and
+// indistinguishable from e.g. the zone between 3 and 4. Also, we cannot store
+// the integer set including the extrema; the set ]1,2[ + ]3,4[ could be
+// coalesced to ]1,3[, although we defined the range [2,3] to be not in the set.
+// Instead, we store the "half-open" integer extrema, including the lower bound,
+// but excluding the upper bound. Examples:
+//
+// * The set { [i] : 1 <= i <= 3 } represents the zone ]0,3[ (which contains the
+//   integer points 1 and 2, but not 0 or 3)
+//
+// * { [1] } represents the zone ]0,1[
+//
+// * { [i] : i = 1 or i = 3 } represents the zone ]0,1[ + ]2,3[
+//
+// Therefore, an integer i in the set represents the zone ]i-1,i[, i.e. strictly
+// speaking the integer points never belong to the zone. However, depending an
+// the interpretation, one might want to include them:
+//
+// * The timepoints adjacent to two unit zones
+//
+// * The timepoints before a unit zone begins
+//
+// * The timepoints that directly follow a unit zone: Reinterpret the zone as
+//   a set of timepoints.
+//
+// It sometimes helps to think about a value of i stored in an isl_set to
+// represent the timepoint i-0.5 between two integer-valued timepoints.
+//
+//
+// The code makes use of maps and sets in many different spaces. To not loose
+// track in which space a set or map is expected to be in, variables holding an
+// isl reference are usually annotated in the comments. They roughly follow isl
+// syntax for spaces, but only the tuples, not the dimensions. The tuples have a
+// meaning as follows:
+//
+// * Space[] - An unspecified tuple. Used for function parameters such that the
+//             function caller can use it for anything they like.
+//
+// * Domain[] - A statement instance as returned by ScopStmt::getDomain()
+//     isl_id_get_name: Stmt_<NameOfBasicBlock>
+//     isl_id_get_user: Pointer to ScopStmt
+//
+// * Element[] - An array element as in the range part of
+//               MemoryAccess::getAccessRelation()
+//     isl_id_get_name: MemRef_<NameOfArrayVariable>
+//     isl_id_get_user: Pointer to ScopArrayInfo
+//
+// * Scatter[] - Scatter space or space of timepoints
+//     Has no tuple id
+//
+// * Zone[] - Range between timepoints as described above
+//     Has no tuple id
+//
+// An annotation "{ Domain[] -> Scatter[] }" therefore means: A map from a
+// statement instance to a timepoint, aka a schedule. There is only one scatter
+// space, but most of the time multiple statements are processed in one set.
+// This is why most of the time isl_union_map has to be used.
+//
+// The basic algorithm works as follows:
+// At first we verify that the SCoP is compatible with this technique. For
+// instance, two writes cannot write to the same location at the same statement
+// instance because we cannot determine within the polyhedral model which one
+// comes first. Once this was verified, we compute zones at which an array
+// element is unused. This computation can fail if it takes too long. Then the
+// main algorithm is executed. Because every store potentially trails an unused
+// zone, we start at stores. We search for a scalar (MemoryKind::Value or
+// MemoryKind::PHI) that we can map to the array element overwritten by the
+// store, preferably one that is used by the store or at least the ScopStmt.
+// When it does not conflict with the lifetime of the values in the array
+// element, the map is applied and the unused zone updated as it is now used. We
+// continue to try to map scalars to the array element until there are no more
+// candidates to map. The algorithm is greedy in the sense that the first scalar
+// not conflicting will be mapped. Other scalars processed later that could have
+// fit the same unused zone will be rejected. As such the result depends on the
+// processing order.
+//
 //===----------------------------------------------------------------------===//
 
 #include "polly/DeLICM.h"
+#include "polly/Options.h"
 #include "polly/ScopInfo.h"
 #include "polly/ScopPass.h"
+#include "polly/Support/ISLTools.h"
+#include "llvm/ADT/Statistic.h"
 #define DEBUG_TYPE "polly-delicm"
 
 using namespace polly;
 using namespace llvm;
 
