2017-10-18 05:27:42 +08:00
|
|
|
//===- VPlan.h - Represent A Vectorizer Plan --------------------*- C++ -*-===//
|
2017-08-27 20:55:46 +08:00
|
|
|
//
|
2019-01-19 16:50:56 +08:00
|
|
|
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
|
|
|
|
// See https://llvm.org/LICENSE.txt for license information.
|
|
|
|
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
|
2017-08-27 20:55:46 +08:00
|
|
|
//
|
|
|
|
//===----------------------------------------------------------------------===//
|
2017-10-18 05:27:42 +08:00
|
|
|
//
|
2017-08-27 20:55:46 +08:00
|
|
|
/// \file
|
|
|
|
/// This file contains the declarations of the Vectorization Plan base classes:
|
|
|
|
/// 1. VPBasicBlock and VPRegionBlock that inherit from a common pure virtual
|
|
|
|
/// VPBlockBase, together implementing a Hierarchical CFG;
|
|
|
|
/// 2. Specializations of GraphTraits that allow VPBlockBase graphs to be
|
|
|
|
/// treated as proper graphs for generic algorithms;
|
|
|
|
/// 3. Pure virtual VPRecipeBase serving as the base class for recipes contained
|
|
|
|
/// within VPBasicBlocks;
|
2017-11-20 20:01:47 +08:00
|
|
|
/// 4. VPInstruction, a concrete Recipe and VPUser modeling a single planned
|
|
|
|
/// instruction;
|
|
|
|
/// 5. The VPlan class holding a candidate for vectorization;
|
|
|
|
/// 6. The VPlanPrinter class providing a way to print a plan in dot format;
|
2017-08-27 20:55:46 +08:00
|
|
|
/// These are documented in docs/VectorizationPlan.rst.
|
2017-10-18 05:27:42 +08:00
|
|
|
//
|
2017-08-27 20:55:46 +08:00
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
|
|
|
#ifndef LLVM_TRANSFORMS_VECTORIZE_VPLAN_H
|
|
|
|
#define LLVM_TRANSFORMS_VECTORIZE_VPLAN_H
|
|
|
|
|
2018-07-31 09:57:29 +08:00
|
|
|
#include "VPlanLoopInfo.h"
|
2017-11-20 20:01:47 +08:00
|
|
|
#include "VPlanValue.h"
|
2017-10-18 05:27:42 +08:00
|
|
|
#include "llvm/ADT/DenseMap.h"
|
2018-07-31 05:33:31 +08:00
|
|
|
#include "llvm/ADT/DepthFirstIterator.h"
|
2017-08-27 20:55:46 +08:00
|
|
|
#include "llvm/ADT/GraphTraits.h"
|
2017-10-18 05:27:42 +08:00
|
|
|
#include "llvm/ADT/Optional.h"
|
2019-10-17 02:39:53 +08:00
|
|
|
#include "llvm/ADT/SmallBitVector.h"
|
2018-06-12 19:16:56 +08:00
|
|
|
#include "llvm/ADT/SmallPtrSet.h"
|
2017-08-27 20:55:46 +08:00
|
|
|
#include "llvm/ADT/SmallSet.h"
|
2017-10-18 05:27:42 +08:00
|
|
|
#include "llvm/ADT/SmallVector.h"
|
|
|
|
#include "llvm/ADT/Twine.h"
|
2017-08-27 20:55:46 +08:00
|
|
|
#include "llvm/ADT/ilist.h"
|
|
|
|
#include "llvm/ADT/ilist_node.h"
|
2018-11-13 23:58:18 +08:00
|
|
|
#include "llvm/Analysis/VectorUtils.h"
|
2017-08-27 20:55:46 +08:00
|
|
|
#include "llvm/IR/IRBuilder.h"
|
2017-10-18 05:27:42 +08:00
|
|
|
#include <algorithm>
|
|
|
|
#include <cassert>
|
|
|
|
#include <cstddef>
|
|
|
|
#include <map>
|
|
|
|
#include <string>
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
namespace llvm {
|
|
|
|
|
|
|
|
class BasicBlock;
|
2017-10-18 05:27:42 +08:00
|
|
|
class DominatorTree;
|
2017-08-27 20:55:46 +08:00
|
|
|
class InnerLoopVectorizer;
|
2018-11-13 23:58:18 +08:00
|
|
|
template <class T> class InterleaveGroup;
|
|
|
|
class LoopInfo;
|
2017-10-18 05:27:42 +08:00
|
|
|
class raw_ostream;
|
2020-08-06 17:10:50 +08:00
|
|
|
class RecurrenceDescriptor;
|
2017-10-18 05:27:42 +08:00
|
|
|
class Value;
|
2017-08-27 20:55:46 +08:00
|
|
|
class VPBasicBlock;
|
2017-10-18 05:27:42 +08:00
|
|
|
class VPRegionBlock;
|
2020-04-23 19:33:57 +08:00
|
|
|
class VPSlotTracker;
|
2018-06-09 01:30:45 +08:00
|
|
|
class VPlan;
|
[VPlan, SLP] Add simple SLP analysis on top of VPlan.
This patch adds an initial implementation of the look-ahead SLP tree
construction described in 'Look-Ahead SLP: Auto-vectorization in the Presence
of Commutative Operations, CGO 2018 by Vasileios Porpodas, Rodrigo C. O. Rocha,
Luís F. W. Góes'.
It returns an SLP tree represented as VPInstructions, with combined
instructions represented as a single, wider VPInstruction.
This initial version does not support instructions with multiple
different users (either inside or outside the SLP tree) or
non-instruction operands; it won't generate any shuffles or
insertelement instructions.
It also just adds the analysis that builds an SLP tree rooted in a set
of stores. It does not include any cost modeling or memory legality
checks. The plan is to integrate it with VPlan based cost modeling, once
available and to only apply it to operations that can be widened.
A follow-up patch will add a support for replacing instructions in a
VPlan with their SLP counter parts.
Reviewers: Ayal, mssimpso, rengolin, mkuper, hfinkel, hsaito, dcaballe, vporpo, RKSimon, ABataev
Reviewed By: rengolin
Differential Revision: https://reviews.llvm.org/D4949
llvm-svn: 346857
2018-11-14 21:11:49 +08:00
|
|
|
class VPlanSlp;
|
2018-06-09 01:30:45 +08:00
|
|
|
|
|
|
|
/// A range of powers-of-2 vectorization factors with fixed start and
|
|
|
|
/// adjustable end. The range includes start and excludes end, e.g.,:
|
|
|
|
/// [1, 9) = {1, 2, 4, 8}
|
|
|
|
struct VFRange {
|
|
|
|
// A power of 2.
|
|
|
|
const unsigned Start;
|
|
|
|
|
|
|
|
// Need not be a power of 2. If End <= Start range is empty.
|
|
|
|
unsigned End;
|
|
|
|
};
|
|
|
|
|
|
|
|
using VPlanPtr = std::unique_ptr<VPlan>;
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
/// In what follows, the term "input IR" refers to code that is fed into the
|
|
|
|
/// vectorizer whereas the term "output IR" refers to code that is generated by
|
|
|
|
/// the vectorizer.
|
|
|
|
|
|
|
|
/// VPIteration represents a single point in the iteration space of the output
|
|
|
|
/// (vectorized and/or unrolled) IR loop.
|
|
|
|
struct VPIteration {
|
2017-10-18 05:27:42 +08:00
|
|
|
/// in [0..UF)
|
|
|
|
unsigned Part;
|
|
|
|
|
|
|
|
/// in [0..VF)
|
|
|
|
unsigned Lane;
|
2017-08-27 20:55:46 +08:00
|
|
|
};
|
|
|
|
|
|
|
|
/// This is a helper struct for maintaining vectorization state. It's used for
|
|
|
|
/// mapping values from the original loop to their corresponding values in
|
|
|
|
/// the new loop. Two mappings are maintained: one for vectorized values and
|
|
|
|
/// one for scalarized values. Vectorized values are represented with UF
|
|
|
|
/// vector values in the new loop, and scalarized values are represented with
|
|
|
|
/// UF x VF scalar values in the new loop. UF and VF are the unroll and
|
|
|
|
/// vectorization factors, respectively.
|
|
|
|
///
|
|
|
|
/// Entries can be added to either map with setVectorValue and setScalarValue,
|
|
|
|
/// which assert that an entry was not already added before. If an entry is to
|
|
|
|
/// replace an existing one, call resetVectorValue and resetScalarValue. This is
|
|
|
|
/// currently needed to modify the mapped values during "fix-up" operations that
|
|
|
|
/// occur once the first phase of widening is complete. These operations include
|
|
|
|
/// type truncation and the second phase of recurrence widening.
|
|
|
|
///
|
|
|
|
/// Entries from either map can be retrieved using the getVectorValue and
|
|
|
|
/// getScalarValue functions, which assert that the desired value exists.
|
|
|
|
struct VectorizerValueMap {
|
2017-11-20 20:01:47 +08:00
|
|
|
friend struct VPTransformState;
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
private:
|
|
|
|
/// The unroll factor. Each entry in the vector map contains UF vector values.
|
|
|
|
unsigned UF;
|
|
|
|
|
|
|
|
/// The vectorization factor. Each entry in the scalar map contains UF x VF
|
|
|
|
/// scalar values.
|
[llvm][LV] Replace `unsigned VF` with `ElementCount VF` [NFCI]
Changes:
* Change `ToVectorTy` to deal directly with `ElementCount` instances.
* `VF == 1` replaced with `VF.isScalar()`.
* `VF > 1` and `VF >=2` replaced with `VF.isVector()`.
* `VF <=1` is replaced with `VF.isZero() || VF.isScalar()`.
* Replaced the uses of `llvm::SmallSet<ElementCount, ...>` with
`llvm::SmallSetVector<ElementCount, ...>`. This avoids the need of an
ordering function for the `ElementCount` class.
* Bits and pieces around printing the `ElementCount` to string streams.
To guarantee that this change is a NFC, `VF.Min` and asserts are used
in the following places:
1. When it doesn't make sense to deal with the scalable property, for
example:
a. When computing unrolling factors.
b. When shuffle masks are built for fixed width vector types
In this cases, an
assert(!VF.Scalable && "<mgs>") has been added to make sure we don't
enter coepaths that don't make sense for scalable vectors.
2. When there is a conscious decision to use `FixedVectorType`. These
uses of `FixedVectorType` will likely be removed in favour of
`VectorType` once the vectorizer is generic enough to deal with both
fixed vector types and scalable vector types.
3. When dealing with building constants out of the value of VF, for
example when computing the vectorization `step`, or building vectors
of indices. These operation _make sense_ for scalable vectors too,
but changing the code in these places to be generic and make it work
for scalable vectors is to be submitted in a separate patch, as it is
a functional change.
4. When building the potential VFs in VPlan. Making the VPlan generic
enough to handle scalable vectorization factors is a functional change
that needs a separate patch. See for example `void
LoopVectorizationPlanner::buildVPlans(unsigned MinVF, unsigned
MaxVF)`.
5. The class `IntrinsicCostAttribute`: this class still uses `unsigned
VF` as updating the field to use `ElementCount` woudl require changes
that could result in changing the behavior of the compiler. Will be done
in a separate patch.
7. When dealing with user input for forcing the vectorization
factor. In this case, adding support for scalable vectorization is a
functional change that migh require changes at command line.
Note that in some places the idiom
```
unsigned VF = ...
auto VTy = FixedVectorType::get(ScalarTy, VF)
```
has been replaced with
```
ElementCount VF = ...
assert(!VF.Scalable && ...);
auto VTy = VectorType::get(ScalarTy, VF)
```
The assertion guarantees that the new code is (at least in debug mode)
functionally equivalent to the old version. Notice that this change had been
possible because none of the methods that are specific to `FixedVectorType`
were used after the instantiation of `VTy`.
