MergeFunctions pass, how it works

Introduction

Sometimes code contains equal functions, or functions that does exactly the same thing even though they are non-equal on the IR level (e.g.: multiplication on 2 and ‘shl 1’). It could happen due to several reasons: mainly, the usage of templates and automatic code generators. Though, sometimes the user itself could write the same thing twice :-)

The main purpose of this pass is to recognize such functions and merge them.

This document is the extension to pass comments and describes the pass logic. It describes the algorithm that is used in order to compare functions and explains how we could combine equal functions correctly to keep the module valid.

Material is brought in a top-down form, so the reader could start to learn pass from high level ideas and end with low-level algorithm details, thus preparing him or her for reading the sources.

The main goal is to describe the algorithm and logic here and the concept. If you don’t want to read the source code, but want to understand pass algorithms, this document is good for you. The author tries not to repeat the source-code and covers only common cases to avoid the cases of needing to update this document after any minor code changes.

What should I know to be able to follow along with this document?

The reader should be familiar with common compile-engineering principles and LLVM code fundamentals. In this article, we assume the reader is familiar with Single Static Assignment concept and has an understanding of IR structure.

We will use terms such as “module”, “function”, “basic block”, “user”, “value”, “instruction”.

As a good starting point, the Kaleidoscope tutorial can be used:

LLVM Tutorial: Table of Contents

It’s especially important to understand chapter 3 of tutorial:

Kaleidoscope Tutorial

The reader should also know how passes work in LLVM. They could use this article as a reference and start point here:

Writing an LLVM Pass (legacy PM version)

What else? Well perhaps the reader should also have some experience in LLVM pass debugging and bug-fixing.

Narrative structure

The article consists of three parts. The first part explains pass functionality on the top-level. The second part describes the comparison procedure itself. The third part describes the merging process.

In every part, the author tries to put the contents in the top-down form. The top-level methods will first be described followed by the terminal ones at the end, in the tail of each part. If the reader sees the reference to the method that wasn’t described yet, they will find its description a bit below.

Basics

How to do it?

Do we need to merge functions? The obvious answer is: Yes, that is quite a possible case. We usually do have duplicates and it would be good to get rid of them. But how do we detect duplicates? This is the idea: we split functions into smaller bricks or parts and compare the “bricks” amount. If equal, we compare the “bricks” themselves, and then do our conclusions about functions themselves.

What could the difference be? For example, on a machine with 64-bit pointers (let’s assume we have only one address space), one function stores a 64-bit integer, while another one stores a pointer. If the target is the machine mentioned above, and if functions are identical, except the parameter type (we could consider it as a part of function type), then we can treat a uint64_t and a void* as equal.

This is just an example; more possible details are described a bit below.

As another example, the reader may imagine two more functions. The first function performs a multiplication by 2, while the second one performs an logical left shift by 1.

Possible solutions

Let’s briefly consider possible options about how and what we have to implement in order to create full-featured functions merging, and also what it would mean for us.

Equal function detection obviously supposes that a “detector” method to be implemented and latter should answer the question “whether functions are equal”. This “detector” method consists of tiny “sub-detectors”, which each answers exactly the same question, but for function parts.

As the second step, we should merge equal functions. So it should be a “merger” method. “Merger” accepts two functions F1 and F2, and produces F1F2 function, the result of merging.

Having such routines in our hands, we can process a whole module, and merge all equal functions.

In this case, we have to compare every function with every another function. As the reader may notice, this way seems to be quite expensive. Of course we could introduce hashing and other helpers, but it is still just an optimization, and thus the level of O(N*N) complexity.

Can we reach another level? Could we introduce logarithmical search, or random access lookup? The answer is: “yes”.

Random-access

How it could this be done? Just convert each function to a number, and gather all of them in a special hash-table. Functions with equal hashes are equal. Good hashing means, that every function part must be taken into account. That means we have to convert every function part into some number, and then add it into the hash. The lookup-up time would be small, but such an approach adds some delay due to the hashing routine.

Present state

Both of the approaches (random-access and logarithmical) have been implemented and tested and both give a very good improvement. What was most surprising is that logarithmical search was faster; sometimes by up to 15%. The hashing method needs some extra CPU time, which is the main reason why it works slower; in most cases, total “hashing” time is greater than total “logarithmical-search” time.

So, preference has been granted to the “logarithmical search”.

Though in the case of need, logarithmical-search (read “total-ordering”) could be used as a milestone on our way to the random-access implementation.