 namespace {
 
+cl::opt<unsigned long>
+    DelicmMaxOps("polly-delicm-max-ops",
+                 cl::desc("Maximum number of isl operations to invest for "
+                          "lifetime analysis; 0=no limit"),
+                 cl::init(1000000), cl::cat(PollyCategory));
+
+STATISTIC(DeLICMAnalyzed, "Number of successfully analyzed SCoPs");
+STATISTIC(DeLICMOutOfQuota,
+          "Analyses aborted because max_operations was reached");
+STATISTIC(DeLICMIncompatible, "Number of SCoPs incompatible for analysis");
+STATISTIC(MappedValueScalars, "Number of mapped Value scalars");
+STATISTIC(MappedPHIScalars, "Number of mapped PHI scalars");
+STATISTIC(TargetsMapped, "Number of stores used for at least one mapping");
+STATISTIC(DeLICMScopsModified, "Number of SCoPs optimized");
+
+/// Class for keeping track of scalar def-use chains in the polyhedral
+/// representation.
+///
+/// MemoryKind::Value: There is one definition in SCoP and an arbitrary number
+/// of reads.
+///
+/// MemoryKind::PHI, MemoryKind::ExitPHI: There is at least one write (the
+/// incoming blocks/stmts) and one (MemoryKind::PHI) or zero
+/// (MemoryKind::ExitPHI) reads.
+class ScalarDefUseChains {
+private:
+  /// The definitions/write MemoryAccess of an MemoryKind::Value scalar.
+  ///
+  /// Note that read-only values have no value-defining write access.
+  DenseMap<const ScopArrayInfo *, MemoryAccess *> ValueDefAccs;
+
+  /// List of all uses/read MemoryAccesses for an MemoryKind::Value scalar.
+  DenseMap<const ScopArrayInfo *, SmallVector<MemoryAccess *, 4>> ValueUseAccs;
+
+  /// The PHI/read MemoryAccess of an MemoryKind::PHI/MemoryKind::ExitPHI
+  /// scalar.
+  DenseMap<const ScopArrayInfo *, MemoryAccess *> PHIReadAccs;
+
+  /// List of all incoming values/writes of an
+  /// MemoryKind::PHI/MemoryKind::ExitPHI scalar.
+  DenseMap<const ScopArrayInfo *, SmallVector<MemoryAccess *, 4>>
+      PHIIncomingAccs;
+
+public:
+  /// Find the MemoryAccesses that access the ScopArrayInfo-represented memory.
+  ///
+  /// @param S The SCoP to analyze.
+  void compute(Scop *S) {
+    // Purge any previous result.
+    reset();
+
+    for (auto &Stmt : *S) {
+      for (auto *MA : Stmt) {
+        if (MA->isOriginalValueKind() && MA->isWrite()) {
+          auto *SAI = MA->getScopArrayInfo();
+          assert(!ValueDefAccs.count(SAI) &&
+                 "There can be at most one "
+                 "definition per MemoryKind::Value "
+                 "scalar");
+          ValueDefAccs[SAI] = MA;
+        }
+
+        if (MA->isOriginalValueKind() && MA->isRead())
+          ValueUseAccs[MA->getScopArrayInfo()].push_back(MA);
+
+        if (MA->isOriginalAnyPHIKind() && MA->isRead()) {
+          auto *SAI = MA->getScopArrayInfo();
+          assert(!PHIReadAccs.count(SAI) &&
+                 "There must be exactly one read "
+                 "per PHI (that's where the PHINode "
+                 "is)");
+          PHIReadAccs[SAI] = MA;
+        }
+
+        if (MA->isOriginalAnyPHIKind() && MA->isWrite())
+          PHIIncomingAccs[MA->getScopArrayInfo()].push_back(MA);
+      }
+    }
+  }
+
+  /// Free all memory used by the analysis.
+  void reset() {
+    ValueDefAccs.clear();
+    ValueUseAccs.clear();
+    PHIReadAccs.clear();
+    PHIIncomingAccs.clear();
+  }
+
+  MemoryAccess *getValueDef(const ScopArrayInfo *SAI) const {
+    return ValueDefAccs.lookup(SAI);
+  }
+
+  ArrayRef<MemoryAccess *> getValueUses(const ScopArrayInfo *SAI) const {
+    auto It = ValueUseAccs.find(SAI);
+    if (It == ValueUseAccs.end())
+      return {};
+    return It->second;
+  }
+
+  MemoryAccess *getPHIRead(const ScopArrayInfo *SAI) const {
+    return PHIReadAccs.lookup(SAI);
+  }
+
+  ArrayRef<MemoryAccess *> getPHIIncomings(const ScopArrayInfo *SAI) const {
+    auto It = PHIIncomingAccs.find(SAI);
+    if (It == PHIIncomingAccs.end())
+      return {};
+    return It->second;
+  }
+};
+
+/// Compute the next overwrite for a scalar.
+///
+/// @param Schedule      { DomainWrite[] -> Scatter[] }
+///                      Schedule of (at least) all writes. Instances not in @p
+///                      Writes are ignored.
+/// @param Writes        { DomainWrite[] }
+///                      The element instances that write to the scalar.
+/// @param InclPrevWrite Whether to extend an overwrite timepoints to include
+///                      the timepoint where the previous write happens.
+/// @param InclOverwrite Whether the reaching overwrite includes the timepoint
+///                      of the overwrite itself.
+///
+/// @return { Scatter[] -> DomainDef[] }
+IslPtr<isl_union_map>
+computeScalarReachingOverwrite(IslPtr<isl_union_map> Schedule,
+                               IslPtr<isl_union_set> Writes, bool InclPrevWrite,
+                               bool InclOverwrite) {
+
+  // { DomainWrite[] }
+  auto WritesMap = give(isl_union_map_from_domain(Writes.take()));
+
+  // { [Element[] -> Scatter[]] -> DomainWrite[] }
+  auto Result = computeReachingOverwrite(
+      std::move(Schedule), std::move(WritesMap), InclPrevWrite, InclOverwrite);
+
+  return give(isl_union_map_domain_factor_range(Result.take()));
+}
+
+/// Overload of computeScalarReachingOverwrite, with only one writing statement.
+/// Consequently, the result consists of only one map space.
+///
+/// @param Schedule      { DomainWrite[] -> Scatter[] }
+/// @param Writes        { DomainWrite[] }
+/// @param InclPrevWrite Include the previous write to result.
+/// @param InclOverwrite Include the overwrite to the result.
+///
+/// @return { Scatter[] -> DomainWrite[] }
+IslPtr<isl_map> computeScalarReachingOverwrite(IslPtr<isl_union_map> Schedule,
+                                               IslPtr<isl_set> Writes,
+                                               bool InclPrevWrite,
+                                               bool InclOverwrite) {
+  auto ScatterSpace = getScatterSpace(Schedule);
+  auto DomSpace = give(isl_set_get_space(Writes.keep()));
+
+  auto ReachOverwrite = computeScalarReachingOverwrite(
+      Schedule, give(isl_union_set_from_set(Writes.take())), InclPrevWrite,
+      InclOverwrite);
+
+  auto ResultSpace = give(isl_space_map_from_domain_and_range(
+      ScatterSpace.take(), DomSpace.take()));
+  return singleton(std::move(ReachOverwrite), ResultSpace);
+}
+
+/// Compute the reaching definition of a scalar.
+///
+/// Compared to computeReachingDefinition, there is just one element which is
+/// accessed and therefore does not need to be specified.
+///
+/// @param Schedule  { DomainWrite[] -> Scatter[] }
+/// @param Writes    { DomainWrite[] }
+/// @param InclDef   Include the timepoint of the definition to the result.
+/// @param InclRedef Include the timepoint of the overwrite into the result.
+///
+/// @return { Scatter[] -> DomainWrite[] }
+IslPtr<isl_union_map>
+computeScalarReachingDefinition(IslPtr<isl_union_map> Schedule,
+                                IslPtr<isl_union_set> Writes, bool InclDef,
+                                bool InclRedef) {
+
+  // { DomainWrite[] -> Element[] }
+  auto Defs = give(isl_union_map_from_domain(Writes.take()));
+
+  // { [Element[] -> Scatter[]] -> DomainWrite[] }
+  auto ReachDefs =
+      computeReachingDefinition(Schedule, Defs, InclDef, InclRedef);
+
+  // { Scatter[] -> DomainWrite[] }
+  return give(isl_union_set_unwrap(
+      isl_union_map_range(isl_union_map_curry(ReachDefs.take()))));
+}
+
+/// Compute the reaching definition of a scalar.
+///
+/// This overload accepts only a single writing statement as an isl_map,
+/// consequently the result also is only a single isl_map.
+///
+/// @param Schedule  { DomainWrite[] -> Scatter[] }
+/// @param Writes    { DomainWrite[] }
+/// @param InclDef   Include the timepoint of the definition to the result.
+/// @param InclRedef Include the timepoint of the overwrite into the result.
+///
+/// @return { Scatter[] -> DomainWrite[] }
+IslPtr<isl_map> computeScalarReachingDefinition( // { Domain[] -> Zone[] }
+    IslPtr<isl_union_map> Schedule, IslPtr<isl_set> Writes, bool InclDef,
+    bool InclRedef) {
+  auto DomainSpace = give(isl_set_get_space(Writes.keep()));
+  auto ScatterSpace = getScatterSpace(Schedule);
+
+  //  { Scatter[] -> DomainWrite[] }
+  auto UMap = computeScalarReachingDefinition(
+      Schedule, give(isl_union_set_from_set(Writes.take())), InclDef,
+      InclRedef);
+
+  auto ResultSpace = give(isl_space_map_from_domain_and_range(
+      ScatterSpace.take(), DomainSpace.take()));
+  return singleton(UMap, ResultSpace);
+}
+
+/// If InputVal is not defined in the stmt itself, return the MemoryAccess that
+/// reads the scalar. Return nullptr otherwise (if the value is defined in the
+/// scop, or is synthesizable).
+MemoryAccess *getInputAccessOf(Value *InputVal, ScopStmt *Stmt) {
+  for (auto *MA : *Stmt) {
+    if (!MA->isRead())
+      continue;
+    if (!MA->isLatestScalarKind())
+      continue;
+
+    assert(MA->getAccessValue() == MA->getBaseAddr());
+    if (MA->getAccessValue() == InputVal)
+      return MA;
+  }
+  return nullptr;
+}
+
+/// Represent the knowledge of the contents of any array elements in any zone or
+/// the knowledge we would add when mapping a scalar to an array element.
+///
+/// Every array element at every zone unit has one of three states:
+/// - Undef: Not occupied by any value so a transformation can change it to
+///          other values.
+/// - Unknown: The element contains a value, but it is not a determinable
+///            llvm::Value instance.
+///
+/// There are two uses for the Knowledge class:
+/// 1) To represent the knowledge of the current state of ScopInfo. The Undef
+///    state means that an element is currently unused: there is no read of it
+///    before the next overwrite. Also called 'existing'.
+/// 2) To represent the requirements for mapping a scalar to array elements. The
+///    undef state means that there is no change/requirement. Also called
+///    'proposed'.
+///
+/// In addition to the these states at unit zones, Knowledge need to know when
+/// values are written. This is because written values may have no lifetime (one
+/// reason is that the value is never read). Such writes would therefore never
+/// conflict, but overwrite values that might still be required. Another source
+/// of problems are multiple writes to the same element at the same timepoint,
+/// because their order is undefined.
+class Knowledge {
+private:
+  /// { [Element[] -> Zone[]] }
+  IslPtr<isl_union_set> Occupied;
+
+  /// { [Element[] -> Zone[]] }
+  IslPtr<isl_union_set> Unused;
+
+  /// { [Element[] -> Scatter[]] }
+  /// The write actions currently in the scop or that would be added when
+  /// mapping a scalar.
+  IslPtr<isl_union_set> Written;
+
+  /// Check whether this Knowledge object is well-formed.
+  void checkConsistency() const {
+    // Default-initialized object
+    if (!Occupied && !Unused && !Written)
+      return;
+
+    assert(Occupied || Unused);
+    assert(!Occupied || !Unused ||
+           (isl_union_set_is_disjoint(Occupied.keep(), Unused.keep()) ==
+            isl_bool_true));
+    assert(Written);
+  }
+
+public:
+  /// Initialize an nullptr-Knowledge. This is only provided for convenience; do
+  /// not use such an object.
+  Knowledge() {}
+
+  /// Create a new object with the given members.
+  Knowledge(IslPtr<isl_union_set> Occupied, IslPtr<isl_union_set> Unused,
+            IslPtr<isl_union_set> Written)
+      : Occupied(std::move(Occupied)), Unused(std::move(Unused)),
+        Written(std::move(Written)) {
+    checkConsistency();
+  }
+
+  /// Alternative constructor taking isl_sets instead isl_union_sets.
+  Knowledge(IslPtr<isl_set> Occupied, IslPtr<isl_set> Unused,
+            IslPtr<isl_set> Written)
+      : Knowledge(give(isl_union_set_from_set(Occupied.take())),
+                  give(isl_union_set_from_set(Unused.take())),
+                  give(isl_union_set_from_set(Written.take()))) {}
+
+  /// Return whether this object was not default-constructed.
+  bool isUsable() const { return (Occupied || Unused) && Written; }
+
+  /// Print the content of this object to @p OS.
+  void print(llvm::raw_ostream &OS, unsigned Indent = 0) const {
+    if (isUsable()) {
+      if (Occupied)
+        OS.indent(Indent) << "Occupied: " << Occupied << "\n";
+      else
+        OS.indent(Indent) << "Occupied: <implicit>\n";
+      if (Unused)
+        OS.indent(Indent) << "Unused:   " << Unused << "\n";
+      else
+        OS.indent(Indent) << "Unused:   <implicit>\n";
+      OS.indent(Indent) << "Written : " << Written << '\n';
+    } else {
+      OS.indent(Indent) << "Invalid knowledge\n";
+    }
+  }
+
+  /// Combine two knowledges, this and @p That.
+  void learnFrom(Knowledge That) {