Reviewed By: rengolin, ctetreau
Differential Revision: https://reviews.llvm.org/D85794
2020-08-08 06:03:24 +08:00
|
|
|
ElementCount VF;
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
/// The vector and scalar map storage. We use std::map and not DenseMap
|
|
|
|
/// because insertions to DenseMap invalidate its iterators.
|
2017-10-18 05:27:42 +08:00
|
|
|
using VectorParts = SmallVector<Value *, 2>;
|
|
|
|
using ScalarParts = SmallVector<SmallVector<Value *, 4>, 2>;
|
2017-08-27 20:55:46 +08:00
|
|
|
std::map<Value *, VectorParts> VectorMapStorage;
|
|
|
|
std::map<Value *, ScalarParts> ScalarMapStorage;
|
|
|
|
|
|
|
|
public:
|
|
|
|
/// Construct an empty map with the given unroll and vectorization factors.
|
[llvm][LV] Replace `unsigned VF` with `ElementCount VF` [NFCI]
Changes:
* Change `ToVectorTy` to deal directly with `ElementCount` instances.
* `VF == 1` replaced with `VF.isScalar()`.
* `VF > 1` and `VF >=2` replaced with `VF.isVector()`.
* `VF <=1` is replaced with `VF.isZero() || VF.isScalar()`.
* Replaced the uses of `llvm::SmallSet<ElementCount, ...>` with
`llvm::SmallSetVector<ElementCount, ...>`. This avoids the need of an
ordering function for the `ElementCount` class.
* Bits and pieces around printing the `ElementCount` to string streams.
To guarantee that this change is a NFC, `VF.Min` and asserts are used
in the following places:
1. When it doesn't make sense to deal with the scalable property, for
example:
a. When computing unrolling factors.
b. When shuffle masks are built for fixed width vector types
In this cases, an
assert(!VF.Scalable && "<mgs>") has been added to make sure we don't
enter coepaths that don't make sense for scalable vectors.
2. When there is a conscious decision to use `FixedVectorType`. These
uses of `FixedVectorType` will likely be removed in favour of
`VectorType` once the vectorizer is generic enough to deal with both
fixed vector types and scalable vector types.
3. When dealing with building constants out of the value of VF, for
example when computing the vectorization `step`, or building vectors
of indices. These operation _make sense_ for scalable vectors too,
but changing the code in these places to be generic and make it work
for scalable vectors is to be submitted in a separate patch, as it is
a functional change.
4. When building the potential VFs in VPlan. Making the VPlan generic
enough to handle scalable vectorization factors is a functional change
that needs a separate patch. See for example `void
LoopVectorizationPlanner::buildVPlans(unsigned MinVF, unsigned
MaxVF)`.
5. The class `IntrinsicCostAttribute`: this class still uses `unsigned
VF` as updating the field to use `ElementCount` woudl require changes
that could result in changing the behavior of the compiler. Will be done
in a separate patch.
7. When dealing with user input for forcing the vectorization
factor. In this case, adding support for scalable vectorization is a
functional change that migh require changes at command line.
Note that in some places the idiom
```
unsigned VF = ...
auto VTy = FixedVectorType::get(ScalarTy, VF)
```
has been replaced with
```
ElementCount VF = ...
assert(!VF.Scalable && ...);
auto VTy = VectorType::get(ScalarTy, VF)
```
The assertion guarantees that the new code is (at least in debug mode)
functionally equivalent to the old version. Notice that this change had been
possible because none of the methods that are specific to `FixedVectorType`
were used after the instantiation of `VTy`.
Reviewed By: rengolin, ctetreau
Differential Revision: https://reviews.llvm.org/D85794
2020-08-08 06:03:24 +08:00
|
|
|
VectorizerValueMap(unsigned UF, ElementCount VF) : UF(UF), VF(VF) {}
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
/// \return True if the map has any vector entry for \p Key.
|
|
|
|
bool hasAnyVectorValue(Value *Key) const {
|
|
|
|
return VectorMapStorage.count(Key);
|
|
|
|
}
|
|
|
|
|
|
|
|
/// \return True if the map has a vector entry for \p Key and \p Part.
|
|
|
|
bool hasVectorValue(Value *Key, unsigned Part) const {
|
|
|
|
assert(Part < UF && "Queried Vector Part is too large.");
|
|
|
|
if (!hasAnyVectorValue(Key))
|
|
|
|
return false;
|
|
|
|
const VectorParts &Entry = VectorMapStorage.find(Key)->second;
|
|
|
|
assert(Entry.size() == UF && "VectorParts has wrong dimensions.");
|
|
|
|
return Entry[Part] != nullptr;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// \return True if the map has any scalar entry for \p Key.
|
|
|
|
bool hasAnyScalarValue(Value *Key) const {
|
|
|
|
return ScalarMapStorage.count(Key);
|
|
|
|
}
|
|
|
|
|
|
|
|
/// \return True if the map has a scalar entry for \p Key and \p Instance.
|
|
|
|
bool hasScalarValue(Value *Key, const VPIteration &Instance) const {
|
|
|
|
assert(Instance.Part < UF && "Queried Scalar Part is too large.");
|
2020-08-14 19:15:59 +08:00
|
|
|
assert(Instance.Lane < VF.getKnownMinValue() &&
|
|
|
|
"Queried Scalar Lane is too large.");
|
|
|
|
assert(!VF.isScalable() && "VF is assumed to be non scalable.");
|
[llvm][LV] Replace `unsigned VF` with `ElementCount VF` [NFCI]
Changes:
* Change `ToVectorTy` to deal directly with `ElementCount` instances.
* `VF == 1` replaced with `VF.isScalar()`.
* `VF > 1` and `VF >=2` replaced with `VF.isVector()`.
* `VF <=1` is replaced with `VF.isZero() || VF.isScalar()`.
* Replaced the uses of `llvm::SmallSet<ElementCount, ...>` with
`llvm::SmallSetVector<ElementCount, ...>`. This avoids the need of an
ordering function for the `ElementCount` class.
* Bits and pieces around printing the `ElementCount` to string streams.
To guarantee that this change is a NFC, `VF.Min` and asserts are used
in the following places:
1. When it doesn't make sense to deal with the scalable property, for
example:
a. When computing unrolling factors.
b. When shuffle masks are built for fixed width vector types
In this cases, an
assert(!VF.Scalable && "<mgs>") has been added to make sure we don't
enter coepaths that don't make sense for scalable vectors.
2. When there is a conscious decision to use `FixedVectorType`. These
uses of `FixedVectorType` will likely be removed in favour of
`VectorType` once the vectorizer is generic enough to deal with both
fixed vector types and scalable vector types.
3. When dealing with building constants out of the value of VF, for
example when computing the vectorization `step`, or building vectors
of indices. These operation _make sense_ for scalable vectors too,
but changing the code in these places to be generic and make it work
for scalable vectors is to be submitted in a separate patch, as it is
a functional change.
4. When building the potential VFs in VPlan. Making the VPlan generic
enough to handle scalable vectorization factors is a functional change
that needs a separate patch. See for example `void
LoopVectorizationPlanner::buildVPlans(unsigned MinVF, unsigned
MaxVF)`.
5. The class `IntrinsicCostAttribute`: this class still uses `unsigned
VF` as updating the field to use `ElementCount` woudl require changes
that could result in changing the behavior of the compiler. Will be done
in a separate patch.
7. When dealing with user input for forcing the vectorization
factor. In this case, adding support for scalable vectorization is a
functional change that migh require changes at command line.
Note that in some places the idiom
```
unsigned VF = ...
auto VTy = FixedVectorType::get(ScalarTy, VF)
```
has been replaced with
```
ElementCount VF = ...
assert(!VF.Scalable && ...);
auto VTy = VectorType::get(ScalarTy, VF)
```
The assertion guarantees that the new code is (at least in debug mode)
functionally equivalent to the old version. Notice that this change had been
possible because none of the methods that are specific to `FixedVectorType`
were used after the instantiation of `VTy`.
Reviewed By: rengolin, ctetreau
Differential Revision: https://reviews.llvm.org/D85794
2020-08-08 06:03:24 +08:00
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
if (!hasAnyScalarValue(Key))
|
|
|
|
return false;
|
|
|
|
const ScalarParts &Entry = ScalarMapStorage.find(Key)->second;
|
|
|
|
assert(Entry.size() == UF && "ScalarParts has wrong dimensions.");
|
2020-08-14 19:15:59 +08:00
|
|
|
assert(Entry[Instance.Part].size() == VF.getKnownMinValue() &&
|
2017-08-27 20:55:46 +08:00
|
|
|
"ScalarParts has wrong dimensions.");
|
|
|
|
return Entry[Instance.Part][Instance.Lane] != nullptr;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Retrieve the existing vector value that corresponds to \p Key and
|
|
|
|
/// \p Part.
|
|
|
|
Value *getVectorValue(Value *Key, unsigned Part) {
|
|
|
|
assert(hasVectorValue(Key, Part) && "Getting non-existent value.");
|
|
|
|
return VectorMapStorage[Key][Part];
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Retrieve the existing scalar value that corresponds to \p Key and
|
|
|
|
/// \p Instance.
|
|
|
|
Value *getScalarValue(Value *Key, const VPIteration &Instance) {
|
|
|
|
assert(hasScalarValue(Key, Instance) && "Getting non-existent value.");
|
|
|
|
return ScalarMapStorage[Key][Instance.Part][Instance.Lane];
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Set a vector value associated with \p Key and \p Part. Assumes such a
|
|
|
|
/// value is not already set. If it is, use resetVectorValue() instead.
|
|
|
|
void setVectorValue(Value *Key, unsigned Part, Value *Vector) {
|
|
|
|
assert(!hasVectorValue(Key, Part) && "Vector value already set for part");
|
|
|
|
if (!VectorMapStorage.count(Key)) {
|
|
|
|
VectorParts Entry(UF);
|
|
|
|
VectorMapStorage[Key] = Entry;
|
|
|
|
}
|
|
|
|
VectorMapStorage[Key][Part] = Vector;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Set a scalar value associated with \p Key and \p Instance. Assumes such a
|
|
|
|
/// value is not already set.
|
|
|
|
void setScalarValue(Value *Key, const VPIteration &Instance, Value *Scalar) {
|
|
|
|
assert(!hasScalarValue(Key, Instance) && "Scalar value already set");
|
|
|
|
if (!ScalarMapStorage.count(Key)) {
|
|
|
|
ScalarParts Entry(UF);
|
|
|
|
// TODO: Consider storing uniform values only per-part, as they occupy
|
|
|
|
// lane 0 only, keeping the other VF-1 redundant entries null.
|
|
|
|
for (unsigned Part = 0; Part < UF; ++Part)
|
2020-08-14 19:15:59 +08:00
|
|
|
Entry[Part].resize(VF.getKnownMinValue(), nullptr);
|
2017-08-27 20:55:46 +08:00
|
|
|
ScalarMapStorage[Key] = Entry;
|
|
|
|
}
|
|
|
|
ScalarMapStorage[Key][Instance.Part][Instance.Lane] = Scalar;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Reset the vector value associated with \p Key for the given \p Part.
|
|
|
|
/// This function can be used to update values that have already been
|
|
|
|
/// vectorized. This is the case for "fix-up" operations including type
|
|
|
|
/// truncation and the second phase of recurrence vectorization.
|
|
|
|
void resetVectorValue(Value *Key, unsigned Part, Value *Vector) {
|
|
|
|
assert(hasVectorValue(Key, Part) && "Vector value not set for part");
|
|
|
|
VectorMapStorage[Key][Part] = Vector;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Reset the scalar value associated with \p Key for \p Part and \p Lane.
|
|
|
|
/// This function can be used to update values that have already been
|
|
|
|
/// scalarized. This is the case for "fix-up" operations including scalar phi
|
|
|
|
/// nodes for scalarized and predicated instructions.
|
|
|
|
void resetScalarValue(Value *Key, const VPIteration &Instance,
|
|
|
|
Value *Scalar) {
|
|
|
|
assert(hasScalarValue(Key, Instance) &&
|
|
|
|
"Scalar value not set for part and lane");
|
|
|
|
ScalarMapStorage[Key][Instance.Part][Instance.Lane] = Scalar;
|
|
|
|
}
|
|
|
|
};
|
|
|
|
|
2017-11-20 20:01:47 +08:00
|
|
|
/// This class is used to enable the VPlan to invoke a method of ILV. This is
|
|
|
|
/// needed until the method is refactored out of ILV and becomes reusable.
|
|
|
|
struct VPCallback {
|
|
|
|
virtual ~VPCallback() {}
|
|
|
|
virtual Value *getOrCreateVectorValues(Value *V, unsigned Part) = 0;
|
2019-11-28 19:54:38 +08:00
|
|
|
virtual Value *getOrCreateScalarValue(Value *V,
|
|
|
|
const VPIteration &Instance) = 0;
|
2017-11-20 20:01:47 +08:00
|
|
|
};
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
/// VPTransformState holds information passed down when "executing" a VPlan,
|
|
|
|
/// needed for generating the output IR.
|
|
|
|
struct VPTransformState {
|
[llvm][LV] Replace `unsigned VF` with `ElementCount VF` [NFCI]
Changes:
* Change `ToVectorTy` to deal directly with `ElementCount` instances.