Every comparison is based either on the numbers or on the flags comparison. In the random-access approach, we could use the same comparison algorithm. During comparison, we exit once we find the difference, but here we might have to scan the whole function body every time (note, it could be slower). Like in “total-ordering”, we will track every number and flag, but instead of comparison, we should get the numbers sequence and then create the hash number. So, once again, total-ordering could be considered as a milestone for even faster (in theory) random-access approach.

MergeFunctions, main fields and runOnModule

There are two main important fields in the class:

FnTree – the set of all unique functions. It keeps items that couldn’t be merged with each other. It is defined as:

std::set<FunctionNode> FnTree;

Here FunctionNode is a wrapper for llvm::Function class, with implemented “<” operator among the functions set (below we explain how it works exactly; this is a key point in fast functions comparison).

Deferred – merging process can affect bodies of functions that are in FnTree already. Obviously, such functions should be rechecked again. In this case, we remove them from FnTree, and mark them to be rescanned, namely put them into Deferred list.

runOnModule

The algorithm is pretty simple:

  1. Put all module’s functions into the worklist.

2. Scan worklist’s functions twice: first enumerate only strong functions and then only weak ones:

2.1. Loop body: take a function from worklist (call it FCur) and try to insert it into FnTree: check whether FCur is equal to one of functions in FnTree. If there is an equal function in FnTree (call it FExists): merge function FCur with FExists. Otherwise add the function from the worklist to FnTree.

3. Once the worklist scanning and merging operations are complete, check the Deferred list. If it is not empty: refill the worklist contents with Deferred list and redo step 2, if the Deferred list is empty, then exit from method.

Functions comparison

At first, let’s define how exactly we compare complex objects.

Complex object comparison (function, basic-block, etc) is mostly based on its sub-object comparison results. It is similar to the next “tree” objects comparison:

  1. For two trees T1 and T2 we perform depth-first-traversal and have two sequences as a product: “T1Items” and “T2Items”.

  2. We then compare chains “T1Items” and “T2Items” in the most-significant-item-first order. The result of items comparison would be the result of T1 and T2 comparison itself.

FunctionComparator::compare(void)

A brief look at the source code tells us that the comparison starts in the “int FunctionComparator::compare(void)” method.

1. The first parts to be compared are the function’s attributes and some properties that is outside the “attributes” term, but still could make the function different without changing its body. This part of the comparison is usually done within simple cmpNumbers or cmpFlags operations (e.g. cmpFlags(F1->hasGC(), F2->hasGC())). Below is a full list of function’s properties to be compared on this stage:

  • Attributes (those are returned by Function::getAttributes() method).

  • GC, for equivalence, RHS and LHS should be both either without GC or with the same one.

  • Section, just like a GC: RHS and LHS should be defined in the same section.

  • Variable arguments. LHS and RHS should be both either with or without var-args.

  • Calling convention should be the same.

2. Function type. Checked by FunctionComparator::cmpType(Type*, Type*) method. It checks return type and parameters type; the method itself will be described later.

3. Associate function formal parameters with each other. Then comparing function bodies, if we see the usage of LHS’s i-th argument in LHS’s body, then, we want to see usage of RHS’s i-th argument at the same place in RHS’s body, otherwise functions are different. On this stage we grant the preference to those we met later in function body (value we met first would be less). This is done by “FunctionComparator::cmpValues(const Value*, const Value*)” method (will be described a bit later).

  1. Function body comparison. As it written in method comments:

“We do a CFG-ordered walk since the actual ordering of the blocks in the linked list is immaterial. Our walk starts at the entry block for both functions, then takes each block from each terminator in order. As an artifact, this also means that unreachable blocks are ignored.”

So, using this walk we get BBs from left and right in the same order, and compare them by “FunctionComparator::compare(const BasicBlock*, const BasicBlock*)” method.

We also associate BBs with each other, like we did it with function formal arguments (see cmpValues method below).

FunctionComparator::cmpType

Consider how type comparison works.

1. Coerce pointer to integer. If left type is a pointer, try to coerce it to the integer type. It could be done if its address space is 0, or if address spaces are ignored at all. Do the same thing for the right type.

2. If left and right types are equal, return 0. Otherwise we need to give preference to one of them. So proceed to the next step.

3. If types are of different kind (different type IDs). Return result of type IDs comparison, treating them as numbers (use cmpNumbers operation).

4. If types are vectors or integers, return result of their pointers comparison, comparing them as numbers.

  1. Check whether type ID belongs to the next group (call it equivalent-group):

    • Void

    • Float

    • Double

    • X86_FP80

    • FP128

    • PPC_FP128

    • Label

    • Metadata.