+    assert(!isConflicting(*this, That));
+
+    if (Unused && That.Occupied)
+      Unused =
+          give(isl_union_set_subtract(Unused.take(), That.Occupied.copy()));
+    if (Unused && That.Unused)
+      Unused = give(isl_union_set_intersect(Unused.take(), That.Unused.copy()));
+
+    if (Occupied && That.Occupied)
+      Occupied =
+          give(isl_union_set_union(Occupied.take(), That.Occupied.copy()));
+    if (Occupied && That.Unused)
+      assert(That.Occupied &&
+             "Cannot derive the total of occupied elements "
+             "knowing only those that are unused");
+
+    Written = give(isl_union_set_union(Written.take(), That.Written.take()));
+
+    checkConsistency();
+  }
+
+  /// Determine whether two Knowledges conflict each other.
+  ///
+  /// In theory @p This and @p That are symmetric, but the implementation is
+  /// constrained by the implicit interpretation.
+  ///
+  /// A conflict is defined as non-preserved semantics when they are merged. For
+  /// instance, when for the same array and zone they assume different
+  /// llvm::Values.
+  ///
+  /// @param Existing One of the knowledges; current implementation requires it
+  ///                 to be 'implicit unknown' (use case 1).
+  /// @param Proposed One of the knowledges; current implementation requires it
+  ///                 to be 'implicit undef' (use case 2).
+  /// @param OS       Dump the conflict reason to this output stream; use
+  ///                 nullptr to not output anything.
+  /// @param Indent   Indention for the conflict reason.
+  ///
+  /// @return True, iff the two knowledges are conflicting.
+  static bool isConflicting(const Knowledge &Existing,
+                            const Knowledge &Proposed,
+                            llvm::raw_ostream *OS = nullptr,
+                            unsigned Indent = 0) {
+    assert(Existing.Unused);
+    assert(Proposed.Occupied);
+
+    // Are the new lifetimes required for Proposed unused in Existing?
+    if (isl_union_set_is_subset(Proposed.Occupied.keep(),
+                                Existing.Unused.keep()) != isl_bool_true) {
+      if (OS) {
+        auto ConflictingLifetimes = give(isl_union_set_subtract(
+            Proposed.Occupied.copy(), Existing.Unused.copy()));
+        OS->indent(Indent) << "Proposed lifetimes are not unused in existing\n";
+        OS->indent(Indent) << "Conflicting lifetimes: " << ConflictingLifetimes
+                           << "\n";
+      }
+      return true;
+    }
+
+    // Do the writes in Existing only overwrite unused values in Proposed?
+    auto ProposedFixedDefs =
+        convertZoneToTimepoints(Proposed.Occupied, true, false);
+    if (isl_union_set_is_disjoint(Existing.Written.keep(),
+                                  ProposedFixedDefs.keep()) != isl_bool_true) {
+      if (OS) {
+        auto ConflictingWrites = give(isl_union_set_intersect(
+            Existing.Written.copy(), ProposedFixedDefs.copy()));
+        OS->indent(Indent) << "Proposed writes into range used by existing\n";
+        OS->indent(Indent) << "Conflicting writes: " << ConflictingWrites
+                           << "\n";
+      }
+      return true;
+    }
+
+    // Do the new writes in Proposed only overwrite unused values in Existing?
+    auto ExistingAvailableDefs =
+        convertZoneToTimepoints(Existing.Unused, true, false);
+    if (isl_union_set_is_subset(Proposed.Written.keep(),
+                                ExistingAvailableDefs.keep()) !=
+        isl_bool_true) {
+      if (OS) {
+        auto ConflictingWrites = give(isl_union_set_subtract(
+            Proposed.Written.copy(), ExistingAvailableDefs.copy()));
+        OS->indent(Indent)
+            << "Proposed a lifetime where there is an Existing write into it\n";
+        OS->indent(Indent) << "Conflicting writes: " << ConflictingWrites
+                           << "\n";
+      }
+      return true;
+    }
+
+    // Does Proposed write at the same time as Existing already does (order of
+    // writes is undefined)?
+    if (isl_union_set_is_disjoint(Existing.Written.keep(),
+                                  Proposed.Written.keep()) != isl_bool_true) {
+      if (OS) {
+        auto ConflictingWrites = give(isl_union_set_intersect(
+            Existing.Written.copy(), Proposed.Written.copy()));
+        OS->indent(Indent) << "Proposed writes at the same time as an already "
+                              "Existing write\n";
+        OS->indent(Indent) << "Conflicting writes: " << ConflictingWrites
+                           << "\n";
+      }
+      return true;
+    }
+
+    return false;
+  }
+};
+
+/// Base class for algorithms based on zones, like DeLICM.
+class ZoneAlgorithm {
+protected:
+  /// Hold a reference to the isl_ctx to avoid it being freed before we released
+  /// all of the isl objects.
+  ///
+  /// This must be the declared before any other member that holds an isl
+  /// object. This guarantees that the shared_ptr and its isl_ctx is destructed
+  /// last, after all other members free'd the isl objects they were holding.
+  std::shared_ptr<isl_ctx> IslCtx;
+
+  /// Cached reaching definitions for each ScopStmt.
+  ///
+  /// Use getScalarReachingDefinition() to get its contents.
+  DenseMap<ScopStmt *, IslPtr<isl_map>> ScalarReachDefZone;
+
+  /// The analyzed Scop.
+  Scop *S;
+
+  /// Parameter space that does not need realignment.
+  IslPtr<isl_space> ParamSpace;
+
+  /// Space the schedule maps to.
+  IslPtr<isl_space> ScatterSpace;
+
+  /// Cached version of the schedule and domains.
+  IslPtr<isl_union_map> Schedule;
+
+  /// Set of all referenced elements.
+  /// { Element[] -> Element[] }
+  IslPtr<isl_union_set> AllElements;
+
+  /// Combined access relations of all MemoryKind::Array READ accesses.
+  /// { DomainRead[] -> Element[] }
+  IslPtr<isl_union_map> AllReads;
+
+  /// Combined access relations of all MemoryKind::Array, MAY_WRITE accesses.
+  /// { DomainMayWrite[] -> Element[] }
+  IslPtr<isl_union_map> AllMayWrites;
+
+  /// Combined access relations of all MemoryKind::Array, MUST_WRITE accesses.
+  /// { DomainMustWrite[] -> Element[] }
+  IslPtr<isl_union_map> AllMustWrites;
+
+  /// Prepare the object before computing the zones of @p S.
+  ZoneAlgorithm(Scop *S)
+      : IslCtx(S->getSharedIslCtx()), S(S), Schedule(give(S->getSchedule())) {
+
+    auto Domains = give(S->getDomains());
+
+    Schedule =
+        give(isl_union_map_intersect_domain(Schedule.take(), Domains.take()));
+    ParamSpace = give(isl_union_map_get_space(Schedule.keep()));
+    ScatterSpace = getScatterSpace(Schedule);
+  }
+
+private:
+  /// Check whether @p Stmt can be accurately analyzed by zones.
+  ///
+  /// What violates our assumptions:
+  /// - A load after a write of the same location; we assume that all reads
+  ///   occur before the writes.
+  /// - Two writes to the same location; we cannot model the order in which
+  ///   these occur.
+  ///
+  /// Scalar reads implicitly always occur before other access therefore never
+  /// violate the first condition. There is also at most one write to a scalar,
+  /// satisfying the second condition.
+  bool isCompatibleStmt(ScopStmt *Stmt) {
+    auto Stores = makeEmptyUnionMap();
+    auto Loads = makeEmptyUnionMap();
+
+    // This assumes that the MemoryKind::Array MemoryAccesses are iterated in
+    // order.
+    for (auto *MA : *Stmt) {
+      if (!MA->isLatestArrayKind())
+        continue;
+
+      auto AccRel =
+          give(isl_union_map_from_map(getAccessRelationFor(MA).take()));
+
+      if (MA->isRead()) {
+        // Reject store after load to same location.
+        if (!isl_union_map_is_disjoint(Stores.keep(), AccRel.keep()))
+          return false;
+
+        Loads = give(isl_union_map_union(Loads.take(), AccRel.take()));
+      }
+
+      if (MA->isWrite()) {
+        if (!isa<StoreInst>(MA->getAccessInstruction())) {
+          DEBUG(dbgs() << "WRITE that is not a StoreInst not supported\n");
+          return false;
+        }
+
+        // In region statements the order is less clear, eg. the load and store
+        // might be in a boxed loop.
+        if (Stmt->isRegionStmt() &&
+            !isl_union_map_is_disjoint(Loads.keep(), AccRel.keep()))
+          return false;
+
+        // Do not allow more than one store to the same location.
+        if (!isl_union_map_is_disjoint(Stores.keep(), AccRel.keep()))
+          return false;
+
+        Stores = give(isl_union_map_union(Stores.take(), AccRel.take()));
+      }
+    }
+
+    return true;
+  }
+
+  void addArrayReadAccess(MemoryAccess *MA) {
+    assert(MA->isLatestArrayKind());
+    assert(MA->isRead());
+
+    // { DomainRead[] -> Element[] }
+    auto AccRel = getAccessRelationFor(MA);
+    AllReads = give(isl_union_map_add_map(AllReads.take(), AccRel.copy()));
+  }
+
+  void addArrayWriteAccess(MemoryAccess *MA) {
+    assert(MA->isLatestArrayKind());
+    assert(MA->isWrite());
+
+    // { Domain[] -> Element[] }
+    auto AccRel = getAccessRelationFor(MA);
+
+    if (MA->isMustWrite()) {
+      AllMustWrites =
+          give(isl_union_map_add_map(AllMustWrites.take(), AccRel.copy()));
+    }
+    if (MA->isMayWrite()) {
+      AllMayWrites =
+          give(isl_union_map_add_map(AllMayWrites.take(), AccRel.copy()));
+    }
+  }
+
+protected:
+  IslPtr<isl_union_set> makeEmptyUnionSet() {
+    return give(isl_union_set_empty(ParamSpace.copy()));
+  }
+
+  IslPtr<isl_union_map> makeEmptyUnionMap() {
+    return give(isl_union_map_empty(ParamSpace.copy()));
+  }
+
+  /// Check whether @p S can be accurately analyzed by zones.
+  bool isCompatibleScop() {
+    for (auto &Stmt : *S) {
+      if (!isCompatibleStmt(&Stmt))
+        return false;
+    }
+    return true;
+  }
+
+  /// Get the schedule for @p Stmt.
+  ///
+  /// The domain of the result is be as narrow as possible.
+  IslPtr<isl_map> getScatterFor(ScopStmt *Stmt) const {
+    auto ResultSpace = give(isl_space_map_from_domain_and_range(
+        Stmt->getDomainSpace(), ScatterSpace.copy()));
+    return give(isl_union_map_extract_map(Schedule.keep(), ResultSpace.take()));
+  }
+
+  /// Get the schedule of @p MA's parent statement.
+  IslPtr<isl_map> getScatterFor(MemoryAccess *MA) const {
+    return getScatterFor(MA->getStatement());
+  }
+
+  /// Get the schedule for the statement instances of @p Domain.
+  IslPtr<isl_union_map> getScatterFor(IslPtr<isl_union_set> Domain) const {
+    return give(isl_union_map_intersect_domain(Schedule.copy(), Domain.take()));
+  }
+
+  /// Get the schedule for the statement instances of @p Domain.
+  IslPtr<isl_map> getScatterFor(IslPtr<isl_set> Domain) const {
+    auto ResultSpace = give(isl_space_map_from_domain_and_range(
+        isl_set_get_space(Domain.keep()), ScatterSpace.copy()));
+    auto UDomain = give(isl_union_set_from_set(Domain.copy()));
+    auto UResult = getScatterFor(std::move(UDomain));
+    auto Result = singleton(std::move(UResult), std::move(ResultSpace));
+    assert(isl_set_is_equal(give(isl_map_domain(Result.copy())).keep(),
+                            Domain.keep()) == isl_bool_true);
+    return Result;
+  }
+
+  /// Get the domain of @p Stmt.
+  IslPtr<isl_set> getDomainFor(ScopStmt *Stmt) const {
+    return give(Stmt->getDomain());
+  }
+
+  /// Get the domain @p MA's parent statement.
+  IslPtr<isl_set> getDomainFor(MemoryAccess *MA) const {
+    return getDomainFor(MA->getStatement());
+  }
+
+  /// Get the access relation of @p MA.
+  ///
+  /// The domain of the result is as narrow as possible.
+  IslPtr<isl_map> getAccessRelationFor(MemoryAccess *MA) const {
+    auto Domain = getDomainFor(MA);
+    auto AccRel = give(MA->getLatestAccessRelation());
+    return give(isl_map_intersect_domain(AccRel.take(), Domain.take()));
+  }
+
+  /// Get the reaching definition of a scalar defined in @p Stmt.
+  ///
+  /// Note that this does not depend on the llvm::Instruction, only on the
+  /// statement it is defined in. Therefore the same computation can be reused.
+  ///
+  /// @param Stmt The statement in which a scalar is defined.
+  ///
+  /// @return { Scatter[] -> DomainDef[] }
+  IslPtr<isl_map> getScalarReachingDefinition(ScopStmt *Stmt) {
+    auto &Result = ScalarReachDefZone[Stmt];
+    if (Result)
+      return Result;
+
+    auto Domain = getDomainFor(Stmt);
+    Result = computeScalarReachingDefinition(Schedule, Domain, false, true);
+    simplify(Result);
+
+    assert(Result);
+    return Result;
+  }
+
+  /// Compute the different zones.
+  void computeCommon() {
+    AllReads = makeEmptyUnionMap();
+    AllMayWrites = makeEmptyUnionMap();
+    AllMustWrites = makeEmptyUnionMap();
+
+    for (auto &Stmt : *S) {
+      for (auto *MA : Stmt) {
+        if (!MA->isLatestArrayKind())