* `VF == 1` replaced with `VF.isScalar()`.
* `VF > 1` and `VF >=2` replaced with `VF.isVector()`.
* `VF <=1` is replaced with `VF.isZero() || VF.isScalar()`.
* Replaced the uses of `llvm::SmallSet<ElementCount, ...>` with
`llvm::SmallSetVector<ElementCount, ...>`. This avoids the need of an
ordering function for the `ElementCount` class.
* Bits and pieces around printing the `ElementCount` to string streams.
To guarantee that this change is a NFC, `VF.Min` and asserts are used
in the following places:
1. When it doesn't make sense to deal with the scalable property, for
example:
a. When computing unrolling factors.
b. When shuffle masks are built for fixed width vector types
In this cases, an
assert(!VF.Scalable && "<mgs>") has been added to make sure we don't
enter coepaths that don't make sense for scalable vectors.
2. When there is a conscious decision to use `FixedVectorType`. These
uses of `FixedVectorType` will likely be removed in favour of
`VectorType` once the vectorizer is generic enough to deal with both
fixed vector types and scalable vector types.
3. When dealing with building constants out of the value of VF, for
example when computing the vectorization `step`, or building vectors
of indices. These operation _make sense_ for scalable vectors too,
but changing the code in these places to be generic and make it work
for scalable vectors is to be submitted in a separate patch, as it is
a functional change.
4. When building the potential VFs in VPlan. Making the VPlan generic
enough to handle scalable vectorization factors is a functional change
that needs a separate patch. See for example `void
LoopVectorizationPlanner::buildVPlans(unsigned MinVF, unsigned
MaxVF)`.
5. The class `IntrinsicCostAttribute`: this class still uses `unsigned
VF` as updating the field to use `ElementCount` woudl require changes
that could result in changing the behavior of the compiler. Will be done
in a separate patch.
7. When dealing with user input for forcing the vectorization
factor. In this case, adding support for scalable vectorization is a
functional change that migh require changes at command line.
Note that in some places the idiom
```
unsigned VF = ...
auto VTy = FixedVectorType::get(ScalarTy, VF)
```
has been replaced with
```
ElementCount VF = ...
assert(!VF.Scalable && ...);
auto VTy = VectorType::get(ScalarTy, VF)
```
The assertion guarantees that the new code is (at least in debug mode)
functionally equivalent to the old version. Notice that this change had been
possible because none of the methods that are specific to `FixedVectorType`
were used after the instantiation of `VTy`.
Reviewed By: rengolin, ctetreau
Differential Revision: https://reviews.llvm.org/D85794
2020-08-08 06:03:24 +08:00
|
|
|
VPTransformState(ElementCount VF, unsigned UF, LoopInfo *LI,
|
|
|
|
DominatorTree *DT, IRBuilder<> &Builder,
|
|
|
|
VectorizerValueMap &ValueMap, InnerLoopVectorizer *ILV,
|
|
|
|
VPCallback &Callback)
|
2017-11-20 20:01:47 +08:00
|
|
|
: VF(VF), UF(UF), Instance(), LI(LI), DT(DT), Builder(Builder),
|
|
|
|
ValueMap(ValueMap), ILV(ILV), Callback(Callback) {}
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
/// The chosen Vectorization and Unroll Factors of the loop being vectorized.
|
[llvm][LV] Replace `unsigned VF` with `ElementCount VF` [NFCI]
Changes:
* Change `ToVectorTy` to deal directly with `ElementCount` instances.
* `VF == 1` replaced with `VF.isScalar()`.
* `VF > 1` and `VF >=2` replaced with `VF.isVector()`.
* `VF <=1` is replaced with `VF.isZero() || VF.isScalar()`.
* Replaced the uses of `llvm::SmallSet<ElementCount, ...>` with
`llvm::SmallSetVector<ElementCount, ...>`. This avoids the need of an
ordering function for the `ElementCount` class.
* Bits and pieces around printing the `ElementCount` to string streams.
To guarantee that this change is a NFC, `VF.Min` and asserts are used
in the following places:
1. When it doesn't make sense to deal with the scalable property, for
example:
a. When computing unrolling factors.
b. When shuffle masks are built for fixed width vector types
In this cases, an
assert(!VF.Scalable && "<mgs>") has been added to make sure we don't
enter coepaths that don't make sense for scalable vectors.
2. When there is a conscious decision to use `FixedVectorType`. These
uses of `FixedVectorType` will likely be removed in favour of
`VectorType` once the vectorizer is generic enough to deal with both
fixed vector types and scalable vector types.
3. When dealing with building constants out of the value of VF, for
example when computing the vectorization `step`, or building vectors
of indices. These operation _make sense_ for scalable vectors too,
but changing the code in these places to be generic and make it work
for scalable vectors is to be submitted in a separate patch, as it is
a functional change.
4. When building the potential VFs in VPlan. Making the VPlan generic
enough to handle scalable vectorization factors is a functional change
that needs a separate patch. See for example `void
LoopVectorizationPlanner::buildVPlans(unsigned MinVF, unsigned
MaxVF)`.
5. The class `IntrinsicCostAttribute`: this class still uses `unsigned
VF` as updating the field to use `ElementCount` woudl require changes
that could result in changing the behavior of the compiler. Will be done
in a separate patch.
7. When dealing with user input for forcing the vectorization
factor. In this case, adding support for scalable vectorization is a
functional change that migh require changes at command line.
Note that in some places the idiom
```
unsigned VF = ...
auto VTy = FixedVectorType::get(ScalarTy, VF)
```
has been replaced with
```
ElementCount VF = ...
assert(!VF.Scalable && ...);
auto VTy = VectorType::get(ScalarTy, VF)
```
The assertion guarantees that the new code is (at least in debug mode)
functionally equivalent to the old version. Notice that this change had been
possible because none of the methods that are specific to `FixedVectorType`
were used after the instantiation of `VTy`.
Reviewed By: rengolin, ctetreau
Differential Revision: https://reviews.llvm.org/D85794
2020-08-08 06:03:24 +08:00
|
|
|
ElementCount VF;
|
2017-08-27 20:55:46 +08:00
|
|
|
unsigned UF;
|
|
|
|
|
|
|
|
/// Hold the indices to generate specific scalar instructions. Null indicates
|
|
|
|
/// that all instances are to be generated, using either scalar or vector
|
|
|
|
/// instructions.
|
|
|
|
Optional<VPIteration> Instance;
|
|
|
|
|
2017-11-20 20:01:47 +08:00
|
|
|
struct DataState {
|
|
|
|
/// A type for vectorized values in the new loop. Each value from the
|
|
|
|
/// original loop, when vectorized, is represented by UF vector values in
|
|
|
|
/// the new unrolled loop, where UF is the unroll factor.
|
|
|
|
typedef SmallVector<Value *, 2> PerPartValuesTy;
|
|
|
|
|
|
|
|
DenseMap<VPValue *, PerPartValuesTy> PerPartOutput;
|
|
|
|
} Data;
|
|
|
|
|
|
|
|
/// Get the generated Value for a given VPValue and a given Part. Note that
|
|
|
|
/// as some Defs are still created by ILV and managed in its ValueMap, this
|
|
|
|
/// method will delegate the call to ILV in such cases in order to provide
|
|
|
|
/// callers a consistent API.
|
|
|
|
/// \see set.
|
|
|
|
Value *get(VPValue *Def, unsigned Part) {
|
|
|
|
// If Values have been set for this Def return the one relevant for \p Part.
|
|
|
|
if (Data.PerPartOutput.count(Def))
|
|
|
|
return Data.PerPartOutput[Def][Part];
|
|
|
|
// Def is managed by ILV: bring the Values from ValueMap.
|
|
|
|
return Callback.getOrCreateVectorValues(VPValue2Value[Def], Part);
|
|
|
|
}
|
|
|
|
|
2020-06-03 18:30:06 +08:00
|
|
|
/// Get the generated Value for a given VPValue and given Part and Lane.
|
2019-11-28 19:54:38 +08:00
|
|
|
Value *get(VPValue *Def, const VPIteration &Instance) {
|
2020-06-03 18:30:06 +08:00
|
|
|
// If the Def is managed directly by VPTransformState, extract the lane from
|
|
|
|
// the relevant part. Note that currently only VPInstructions and external
|
|
|
|
// defs are managed by VPTransformState. Other Defs are still created by ILV
|
|
|
|
// and managed in its ValueMap. For those this method currently just
|
|
|
|
// delegates the call to ILV below.
|
|
|
|
if (Data.PerPartOutput.count(Def)) {
|
|
|
|
auto *VecPart = Data.PerPartOutput[Def][Instance.Part];
|
|
|
|
// TODO: Cache created scalar values.
|
|
|
|
return Builder.CreateExtractElement(VecPart,
|
|
|
|
Builder.getInt32(Instance.Lane));
|
|
|
|
}
|
|
|
|
|
2019-11-28 19:54:38 +08:00
|
|
|
return Callback.getOrCreateScalarValue(VPValue2Value[Def], Instance);
|
|
|
|
}
|
|
|
|
|
2017-11-20 20:01:47 +08:00
|
|
|
/// Set the generated Value for a given VPValue and a given Part.
|
|
|
|
void set(VPValue *Def, Value *V, unsigned Part) {
|
|
|
|
if (!Data.PerPartOutput.count(Def)) {
|
|
|
|
DataState::PerPartValuesTy Entry(UF);
|
|
|
|
Data.PerPartOutput[Def] = Entry;
|
|
|
|
}
|
|
|
|
Data.PerPartOutput[Def][Part] = V;
|
|
|
|
}
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
/// Hold state information used when constructing the CFG of the output IR,
|
|
|
|
/// traversing the VPBasicBlocks and generating corresponding IR BasicBlocks.
|
|
|
|
struct CFGState {
|
|
|
|
/// The previous VPBasicBlock visited. Initially set to null.
|
2017-10-18 05:27:42 +08:00
|
|
|
VPBasicBlock *PrevVPBB = nullptr;
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
/// The previous IR BasicBlock created or used. Initially set to the new
|
|
|
|
/// header BasicBlock.
|
2017-10-18 05:27:42 +08:00
|
|
|
BasicBlock *PrevBB = nullptr;
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
/// The last IR BasicBlock in the output IR. Set to the new latch
|
|
|
|
/// BasicBlock, used for placing the newly created BasicBlocks.
|
2017-10-18 05:27:42 +08:00
|
|
|
BasicBlock *LastBB = nullptr;
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
/// A mapping of each VPBasicBlock to the corresponding BasicBlock. In case
|
|
|
|
/// of replication, maps the BasicBlock of the last replica created.
|
|
|
|
SmallDenseMap<VPBasicBlock *, BasicBlock *> VPBB2IRBB;
|
|
|
|
|
[VPlan] Implement initial vector code generation support for simple outer loops.