    If ID belongs to group above, return 0. Since it’s enough to see that types has the same TypeID. No additional information is required.

6. Left and right are pointers. Return result of address space comparison (numbers comparison).

7. Complex types (structures, arrays, etc.). Follow complex objects comparison technique (see the very first paragraph of this chapter). Both left and right are to be expanded and their element types will be checked the same way. If we get -1 or 1 on some stage, return it. Otherwise return 0.

8. Steps 1-6 describe all the possible cases, if we passed steps 1-6 and didn’t get any conclusions, then invoke llvm_unreachable, since it’s quite an unexpectable case.

cmpValues(const Value*, const Value*)

Method that compares local values.

This method gives us an answer to a very curious question: whether we could treat local values as equal, and which value is greater otherwise. It’s better to start from example:

Consider the situation when we’re looking at the same place in left function “FL” and in right function “FR”. Every part of left place is equal to the corresponding part of right place, and (!) both parts use Value instances, for example:

instr0 i32 %LV   ; left side, function FL
instr0 i32 %RV   ; right side, function FR

So, now our conclusion depends on Value instances comparison.

The main purpose of this method is to determine relation between such values.

What can we expect from equal functions? At the same place, in functions “FL” and “FR” we expect to see equal values, or values defined at the same place in “FL” and “FR”.

Consider a small example here:

define void %f(i32 %pf0, i32 %pf1) {
  instr0 i32 %pf0 instr1 i32 %pf1 instr2 i32 123
}
define void %g(i32 %pg0, i32 %pg1) {
  instr0 i32 %pg0 instr1 i32 %pg0 instr2 i32 123
}

In this example, pf0 is associated with pg0, pf1 is associated with pg1, and we also declare that pf0 < pf1, and thus pg0 < pf1.

Instructions with opcode “instr0” would be equal, since their types and opcodes are equal, and values are associated.

Instructions with opcode “instr1” from f is greater than instructions with opcode “instr1” from g; here we have equal types and opcodes, but “pf1 is greater than “pg0”.

Instructions with opcode “instr2” are equal, because their opcodes and types are equal, and the same constant is used as a value.

What we associate in cmpValues?

  • Function arguments. i-th argument from left function associated with i-th argument from right function.

  • BasicBlock instances. In basic-block enumeration loop we associate i-th BasicBlock from the left function with i-th BasicBlock from the right function.

  • Instructions.

  • Instruction operands. Note, we can meet Value here we have never seen before. In this case it is not a function argument, nor BasicBlock, nor Instruction. It is a global value. It is a constant, since it’s the only supposed global here. The method also compares: Constants that are of the same type and if right constant can be losslessly bit-casted to the left one, then we also compare them.

How to implement cmpValues?

Association is a case of equality for us. We just treat such values as equal, but, in general, we need to implement antisymmetric relation. As mentioned above, to understand what is less, we can use order in which we meet values. If both values have the same order in a function (met at the same time), we then treat values as associated. Otherwise – it depends on who was first.

Every time we run the top-level compare method, we initialize two identical maps (one for the left side, another one for the right side):

map<Value, int> sn_mapL, sn_mapR;

The key of the map is the Value itself, the value – is its order (call it serial number).

To add value V we need to perform the next procedure:

sn_map.insert(std::make_pair(V, sn_map.size()));

For the first Value, map will return 0, for the second Value map will return 1, and so on.

We can then check whether left and right values met at the same time with a simple comparison:

cmpNumbers(sn_mapL[Left], sn_mapR[Right]);

Of course, we can combine insertion and comparison:

std::pair<iterator, bool>
  LeftRes = sn_mapL.insert(std::make_pair(Left, sn_mapL.size())), RightRes
  = sn_mapR.insert(std::make_pair(Right, sn_mapR.size()));
return cmpNumbers(LeftRes.first->second, RightRes.first->second);

Let’s look, how whole method could be implemented.

1. We have to start with the bad news. Consider function self and cross-referencing cases:

// self-reference unsigned fact0(unsigned n) { return n > 1 ? n
* fact0(n-1) : 1; } unsigned fact1(unsigned n) { return n > 1 ? n *
fact1(n-1) : 1; }

// cross-reference unsigned ping(unsigned n) { return n!= 0 ? pong(n-1) : 0;
} unsigned pong(unsigned n) { return n!= 0 ? ping(n-1) : 0; }

This comparison has been implemented in initial MergeFunctions pass version. But, unfortunately, it is not transitive. And this is the only case we can’t convert to less-equal-greater comparison. It is a seldom case, 4-5 functions of 10000 (checked in test-suite), and, we hope, the reader would forgive us for such a sacrifice in order to get the O(log(N)) pass time.