+          continue;
+
+        if (MA->isRead())
+          addArrayReadAccess(MA);
+
+        if (MA->isWrite())
+          addArrayWriteAccess(MA);
+      }
+    }
+
+    // { DomainWrite[] -> Element[] }
+    auto AllWrites =
+        give(isl_union_map_union(AllMustWrites.copy(), AllMayWrites.copy()));
+
+    // { Element[] }
+    AllElements = makeEmptyUnionSet();
+    foreachElt(AllWrites, [this](IslPtr<isl_map> Write) {
+      auto Space = give(isl_map_get_space(Write.keep()));
+      auto EltSpace = give(isl_space_range(Space.take()));
+      auto EltUniv = give(isl_set_universe(EltSpace.take()));
+      AllElements =
+          give(isl_union_set_add_set(AllElements.take(), EltUniv.take()));
+    });
+
+    // { Element[] -> Element[] }
+    auto AllElementsId = makeIdentityMap(AllElements, false);
+
+    // { Element[] -> Zone[] }
+    auto UniverseZone = give(isl_union_map_from_domain_and_range(
+        AllElements.copy(),
+        isl_union_set_from_set(isl_set_universe(ScatterSpace.copy()))));
+  }
+
+  /// Print the current state of all MemoryAccesses to @p.
+  void printAccesses(llvm::raw_ostream &OS, int Indent = 0) const {
+    OS.indent(Indent) << "After accesses {\n";
+    for (auto &Stmt : *S) {
+      OS.indent(Indent + 4) << Stmt.getBaseName() << "\n";
+      for (auto *MA : Stmt)
+        MA->print(OS);
+    }
+    OS.indent(Indent) << "}\n";
+  }
+
+public:
+  /// Return the SCoP this object is analyzing.
+  Scop *getScop() const { return S; }
+};
+
+/// Implementation of the DeLICM/DePRE transformation.
+class DeLICMImpl : public ZoneAlgorithm {
+private:
+  /// Knowledge before any transformation took place.
+  Knowledge OriginalZone;
+
+  /// Current knowledge of the SCoP including all already applied
+  /// transformations.
+  Knowledge Zone;
+
+  ScalarDefUseChains DefUse;
+
+  /// Determine whether two knowledges are conflicting with each other.
+  ///
+  /// @see Knowledge::isConflicting
+  bool isConflicting(const Knowledge &Proposed) {
+    raw_ostream *OS = nullptr;
+    DEBUG(OS = &llvm::dbgs());
+    return Knowledge::isConflicting(Zone, Proposed, OS, 4);
+  }
+
+  /// Determine whether @p SAI is a scalar that can be mapped to an array
+  /// element.
+  bool isMappable(const ScopArrayInfo *SAI) {
+    assert(SAI);
+
+    if (SAI->isValueKind()) {
+      auto *MA = DefUse.getValueDef(SAI);
+      if (!MA) {
+        DEBUG(dbgs()
+              << "    Reject because value is read-only within the scop\n");
+        return false;
+      }
+
+      // Mapping if value is used after scop is not supported. The code
+      // generator would need to reload the scalar after the scop, but it
+      // does not have the information to where it is mapped to. Only the
+      // MemoryAccesses have that information, not the ScopArrayInfo.
+      auto Inst = MA->getAccessInstruction();
+      for (auto User : Inst->users()) {
+        if (!isa<Instruction>(User))
+          return false;
+        auto UserInst = cast<Instruction>(User);
+
+        if (!S->contains(UserInst)) {
+          DEBUG(dbgs() << "    Reject because value is escaping\n");
+          return false;
+        }
+      }
+
+      return true;
+    }
+
+    if (SAI->isPHIKind()) {
+      auto *MA = DefUse.getPHIRead(SAI);
+      assert(MA);
+
+      // Mapping of an incoming block from before the SCoP is not supported by
+      // the code generator.
+      auto PHI = cast<PHINode>(MA->getAccessInstruction());
+      for (auto Incoming : PHI->blocks()) {
+        if (!S->contains(Incoming)) {
+          DEBUG(dbgs() << "    Reject because at least one incoming block is "
+                          "not in the scop region\n");
+          return false;
+        }
+      }
+
+      return true;
+    }
+
+    DEBUG(dbgs() << "    Reject ExitPHI or other non-value\n");
+    return false;
+  }
+
+  /// Compute the uses of a MemoryKind::Value and its lifetime (from its
+  /// definition to the last use).
+  ///
+  /// @param SAI The ScopArrayInfo representing the value's storage.
+  ///
+  /// @return { DomainDef[] -> DomainUse[] }, { DomainDef[] -> Zone[] }
+  ///         First element is the set of uses for each definition.
+  ///         The second is the lifetime of each definition.
+  std::tuple<IslPtr<isl_union_map>, IslPtr<isl_map>>
+  computeValueUses(const ScopArrayInfo *SAI) {
+    assert(SAI->isValueKind());
+
+    // { DomainRead[] }
+    auto Reads = makeEmptyUnionSet();
+
+    // Find all uses.
+    for (auto *MA : DefUse.getValueUses(SAI))
+      Reads =
+          give(isl_union_set_add_set(Reads.take(), getDomainFor(MA).take()));
+
+    // { DomainRead[] -> Scatter[] }
+    auto ReadSchedule = getScatterFor(Reads);
+
+    auto *DefMA = DefUse.getValueDef(SAI);
+    assert(DefMA);
+
+    // { DomainDef[] }
+    auto Writes = getDomainFor(DefMA);
+
+    // { DomainDef[] -> Scatter[] }
+    auto WriteScatter = getScatterFor(Writes);
+
+    // { Scatter[] -> DomainDef[] }
+    auto ReachDef = getScalarReachingDefinition(DefMA->getStatement());
+
+    // { [DomainDef[] -> Scatter[]] -> DomainUse[] }
+    auto Uses = give(
+        isl_union_map_apply_range(isl_union_map_from_map(isl_map_range_map(
+                                      isl_map_reverse(ReachDef.take()))),
+                                  isl_union_map_reverse(ReadSchedule.take())));
+
+    // { DomainDef[] -> Scatter[] }
+    auto UseScatter =
+        singleton(give(isl_union_set_unwrap(isl_union_map_domain(Uses.copy()))),
+                  give(isl_space_map_from_domain_and_range(
+                      isl_set_get_space(Writes.keep()), ScatterSpace.copy())));
+
+    // { DomainDef[] -> Zone[] }
+    auto Lifetime = betweenScatter(WriteScatter, UseScatter, false, true);
+
+    // { DomainDef[] -> DomainRead[] }
+    auto DefUses = give(isl_union_map_domain_factor_domain(Uses.take()));
+
+    return std::make_pair(DefUses, Lifetime);
+  }
+
+  /// For each 'execution' of a PHINode, get the incoming block that was
+  /// executed before.
+  ///
+  /// For each PHI instance we can directly determine which was the incoming
+  /// block, and hence derive which value the PHI has.
+  ///
+  /// @param SAI The ScopArrayInfo representing the PHI's storage.
+  ///
+  /// @return { DomainPHIRead[] -> DomainPHIWrite[] }
+  IslPtr<isl_union_map> computePerPHI(const ScopArrayInfo *SAI) {
+    assert(SAI->isPHIKind());
+
+    // { DomainPHIWrite[] -> Scatter[] }
+    auto PHIWriteScatter = makeEmptyUnionMap();
+
+    // Collect all incoming block timepoint.
+    for (auto *MA : DefUse.getPHIIncomings(SAI)) {
+      auto Scatter = getScatterFor(MA);
+      PHIWriteScatter =
+          give(isl_union_map_add_map(PHIWriteScatter.take(), Scatter.take()));
+    }
+
+    // { DomainPHIRead[] -> Scatter[] }
+    auto PHIReadScatter = getScatterFor(DefUse.getPHIRead(SAI));
+
+    // { DomainPHIRead[] -> Scatter[] }
+    auto BeforeRead = beforeScatter(PHIReadScatter, true);
+
+    // { Scatter[] }
+    auto WriteTimes = singleton(
+        give(isl_union_map_range(PHIWriteScatter.copy())), ScatterSpace);
+
+    // { DomainPHIRead[] -> Scatter[] }
+    auto PHIWriteTimes =
+        give(isl_map_intersect_range(BeforeRead.take(), WriteTimes.take()));
+    auto LastPerPHIWrites = give(isl_map_lexmax(PHIWriteTimes.take()));
+
+    // { DomainPHIRead[] -> DomainPHIWrite[] }
+    auto Result = give(isl_union_map_apply_range(
+        isl_union_map_from_map(LastPerPHIWrites.take()),
+        isl_union_map_reverse(PHIWriteScatter.take())));
+    assert(isl_union_map_is_single_valued(Result.keep()) == isl_bool_true);
+    assert(isl_union_map_is_injective(Result.keep()) == isl_bool_true);
+    return Result;
+  }
+
+  /// Try to map a MemoryKind::Value to a given array element.
+  ///
+  /// @param SAI       Representation of the scalar's memory to map.
+  /// @param TargetElt { Scatter[] -> Element[] }
+  ///                  Suggestion where to map a scalar to when at a timepoint.
+  ///
+  /// @return true if the scalar was successfully mapped.
+  bool tryMapValue(const ScopArrayInfo *SAI, IslPtr<isl_map> TargetElt) {
+    assert(SAI->isValueKind());
+
+    auto *DefMA = DefUse.getValueDef(SAI);
+    assert(DefMA->isValueKind());
+    assert(DefMA->isMustWrite());
+
+    // Stop if the scalar has already been mapped.
+    if (!DefMA->getLatestScopArrayInfo()->isValueKind())
+      return false;
+
+    // { DomainDef[] -> Scatter[] }
+    auto DefSched = getScatterFor(DefMA);
+
+    // Where each write is mapped to, according to the suggestion.
+    // { DomainDef[] -> Element[] }
+    auto DefTarget = give(isl_map_apply_domain(
+        TargetElt.copy(), isl_map_reverse(DefSched.copy())));
+    simplify(DefTarget);
+    DEBUG(dbgs() << "    Def Mapping: " << DefTarget << '\n');
+
+    auto OrigDomain = getDomainFor(DefMA);
+    auto MappedDomain = give(isl_map_domain(DefTarget.copy()));
+    if (!isl_set_is_subset(OrigDomain.keep(), MappedDomain.keep())) {
+      DEBUG(dbgs()
+            << "    Reject because mapping does not encompass all instances\n");
+      return false;
+    }
+
+    // { DomainDef[] -> Zone[] }
+    IslPtr<isl_map> Lifetime;
+
+    // { DomainDef[] -> DomainUse[] }
+    IslPtr<isl_union_map> DefUses;
+
+    std::tie(DefUses, Lifetime) = computeValueUses(SAI);
+    DEBUG(dbgs() << "    Lifetime: " << Lifetime << '\n');
+
+    /// { [Element[] -> Zone[]] }
+    auto EltZone = give(
+        isl_map_wrap(isl_map_apply_domain(Lifetime.copy(), DefTarget.copy())));
+    simplify(EltZone);
+
+    // { [Element[] -> Scatter[]] }
+    auto DefEltSched = give(isl_map_wrap(isl_map_reverse(
+        isl_map_apply_domain(DefTarget.copy(), DefSched.copy()))));
+    simplify(DefEltSched);
+
+    Knowledge Proposed(EltZone, nullptr, DefEltSched);
+    if (isConflicting(Proposed))
+      return false;
+
+    // { DomainUse[] -> Element[] }
+    auto UseTarget = give(
+        isl_union_map_apply_range(isl_union_map_reverse(DefUses.take()),
+                                  isl_union_map_from_map(DefTarget.copy())));
+
+    mapValue(SAI, std::move(DefTarget), std::move(UseTarget),
+             std::move(Lifetime), std::move(Proposed));
+    return true;
+  }
+
+  /// After a scalar has been mapped, update the global knowledge.
+  void applyLifetime(Knowledge Proposed) {
+    Zone.learnFrom(std::move(Proposed));
+  }
+
+  /// Map a MemoryKind::Value scalar to an array element.
+  ///
+  /// Callers must have ensured that the mapping is valid and not conflicting.
+  ///
+  /// @param SAI       The ScopArrayInfo representing the scalar's memory to
+  ///                  map.
+  /// @param DefTarget { DomainDef[] -> Element[] }
+  ///                  The array element to map the scalar to.
+  /// @param UseTarget { DomainUse[] -> Element[] }
+  ///                  The array elements the uses are mapped to.
+  /// @param Lifetime  { DomainDef[] -> Zone[] }
+  ///                  The lifetime of each llvm::Value definition for
+  ///                  reporting.
+  /// @param Proposed  Mapping constraints for reporting.
+  void mapValue(const ScopArrayInfo *SAI, IslPtr<isl_map> DefTarget,
+                IslPtr<isl_union_map> UseTarget, IslPtr<isl_map> Lifetime,
+                Knowledge Proposed) {
+    // Redirect the read accesses.
+    for (auto *MA : DefUse.getValueUses(SAI)) {
+      // { DomainUse[] }
+      auto Domain = getDomainFor(MA);
+
+      // { DomainUse[] -> Element[] }
+      auto NewAccRel = give(isl_union_map_intersect_domain(
+          UseTarget.copy(), isl_union_set_from_set(Domain.take())));
+      simplify(NewAccRel);
+
+      assert(isl_union_map_n_map(NewAccRel.keep()) == 1);
+      MA->setNewAccessRelation(isl_map_from_union_map(NewAccRel.take()));
+    }
+
+    auto *WA = DefUse.getValueDef(SAI);
+    WA->setNewAccessRelation(DefTarget.copy());
+    applyLifetime(Proposed);
+
+    MappedValueScalars++;
+  }
+
+  /// Try to map a MemoryKind::PHI scalar to a given array element.
+  ///
+  /// @param SAI       Representation of the scalar's memory to map.
+  /// @param TargetElt { Scatter[] -> Element[] }
+  ///                  Suggestion where to map the scalar to when at a
+  ///                  timepoint.
+  ///
+  /// @return true if the PHI scalar has been mapped.
+  bool tryMapPHI(const ScopArrayInfo *SAI, IslPtr<isl_map> TargetElt) {
+    auto *PHIRead = DefUse.getPHIRead(SAI);
+    assert(PHIRead->isPHIKind());
+    assert(PHIRead->isRead());
+
+    // Skip if already been mapped.
+    if (!PHIRead->getLatestScopArrayInfo()->isPHIKind())
+      return false;
+
+    // { DomainRead[] -> Scatter[] }
+    auto PHISched = getScatterFor(PHIRead);
+
+    // { DomainRead[] -> Element[] }
+    auto PHITarget =
+        give(isl_map_apply_range(PHISched.copy(), TargetElt.copy()));