Summary:
[VPlan] Implement vector code generation support for simple outer loops.
Context: Patch Series #1 for outer loop vectorization support in LV using VPlan. (RFC: http://lists.llvm.org/pipermail/llvm-dev/2017-December/119523.html).
This patch introduces vector code generation support for simple outer loops that are currently supported in the VPlanNativePath. Changes here essentially do the following:
- force vector code generation using explicit vectorize_width
- add conservative early returns in cost model and other places for VPlanNativePath
- add code for setting up outer loop inductions
- support for widening non-induction PHIs that can result from inner loops and uniform conditional branches
- support for generating uniform inner branches
We plan to add a handful C outer loop executable tests once the initial code generation support is committed. This patch is expected to be NFC for the inner loop vectorizer path. Since we are moving in the direction of supporting outer loop vectorization in LV, it may also be time to rename classes such as InnerLoopVectorizer.
Reviewers: fhahn, rengolin, hsaito, dcaballe, mkuper, hfinkel, Ayal
Reviewed By: fhahn, hsaito
Subscribers: dmgreen, bollu, tschuett, rkruppe, rogfer01, llvm-commits
Differential Revision: https://reviews.llvm.org/D50820
llvm-svn: 342197
2018-09-14 08:36:00 +08:00
|
|
|
/// Vector of VPBasicBlocks whose terminator instruction needs to be fixed
|
|
|
|
/// up at the end of vector code generation.
|
|
|
|
SmallVector<VPBasicBlock *, 8> VPBBsToFix;
|
|
|
|
|
2017-10-18 05:27:42 +08:00
|
|
|
CFGState() = default;
|
2017-08-27 20:55:46 +08:00
|
|
|
} CFG;
|
|
|
|
|
|
|
|
/// Hold a pointer to LoopInfo to register new basic blocks in the loop.
|
2017-10-18 05:27:42 +08:00
|
|
|
LoopInfo *LI;
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
/// Hold a pointer to Dominator Tree to register new basic blocks in the loop.
|
2017-10-18 05:27:42 +08:00
|
|
|
DominatorTree *DT;
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
/// Hold a reference to the IRBuilder used to generate output IR code.
|
|
|
|
IRBuilder<> &Builder;
|
|
|
|
|
|
|
|
/// Hold a reference to the Value state information used when generating the
|
|
|
|
/// Values of the output IR.
|
|
|
|
VectorizerValueMap &ValueMap;
|
|
|
|
|
2017-11-20 20:01:47 +08:00
|
|
|
/// Hold a reference to a mapping between VPValues in VPlan and original
|
|
|
|
/// Values they correspond to.
|
|
|
|
VPValue2ValueTy VPValue2Value;
|
|
|
|
|
2020-04-07 06:53:59 +08:00
|
|
|
/// Hold the canonical scalar IV of the vector loop (start=0, step=VF*UF).
|
|
|
|
Value *CanonicalIV = nullptr;
|
|
|
|
|
2018-10-18 23:03:15 +08:00
|
|
|
/// Hold the trip count of the scalar loop.
|
|
|
|
Value *TripCount = nullptr;
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
/// Hold a pointer to InnerLoopVectorizer to reuse its IR generation methods.
|
2017-10-18 05:27:42 +08:00
|
|
|
InnerLoopVectorizer *ILV;
|
2017-11-20 20:01:47 +08:00
|
|
|
|
|
|
|
VPCallback &Callback;
|
2017-08-27 20:55:46 +08:00
|
|
|
};
|
|
|
|
|
|
|
|
/// VPBlockBase is the building block of the Hierarchical Control-Flow Graph.
|
|
|
|
/// A VPBlockBase can be either a VPBasicBlock or a VPRegionBlock.
|
|
|
|
class VPBlockBase {
|
2018-05-22 02:14:23 +08:00
|
|
|
friend class VPBlockUtils;
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
const unsigned char SubclassID; ///< Subclass identifier (for isa/dyn_cast).
|
|
|
|
|
|
|
|
/// An optional name for the block.
|
|
|
|
std::string Name;
|
|
|
|
|
|
|
|
/// The immediate VPRegionBlock which this VPBlockBase belongs to, or null if
|
|
|
|
/// it is a topmost VPBlockBase.
|
2017-10-18 05:27:42 +08:00
|
|
|
VPRegionBlock *Parent = nullptr;
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
/// List of predecessor blocks.
|
|
|
|
SmallVector<VPBlockBase *, 1> Predecessors;
|
|
|
|
|
|
|
|
/// List of successor blocks.
|
|
|
|
SmallVector<VPBlockBase *, 1> Successors;
|
|
|
|
|
2018-07-09 23:57:09 +08:00
|
|
|
/// Successor selector, null for zero or single successor blocks.
|
|
|
|
VPValue *CondBit = nullptr;
|
|
|
|
|
2019-01-24 06:43:12 +08:00
|
|
|
/// Current block predicate - null if the block does not need a predicate.
|
|
|
|
VPValue *Predicate = nullptr;
|
|
|
|
|
2020-03-03 18:28:53 +08:00
|
|
|
/// VPlan containing the block. Can only be set on the entry block of the
|
|
|
|
/// plan.
|
|
|
|
VPlan *Plan = nullptr;
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
/// Add \p Successor as the last successor to this block.
|
|
|
|
void appendSuccessor(VPBlockBase *Successor) {
|
|
|
|
assert(Successor && "Cannot add nullptr successor!");
|
|
|
|
Successors.push_back(Successor);
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Add \p Predecessor as the last predecessor to this block.
|
|
|
|
void appendPredecessor(VPBlockBase *Predecessor) {
|
|
|
|
assert(Predecessor && "Cannot add nullptr predecessor!");
|
|
|
|
Predecessors.push_back(Predecessor);
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Remove \p Predecessor from the predecessors of this block.
|
|
|
|
void removePredecessor(VPBlockBase *Predecessor) {
|
|
|
|
auto Pos = std::find(Predecessors.begin(), Predecessors.end(), Predecessor);
|
|
|
|
assert(Pos && "Predecessor does not exist");
|
|
|
|
Predecessors.erase(Pos);
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Remove \p Successor from the successors of this block.
|
|
|
|
void removeSuccessor(VPBlockBase *Successor) {
|
|
|
|
auto Pos = std::find(Successors.begin(), Successors.end(), Successor);
|
|
|
|
assert(Pos && "Successor does not exist");
|
|
|
|
Successors.erase(Pos);
|
|
|
|
}
|
|
|
|
|
|
|
|
protected:
|
|
|
|
VPBlockBase(const unsigned char SC, const std::string &N)
|
2017-10-18 05:27:42 +08:00
|
|
|
: SubclassID(SC), Name(N) {}
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
public:
|
|
|
|
/// An enumeration for keeping track of the concrete subclass of VPBlockBase
|
|
|
|
/// that are actually instantiated. Values of this enumeration are kept in the
|
|
|
|
/// SubclassID field of the VPBlockBase objects. They are used for concrete
|
|
|
|
/// type identification.
|
2017-10-18 05:27:42 +08:00
|
|
|
using VPBlockTy = enum { VPBasicBlockSC, VPRegionBlockSC };
|
2017-08-27 20:55:46 +08:00
|
|
|
|
2017-10-18 05:27:42 +08:00
|
|
|
using VPBlocksTy = SmallVectorImpl<VPBlockBase *>;
|
2017-08-27 20:55:46 +08:00
|
|
|
|
2017-10-18 05:27:42 +08:00
|
|
|
virtual ~VPBlockBase() = default;
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
const std::string &getName() const { return Name; }
|
|
|
|
|
|
|
|
void setName(const Twine &newName) { Name = newName.str(); }
|
|
|
|
|
|
|
|
/// \return an ID for the concrete type of this object.
|
|
|
|
/// This is used to implement the classof checks. This should not be used
|
|
|
|
/// for any other purpose, as the values may change as LLVM evolves.
|
|
|
|
unsigned getVPBlockID() const { return SubclassID; }
|
|
|
|
|
2018-05-22 02:14:23 +08:00
|
|
|
VPRegionBlock *getParent() { return Parent; }
|
2017-08-27 20:55:46 +08:00
|
|
|
const VPRegionBlock *getParent() const { return Parent; }
|
|
|
|
|
2020-03-03 18:28:53 +08:00
|
|
|
/// \return A pointer to the plan containing the current block.
|
|
|
|
VPlan *getPlan();
|
|
|
|
const VPlan *getPlan() const;
|
|
|
|
|
|
|
|
/// Sets the pointer of the plan containing the block. The block must be the
|
|
|
|
/// entry block into the VPlan.
|
|
|
|
void setPlan(VPlan *ParentPlan);
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
void setParent(VPRegionBlock *P) { Parent = P; }
|
|
|
|
|
|
|
|
/// \return the VPBasicBlock that is the entry of this VPBlockBase,
|
|
|
|
/// recursively, if the latter is a VPRegionBlock. Otherwise, if this
|
|
|
|
/// VPBlockBase is a VPBasicBlock, it is returned.
|
|
|
|
const VPBasicBlock *getEntryBasicBlock() const;
|
|
|
|
VPBasicBlock *getEntryBasicBlock();
|
|
|
|
|
|
|
|
/// \return the VPBasicBlock that is the exit of this VPBlockBase,
|
|
|
|
/// recursively, if the latter is a VPRegionBlock. Otherwise, if this
|
|
|
|
/// VPBlockBase is a VPBasicBlock, it is returned.
|
|
|
|
const VPBasicBlock *getExitBasicBlock() const;
|
|
|
|
VPBasicBlock *getExitBasicBlock();
|
|
|
|
|
|
|
|
const VPBlocksTy &getSuccessors() const { return Successors; }
|
|
|
|
VPBlocksTy &getSuccessors() { return Successors; }
|
|
|
|
|
|
|
|
const VPBlocksTy &getPredecessors() const { return Predecessors; }
|
|
|
|
VPBlocksTy &getPredecessors() { return Predecessors; }
|
|
|
|
|
|
|
|
/// \return the successor of this VPBlockBase if it has a single successor.
|
|
|
|
/// Otherwise return a null pointer.
|
|
|
|
VPBlockBase *getSingleSuccessor() const {
|
|
|
|
return (Successors.size() == 1 ? *Successors.begin() : nullptr);
|
|
|
|
}
|
|
|
|
|
|
|
|
/// \return the predecessor of this VPBlockBase if it has a single
|
|
|
|
/// predecessor. Otherwise return a null pointer.
|
|
|
|
VPBlockBase *getSinglePredecessor() const {
|
|
|
|
return (Predecessors.size() == 1 ? *Predecessors.begin() : nullptr);
|
|
|
|
}
|
|
|
|
|
2018-05-22 02:14:23 +08:00
|
|
|
size_t getNumSuccessors() const { return Successors.size(); }
|
|
|
|
size_t getNumPredecessors() const { return Predecessors.size(); }
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
/// An Enclosing Block of a block B is any block containing B, including B
|
|
|
|
/// itself. \return the closest enclosing block starting from "this", which
|
|
|
|
/// has successors. \return the root enclosing block if all enclosing blocks
|
|
|
|
/// have no successors.
|
|
|
|
VPBlockBase *getEnclosingBlockWithSuccessors();
|
|
|
|
|
|
|
|
/// \return the closest enclosing block starting from "this", which has
|
|
|
|
/// predecessors. \return the root enclosing block if all enclosing blocks
|
|
|
|
/// have no predecessors.
|
|
|
|
VPBlockBase *getEnclosingBlockWithPredecessors();
|
|
|
|
|
|
|
|
/// \return the successors either attached directly to this VPBlockBase or, if
|
|
|
|
/// this VPBlockBase is the exit block of a VPRegionBlock and has no
|
|
|
|
/// successors of its own, search recursively for the first enclosing
|
|
|
|
/// VPRegionBlock that has successors and return them. If no such
|
|
|
|
/// VPRegionBlock exists, return the (empty) successors of the topmost
|
|
|
|
/// VPBlockBase reached.