2. If left/right Value is a constant, we have to compare them. Return 0 if it is the same constant, or use cmpConstants method otherwise.

3. If left/right is InlineAsm instance. Return result of Value pointers comparison.

4. Explicit association of L (left value) and R (right value). We need to find out whether values met at the same time, and thus are associated. Or we need to put the rule: when we treat L < R. Now it is easy: we just return the result of numbers comparison:

std::pair<iterator, bool>
  LeftRes = sn_mapL.insert(std::make_pair(Left, sn_mapL.size())),
  RightRes = sn_mapR.insert(std::make_pair(Right, sn_mapR.size()));
if (LeftRes.first->second == RightRes.first->second) return 0;
if (LeftRes.first->second < RightRes.first->second) return -1;
return 1;

Now when cmpValues returns 0, we can proceed the comparison procedure. Otherwise, if we get (-1 or 1), we need to pass this result to the top level, and finish comparison procedure.

cmpConstants

Performs constants comparison as follows:

1. Compare constant types using cmpType method. If the result is -1 or 1, goto step 2, otherwise proceed to step 3.

2. If types are different, we still can check whether constants could be losslessly bitcasted to each other. The further explanation is modification of canLosslesslyBitCastTo method.

2.1 Check whether constants are of the first class types (isFirstClassType check):

2.1.1. If both constants are not of the first class type: return result of cmpType.

2.1.2. Otherwise, if left type is not of the first class, return -1. If right type is not of the first class, return 1.

2.1.3. If both types are of the first class type, proceed to the next step (2.1.3.1).

2.1.3.1. If types are vectors, compare their bitwidth using the cmpNumbers. If result is not 0, return it.

2.1.3.2. Different types, but not a vectors:

  • if both of them are pointers, good for us, we can proceed to step 3.

  • if one of types is pointer, return result of isPointer flags comparison (cmpFlags operation).

  • otherwise we have no methods to prove bitcastability, and thus return result of types comparison (-1 or 1).

Steps below are for the case when types are equal, or case when constants are bitcastable:

3. One of constants is a “null” value. Return the result of cmpFlags(L->isNullValue, R->isNullValue) comparison.

  1. Compare value IDs, and return result if it is not 0:

if (int Res = cmpNumbers(L->getValueID(), R->getValueID()))
  return Res;

5. Compare the contents of constants. The comparison depends on the kind of constants, but on this stage it is just a lexicographical comparison. Just see how it was described in the beginning of “Functions comparison” paragraph. Mathematically, it is equal to the next case: we encode left constant and right constant (with similar way bitcode-writer does). Then compare left code sequence and right code sequence.

compare(const BasicBlock*, const BasicBlock*)

Compares two BasicBlock instances.

It enumerates instructions from left BB and right BB.

1. It assigns serial numbers to the left and right instructions, using cmpValues method.

2. If one of left or right is GEP (GetElementPtr), then treat GEP as greater than other instructions. If both instructions are GEPs use cmpGEP method for comparison. If result is -1 or 1, pass it to the top-level comparison (return it).

3.1. Compare operations. Call cmpOperation method. If result is -1 or 1, return it.

3.2. Compare number of operands, if result is -1 or 1, return it.

3.3. Compare operands themselves, use cmpValues method. Return result if it is -1 or 1.

3.4. Compare type of operands, using cmpType method. Return result if it is -1 or 1.

3.5. Proceed to the next instruction.

  1. We can finish instruction enumeration in 3 cases:

    4.1. We reached the end of both left and right basic-blocks. We didn’t exit on steps 1-3, so contents are equal, return 0.

    4.2. We have reached the end of the left basic-block. Return -1.

    4.3. Return 1 (we reached the end of the right basic block).

cmpGEP

Compares two GEPs (getelementptr instructions).

It differs from regular operations comparison with the only thing: possibility to use accumulateConstantOffset method.

So, if we get constant offset for both left and right GEPs, then compare it as numbers, and return comparison result.

Otherwise treat it like a regular operation (see previous paragraph).

cmpOperation

Compares instruction opcodes and some important operation properties.

  1. Compare opcodes, if it differs return the result.

  2. Compare number of operands. If it differs – return the result.

3. Compare operation types, use cmpType. All the same – if types are different, return result.

4. Compare subclassOptionalData, get it with getRawSubclassOptionalData method, and compare it like a numbers.

  1. Compare operand types.

6. For some particular instructions, check equivalence (relation in our case) of some significant attributes. For example, we have to compare alignment for load instructions.