+    simplify(PHITarget);
+    DEBUG(dbgs() << "    Mapping: " << PHITarget << '\n');
+
+    auto OrigDomain = getDomainFor(PHIRead);
+    auto MappedDomain = give(isl_map_domain(PHITarget.copy()));
+    if (!isl_set_is_subset(OrigDomain.keep(), MappedDomain.keep())) {
+      DEBUG(dbgs()
+            << "    Reject because mapping does not encompass all instances\n");
+      return false;
+    }
+
+    // { DomainRead[] -> DomainWrite[] }
+    auto PerPHIWrites = computePerPHI(SAI);
+
+    // { DomainWrite[] -> Element[] }
+    auto WritesTarget = give(isl_union_map_reverse(isl_union_map_apply_domain(
+        PerPHIWrites.copy(), isl_union_map_from_map(PHITarget.copy()))));
+    simplify(WritesTarget);
+
+    // { DomainWrite[] }
+    auto ExpandedWritesDom = give(isl_union_map_domain(WritesTarget.copy()));
+    auto UniverseWritesDom = give(isl_union_set_empty(ParamSpace.copy()));
+
+    for (auto *MA : DefUse.getPHIIncomings(SAI))
+      UniverseWritesDom = give(isl_union_set_add_set(UniverseWritesDom.take(),
+                                                     getDomainFor(MA).take()));
+
+    if (!isl_union_set_is_subset(UniverseWritesDom.keep(),
+                                 ExpandedWritesDom.keep())) {
+      DEBUG(dbgs() << "    Reject because did not find PHI write mapping for "
+                      "all instances\n");
+      DEBUG(dbgs() << "      Deduced Mapping:     " << WritesTarget << '\n');
+      DEBUG(dbgs() << "      Missing instances:    "
+                   << give(isl_union_set_subtract(UniverseWritesDom.copy(),
+                                                  ExpandedWritesDom.copy()))
+                   << '\n');
+      return false;
+    }
+
+    //  { DomainRead[] -> Scatter[] }
+    auto PerPHIWriteScatter = give(isl_map_from_union_map(
+        isl_union_map_apply_range(PerPHIWrites.copy(), Schedule.copy())));
+
+    // { DomainRead[] -> Zone[] }
+    auto Lifetime = betweenScatter(PerPHIWriteScatter, PHISched, false, true);
+    simplify(Lifetime);
+    DEBUG(dbgs() << "    Lifetime: " << Lifetime << "\n");
+
+    // { DomainWrite[] -> Zone[] }
+    auto WriteLifetime = give(isl_union_map_apply_domain(
+        isl_union_map_from_map(Lifetime.copy()), PerPHIWrites.copy()));
+
+    // { DomainWrite[] -> [Element[] -> Scatter[]] }
+    auto WrittenTranslator =
+        give(isl_union_map_range_product(WritesTarget.copy(), Schedule.copy()));
+
+    // { [Element[] -> Scatter[]] }
+    auto Written = give(isl_union_map_range(WrittenTranslator.copy()));
+    simplify(Written);
+
+    // { DomainWrite[] -> [Element[] -> Zone[]] }
+    auto LifetimeTranslator = give(
+        isl_union_map_range_product(WritesTarget.copy(), WriteLifetime.take()));
+
+    // { [Element[] -> Zone[] }
+    auto Occupied = give(isl_union_map_range(LifetimeTranslator.copy()));
+    simplify(Occupied);
+
+    Knowledge Proposed(Occupied, nullptr, Written);
+    if (isConflicting(Proposed))
+      return false;
+
+    mapPHI(SAI, std::move(PHITarget), std::move(WritesTarget),
+           std::move(Lifetime), std::move(Proposed));
+    return true;
+  }
+
+  /// Map a MemoryKind::PHI scalar to an array element.
+  ///
+  /// Callers must have ensured that the mapping is valid and not conflicting
+  /// with the common knowledge.
+  ///
+  /// @param SAI         The ScopArrayInfo representing the scalar's memory to
+  ///                    map.
+  /// @param ReadTarget  { DomainRead[] -> Element[] }
+  ///                    The array element to map the scalar to.
+  /// @param WriteTarget { DomainWrite[] -> Element[] }
+  ///                    New access target for each PHI incoming write.
+  /// @param Lifetime    { DomainRead[] -> Zone[] }
+  ///                    The lifetime of each PHI for reporting.
+  /// @param Proposed    Mapping constraints for reporting.
+  void mapPHI(const ScopArrayInfo *SAI, IslPtr<isl_map> ReadTarget,
+              IslPtr<isl_union_map> WriteTarget, IslPtr<isl_map> Lifetime,
+              Knowledge Proposed) {
+    // Redirect the PHI incoming writes.
+    for (auto *MA : DefUse.getPHIIncomings(SAI)) {
+      // { DomainWrite[] }
+      auto Domain = getDomainFor(MA);
+
+      // { DomainWrite[] -> Element[] }
+      auto NewAccRel = give(isl_union_map_intersect_domain(
+          WriteTarget.copy(), isl_union_set_from_set(Domain.take())));
+      simplify(NewAccRel);
+
+      assert(isl_union_map_n_map(NewAccRel.keep()) == 1);
+      MA->setNewAccessRelation(isl_map_from_union_map(NewAccRel.take()));
+    }
+
+    // Redirect the PHI read.
+    auto *PHIRead = DefUse.getPHIRead(SAI);
+    PHIRead->setNewAccessRelation(ReadTarget.copy());
+    applyLifetime(Proposed);
+
+    MappedPHIScalars++;
+  }
+
+  /// Search and map scalars to memory overwritten by @p TargetStoreMA.
+  ///
+  /// Start trying to map scalars that are used in the same statement as the
+  /// store. For every successful mapping, try to also map scalars of the
+  /// statements where those are written. Repeat, until no more mapping
+  /// opportunity is found.
+  ///
+  /// There is currently no preference in which order scalars are tried.
+  /// Ideally, we would direct it towards a load instruction of the same array
+  /// element.
+  bool collapseScalarsToStore(MemoryAccess *TargetStoreMA) {
+    assert(TargetStoreMA->isLatestArrayKind());
+    assert(TargetStoreMA->isMustWrite());
+
+    auto TargetStmt = TargetStoreMA->getStatement();
+
+    // { DomTarget[] }
+    auto TargetDom = getDomainFor(TargetStmt);
+
+    // { DomTarget[] -> Element[] }
+    auto TargetAccRel = getAccessRelationFor(TargetStoreMA);
+
+    // { Zone[] -> DomTarget[] }
+    // For each point in time, find the next target store instance.
+    auto Target =
+        computeScalarReachingOverwrite(Schedule, TargetDom, false, true);
+
+    // { Zone[] -> Element[] }
+    // Use the target store's write location as a suggestion to map scalars to.
+    auto EltTarget =
+        give(isl_map_apply_range(Target.take(), TargetAccRel.take()));
+    simplify(EltTarget);
+    DEBUG(dbgs() << "    Target mapping is " << EltTarget << '\n');
+
+    // Stack of elements no yet processed.
+    SmallVector<MemoryAccess *, 16> Worklist;
+
+    // Set of scalars already tested.
+    SmallPtrSet<const ScopArrayInfo *, 16> Closed;
+
+    // Lambda to add all scalar reads to the work list.
+    auto ProcessAllIncoming = [&](ScopStmt *Stmt) {
+      for (auto *MA : *Stmt) {
+        if (!MA->isLatestScalarKind())
+          continue;
+        if (!MA->isRead())
+          continue;
+
+        Worklist.push_back(MA);
+      }
+    };
+
+    // Add initial scalar. Either the value written by the store, or all inputs
+    // of its statement.
+    auto WrittenVal = TargetStoreMA->getAccessValue();
+    if (auto InputAcc = getInputAccessOf(WrittenVal, TargetStmt))
+      Worklist.push_back(InputAcc);
+    else
+      ProcessAllIncoming(TargetStmt);
+
+    auto AnyMapped = false;
+    auto &DL =
+        S->getRegion().getEntry()->getParent()->getParent()->getDataLayout();
+    auto StoreSize =
+        DL.getTypeAllocSize(TargetStoreMA->getAccessValue()->getType());
+
+    while (!Worklist.empty()) {
+      auto *MA = Worklist.pop_back_val();
+
+      auto *SAI = MA->getScopArrayInfo();
+      if (Closed.count(SAI))
+        continue;
+      Closed.insert(SAI);
+      DEBUG(dbgs() << "\n    Trying to map " << MA << " (SAI: " << SAI
+                   << ")\n");
+
+      // Skip non-mappable scalars.
+      if (!isMappable(SAI))
+        continue;
+
+      auto MASize = DL.getTypeAllocSize(MA->getAccessValue()->getType());
+      if (MASize > StoreSize) {
+        DEBUG(dbgs() << "    Reject because storage size is insufficient\n");
+        continue;
+      }
+
+      // Try to map MemoryKind::Value scalars.
+      if (SAI->isValueKind()) {
+        if (!tryMapValue(SAI, EltTarget))
+          continue;
+
+        auto *DefAcc = DefUse.getValueDef(SAI);
+        ProcessAllIncoming(DefAcc->getStatement());
+
+        AnyMapped = true;
+        continue;
+      }
+
+      // Try to map MemoryKind::PHI scalars.
+      if (SAI->isPHIKind()) {
+        if (!tryMapPHI(SAI, EltTarget))
+          continue;
+        // Add inputs of all incoming statements to the worklist.
+        for (auto *PHIWrite : DefUse.getPHIIncomings(SAI))
+          ProcessAllIncoming(PHIWrite->getStatement());
+
+        AnyMapped = true;
+        continue;
+      }
+    }
+
+    if (AnyMapped)
+      TargetsMapped++;
+    return AnyMapped;
+  }
+
+  /// Compute when an array element is unused.
+  ///
+  /// @return { [Element[] -> Zone[]] }
+  IslPtr<isl_union_set> computeLifetime() const {
+    // { Element[] -> Zone[] }
+    auto ArrayUnused = computeArrayUnused(Schedule, AllMustWrites, AllReads,
+                                          false, false, true);
+
+    auto Result = give(isl_union_map_wrap(ArrayUnused.copy()));
+
+    simplify(Result);
+    return Result;
+  }
+
+  /// Determine when an array element is written to.
+  ///
+  /// @return { [Element[] -> Scatter[]] }
+  IslPtr<isl_union_set> computeWritten() const {
+    // { WriteDomain[] -> Element[] }
+    auto AllWrites =
+        give(isl_union_map_union(AllMustWrites.copy(), AllMayWrites.copy()));
+
+    // { Scatter[] -> Element[] }
+    auto WriteTimepoints =
+        give(isl_union_map_apply_domain(AllWrites.copy(), Schedule.copy()));
+
+    auto Result =
+        give(isl_union_map_wrap(isl_union_map_reverse(WriteTimepoints.copy())));
+
+    simplify(Result);
+    return Result;
+  }
+
+  /// Determine whether an access touches at most one element.
+  ///
+  /// The accessed element could be a scalar or accessing an array with constant
+  /// subscript, such that all instances access only that element.
+  ///
+  /// @param MA The access to test.
+  ///
+  /// @return True, if zero or one elements are accessed; False if at least two
+  ///         different elements are accessed.
+  bool isScalarAccess(MemoryAccess *MA) {
+    auto Map = getAccessRelationFor(MA);
+    auto Set = give(isl_map_range(Map.take()));
+    return isl_set_is_singleton(Set.keep()) == isl_bool_true;
+  }
+
+public:
+  DeLICMImpl(Scop *S) : ZoneAlgorithm(S) {}
+
+  /// Calculate the lifetime (definition to last use) of every array element.
+  ///
+  /// @return True if the computed lifetimes (#Zone) is usable.
+  bool computeZone() {
+    // Check that nothing strange occurs.
+    if (!isCompatibleScop()) {
+      DeLICMIncompatible++;
+      return false;
+    }
+
+    DefUse.compute(S);
+    IslPtr<isl_union_set> EltUnused, EltWritten;
+
+    {
+      IslMaxOperationsGuard MaxOpGuard(IslCtx.get(), DelicmMaxOps);
+
+      computeCommon();
+
+      EltUnused = computeLifetime();
+      EltWritten = computeWritten();
+    }
+
+    if (isl_ctx_last_error(IslCtx.get()) == isl_error_quota) {
+      DeLICMOutOfQuota++;
+      DEBUG(dbgs() << "DeLICM analysis exceeded max_operations\n");
+    }
+
+    DeLICMAnalyzed++;
+    OriginalZone = Knowledge(nullptr, EltUnused, EltWritten);
+    DEBUG(dbgs() << "Computed Zone:\n"; OriginalZone.print(dbgs(), 4));
+
+    Zone = OriginalZone;
+
+    return DelicmMaxOps == 0 || Zone.isUsable();
+  }
+
+  /// Try to map as many scalars to unused array elements as possible.
+  ///
+  /// Multiple scalars might be mappable to intersecting unused array element
+  /// zones, but we can only chose one. This is a greedy algorithm, therefore
+  /// the first processed claims it.
+  void greedyCollapse() {
+    bool Modified = false;
+
+    for (auto &Stmt : *S) {
+      for (auto *MA : Stmt) {
+        if (!MA->isLatestArrayKind())
+          continue;
+        if (!MA->isWrite())
+          continue;
+
+        if (MA->isMayWrite()) {
+          DEBUG(dbgs() << "Access " << MA
+                       << " pruned because it is a MAY_WRITE\n");
+          continue;
+        }
+
+        if (Stmt.getNumIterators() == 0) {
+          DEBUG(dbgs() << "Access " << MA
+                       << " pruned because it is not in a loop\n");
+          continue;
+        }
+
+        if (isScalarAccess(MA)) {
+          DEBUG(dbgs() << "Access " << MA
+                       << " pruned because it writes only a single element\n");
+          continue;
+        }
+
+        DEBUG(dbgs() << "Analyzing target access " << MA << "\n");
+        if (collapseScalarsToStore(MA))
+          Modified = true;
+      }
+    }
+
+    if (Modified)
+      DeLICMScopsModified++;
+  }
+
+  /// Dump the internal information about a performed DeLICM to @p OS.
+  void print(llvm::raw_ostream &OS, int indent = 0) {
+    printAccesses(OS, indent);
+  }
+};
+
 class DeLICM : public ScopPass {
 private:
   DeLICM(const DeLICM &) = delete;
   const DeLICM &operator=(const DeLICM &) = delete;
 