|
|
|
|
const VPBlocksTy &getHierarchicalSuccessors() {
|
|
|
|
return getEnclosingBlockWithSuccessors()->getSuccessors();
|
|
|
|
}
|
|
|
|
|
|
|
|
/// \return the hierarchical successor of this VPBlockBase if it has a single
|
|
|
|
/// hierarchical successor. Otherwise return a null pointer.
|
|
|
|
VPBlockBase *getSingleHierarchicalSuccessor() {
|
|
|
|
return getEnclosingBlockWithSuccessors()->getSingleSuccessor();
|
|
|
|
}
|
|
|
|
|
|
|
|
/// \return the predecessors either attached directly to this VPBlockBase or,
|
|
|
|
/// if this VPBlockBase is the entry block of a VPRegionBlock and has no
|
|
|
|
/// predecessors of its own, search recursively for the first enclosing
|
|
|
|
/// VPRegionBlock that has predecessors and return them. If no such
|
|
|
|
/// VPRegionBlock exists, return the (empty) predecessors of the topmost
|
|
|
|
/// VPBlockBase reached.
|
|
|
|
const VPBlocksTy &getHierarchicalPredecessors() {
|
|
|
|
return getEnclosingBlockWithPredecessors()->getPredecessors();
|
|
|
|
}
|
|
|
|
|
|
|
|
/// \return the hierarchical predecessor of this VPBlockBase if it has a
|
|
|
|
/// single hierarchical predecessor. Otherwise return a null pointer.
|
|
|
|
VPBlockBase *getSingleHierarchicalPredecessor() {
|
|
|
|
return getEnclosingBlockWithPredecessors()->getSinglePredecessor();
|
|
|
|
}
|
|
|
|
|
2018-07-09 23:57:09 +08:00
|
|
|
/// \return the condition bit selecting the successor.
|
|
|
|
VPValue *getCondBit() { return CondBit; }
|
|
|
|
|
|
|
|
const VPValue *getCondBit() const { return CondBit; }
|
|
|
|
|
|
|
|
void setCondBit(VPValue *CV) { CondBit = CV; }
|
|
|
|
|
2019-01-24 06:43:12 +08:00
|
|
|
VPValue *getPredicate() { return Predicate; }
|
|
|
|
|
|
|
|
const VPValue *getPredicate() const { return Predicate; }
|
|
|
|
|
|
|
|
void setPredicate(VPValue *Pred) { Predicate = Pred; }
|
|
|
|
|
2018-05-22 02:14:23 +08:00
|
|
|
/// Set a given VPBlockBase \p Successor as the single successor of this
|
|
|
|
/// VPBlockBase. This VPBlockBase is not added as predecessor of \p Successor.
|
|
|
|
/// This VPBlockBase must have no successors.
|
|
|
|
void setOneSuccessor(VPBlockBase *Successor) {
|
2017-08-27 20:55:46 +08:00
|
|
|
assert(Successors.empty() && "Setting one successor when others exist.");
|
|
|
|
appendSuccessor(Successor);
|
|
|
|
}
|
|
|
|
|
2018-05-22 02:14:23 +08:00
|
|
|
/// Set two given VPBlockBases \p IfTrue and \p IfFalse to be the two
|
2018-07-09 23:57:09 +08:00
|
|
|
/// successors of this VPBlockBase. \p Condition is set as the successor
|
|
|
|
/// selector. This VPBlockBase is not added as predecessor of \p IfTrue or \p
|
|
|
|
/// IfFalse. This VPBlockBase must have no successors.
|
|
|
|
void setTwoSuccessors(VPBlockBase *IfTrue, VPBlockBase *IfFalse,
|
|
|
|
VPValue *Condition) {
|
2017-08-27 20:55:46 +08:00
|
|
|
assert(Successors.empty() && "Setting two successors when others exist.");
|
2018-07-09 23:57:09 +08:00
|
|
|
assert(Condition && "Setting two successors without condition!");
|
|
|
|
CondBit = Condition;
|
2017-08-27 20:55:46 +08:00
|
|
|
appendSuccessor(IfTrue);
|
|
|
|
appendSuccessor(IfFalse);
|
|
|
|
}
|
|
|
|
|
2018-05-22 02:14:23 +08:00
|
|
|
/// Set each VPBasicBlock in \p NewPreds as predecessor of this VPBlockBase.
|
|
|
|
/// This VPBlockBase must have no predecessors. This VPBlockBase is not added
|
|
|
|
/// as successor of any VPBasicBlock in \p NewPreds.
|
|
|
|
void setPredecessors(ArrayRef<VPBlockBase *> NewPreds) {
|
|
|
|
assert(Predecessors.empty() && "Block predecessors already set.");
|
|
|
|
for (auto *Pred : NewPreds)
|
|
|
|
appendPredecessor(Pred);
|
2017-08-27 20:55:46 +08:00
|
|
|
}
|
|
|
|
|
2019-01-24 06:43:12 +08:00
|
|
|
/// Remove all the predecessor of this block.
|
|
|
|
void clearPredecessors() { Predecessors.clear(); }
|
|
|
|
|
|
|
|
/// Remove all the successors of this block and set to null its condition bit
|
|
|
|
void clearSuccessors() {
|
|
|
|
Successors.clear();
|
|
|
|
CondBit = nullptr;
|
|
|
|
}
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
/// The method which generates the output IR that correspond to this
|
|
|
|
/// VPBlockBase, thereby "executing" the VPlan.
|
|
|
|
virtual void execute(struct VPTransformState *State) = 0;
|
|
|
|
|
|
|
|
/// Delete all blocks reachable from a given VPBlockBase, inclusive.
|
|
|
|
static void deleteCFG(VPBlockBase *Entry);
|
2018-07-31 05:33:31 +08:00
|
|
|
|
|
|
|
void printAsOperand(raw_ostream &OS, bool PrintType) const {
|
|
|
|
OS << getName();
|
|
|
|
}
|
|
|
|
|
|
|
|
void print(raw_ostream &OS) const {
|
|
|
|
// TODO: Only printing VPBB name for now since we only have dot printing
|
|
|
|
// support for VPInstructions/Recipes.
|
|
|
|
printAsOperand(OS, false);
|
|
|
|
}
|
2018-07-31 09:57:29 +08:00
|
|
|
|
|
|
|
/// Return true if it is legal to hoist instructions into this block.
|
|
|
|
bool isLegalToHoistInto() {
|
|
|
|
// There are currently no constraints that prevent an instruction to be
|
|
|
|
// hoisted into a VPBlockBase.
|
|
|
|
return true;
|
|
|
|
}
|
2017-08-27 20:55:46 +08:00
|
|
|
};
|
|
|
|
|
|
|
|
/// VPRecipeBase is a base class modeling a sequence of one or more output IR
|
|
|
|
/// instructions.
|
|
|
|
class VPRecipeBase : public ilist_node_with_parent<VPRecipeBase, VPBasicBlock> {
|
|
|
|
friend VPBasicBlock;
|
2019-10-07 22:24:33 +08:00
|
|
|
friend class VPBlockUtils;
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
const unsigned char SubclassID; ///< Subclass identifier (for isa/dyn_cast).
|
|
|
|
|
|
|
|
/// Each VPRecipe belongs to a single VPBasicBlock.
|
2017-10-18 05:27:42 +08:00
|
|
|
VPBasicBlock *Parent = nullptr;
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
public:
|
|
|
|
/// An enumeration for keeping track of the concrete subclass of VPRecipeBase
|
|
|
|
/// that is actually instantiated. Values of this enumeration are kept in the
|
|
|
|
/// SubclassID field of the VPRecipeBase objects. They are used for concrete
|
|
|
|
/// type identification.
|
2017-10-18 05:27:42 +08:00
|
|
|
using VPRecipeTy = enum {
|
2017-11-14 20:09:30 +08:00
|
|
|
VPBlendSC,
|
2017-08-27 20:55:46 +08:00
|
|
|
VPBranchOnMaskSC,
|
2017-11-20 20:01:47 +08:00
|
|
|
VPInstructionSC,
|
2017-08-27 20:55:46 +08:00
|
|
|
VPInterleaveSC,
|
|
|
|
VPPredInstPHISC,
|
2020-08-06 17:10:50 +08:00
|
|
|
VPReductionSC,
|
2017-08-27 20:55:46 +08:00
|
|
|
VPReplicateSC,
|
2020-04-06 22:59:45 +08:00
|
|
|
VPWidenCallSC,
|
2020-04-07 06:53:59 +08:00
|
|
|
VPWidenCanonicalIVSC,
|
2019-10-17 02:39:53 +08:00
|
|
|
VPWidenGEPSC,
|
2017-08-27 20:55:46 +08:00
|
|
|
VPWidenIntOrFpInductionSC,
|
2017-11-14 20:09:30 +08:00
|
|
|
VPWidenMemoryInstructionSC,
|
2017-08-27 20:55:46 +08:00
|
|
|
VPWidenPHISC,
|
|
|
|
VPWidenSC,
|
2020-04-13 15:28:26 +08:00
|
|
|
VPWidenSelectSC
|
2017-10-18 05:27:42 +08:00
|
|
|
};
|
2017-08-27 20:55:46 +08:00
|
|
|
|
2017-10-18 05:27:42 +08:00
|
|
|
VPRecipeBase(const unsigned char SC) : SubclassID(SC) {}
|
|
|
|
virtual ~VPRecipeBase() = default;
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
/// \return an ID for the concrete type of this object.
|
|
|
|
/// This is used to implement the classof checks. This should not be used
|
|
|
|
/// for any other purpose, as the values may change as LLVM evolves.
|
|
|
|
unsigned getVPRecipeID() const { return SubclassID; }
|
|
|
|
|
|
|
|
/// \return the VPBasicBlock which this VPRecipe belongs to.
|
|
|
|
VPBasicBlock *getParent() { return Parent; }
|
|
|
|
const VPBasicBlock *getParent() const { return Parent; }
|
|
|
|
|
|
|
|
/// The method which generates the output IR instructions that correspond to
|
|
|
|
/// this VPRecipe, thereby "executing" the VPlan.
|
|
|
|
virtual void execute(struct VPTransformState &State) = 0;
|
|
|
|
|
|
|
|
/// Each recipe prints itself.
|
2020-03-05 22:55:08 +08:00
|
|
|
virtual void print(raw_ostream &O, const Twine &Indent,
|
|
|
|
VPSlotTracker &SlotTracker) const = 0;
|
2018-06-18 19:34:17 +08:00
|
|
|
|
|
|
|
/// Insert an unlinked recipe into a basic block immediately before
|
|
|
|
/// the specified recipe.
|
|
|
|
void insertBefore(VPRecipeBase *InsertPos);
|
2018-06-18 23:18:48 +08:00
|
|
|
|
2019-10-07 22:24:33 +08:00
|
|
|
/// Insert an unlinked Recipe into a basic block immediately after
|
|
|
|
/// the specified Recipe.
|
|
|
|
void insertAfter(VPRecipeBase *InsertPos);
|
|
|
|
|
2019-10-11 23:36:55 +08:00
|
|
|
/// Unlink this recipe from its current VPBasicBlock and insert it into
|
|
|
|
/// the VPBasicBlock that MovePos lives in, right after MovePos.
|
|
|
|
void moveAfter(VPRecipeBase *MovePos);
|
|
|
|
|
2019-10-07 22:24:33 +08:00
|
|
|
/// This method unlinks 'this' from the containing basic block, but does not
|
|
|
|
/// delete it.
|
|
|
|
void removeFromParent();
|
|
|
|
|
2018-06-18 23:18:48 +08:00
|
|
|
/// This method unlinks 'this' from the containing basic block and deletes it.
|
|
|
|
///
|
|
|
|
/// \returns an iterator pointing to the element after the erased one
|
|
|
|
iplist<VPRecipeBase>::iterator eraseFromParent();
|
2017-08-27 20:55:46 +08:00
|
|
|
};
|
|
|
|
|
2017-11-20 20:01:47 +08:00
|
|
|
/// This is a concrete Recipe that models a single VPlan-level instruction.
|
|
|
|
/// While as any Recipe it may generate a sequence of IR instructions when
|
|
|
|
/// executed, these instructions would always form a single-def expression as
|
|
|
|
/// the VPInstruction is also a single def-use vertex.
|
|
|
|
class VPInstruction : public VPUser, public VPRecipeBase {
|
[VPlan, SLP] Add simple SLP analysis on top of VPlan.