O(log(N))

Methods described above implement order relationship. And latter, could be used for nodes comparison in a binary tree. So we can organize functions set into the binary tree and reduce the cost of lookup procedure from O(N*N) to O(log(N)).

Merging process, mergeTwoFunctions

Once MergeFunctions detected that current function (G) is equal to one that were analyzed before (function F) it calls mergeTwoFunctions(Function*, Function*).

Operation affects FnTree contents with next way: F will stay in FnTree. G being equal to F will not be added to FnTree. Calls of G would be replaced with something else. It changes bodies of callers. So, functions that calls G would be put into Deferred set and removed from FnTree, and analyzed again.

The approach is next:

1. Most wished case: when we can use alias and both of F and G are weak. We make both of them with aliases to the third strong function H. Actually H is F. See below how it’s made (but it’s better to look straight into the source code). Well, this is a case when we can just replace G with F everywhere, we use replaceAllUsesWith operation here (RAUW).

2. F could not be overridden, while G could. It would be good to do the next: after merging the places where overridable function were used, still use overridable stub. So try to make G alias to F, or create overridable tail call wrapper around F and replace G with that call.

3. Neither F nor G could be overridden. We can’t use RAUW. We can just change the callers: call F instead of G. That’s what replaceDirectCallers does.

Below is a detailed body description.

If “F” may be overridden

As follows from mayBeOverridden comments: “whether the definition of this global may be replaced by something non-equivalent at link time”. If so, that’s ok: we can use alias to F instead of G or change call instructions itself.

HasGlobalAliases, removeUsers

First consider the case when we have global aliases of one function name to another. Our purpose is make both of them with aliases to the third strong function. Though if we keep F alive and without major changes we can leave it in FnTree. Try to combine these two goals.

Do stub replacement of F itself with an alias to F.

1. Create stub function H, with the same name and attributes like function F. It takes maximum alignment of F and G.

2. Replace all uses of function F with uses of function H. It is the two steps procedure instead. First of all, we must take into account, all functions from whom F is called would be changed: since we change the call argument (from F to H). If so we must to review these caller functions again after this procedure. We remove callers from FnTree, method with name removeUsers(F) does that (don’t confuse with replaceAllUsesWith):

2.1. Inside removeUsers(Value* V) we go through the all values that use value V (or F in our context). If value is instruction, we go to function that holds this instruction and mark it as to-be-analyzed-again (put to Deferred set), we also remove caller from FnTree.

2.2. Now we can do the replacement: call F->replaceAllUsesWith(H).

3. H (that now “officially” plays F’s role) is replaced with alias to F. Do the same with G: replace it with alias to F. So finally everywhere F was used, we use H and it is alias to F, and everywhere G was used we also have alias to F.

  1. Set F linkage to private. Make it strong :-)

No global aliases, replaceDirectCallers

If global aliases are not supported. We call replaceDirectCallers. Just go through all calls of G and replace it with calls of F. If you look into the method you will see that it scans all uses of G too, and if use is callee (if user is call instruction and G is used as what to be called), we replace it with use of F.

If “F” could not be overridden, fix it!

We call writeThunkOrAlias(Function *F, Function *G). Here we try to replace G with alias to F first. The next conditions are essential:

  • target should support global aliases,

  • the address itself of G should be not significant, not named and not referenced anywhere,

  • function should come with external, local or weak linkage.

Otherwise we write thunk: some wrapper that has G’s interface and calls F, so G could be replaced with this wrapper.

writeAlias

As follows from llvm reference:

“Aliases act as second name for the aliasee value”. So we just want to create a second name for F and use it instead of G:

  1. create global alias itself (GA),

  2. adjust alignment of F so it must be maximum of current and G’s alignment;

  3. replace uses of G:

    3.1. first mark all callers of G as to-be-analyzed-again, using removeUsers method (see chapter above),

    3.2. call G->replaceAllUsesWith(GA).

  4. Get rid of G.

writeThunk

As it written in method comments:

“Replace G with a simple tail call to bitcast(F). Also replace direct uses of G with bitcast(F). Deletes G.”

In general it does the same as usual when we want to replace callee, except the first point:

1. We generate tail call wrapper around F, but with interface that allows use it instead of G.

  1. “As-usual”: removeUsers and replaceAllUsesWith then.

  2. Get rid of G.