+  /// The pass implementation, also holding per-scop data.
+  std::unique_ptr<DeLICMImpl> Impl;
+
+  void collapseToUnused(Scop &S) {
+    Impl = make_unique<DeLICMImpl>(&S);
+
+    if (!Impl->computeZone()) {
+      DEBUG(dbgs() << "Abort because cannot reliably compute lifetimes\n");
+      return;
+    }
+
+    DEBUG(dbgs() << "Collapsing scalars to unused array elements...\n");
+    Impl->greedyCollapse();
+
+    DEBUG(dbgs() << "\nFinal Scop:\n");
+    DEBUG(S.print(dbgs()));
+  }
+
 public:
   static char ID;
   explicit DeLICM() : ScopPass(ID) {}
 
   virtual void getAnalysisUsage(AnalysisUsage &AU) const override {
     AU.addRequiredTransitive<ScopInfoRegionPass>();
     AU.setPreservesAll();
   }
 
   virtual bool runOnScop(Scop &S) override {
     // Free resources for previous scop's computation, if not yet done.
     releaseMemory();
 
-    // TODO: Run DeLICM algorithm
+    collapseToUnused(S);
 
     return false;
   }
 
   virtual void printScop(raw_ostream &OS, Scop &S) const override {
+    if (!Impl)
+      return;
+    assert(Impl->getScop() == &S);
+
     OS << "DeLICM result:\n";
-    // TODO: Print analysis results and performed transformation details
+    Impl->print(OS);
   }
 
-  virtual void releaseMemory() override {
-    // TODO: Release resources (eg. shared_ptr to isl_ctx)
-  }
+  virtual void releaseMemory() override { Impl.reset(); }
 };
 