This patch adds an initial implementation of the look-ahead SLP tree
construction described in 'Look-Ahead SLP: Auto-vectorization in the Presence
of Commutative Operations, CGO 2018 by Vasileios Porpodas, Rodrigo C. O. Rocha,
Luís F. W. Góes'.
It returns an SLP tree represented as VPInstructions, with combined
instructions represented as a single, wider VPInstruction.
This initial version does not support instructions with multiple
different users (either inside or outside the SLP tree) or
non-instruction operands; it won't generate any shuffles or
insertelement instructions.
It also just adds the analysis that builds an SLP tree rooted in a set
of stores. It does not include any cost modeling or memory legality
checks. The plan is to integrate it with VPlan based cost modeling, once
available and to only apply it to operations that can be widened.
A follow-up patch will add a support for replacing instructions in a
VPlan with their SLP counter parts.
Reviewers: Ayal, mssimpso, rengolin, mkuper, hfinkel, hsaito, dcaballe, vporpo, RKSimon, ABataev
Reviewed By: rengolin
Differential Revision: https://reviews.llvm.org/D4949
llvm-svn: 346857
2018-11-14 21:11:49 +08:00
|
|
|
friend class VPlanSlp;
|
2018-06-19 02:28:49 +08:00
|
|
|
|
2017-11-20 20:01:47 +08:00
|
|
|
public:
|
|
|
|
/// VPlan opcodes, extending LLVM IR with idiomatics instructions.
|
[VPlan, SLP] Add simple SLP analysis on top of VPlan.
This patch adds an initial implementation of the look-ahead SLP tree
construction described in 'Look-Ahead SLP: Auto-vectorization in the Presence
of Commutative Operations, CGO 2018 by Vasileios Porpodas, Rodrigo C. O. Rocha,
Luís F. W. Góes'.
It returns an SLP tree represented as VPInstructions, with combined
instructions represented as a single, wider VPInstruction.
This initial version does not support instructions with multiple
different users (either inside or outside the SLP tree) or
non-instruction operands; it won't generate any shuffles or
insertelement instructions.
It also just adds the analysis that builds an SLP tree rooted in a set
of stores. It does not include any cost modeling or memory legality
checks. The plan is to integrate it with VPlan based cost modeling, once
available and to only apply it to operations that can be widened.
A follow-up patch will add a support for replacing instructions in a
VPlan with their SLP counter parts.
Reviewers: Ayal, mssimpso, rengolin, mkuper, hfinkel, hsaito, dcaballe, vporpo, RKSimon, ABataev
Reviewed By: rengolin
Differential Revision: https://reviews.llvm.org/D4949
llvm-svn: 346857
2018-11-14 21:11:49 +08:00
|
|
|
enum {
|
|
|
|
Not = Instruction::OtherOpsEnd + 1,
|
|
|
|
ICmpULE,
|
|
|
|
SLPLoad,
|
|
|
|
SLPStore,
|
2020-06-17 17:48:20 +08:00
|
|
|
ActiveLaneMask,
|
[VPlan, SLP] Add simple SLP analysis on top of VPlan.
This patch adds an initial implementation of the look-ahead SLP tree
construction described in 'Look-Ahead SLP: Auto-vectorization in the Presence
of Commutative Operations, CGO 2018 by Vasileios Porpodas, Rodrigo C. O. Rocha,
Luís F. W. Góes'.
It returns an SLP tree represented as VPInstructions, with combined
instructions represented as a single, wider VPInstruction.
This initial version does not support instructions with multiple
different users (either inside or outside the SLP tree) or
non-instruction operands; it won't generate any shuffles or
insertelement instructions.
It also just adds the analysis that builds an SLP tree rooted in a set
of stores. It does not include any cost modeling or memory legality
checks. The plan is to integrate it with VPlan based cost modeling, once
available and to only apply it to operations that can be widened.
A follow-up patch will add a support for replacing instructions in a
VPlan with their SLP counter parts.
Reviewers: Ayal, mssimpso, rengolin, mkuper, hfinkel, hsaito, dcaballe, vporpo, RKSimon, ABataev
Reviewed By: rengolin
Differential Revision: https://reviews.llvm.org/D4949
llvm-svn: 346857
2018-11-14 21:11:49 +08:00
|
|
|
};
|
2017-11-20 20:01:47 +08:00
|
|
|
|
|
|
|
private:
|
|
|
|
typedef unsigned char OpcodeTy;
|
|
|
|
OpcodeTy Opcode;
|
|
|
|
|
|
|
|
/// Utility method serving execute(): generates a single instance of the
|
|
|
|
/// modeled instruction.
|
|
|
|
void generateInstruction(VPTransformState &State, unsigned Part);
|
|
|
|
|
[VPlan, SLP] Add simple SLP analysis on top of VPlan.
This patch adds an initial implementation of the look-ahead SLP tree
construction described in 'Look-Ahead SLP: Auto-vectorization in the Presence
of Commutative Operations, CGO 2018 by Vasileios Porpodas, Rodrigo C. O. Rocha,
Luís F. W. Góes'.
It returns an SLP tree represented as VPInstructions, with combined
instructions represented as a single, wider VPInstruction.
This initial version does not support instructions with multiple
different users (either inside or outside the SLP tree) or
non-instruction operands; it won't generate any shuffles or
insertelement instructions.
It also just adds the analysis that builds an SLP tree rooted in a set
of stores. It does not include any cost modeling or memory legality
checks. The plan is to integrate it with VPlan based cost modeling, once
available and to only apply it to operations that can be widened.
A follow-up patch will add a support for replacing instructions in a
VPlan with their SLP counter parts.
Reviewers: Ayal, mssimpso, rengolin, mkuper, hfinkel, hsaito, dcaballe, vporpo, RKSimon, ABataev
Reviewed By: rengolin
Differential Revision: https://reviews.llvm.org/D4949
llvm-svn: 346857
2018-11-14 21:11:49 +08:00
|
|
|
protected:
|
|
|
|
Instruction *getUnderlyingInstr() {
|
|
|
|
return cast_or_null<Instruction>(getUnderlyingValue());
|
|
|
|
}
|
|
|
|
|
|
|
|
void setUnderlyingInstr(Instruction *I) { setUnderlyingValue(I); }
|
|
|
|
|
2017-11-20 20:01:47 +08:00
|
|
|
public:
|
2018-05-22 02:14:23 +08:00
|
|
|
VPInstruction(unsigned Opcode, ArrayRef<VPValue *> Operands)
|
2017-11-20 20:01:47 +08:00
|
|
|
: VPUser(VPValue::VPInstructionSC, Operands),
|
|
|
|
VPRecipeBase(VPRecipeBase::VPInstructionSC), Opcode(Opcode) {}
|
|
|
|
|
2018-05-22 02:14:23 +08:00
|
|
|
VPInstruction(unsigned Opcode, std::initializer_list<VPValue *> Operands)
|
|
|
|
: VPInstruction(Opcode, ArrayRef<VPValue *>(Operands)) {}
|
|
|
|
|
2017-11-20 20:01:47 +08:00
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPValue *V) {
|
|
|
|
return V->getVPValueID() == VPValue::VPInstructionSC;
|
|
|
|
}
|
|
|
|
|
[VPlan, SLP] Add simple SLP analysis on top of VPlan.
This patch adds an initial implementation of the look-ahead SLP tree
construction described in 'Look-Ahead SLP: Auto-vectorization in the Presence
of Commutative Operations, CGO 2018 by Vasileios Porpodas, Rodrigo C. O. Rocha,
Luís F. W. Góes'.
It returns an SLP tree represented as VPInstructions, with combined
instructions represented as a single, wider VPInstruction.
This initial version does not support instructions with multiple
different users (either inside or outside the SLP tree) or
non-instruction operands; it won't generate any shuffles or
insertelement instructions.
It also just adds the analysis that builds an SLP tree rooted in a set
of stores. It does not include any cost modeling or memory legality
checks. The plan is to integrate it with VPlan based cost modeling, once
available and to only apply it to operations that can be widened.
A follow-up patch will add a support for replacing instructions in a
VPlan with their SLP counter parts.
Reviewers: Ayal, mssimpso, rengolin, mkuper, hfinkel, hsaito, dcaballe, vporpo, RKSimon, ABataev
Reviewed By: rengolin
Differential Revision: https://reviews.llvm.org/D4949
llvm-svn: 346857
2018-11-14 21:11:49 +08:00
|
|
|
VPInstruction *clone() const {
|
|
|
|
SmallVector<VPValue *, 2> Operands(operands());
|
|
|
|
return new VPInstruction(Opcode, Operands);
|
|
|
|
}
|
|
|
|
|
2017-11-20 20:01:47 +08:00
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPRecipeBase *R) {
|
|
|
|
return R->getVPRecipeID() == VPRecipeBase::VPInstructionSC;
|
|
|
|
}
|
|
|
|
|
|
|
|
unsigned getOpcode() const { return Opcode; }
|
|
|
|
|
|
|
|
/// Generate the instruction.
|
|
|
|
/// TODO: We currently execute only per-part unless a specific instance is
|
|
|
|
/// provided.
|
|
|
|
void execute(VPTransformState &State) override;
|
|
|
|
|
|
|
|
/// Print the Recipe.
|
2020-03-05 22:55:08 +08:00
|
|
|
void print(raw_ostream &O, const Twine &Indent,
|
|
|
|
VPSlotTracker &SlotTracker) const override;
|
2017-11-20 20:01:47 +08:00
|
|
|
|
|
|
|
/// Print the VPInstruction.
|
|
|
|
void print(raw_ostream &O) const;
|
2020-03-05 22:55:08 +08:00
|
|
|
void print(raw_ostream &O, VPSlotTracker &SlotTracker) const;
|
[VPlan, SLP] Add simple SLP analysis on top of VPlan.
This patch adds an initial implementation of the look-ahead SLP tree
construction described in 'Look-Ahead SLP: Auto-vectorization in the Presence
of Commutative Operations, CGO 2018 by Vasileios Porpodas, Rodrigo C. O. Rocha,
Luís F. W. Góes'.
It returns an SLP tree represented as VPInstructions, with combined
instructions represented as a single, wider VPInstruction.
This initial version does not support instructions with multiple
different users (either inside or outside the SLP tree) or
non-instruction operands; it won't generate any shuffles or
insertelement instructions.
It also just adds the analysis that builds an SLP tree rooted in a set
of stores. It does not include any cost modeling or memory legality
checks. The plan is to integrate it with VPlan based cost modeling, once
available and to only apply it to operations that can be widened.
A follow-up patch will add a support for replacing instructions in a
VPlan with their SLP counter parts.
Reviewers: Ayal, mssimpso, rengolin, mkuper, hfinkel, hsaito, dcaballe, vporpo, RKSimon, ABataev
Reviewed By: rengolin
Differential Revision: https://reviews.llvm.org/D4949
llvm-svn: 346857
2018-11-14 21:11:49 +08:00
|
|
|
|
|
|
|
/// Return true if this instruction may modify memory.
|
|
|
|
bool mayWriteToMemory() const {
|
|
|
|
// TODO: we can use attributes of the called function to rule out memory
|
|
|
|
// modifications.
|
|
|
|
return Opcode == Instruction::Store || Opcode == Instruction::Call ||
|
|
|
|
Opcode == Instruction::Invoke || Opcode == SLPStore;
|
|
|
|
}
|
2020-03-19 01:39:11 +08:00
|
|
|
|
|
|
|
bool hasResult() const {
|
|
|
|
// CallInst may or may not have a result, depending on the called function.
|
|
|
|
// Conservatively return calls have results for now.
|
|
|
|
switch (getOpcode()) {
|
|
|
|
case Instruction::Ret:
|
|
|
|
case Instruction::Br:
|
|
|
|
case Instruction::Store:
|
|
|
|
case Instruction::Switch:
|
|
|
|
case Instruction::IndirectBr:
|
|
|
|
case Instruction::Resume:
|
|
|
|
case Instruction::CatchRet:
|
|
|
|
case Instruction::Unreachable:
|
|
|
|
case Instruction::Fence:
|
|
|
|
case Instruction::AtomicRMW:
|
|
|
|
return false;
|
|
|
|
default:
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
}
|
2017-11-20 20:01:47 +08:00
|
|
|
};
|
|
|
|
|
2020-03-29 18:35:17 +08:00
|
|
|
/// VPWidenRecipe is a recipe for producing a copy of vector type its
|
|
|
|
/// ingredient. This recipe covers most of the traditional vectorization cases
|
|
|
|
/// where each ingredient transforms into a vectorized version of itself.
|
2017-12-16 09:12:50 +08:00
|
|
|
class VPWidenRecipe : public VPRecipeBase {
|
2020-03-29 18:35:17 +08:00
|
|
|
/// Hold the instruction to be widened.
|
|
|
|
Instruction &Ingredient;
|
2017-12-16 09:12:50 +08:00
|
|
|
|
2020-04-29 17:22:30 +08:00
|
|
|
/// Hold VPValues for the operands of the ingredient.
|
|
|
|
VPUser User;
|
|
|
|
|
2017-12-16 09:12:50 +08:00
|
|
|
public:
|
2020-04-29 17:22:30 +08:00
|
|
|
template <typename IterT>
|
|
|
|
VPWidenRecipe(Instruction &I, iterator_range<IterT> Operands)
|
|
|
|
: VPRecipeBase(VPWidenSC), Ingredient(I), User(Operands) {}
|
2017-12-16 09:12:50 +08:00
|
|
|
|
|
|
|
~VPWidenRecipe() override = default;
|
|
|
|
|
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPRecipeBase *V) {
|
|
|
|
return V->getVPRecipeID() == VPRecipeBase::VPWidenSC;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Produce widened copies of all Ingredients.
|
|
|
|
void execute(VPTransformState &State) override;
|
|
|
|
|
|
|
|
/// Print the recipe.
|
2020-03-05 22:55:08 +08:00
|
|
|
void print(raw_ostream &O, const Twine &Indent,
|
|
|
|
VPSlotTracker &SlotTracker) const override;
|
2017-12-16 09:12:50 +08:00
|
|
|
};
|
|
|
|
|
2020-04-06 22:59:45 +08:00
|
|
|
/// A recipe for widening Call instructions.
|
|
|
|
class VPWidenCallRecipe : public VPRecipeBase {
|
|
|
|
/// Hold the call to be widened.
|
|
|
|
CallInst &Ingredient;
|
|
|
|
|
2020-04-09 17:23:21 +08:00
|
|
|
/// Hold VPValues for the arguments of the call.
|
|
|
|
VPUser User;
|
|
|
|
|
2020-04-06 22:59:45 +08:00
|
|
|
public:
|
2020-04-09 17:23:21 +08:00
|
|
|
template <typename IterT>
|
|
|
|
VPWidenCallRecipe(CallInst &I, iterator_range<IterT> CallArguments)
|
|
|
|
: VPRecipeBase(VPWidenCallSC), Ingredient(I), User(CallArguments) {}
|
2020-04-06 22:59:45 +08:00
|
|
|
|
|
|
|
~VPWidenCallRecipe() override = default;
|
|
|
|
|
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPRecipeBase *V) {
|
|
|
|
return V->getVPRecipeID() == VPRecipeBase::VPWidenCallSC;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Produce a widened version of the call instruction.
|
|
|
|
void execute(VPTransformState &State) override;
|
2020-04-13 15:28:26 +08:00
|
|
|
|
|
|
|
/// Print the recipe.
|
|
|
|
void print(raw_ostream &O, const Twine &Indent,
|
|
|
|
VPSlotTracker &SlotTracker) const override;
|
|
|
|
};
|
|
|
|
|
|
|
|
/// A recipe for widening select instructions.
|
|
|
|
class VPWidenSelectRecipe : public VPRecipeBase {
|
|
|
|
private:
|
|
|
|
/// Hold the select to be widened.
|
|
|
|
SelectInst &Ingredient;
|
|
|
|
|
2020-05-19 22:29:40 +08:00
|
|
|
/// Hold VPValues for the operands of the select.
|
2020-04-13 15:28:26 +08:00
|
|
|
VPUser User;
|
|
|
|
|
2020-05-24 20:36:42 +08:00
|
|
|
/// Is the condition of the select loop invariant?
|
|
|
|
bool InvariantCond;
|
|
|
|
|
2020-04-13 15:28:26 +08:00
|
|
|
public:
|
2020-05-24 20:36:42 +08:00
|
|
|
template <typename IterT>
|
|
|
|
VPWidenSelectRecipe(SelectInst &I, iterator_range<IterT> Operands,
|
|
|
|
bool InvariantCond)
|
|
|
|
: VPRecipeBase(VPWidenSelectSC), Ingredient(I), User(Operands),
|
2020-04-13 15:28:26 +08:00
|
|
|
InvariantCond(InvariantCond) {}
|
|
|
|
|
|
|
|
~VPWidenSelectRecipe() override = default;
|
|
|
|
|
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPRecipeBase *V) {
|
|
|
|
return V->getVPRecipeID() == VPRecipeBase::VPWidenSelectSC;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Produce a widened version of the select instruction.
|
|
|
|
void execute(VPTransformState &State) override;
|
2020-04-06 22:59:45 +08:00
|
|
|
|
|
|
|
/// Print the recipe.
|
|
|
|
void print(raw_ostream &O, const Twine &Indent,
|
|
|
|
VPSlotTracker &SlotTracker) const override;
|
|
|
|
};
|
|
|
|
|
2019-10-17 02:39:53 +08:00
|
|
|
/// A recipe for handling GEP instructions.
|
|
|
|
class VPWidenGEPRecipe : public VPRecipeBase {
|
|
|
|
GetElementPtrInst *GEP;
|
2020-06-27 00:51:51 +08:00
|
|
|
|
|
|
|
/// Hold VPValues for the base and indices of the GEP.
|
|
|
|
VPUser User;
|
|
|
|
|
2019-10-17 02:39:53 +08:00
|
|
|
bool IsPtrLoopInvariant;
|
|
|
|
SmallBitVector IsIndexLoopInvariant;
|
|
|
|
|
|
|
|
public:
|
2020-06-27 00:51:51 +08:00
|
|
|
template <typename IterT>
|
|
|
|
VPWidenGEPRecipe(GetElementPtrInst *GEP, iterator_range<IterT> Operands,
|
|
|
|
Loop *OrigLoop)
|
|
|
|
: VPRecipeBase(VPWidenGEPSC), GEP(GEP), User(Operands),
|
2019-10-17 02:39:53 +08:00
|
|
|
IsIndexLoopInvariant(GEP->getNumIndices(), false) {
|
|
|
|
IsPtrLoopInvariant = OrigLoop->isLoopInvariant(GEP->getPointerOperand());
|
|
|
|
for (auto Index : enumerate(GEP->indices()))
|
|
|
|
IsIndexLoopInvariant[Index.index()] =
|
|
|
|
OrigLoop->isLoopInvariant(Index.value().get());
|
|
|
|
}
|
|
|
|
~VPWidenGEPRecipe() override = default;
|
|
|
|
|
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPRecipeBase *V) {
|
|
|
|
return V->getVPRecipeID() == VPRecipeBase::VPWidenGEPSC;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Generate the gep nodes.
|
|
|
|
void execute(VPTransformState &State) override;
|
|
|
|
|
|
|
|
/// Print the recipe.
|
2020-03-05 22:55:08 +08:00
|
|
|
void print(raw_ostream &O, const Twine &Indent,
|
|
|
|
VPSlotTracker &SlotTracker) const override;
|
2019-10-17 02:39:53 +08:00
|
|
|
};
|
|
|
|
|
2017-12-16 09:12:50 +08:00
|
|
|
/// A recipe for handling phi nodes of integer and floating-point inductions,
|
|
|
|
/// producing their vector and scalar values.
|
|
|
|
class VPWidenIntOrFpInductionRecipe : public VPRecipeBase {
|
|
|
|
PHINode *IV;
|
|
|
|
TruncInst *Trunc;
|
|
|
|
|
|
|
|
public:
|
|
|
|
VPWidenIntOrFpInductionRecipe(PHINode *IV, TruncInst *Trunc = nullptr)
|
|
|
|
: VPRecipeBase(VPWidenIntOrFpInductionSC), IV(IV), Trunc(Trunc) {}
|
|
|
|
~VPWidenIntOrFpInductionRecipe() override = default;
|
|
|
|
|
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPRecipeBase *V) {
|
|
|
|
return V->getVPRecipeID() == VPRecipeBase::VPWidenIntOrFpInductionSC;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Generate the vectorized and scalarized versions of the phi node as
|
|
|
|
/// needed by their users.
|
|
|
|
void execute(VPTransformState &State) override;
|
|
|
|
|
|
|
|
/// Print the recipe.
|
2020-03-05 22:55:08 +08:00
|
|
|
void print(raw_ostream &O, const Twine &Indent,
|
|
|
|
VPSlotTracker &SlotTracker) const override;
|
2017-12-16 09:12:50 +08:00
|
|
|
};
|
|
|
|
|
|
|
|
/// A recipe for handling all phi nodes except for integer and FP inductions.
|
|
|
|
class VPWidenPHIRecipe : public VPRecipeBase {
|
|
|
|
PHINode *Phi;
|
|
|
|
|
|
|
|
public:
|
|
|
|
VPWidenPHIRecipe(PHINode *Phi) : VPRecipeBase(VPWidenPHISC), Phi(Phi) {}
|
|
|
|
~VPWidenPHIRecipe() override = default;
|
|
|
|
|
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPRecipeBase *V) {
|
|
|
|
return V->getVPRecipeID() == VPRecipeBase::VPWidenPHISC;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Generate the phi/select nodes.
|
|
|
|
void execute(VPTransformState &State) override;
|
|
|
|
|
|
|
|
/// Print the recipe.
|
2020-03-05 22:55:08 +08:00
|
|
|
void print(raw_ostream &O, const Twine &Indent,
|
|
|
|
VPSlotTracker &SlotTracker) const override;
|
2017-12-16 09:12:50 +08:00
|
|
|
};
|
|
|
|
|
|
|
|
/// A recipe for vectorizing a phi-node as a sequence of mask-based select
|
|
|
|
/// instructions.
|
|
|
|
class VPBlendRecipe : public VPRecipeBase {
|
|
|
|
PHINode *Phi;
|
|
|
|
|
2020-04-06 19:02:34 +08:00
|
|
|
/// The blend operation is a User of the incoming values and of their
|
|
|
|
/// respective masks, ordered [I0, M0, I1, M1, ...]. Note that a single value
|
2020-04-14 22:20:41 +08:00
|
|
|
/// might be incoming with a full mask for which there is no VPValue.