 char DeLICM::ID;
 } // anonymous namespace
 
 Pass *polly::createDeLICMPass() { return new DeLICM(); }
 
 INITIALIZE_PASS_BEGIN(DeLICM, "polly-delicm", "Polly - DeLICM/DePRE", false,
                       false)
 INITIALIZE_PASS_DEPENDENCY(ScopInfoWrapperPass)
 INITIALIZE_PASS_END(DeLICM, "polly-delicm", "Polly - DeLICM/DePRE", false,
                     false)
+
+IslPtr<isl_union_set> polly::convertZoneToTimepoints(IslPtr<isl_union_set> Zone,
+                                                     bool InclStart,
+                                                     bool InclEnd) {
+  if (!InclStart && InclEnd)
+    return Zone;
+
+  auto ShiftedZone = shiftDim(Zone, -1, -1);
+  if (InclStart && !InclEnd)
+    return ShiftedZone;
+  else if (!InclStart && !InclEnd)
+    return give(isl_union_set_intersect(Zone.take(), ShiftedZone.take()));
+
+  assert(InclStart && InclEnd);
+  return give(isl_union_set_union(Zone.take(), ShiftedZone.take()));
+}
+
+IslPtr<isl_union_map>
+polly::computeReachingDefinition(IslPtr<isl_union_map> Schedule,
+                                 IslPtr<isl_union_map> Writes, bool InclDef,
+                                 bool InclRedef) {
+  // { Scatter[] }
+  auto ScatterSpace = getScatterSpace(Schedule);
+
+  // { Element[] -> ScatterWrite[] }
+  auto DefSched =
+      give(isl_union_map_apply_domain(Schedule.copy(), Writes.take()));
+
+  // { ScatterRead[] -> ScatterWrite[] }
+  auto Before = give(InclRedef ? isl_map_lex_gt(ScatterSpace.take())
+                               : isl_map_lex_ge(ScatterSpace.take()));
+
+  // { ScatterWrite[] -> [ScatterRead[] -> ScatterWrite[]] }
+  auto BeforeMap = give(isl_map_reverse(isl_map_range_map(Before.take())));
+
+  // { Element[] -> [ScatterUse[] -> ScatterWrite[]] }
+  auto DefSchedBefore =
+      give(isl_union_map_apply_domain(isl_union_map_from_map(BeforeMap.take()),
+                                      isl_union_map_reverse(DefSched.copy())));
+
+  // For each element, at every point in time, map to the times of previous
+  // definitions.
+  // { [Element[] -> ScatterRead[]] -> ScatterWrite[] }
+  auto ReachableDefs = give(isl_union_map_uncurry(DefSchedBefore.take()));
+  auto LastReachableDef = give(isl_union_map_lexmax(ReachableDefs.copy()));
+
+  // { [Element[] -> ScatterWrite[]] -> ScatterWrite[] }
+  auto SelfUse = give(isl_union_map_range_map(DefSched.take()));
+
+  if (InclDef && InclRedef) {
+    // Add the Def itself to the solution.
+    LastReachableDef =
+        give(isl_union_map_union(LastReachableDef.take(), SelfUse.take()));
+    LastReachableDef = give(isl_union_map_coalesce(LastReachableDef.take()));
+  } else if (!InclDef && !InclRedef) {
+    // Remove Def itself from the solution.
+    LastReachableDef =
+        give(isl_union_map_subtract(LastReachableDef.take(), SelfUse.take()));
+  }
+
+  // { [Element[] -> ScatterRead[]] -> Domain[] }
+  return give(isl_union_map_apply_range(
+      LastReachableDef.take(), isl_union_map_reverse(Schedule.take())));
+}
+
+IslPtr<isl_union_map>
+polly::computeReachingOverwrite(IslPtr<isl_union_map> Schedule,
+                                IslPtr<isl_union_map> Writes,
+                                bool InclPrevWrite, bool IncludeOverwrite) {
+  assert(isl_union_map_is_bijective(Schedule.keep()) != isl_bool_false);
+
+  // { Scatter[] }
+  auto ScatterSpace = getScatterSpace(Schedule);
+
+  // { Element[] -> ScatterWrite[] }
+  auto WriteAction =
+      give(isl_union_map_apply_domain(Schedule.copy(), Writes.take()));
+
+  // { ScatterWrite[] -> Element[] }
+  auto WriteActionRev = give(isl_union_map_reverse(WriteAction.copy()));
+
+  // { ScatterRead[] -> ScatterWrite[] }
+  auto After = give(InclPrevWrite ? isl_map_lex_lt(ScatterSpace.take())
+                                  : isl_map_lex_le(ScatterSpace.take()));
+
+  // { ScatterWrite[] -> [ScatterRead[] -> ScatterWrite[]] }
+  auto BeforeMap = give(isl_map_reverse(isl_map_range_map(After.take())));
+
+  // { Element[] -> [ScatterRead[] -> ScatterWrite[]] }
+  auto DefSchedBefore = give(isl_union_map_apply_domain(
+      isl_union_map_from_map(BeforeMap.take()), WriteActionRev.take()));
+
+  // For each element, at every point in time, map to the times of previous
+  // definitions.
+  // { [Element[] -> ScatterRead[]] -> ScatterWrite[] }
+  auto ReachableDefs = give(isl_union_map_uncurry(DefSchedBefore.take()));
+  auto LastReachableDef = give(isl_union_map_lexmin(ReachableDefs.take()));
+
+  if (InclPrevWrite && IncludeOverwrite) {
+    // Add the def itself to the solution
+
+    // { [Element[] -> ScatterWrite[]] -> ScatterWrite[] }
+    auto SelfUse = give(isl_union_map_range_map(WriteAction.take()));
+    LastReachableDef =
+        give(isl_union_map_union(LastReachableDef.take(), SelfUse.take()));
+    LastReachableDef = give(isl_union_map_coalesce(LastReachableDef.take()));
+  } else if (!InclPrevWrite && !IncludeOverwrite) {
+    // Remove def itself from the solution
+
+    // { [Element[] -> ScatterWrite[]] -> ScatterWrite[] }
+    auto SelfUse = give(isl_union_map_range_map(WriteAction.take()));
+    LastReachableDef =
+        give(isl_union_map_subtract(LastReachableDef.take(), SelfUse.take()));
+  }
+
+  // { [Element[] -> ScatterRead[]] -> Domain[] }
+  auto LastReachableDefDomain = give(isl_union_map_apply_range(
+      LastReachableDef.take(), isl_union_map_reverse(Schedule.take())));
+
+  return LastReachableDefDomain;
+}
+
+IslPtr<isl_union_map> polly::computeArrayUnused(IslPtr<isl_union_map> Schedule,
+                                                IslPtr<isl_union_map> Writes,
+                                                IslPtr<isl_union_map> Reads,
+                                                bool ReadEltInSameInst,
+                                                bool IncludeLastRead,
+                                                bool IncludeWrite) {
+  // { Element[] -> Scatter[] }
+  auto ReadActions =
+      give(isl_union_map_apply_domain(Schedule.copy(), Reads.take()));
+  auto WriteActions =
+      give(isl_union_map_apply_domain(Schedule.copy(), Writes.copy()));
+
+  // { [Element[] -> Scatter[] }
+  auto AfterReads = afterScatter(ReadActions, ReadEltInSameInst);
+  auto WritesBeforeAnyReads =
+      give(isl_union_map_subtract(WriteActions.take(), AfterReads.take()));
+  auto BeforeWritesBeforeAnyReads =
+      beforeScatter(WritesBeforeAnyReads, !IncludeWrite);
+
+  // { [Element[] -> DomainWrite[]] -> Scatter[] }
+  auto EltDomWrites = give(isl_union_map_apply_range(
+      isl_union_map_range_map(isl_union_map_reverse(Writes.copy())),
+      Schedule.copy()));
+
+  // { [Element[] -> Scatter[]] -> DomainWrite[] }
+  auto ReachingOverwrite = computeReachingOverwrite(
+      Schedule, Writes, ReadEltInSameInst, !ReadEltInSameInst);
+
+  // { [Element[] -> Scatter[]] -> DomainWrite[] }
+  auto ReadsOverwritten = give(isl_union_map_intersect_domain(
+      ReachingOverwrite.take(), isl_union_map_wrap(ReadActions.take())));
+
+  // { [Element[] -> DomainWrite[]] -> Scatter[] }
+  auto ReadsOverwrittenRotated = give(isl_union_map_reverse(
+      isl_union_map_curry(reverseDomain(ReadsOverwritten).take())));
+  auto LastOverwrittenRead =
+      give(isl_union_map_lexmax(ReadsOverwrittenRotated.take()));
+
+  // { [Element[] -> DomainWrite[]] -> Scatter[] }
+  auto BetweenLastReadOverwrite = betweenScatter(
+      LastOverwrittenRead, EltDomWrites, IncludeLastRead, IncludeWrite);
+
+  return give(isl_union_map_union(
+      BeforeWritesBeforeAnyReads.take(),
+      isl_union_map_domain_factor_domain(BetweenLastReadOverwrite.take())));
+}
+
+bool polly::isConflicting(IslPtr<isl_union_set> ExistingOccupied,
+                          IslPtr<isl_union_set> ExistingUnused,
+                          IslPtr<isl_union_set> ExistingWrites,
+                          IslPtr<isl_union_set> ProposedOccupied,
+                          IslPtr<isl_union_set> ProposedUnused,
+                          IslPtr<isl_union_set> ProposedWrites,
+                          llvm::raw_ostream *OS, unsigned Indent) {
+  Knowledge Existing(std::move(ExistingOccupied), std::move(ExistingUnused),
+                     std::move(ExistingWrites));
+  Knowledge Proposed(std::move(ProposedOccupied), std::move(ProposedUnused),
+                     std::move(ProposedWrites));
+
+  return Knowledge::isConflicting(Existing, Proposed, OS, Indent);
+}

test/DeLICM/reduction_preheader.ll

+; RUN: opt %loadPolly -polly-flatten-schedule -polly-delicm -analyze < %s | FileCheck %s
+;
+;    void func(double *A) {
+;      for (int j = 0; j < 2; j += 1) { /* outer */
+;        double phi = 0.0;
+;        for (int i = 0; i < 4; i += 1) /* reduction */
+;          phi += 4.2;
+;        A[j] = phi;
+;      }
+;    }
+;
+define void @func(double* noalias nonnull %A) {
+entry:
+  br label %outer.preheader
+
+outer.preheader:
+  br label %outer.for
+
+outer.for:
+  %j = phi i32 [0, %outer.preheader], [%j.inc, %outer.inc]
+  %j.cmp = icmp slt i32 %j, 2
+  br i1 %j.cmp, label %reduction.preheader, label %outer.exit
+
+
+    reduction.preheader:
+      br label %reduction.for
+
+    reduction.for:
+      %i = phi i32 [0, %reduction.preheader], [%i.inc, %reduction.inc]
+      %phi = phi double [0.0, %reduction.preheader], [%add, %reduction.inc]
+      %i.cmp = icmp slt i32 %i, 4
+      br i1 %i.cmp, label %body, label %reduction.exit
+
+
+
+        body:
+          %add = fadd double %phi, 4.2
+          br label %reduction.inc
+
+
+
+    reduction.inc:
+      %i.inc = add nuw nsw i32 %i, 1
+      br label %reduction.for
+
+    reduction.exit:
+      %A_idx = getelementptr inbounds double, double* %A, i32 %j
+      store double %phi, double* %A_idx
+      br label %outer.inc
+
+
+
+outer.inc:
+  %j.inc = add nuw nsw i32 %j, 1
+  br label %outer.for
+
+outer.exit:
+  br label %return
+
+return:
+  ret void
+}
+
+
+; Unrolled flattened schedule:
+; [0] Stmt_reduction_preheader[0]
+; [1] Stmt_reduction_for[0, 0]
+; [2] Stmt_body[0, 0]
+; [3] Stmt_reduction_inc[0, 0]
+; [4] Stmt_reduction_for[0, 1]
+; [5] Stmt_body[0, 1]
+; [6] Stmt_reduction_inc[0, 1]
+; [7] Stmt_reduction_for[0, 2]
+; [8] Stmt_body[0, 2]
+; [9] Stmt_reduction_inc[0, 2]
+; [10] Stmt_reduction_for[0, 3]
+; [11] Stmt_body[0, 3]
+; [12] Stmt_reduction_inc[0, 3]
+; [13] Stmt_reduction_for[0, 4]
+; [14] Stmt_reduction_exit[0]
+; [15] Stmt_reduction_preheader[0]
+; [16] Stmt_reduction_for[1, 0]
+; [17] Stmt_body[1, 0]
+; [18] Stmt_reduction_inc[1, 0]
+; [19] Stmt_reduction_for[1, 1]
+; [20] Stmt_body[1, 1]
+; [21] Stmt_reduction_inc[1, 1]
+; [22] Stmt_reduction_for[1, 2]
+; [23] Stmt_body[1, 2]
+; [24] Stmt_reduction_inc[1, 2]
+; [25] Stmt_reduction_for[1, 3]
+; [26] Stmt_body[1, 3]
+; [27] Stmt_reduction_inc[1, 3]
+; [28] Stmt_reduction_for[1, 4]
+; [29] Stmt_reduction_exit[1]
+
+; CHECK:      After accesses {
+; CHECK-NEXT:     Stmt_reduction_preheader
+; CHECK-NEXT:             MustWriteAccess :=    [Reduction Type: NONE] [Scalar: 1]
+; CHECK-NEXT:                 { Stmt_reduction_preheader[i0] -> MemRef_phi__phi[] };
+; CHECK-NEXT:            new: { Stmt_reduction_preheader[i0] -> MemRef_A[i0] : 0 <= i0 <= 1 };
+; CHECK-NEXT:     Stmt_reduction_for
+; CHECK-NEXT:             ReadAccess :=    [Reduction Type: NONE] [Scalar: 1]
+; CHECK-NEXT:                 { Stmt_reduction_for[i0, i1] -> MemRef_phi__phi[] };
+; CHECK-NEXT:            new: { Stmt_reduction_for[i0, i1] -> MemRef_A[i0] : 0 <= i0 <= 1 and 0 <= i1 <= 4 };
+; CHECK-NEXT:             MustWriteAccess :=    [Reduction Type: NONE] [Scalar: 1]
+; CHECK-NEXT:                 { Stmt_reduction_for[i0, i1] -> MemRef_phi[] };
+; CHECK-NEXT:            new: { Stmt_reduction_for[i0, i1] -> MemRef_A[i0] : 0 <= i0 <= 1 and 0 <= i1 <= 4 };
+; CHECK-NEXT:     Stmt_body
+; CHECK-NEXT:             MustWriteAccess :=    [Reduction Type: NONE] [Scalar: 1]
+; CHECK-NEXT:                 { Stmt_body[i0, i1] -> MemRef_add[] };
+; CHECK-NEXT:            new: { Stmt_body[i0, i1] -> MemRef_A[i0] : 0 <= i0 <= 1 and 0 <= i1 <= 3 };
+; CHECK-NEXT:             ReadAccess :=    [Reduction Type: NONE] [Scalar: 1]
+; CHECK-NEXT:                 { Stmt_body[i0, i1] -> MemRef_phi[] };
+; CHECK-NEXT:            new: { Stmt_body[i0, i1] -> MemRef_A[i0] : 0 <= i0 <= 1 and i1 >= 0 and -5i0 <= i1 <= 8 - 5i0 and i1 <= 3 };
+; CHECK-NEXT:     Stmt_reduction_inc
+; CHECK-NEXT:             ReadAccess :=    [Reduction Type: NONE] [Scalar: 1]
+; CHECK-NEXT:                 { Stmt_reduction_inc[i0, i1] -> MemRef_add[] };
+; CHECK-NEXT:            new: { Stmt_reduction_inc[i0, i1] -> MemRef_A[i0] : i1 >= 0 and -5i0 <= i1 <= 7 - 5i0 and i1 <= 3; Stmt_reduction_inc[1, 3] -> MemRef_A[1] };
+; CHECK-NEXT:             MustWriteAccess :=    [Reduction Type: NONE] [Scalar: 1]
+; CHECK-NEXT:                 { Stmt_reduction_inc[i0, i1] -> MemRef_phi__phi[] };
+; CHECK-NEXT:            new: { Stmt_reduction_inc[i0, i1] -> MemRef_A[i0] : 0 <= i0 <= 1 and 0 <= i1 <= 3 };
+; CHECK-NEXT:     Stmt_reduction_exit
+; CHECK-NEXT:             MustWriteAccess :=    [Reduction Type: NONE] [Scalar: 0]
+; CHECK-NEXT:                 { Stmt_reduction_exit[i0] -> MemRef_A[i0] };
+; CHECK-NEXT:             ReadAccess :=    [Reduction Type: NONE] [Scalar: 1]
+; CHECK-NEXT:                 { Stmt_reduction_exit[i0] -> MemRef_phi[] };
+; CHECK-NEXT:            new: { Stmt_reduction_exit[i0] -> MemRef_A[i0] : 0 <= i0 <= 1 };
+; CHECK-NEXT: }
+

unittests/CMakeLists.txt

else()
     add_executable(${test_name} EXCLUDE_FROM_ALL ${ARGN})
     set_target_properties(${test_name} PROPERTIES RUNTIME_OUTPUT_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR})
 
     target_link_libraries(${test_name} gtest_main gtest)
     add_dependencies(PollyUnitTests ${test_name})
 
     set_property(TARGET ${test_name} PROPERTY FOLDER "Polly")
   endif()
-  target_link_libraries(${test_name} Polly LLVMCore LLVMSupport LLVMDemangle)
+  target_link_libraries(${test_name} Polly LLVMCore LLVMSupport LLVMDemangle LLVMipo)
 endfunction()
 
 add_subdirectory(Isl)
 add_subdirectory(Flatten)
+add_subdirectory(DeLICM)

unittests/DeLICM/CMakeLists.txt

+add_polly_unittest(DeLICMTests
+  DeLICMTest.cpp
+  )

unittests/DeLICM/DeLICMTest.cpp

+//===- DeLICMTest.cpp ----------------------------------------------------===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+
+#include "polly/DeLICM.h"
+#include "gtest/gtest.h"
+#include <isl/map.h>
+#include <isl/set.h>
+#include <isl/stream.h>
+#include <isl/union_map.h>
+#include <isl/union_set.h>
+#include <memory>
+
+using namespace llvm;
+using namespace polly;
+
+/// Loop-like control structure to run body for true and false if passed indef.
+///
+/// Takes an int variable that can be either 0 (false), 1 (true), or -1 (indef).
+/// It then shadows that variable with a new variable of type bool.
+///
+/// - If the int variable has the value of 0, the body is executed with the bool
+///   variable set to false.
+/// - If the int variable has the value of 1, the body is executed with the bool
+///   variable set to true.
+/// - If the int variable has the value of -1, the body is executed twice: once
+///   with the bool variable set to false and a second time with true.
+///
+/// For use in tests where the value of a boolean variable should not change the
+/// result, such that both settings are tested at once.
+#define BOOL_FOR(VAR)                                                          \
+  for (int VAR##_iter = (VAR == indef) ? 0 : VAR;                              \
+       VAR##_iter <= ((VAR == indef) ? 1 : VAR); VAR##_iter += 1)              \
+    for (bool VAR = VAR##_iter, VAR##_repeat = true; VAR##_repeat;             \
+         VAR##_repeat = false)
+
+#define indef (-1)
+
+namespace {
+
+bool checkConvertZoneToTimepoints(const char *ZoneStr,
+                                  const char *TimepointsStr,
+                                  int IncludeZoneStart, int IncludeZoneEnd) {
+  BOOL_FOR(IncludeZoneStart) BOOL_FOR(IncludeZoneEnd) {
+    std::unique_ptr<isl_ctx, decltype(&isl_ctx_free)> Ctx(isl_ctx_alloc(),
+                                                          &isl_ctx_free);
+
+    auto Zone = give(isl_union_set_read_from_str(Ctx.get(), ZoneStr));
+    auto Timepoints =
+        give(isl_union_set_read_from_str(Ctx.get(), TimepointsStr));
+
+    auto Result =
+        convertZoneToTimepoints(Zone, IncludeZoneStart, IncludeZoneEnd);
+    auto Success = isl_union_set_is_equal(Result.keep(), Timepoints.keep());
+    if (Success != isl_bool_true)
+      return false;
+  }
+  return true;
+}
+
+TEST(DeLICM, ConvertZoneToTimepoints) {
+  EXPECT_TRUE(checkConvertZoneToTimepoints("{}", "{}", indef, indef));
+
+  EXPECT_TRUE(checkConvertZoneToTimepoints("{ [1] }", "{ }", false, false));
+  EXPECT_TRUE(checkConvertZoneToTimepoints("{ [1] }", "{ [0] }", true, false));
+  EXPECT_TRUE(checkConvertZoneToTimepoints("{ [1] }", "{ [1] }", false, true));
+  EXPECT_TRUE(
+      checkConvertZoneToTimepoints("{ [1] }", "{ [0]; [1] }", true, true));
+
+  EXPECT_TRUE(
+      checkConvertZoneToTimepoints("{ [1]; [11] }", "{ }", false, false));
+  EXPECT_TRUE(checkConvertZoneToTimepoints("{ [1]; [11] }", "{ [0]; [10] }",
+                                           true, false));
+  EXPECT_TRUE(checkConvertZoneToTimepoints("{ [1]; [11] }", "{ [1]; [11] }",
+                                           false, true));
+  EXPECT_TRUE(checkConvertZoneToTimepoints(
+      "{ [1]; [11] }", "{ [0]; [1]; [10]; [11] }", true, true));
+
+  EXPECT_TRUE(checkConvertZoneToTimepoints(
+      "{ [i] : 0 < i <= 10 }", "{ [i] : 0 < i < 10 }", false, false));
+  EXPECT_TRUE(checkConvertZoneToTimepoints(
+      "{ [i] : 0 < i <= 10 }", "{ [i] : 0 <= i < 10 }", true, false));
+  EXPECT_TRUE(checkConvertZoneToTimepoints(
+      "{ [i] : 0 < i <= 10 }", "{ [i] : 0 < i <= 10 }", false, true));
+  EXPECT_TRUE(checkConvertZoneToTimepoints(
+      "{ [i] : 0 < i <= 10 }", "{ [i] : 0 <= i <= 10 }", true, true));
+
+  EXPECT_TRUE(checkConvertZoneToTimepoints("{ [0,1] }", "{ }", false, false));
+  EXPECT_TRUE(
+      checkConvertZoneToTimepoints("{ [0,1] }", "{ [0,0] }", true, false));
+  EXPECT_TRUE(
+      checkConvertZoneToTimepoints("{ [0,1] }", "{ [0,1] }", false, true));
+  EXPECT_TRUE(checkConvertZoneToTimepoints("{ [0,1] }", "{ [0,0]; [0,1] }",
+                                           true, true));
+}
+
+bool checkComputeReachingDefinition(const char *ScheduleStr,
+                                    const char *DefsStr, int IncludeDef,
+                                    int IncludeRedef, const char *ExpectedStr) {
+  BOOL_FOR(IncludeDef) BOOL_FOR(IncludeRedef) {
+    std::unique_ptr<isl_ctx, decltype(&isl_ctx_free)> Ctx(isl_ctx_alloc(),
+                                                          &isl_ctx_free);
+
+    auto Schedule = give(isl_union_map_read_from_str(Ctx.get(), ScheduleStr));
+    auto Defs = give(isl_union_map_read_from_str(Ctx.get(), DefsStr));
+    auto Expected = give(isl_union_map_read_from_str(Ctx.get(), ExpectedStr));
+
+    auto Result =
+        computeReachingDefinition(Schedule, Defs, IncludeDef, IncludeRedef);
+
+    auto Success = isl_union_map_is_equal(Result.keep(), Expected.keep());
+    if (Success != isl_bool_true)
+      return false;
+  }
+
+  return true;
+}
+
+TEST(DeLICM, ReachingDefinitionZone) {
+  EXPECT_FALSE(computeReachingDefinition(nullptr, nullptr, false, false));
+
+  EXPECT_TRUE(checkComputeReachingDefinition(
+      "{ Dom[] -> [0] }", "{ Dom[] -> Elt[] }", true, indef,
+      "{ [Elt[] -> [i]] -> Dom[] : 0 <= i }"));
+  EXPECT_TRUE(checkComputeReachingDefinition(
+      "{ Dom[] -> [0] }", "{ Dom[] -> Elt[] }", false, indef,
+      "{ [Elt[] -> [i]] -> Dom[] : 0 < i }"));
+
+  EXPECT_TRUE(checkComputeReachingDefinition(
+      "{ Dom1[] -> [0]; Dom2[] -> [10] }",
+      "{ Dom1[] -> Elt[]; Dom2[] -> Elt[] }", true, false,
+      "{ [Elt[] -> [i]] -> Dom1[] : 0 <= i < 10; [Elt[] -> [i]] -> Dom2[] : 10 "
+      "<= i }"));
+  EXPECT_TRUE(checkComputeReachingDefinition(
+      "{ Dom1[] -> [0]; Dom2[] -> [10] }",
+      "{ Dom1[] -> Elt[]; Dom2[] -> Elt[] }", false, false,
+      "{ [Elt[] -> [i]] -> Dom1[] : 0 < i < 10; [Elt[] -> [i]] -> Dom2[] : 10 "
+      "< i }"));
+  EXPECT_TRUE(checkComputeReachingDefinition(
+      "{ Dom1[] -> [0]; Dom2[] -> [10] }",
+      "{ Dom1[] -> Elt[]; Dom2[] -> Elt[] }", false, true,
+      "{ [Elt[] -> [i]] -> Dom1[] "
+      ": 0 < i <= 10; [Elt[] -> "
+      "[i]] -> Dom2[] : 10 < i }"));
+  EXPECT_TRUE(checkComputeReachingDefinition(
+      "{ Dom1[] -> [0]; Dom2[] -> [10] }",
+      "{ Dom1[] -> Elt[]; Dom2[] -> Elt[] }", true, true,
+      "{ [Elt[] -> [i]] -> Dom1[] "
+      ": 0 <= i <= 10; [Elt[] -> "
+      "[i]] -> Dom2[] : 10 <= i }"));
+
+  EXPECT_TRUE(checkComputeReachingDefinition(
+      "{ Dom1[] -> [0]; Dom2[] -> [10] }",
+      "{ Dom1[] -> Elt1[]; Dom2[] -> Elt2[] }", true, indef,
+      "{ [Elt1[] -> [i]] -> Dom1[] : 0 <= i; "
+      "[Elt2[] -> [i]] -> Dom2[] : 10 <= i }"));
+
+  EXPECT_TRUE(checkComputeReachingDefinition(
+      "{ Dom[i] -> [i] }", "{ Dom[i] -> Elt[]; Dom2[] -> Elt[] }", true, false,
+      "{ [Elt[] -> [i]] -> Dom[i] }"));
+
+  EXPECT_TRUE(checkComputeReachingDefinition(
+      "{ Dom[1] -> [0]; Dom[2] -> [10] }",
+      "{ Dom[1] -> Elt[]; Dom[2] -> Elt[] }", false, true,
+      "{ [Elt[] -> [i]] -> Dom[1] "
+      ": 0 < i <= 10; [Elt[] -> "
+      "[i]] -> Dom[2] : 10 < i }"));
+
+  EXPECT_TRUE(checkComputeReachingDefinition(
+      "{ Stmt_reduction_for[i] -> [3i] : 0 <= i <= 4 }",
+      "{ Stmt_reduction_for[i] -> Elt[] : 0 <= i <= 4 }", false, true,
+      "{ [Elt[] -> [i]] -> Stmt_reduction_for[0] : 0 < i <= 3; [Elt[] -> [i]] "
+      "-> Stmt_reduction_for[1] : 3 < i <= 6; [Elt[] -> [i]] -> "
+      "Stmt_reduction_for[2] : 6 < i <= 9; [Elt[] -> [i]] -> "
+      "Stmt_reduction_for[3] : 9 < i <= 12; [Elt[] -> [i]] -> "
+      "Stmt_reduction_for[4] : 12 < i }"));
+}
+
+bool checkComputeReachingOverwrite(const char *ScheduleStr, const char *DefsStr,
+                                   int IncludePrevWrite, int IncludeOverwrite,
+                                   const char *ExpectedStr) {
+  BOOL_FOR(IncludePrevWrite) BOOL_FOR(IncludeOverwrite) {
+    std::unique_ptr<isl_ctx, decltype(&isl_ctx_free)> Ctx(isl_ctx_alloc(),
+                                                          &isl_ctx_free);
+
+    auto Schedule = give(isl_union_map_read_from_str(Ctx.get(), ScheduleStr));
+    auto Defs = give(isl_union_map_read_from_str(Ctx.get(), DefsStr));
+    auto Expected = give(isl_union_map_read_from_str(Ctx.get(), ExpectedStr));
+
+    auto Result = computeReachingOverwrite(Schedule, Defs, IncludePrevWrite,
+                                           IncludeOverwrite);
+    auto Success = isl_union_map_is_equal(Result.keep(), Expected.keep());
+    if (Success != isl_bool_true)
+      return false;
+  }
+
+  return true;
+}
+
+TEST(DeLICM, ReachingOverwrite) {
+  EXPECT_TRUE(checkComputeReachingOverwrite(
+      "{ Write[] -> [0] }", "{ Write[] -> Elt[] }", indef, false,
+      "{ [Elt[] -> [i]] -> Write[] :  i < 0 }"));
+  EXPECT_TRUE(checkComputeReachingOverwrite(
+      "{ Write[] -> [0] }", "{ Write[] -> Elt[] }", indef, true,
+      "{ [Elt[] -> [i]] -> Write[] : i <= 0 }"));
+
+  EXPECT_TRUE(checkComputeReachingOverwrite(
+      "{ Write[0] -> [0]; Write[1] -> [10] }", "{ Write[i] -> Elt[]; }", false,
+      false,
+      "{ [Elt[] -> [i]] -> Write[0] : i < 0 ; [Elt[] -> [i]] -> "
+      "Write[1] : 0 < i < 10 }"));
+  EXPECT_TRUE(checkComputeReachingOverwrite(
+      "{ Write[0] -> [0]; Write[1] -> [10] }", "{ Write[i] -> Elt[]; }", false,
+      true,
+      "{ [Elt[] -> [i]] -> Write[0] : i <= 0 ; [Elt[] -> [i]] -> "
+      "Write[1] : 0 < i <= 10 }"));
+  EXPECT_TRUE(checkComputeReachingOverwrite(
+      "{ Write[0] -> [0]; Write[1] -> [10] }", "{ Write[i] -> Elt[]; }", true,
+      false,
+      "{ [Elt[] -> [i]] -> Write[0] : i < 0 ; [Elt[] -> [i]] -> "
+      "Write[1] : 0 <= i < 10 }"));
+  EXPECT_TRUE(checkComputeReachingOverwrite(
+      "{ Write[0] -> [0]; Write[1] -> [10] }", "{ Write[i] -> Elt[]; }", true,
+      true,
+      "{ [Elt[] -> [i]] -> Write[0] : i <= 0 ; [Elt[] -> [i]] -> "
+      "Write[1] : 0 <= i <= 10 }"));
+}
+
+bool checkComputeArrayUnused(const char *ScheduleStr, const char *WritesStr,
+                             const char *ReadsStr, int ReadEltInSameInst,
+                             int IncludeLastRead, int IncludeWrite,
+                             const char *ExpectedStr) {
+  BOOL_FOR(ReadEltInSameInst) BOOL_FOR(IncludeLastRead) BOOL_FOR(IncludeWrite) {
+    std::unique_ptr<isl_ctx, decltype(&isl_ctx_free)> Ctx(isl_ctx_alloc(),
+                                                          &isl_ctx_free);
+
+    auto Schedule = give(isl_union_map_read_from_str(Ctx.get(), ScheduleStr));
+    auto Writes = give(isl_union_map_read_from_str(Ctx.get(), WritesStr));
+    auto Reads = give(isl_union_map_read_from_str(Ctx.get(), ReadsStr));
+    auto Expected = give(isl_union_map_read_from_str(Ctx.get(), ExpectedStr));
+
+    auto Result = computeArrayUnused(Schedule, Writes, Reads, ReadEltInSameInst,
+                                     IncludeLastRead, IncludeWrite);
+    auto Success = isl_union_map_is_equal(Result.keep(), Expected.keep());
+    if (Success != isl_bool_true)
+      return false;
+  }
+
+  return isl_bool_true;
+}
+
+TEST(DeLICM, ArrayUnused) {
+  EXPECT_TRUE(checkComputeArrayUnused("{ Read[] -> [0]; Write[] -> [10] }",
+                                      "{ Write[] -> Elt[] }",
+                                      "{ Read[] -> Elt[] }", indef, false,
+                                      false, "{ Elt[] -> [i] : 0 < i < 10 }"));
+  EXPECT_TRUE(checkComputeArrayUnused("{ Read[] -> [0]; Write[] -> [10] }",
+                                      "{ Write[] -> Elt[] }",
+                                      "{ Read[] -> Elt[] }", indef, false, true,
+                                      "{ Elt[] -> [i] : 0 < i <= 10 }"));
+  EXPECT_TRUE(checkComputeArrayUnused("{ Read[] -> [0]; Write[] -> [10] }",
+                                      "{ Write[] -> Elt[] }",
+                                      "{ Read[] -> Elt[] }", indef, true, false,
+                                      "{ Elt[] -> [i] : 0 <= i < 10 }"));
+  EXPECT_TRUE(checkComputeArrayUnused("{ Read[] -> [0]; Write[] -> [10] }",
+                                      "{ Write[] -> Elt[] }",
+                                      "{ Read[] -> Elt[] }", indef, true, true,
+                                      "{ Elt[] -> [i] : 0 <= i <= 10 }"));
+
+  EXPECT_TRUE(checkComputeArrayUnused(
+      "{ Read[0] -> [-10]; Read[1] -> [0]; Write[] -> [10] }",
+      "{ Write[] -> Elt[] }", "{ Read[i] -> Elt[] }", indef, false, true,
+      "{ Elt[] -> [i] : 0 < i <= 10 }"));
+  EXPECT_TRUE(checkComputeArrayUnused("{ Read[] -> [0];  }", "{ }",
+                                      "{ Read[] -> Elt[] }", indef, indef,
+                                      indef, "{ }"));
+  EXPECT_TRUE(checkComputeArrayUnused(
+      "{ Write[] -> [0];  }", "{ Write[] -> Elt[] }", "{ }", indef, indef, true,
+      "{ Elt[] -> [i] : i <= 0  }"));
+
+  EXPECT_TRUE(checkComputeArrayUnused("{ RW[] -> [0] }", "{ RW[] -> Elt[] }",
+                                      "{ RW[] -> Elt[] }", true, indef, false,
+                                      "{ Elt[] -> [i] : i < 0 }"));
+  EXPECT_TRUE(checkComputeArrayUnused("{ RW[] -> [0] }", "{ RW[] -> Elt[] }",
+                                      "{ RW[] -> Elt[] }", true, indef, true,
+                                      "{ Elt[] -> [i] : i <= 0 }"));
+  EXPECT_TRUE(checkComputeArrayUnused("{ RW[] -> [0] }", "{ RW[] -> Elt[] }",
+                                      "{ RW[] -> Elt[] }", false, true, true,
+                                      "{ Elt[] -> [0] }"));
+}
+
+/// Get the universes of all spaces in @p USet.
+IslPtr<isl_union_set> unionSpace(NonowningIslPtr<isl_union_set> USet) {
+  auto Result = give(isl_union_set_empty(isl_union_set_get_space(USet.keep())));
+  foreachElt(USet, [=, &Result](IslPtr<isl_set> Set) {
+    auto Space = give(isl_set_get_space(Set.keep()));
+    auto Universe = give(isl_set_universe(Space.take()));
+    Result = give(isl_union_set_add_set(Result.take(), Universe.take()));
+  });
+  return Result;
+}
+
+void completeLifetime(IslPtr<isl_union_set> Universe,
+                      IslPtr<isl_union_set> &Unknown,
+                      IslPtr<isl_union_set> &Undef) {
+  if (!Unknown) {
+    assert(Undef);
+    Unknown = give(isl_union_set_subtract(Universe.copy(), Undef.copy()));
+  }
+
+  if (!Undef) {
+    assert(Unknown);
+    Undef = give(isl_union_set_subtract(Universe.copy(), Unknown.copy()));
+  }
+}
+
+typedef struct {
+  const char *OccupiedStr;
+  const char *UndefStr;
+  const char *WrittenStr;
+} Knowledge;
+
+bool checkIsConflictingNonsymmetric(Knowledge Existing, Knowledge Proposed) {
+  std::unique_ptr<isl_ctx, decltype(&isl_ctx_free)> Ctx(isl_ctx_alloc(),
+                                                        &isl_ctx_free);
+
+  // Parse knowledge
+  auto ExistingOccupied =
+      Existing.OccupiedStr
+          ? give(isl_union_set_read_from_str(Ctx.get(), Existing.OccupiedStr))
+          : nullptr;
+  auto ExistingUndef =
+      Existing.UndefStr
+          ? give(isl_union_set_read_from_str(Ctx.get(), Existing.UndefStr))
+          : nullptr;
+  auto ExistingWritten =
+      give(isl_union_set_read_from_str(Ctx.get(), Existing.WrittenStr));
+
+  auto ProposedOccupied =
+      Proposed.OccupiedStr
+          ? give(isl_union_set_read_from_str(Ctx.get(), Proposed.OccupiedStr))
+          : nullptr;
+  auto ProposedUndef =
+      Proposed.UndefStr
+          ? give(isl_union_set_read_from_str(Ctx.get(), Proposed.UndefStr))
+          : nullptr;
+  auto ProposedWritten =
+      give(isl_union_set_read_from_str(Ctx.get(), Proposed.WrittenStr));
+
+  // Determine universe (set of all possible domains)
+  auto Universe =
+      give(isl_union_set_empty(isl_space_params_alloc(Ctx.get(), 0)));
+  if (ExistingOccupied)
+    Universe =
+        give(isl_union_set_union(Universe.take(), ExistingOccupied.copy()));
+  if (ExistingUndef)
+    Universe = give(isl_union_set_union(Universe.take(), ExistingUndef.copy()));
+  if (ExistingWritten)
+    Universe =
+        give(isl_union_set_union(Universe.take(), ExistingWritten.copy()));
+  if (ProposedOccupied)
+    Universe =
+        give(isl_union_set_union(Universe.take(), ProposedOccupied.copy()));
+  if (ProposedUndef)
+    Universe = give(isl_union_set_union(Universe.take(), ProposedUndef.copy()));
+  if (ProposedWritten)
+    Universe =
+        give(isl_union_set_union(Universe.take(), ProposedWritten.copy()));
+  Universe = unionSpace(Universe);
+
+  // Using the universe, fill missing data
+  completeLifetime(Universe, ExistingOccupied, ExistingUndef);
+  completeLifetime(Universe, ProposedOccupied, ProposedUndef);
+
+  return isConflicting(ExistingOccupied, ExistingUndef, ExistingWritten,
+                       ProposedOccupied, ProposedUndef, ProposedWritten);
+}
+
+bool checkIsConflicting(Knowledge Existing, Knowledge Proposed) {
+  auto Forward = checkIsConflictingNonsymmetric(Existing, Proposed);
+  auto Backward = checkIsConflictingNonsymmetric(Proposed, Existing);
+
+  // isConflicting should be symmetric.
+  EXPECT_EQ(Forward, Backward);
+
+  return Forward || Backward;
+}
+
+TEST(DeLICM, isConflicting) {
+
+  // Check occupied vs occupied
+  EXPECT_TRUE(
+      checkIsConflicting({"{ Dom[i] }", nullptr, "{}"}, {nullptr, "{}", "{}"}));
+  EXPECT_TRUE(checkIsConflicting({"{ Dom[i] }", nullptr, "{}"},
+                                 {"{ Dom[i] }", nullptr, "{}"}));
+  EXPECT_FALSE(checkIsConflicting({"{ Dom[0] }", nullptr, "{}"},
+                                  {nullptr, "{ Dom[0] }", "{}"}));
+  EXPECT_FALSE(checkIsConflicting({"{ Dom[i] : i != 0 }", nullptr, "{}"},
+                                  {"{ Dom[0] }", nullptr, "{}"}));
+
+  // Check occupied vs. written
+  EXPECT_TRUE(
+      checkIsConflicting({nullptr, "{}", "{}"}, {"{}", nullptr, "{ Dom[0] }"}));
+  EXPECT_FALSE(
+      checkIsConflicting({"{}", nullptr, "{}"}, {"{}", nullptr, "{ Dom[0] }"}));
+
+  EXPECT_TRUE(checkIsConflicting({"{ Dom[i] }", nullptr, "{}"},
+                                 {"{}", nullptr, "{ Dom[0] }"}));
+  EXPECT_FALSE(checkIsConflicting({"{ DomA[i] }", nullptr, "{}"},
+                                  {"{}", nullptr, "{ DomB[0] }"}));
+
+  EXPECT_TRUE(checkIsConflicting({"{ Dom[1] }", nullptr, "{}"},
+                                 {"{}", nullptr, "{ Dom[0] }"}));
+  EXPECT_FALSE(checkIsConflicting({"{ Dom[i] : i != 1 }", nullptr, "{}"},
+                                  {"{}", nullptr, "{ Dom[0] }"}));
+
+  // Check written vs. written
+  EXPECT_TRUE(checkIsConflicting({"{}", nullptr, "{ Dom[0] }"},
+                                 {"{}", nullptr, "{ Dom[0] }"}));
+  EXPECT_FALSE(checkIsConflicting({"{}", nullptr, "{ Dom[-1] }"},
+                                  {"{}", nullptr, "{ Dom[0] }"}));
+  EXPECT_FALSE(checkIsConflicting({"{}", nullptr, "{ Dom[1] }"},
+                                  {"{}", nullptr, "{ Dom[0] }"}));
+}
+} // anonymous namespace