|
2020-04-06 19:02:34 +08:00
|
|
|
VPUser User;
|
2017-12-16 09:12:50 +08:00
|
|
|
|
|
|
|
public:
|
2020-04-06 19:02:34 +08:00
|
|
|
VPBlendRecipe(PHINode *Phi, ArrayRef<VPValue *> Operands)
|
|
|
|
: VPRecipeBase(VPBlendSC), Phi(Phi), User(Operands) {
|
2020-04-14 22:20:41 +08:00
|
|
|
assert(Operands.size() > 0 &&
|
|
|
|
((Operands.size() == 1) || (Operands.size() % 2 == 0)) &&
|
|
|
|
"Expected either a single incoming value or a positive even number "
|
|
|
|
"of operands");
|
2017-12-16 09:12:50 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPRecipeBase *V) {
|
|
|
|
return V->getVPRecipeID() == VPRecipeBase::VPBlendSC;
|
|
|
|
}
|
|
|
|
|
2020-04-06 19:02:34 +08:00
|
|
|
/// Return the number of incoming values, taking into account that a single
|
|
|
|
/// incoming value has no mask.
|
|
|
|
unsigned getNumIncomingValues() const {
|
|
|
|
return (User.getNumOperands() + 1) / 2;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Return incoming value number \p Idx.
|
|
|
|
VPValue *getIncomingValue(unsigned Idx) const {
|
|
|
|
return User.getOperand(Idx * 2);
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Return mask number \p Idx.
|
|
|
|
VPValue *getMask(unsigned Idx) const { return User.getOperand(Idx * 2 + 1); }
|
|
|
|
|
2017-12-16 09:12:50 +08:00
|
|
|
/// Generate the phi/select nodes.
|
|
|
|
void execute(VPTransformState &State) override;
|
|
|
|
|
|
|
|
/// Print the recipe.
|
2020-03-05 22:55:08 +08:00
|
|
|
void print(raw_ostream &O, const Twine &Indent,
|
|
|
|
VPSlotTracker &SlotTracker) const override;
|
2017-12-16 09:12:50 +08:00
|
|
|
};
|
|
|
|
|
|
|
|
/// VPInterleaveRecipe is a recipe for transforming an interleave group of load
|
|
|
|
/// or stores into one wide load/store and shuffles.
|
|
|
|
class VPInterleaveRecipe : public VPRecipeBase {
|
2018-11-13 23:58:18 +08:00
|
|
|
const InterleaveGroup<Instruction> *IG;
|
2019-11-28 19:54:38 +08:00
|
|
|
VPUser User;
|
2017-12-16 09:12:50 +08:00
|
|
|
|
|
|
|
public:
|
2019-11-28 19:54:38 +08:00
|
|
|
VPInterleaveRecipe(const InterleaveGroup<Instruction> *IG, VPValue *Addr,
|
|
|
|
VPValue *Mask)
|
|
|
|
: VPRecipeBase(VPInterleaveSC), IG(IG), User({Addr}) {
|
|
|
|
if (Mask)
|
|
|
|
User.addOperand(Mask);
|
2018-10-14 16:50:06 +08:00
|
|
|
}
|
2017-12-16 09:12:50 +08:00
|
|
|
~VPInterleaveRecipe() override = default;
|
|
|
|
|
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPRecipeBase *V) {
|
|
|
|
return V->getVPRecipeID() == VPRecipeBase::VPInterleaveSC;
|
|
|
|
}
|
|
|
|
|
2019-11-28 19:54:38 +08:00
|
|
|
/// Return the address accessed by this recipe.
|
|
|
|
VPValue *getAddr() const {
|
|
|
|
return User.getOperand(0); // Address is the 1st, mandatory operand.
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Return the mask used by this recipe. Note that a full mask is represented
|
|
|
|
/// by a nullptr.
|
|
|
|
VPValue *getMask() const {
|
|
|
|
// Mask is optional and therefore the last, currently 2nd operand.
|
|
|
|
return User.getNumOperands() == 2 ? User.getOperand(1) : nullptr;
|
|
|
|
}
|
|
|
|
|
2017-12-16 09:12:50 +08:00
|
|
|
/// Generate the wide load or store, and shuffles.
|
|
|
|
void execute(VPTransformState &State) override;
|
|
|
|
|
|
|
|
/// Print the recipe.
|
2020-03-05 22:55:08 +08:00
|
|
|
void print(raw_ostream &O, const Twine &Indent,
|
|
|
|
VPSlotTracker &SlotTracker) const override;
|
2017-12-16 09:12:50 +08:00
|
|
|
|
2018-11-13 23:58:18 +08:00
|
|
|
const InterleaveGroup<Instruction> *getInterleaveGroup() { return IG; }
|
2017-12-16 09:12:50 +08:00
|
|
|
};
|
|
|
|
|
2020-08-06 17:10:50 +08:00
|
|
|
/// A recipe to represent inloop reduction operations, performing a reduction on
|
|
|
|
/// a vector operand into a scalar value, and adding the result to a chain.
|
|
|
|
class VPReductionRecipe : public VPRecipeBase {
|
|
|
|
/// The recurrence decriptor for the reduction in question.
|
|
|
|
RecurrenceDescriptor *RdxDesc;
|
|
|
|
/// The original instruction being converted to a reduction.
|
|
|
|
Instruction *I;
|
|
|
|
/// The VPValue of the vector value to be reduced.
|
|
|
|
VPValue *VecOp;
|
|
|
|
/// The VPValue of the scalar Chain being accumulated.
|
|
|
|
VPValue *ChainOp;
|
|
|
|
/// Fast math flags to use for the resulting reduction operation.
|
|
|
|
bool NoNaN;
|
|
|
|
/// Pointer to the TTI, needed to create the target reduction
|
|
|
|
const TargetTransformInfo *TTI;
|
|
|
|
|
|
|
|
public:
|
|
|
|
VPReductionRecipe(RecurrenceDescriptor *R, Instruction *I, VPValue *ChainOp,
|
|
|
|
VPValue *VecOp, bool NoNaN, const TargetTransformInfo *TTI)
|
|
|
|
: VPRecipeBase(VPReductionSC), RdxDesc(R), I(I), VecOp(VecOp),
|
|
|
|
ChainOp(ChainOp), NoNaN(NoNaN), TTI(TTI) {}
|
|
|
|
|
|
|
|
~VPReductionRecipe() override = default;
|
|
|
|
|
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPRecipeBase *V) {
|
|
|
|
return V->getVPRecipeID() == VPRecipeBase::VPReductionSC;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Generate the reduction in the loop
|
|
|
|
void execute(VPTransformState &State) override;
|
|
|
|
|
|
|
|
/// Print the recipe.
|
|
|
|
void print(raw_ostream &O, const Twine &Indent,
|
|
|
|
VPSlotTracker &SlotTracker) const override;
|
|
|
|
};
|
|
|
|
|
2017-12-16 09:12:50 +08:00
|
|
|
/// VPReplicateRecipe replicates a given instruction producing multiple scalar
|
|
|
|
/// copies of the original scalar type, one per lane, instead of producing a
|
|
|
|
/// single copy of widened type for all lanes. If the instruction is known to be
|
|
|
|
/// uniform only one copy, per lane zero, will be generated.
|
|
|
|
class VPReplicateRecipe : public VPRecipeBase {
|
|
|
|
/// The instruction being replicated.
|
|
|
|
Instruction *Ingredient;
|
|
|
|
|
2020-05-19 22:04:48 +08:00
|
|
|
/// Hold VPValues for the operands of the ingredient.
|
|
|
|
VPUser User;
|
|
|
|
|
2017-12-16 09:12:50 +08:00
|
|
|
/// Indicator if only a single replica per lane is needed.
|
|
|
|
bool IsUniform;
|
|
|
|
|
|
|
|
/// Indicator if the replicas are also predicated.
|
|
|
|
bool IsPredicated;
|
|
|
|
|
|
|
|
/// Indicator if the scalar values should also be packed into a vector.
|
|
|
|
bool AlsoPack;
|
|
|
|
|
|
|
|
public:
|
2020-05-19 22:04:48 +08:00
|
|
|
template <typename IterT>
|
|
|
|
VPReplicateRecipe(Instruction *I, iterator_range<IterT> Operands,
|
|
|
|
bool IsUniform, bool IsPredicated = false)
|
|
|
|
: VPRecipeBase(VPReplicateSC), Ingredient(I), User(Operands),
|
|
|
|
IsUniform(IsUniform), IsPredicated(IsPredicated) {
|
2017-12-16 09:12:50 +08:00
|
|
|
// Retain the previous behavior of predicateInstructions(), where an
|
|
|
|
// insert-element of a predicated instruction got hoisted into the
|
|
|
|
// predicated basic block iff it was its only user. This is achieved by
|
|
|
|
// having predicated instructions also pack their values into a vector by
|
|
|
|
// default unless they have a replicated user which uses their scalar value.
|
|
|
|
AlsoPack = IsPredicated && !I->use_empty();
|
|
|
|
}
|
|
|
|
|
|
|
|
~VPReplicateRecipe() override = default;
|
|
|
|
|
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPRecipeBase *V) {
|
|
|
|
return V->getVPRecipeID() == VPRecipeBase::VPReplicateSC;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Generate replicas of the desired Ingredient. Replicas will be generated
|
|
|
|
/// for all parts and lanes unless a specific part and lane are specified in
|
|
|
|
/// the \p State.
|
|
|
|
void execute(VPTransformState &State) override;
|
|
|
|
|
|
|
|
void setAlsoPack(bool Pack) { AlsoPack = Pack; }
|
|
|
|
|
|
|
|
/// Print the recipe.
|
2020-03-05 22:55:08 +08:00
|
|
|
void print(raw_ostream &O, const Twine &Indent,
|
|
|
|
VPSlotTracker &SlotTracker) const override;
|
2017-12-16 09:12:50 +08:00
|
|
|
};
|
|
|
|
|
|
|
|
/// A recipe for generating conditional branches on the bits of a mask.
|
|
|
|
class VPBranchOnMaskRecipe : public VPRecipeBase {
|
2020-05-19 21:03:18 +08:00
|
|
|
VPUser User;
|
2017-12-16 09:12:50 +08:00
|
|
|
|
|
|
|
public:
|
|
|
|
VPBranchOnMaskRecipe(VPValue *BlockInMask) : VPRecipeBase(VPBranchOnMaskSC) {
|
|
|
|
if (BlockInMask) // nullptr means all-one mask.
|
2020-05-19 21:03:18 +08:00
|
|
|
User.addOperand(BlockInMask);
|
2017-12-16 09:12:50 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPRecipeBase *V) {
|
|
|
|
return V->getVPRecipeID() == VPRecipeBase::VPBranchOnMaskSC;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Generate the extraction of the appropriate bit from the block mask and the
|
|
|
|
/// conditional branch.
|
|
|
|
void execute(VPTransformState &State) override;
|
|
|
|
|
|
|
|
/// Print the recipe.
|
2020-03-05 22:55:08 +08:00
|
|
|
void print(raw_ostream &O, const Twine &Indent,
|
|
|
|
VPSlotTracker &SlotTracker) const override {
|
2017-12-16 09:12:50 +08:00
|
|
|
O << " +\n" << Indent << "\"BRANCH-ON-MASK ";
|
2020-05-19 21:03:18 +08:00
|
|
|
if (VPValue *Mask = getMask())
|
|
|
|
Mask->print(O, SlotTracker);
|
2017-12-16 09:12:50 +08:00
|
|
|
else
|
|
|
|
O << " All-One";
|
|
|
|
O << "\\l\"";
|
|
|
|
}
|
2020-05-19 21:03:18 +08:00
|
|
|
|
|
|
|
/// Return the mask used by this recipe. Note that a full mask is represented
|
|
|
|
/// by a nullptr.
|
|
|
|
VPValue *getMask() const {
|
|
|
|
assert(User.getNumOperands() <= 1 && "should have either 0 or 1 operands");
|
|
|
|
// Mask is optional.
|
|
|
|
return User.getNumOperands() == 1 ? User.getOperand(0) : nullptr;
|
|
|
|
}
|
2017-12-16 09:12:50 +08:00
|
|
|
};
|
|
|
|
|
|
|
|
/// VPPredInstPHIRecipe is a recipe for generating the phi nodes needed when
|
|
|
|
/// control converges back from a Branch-on-Mask. The phi nodes are needed in
|
|
|
|
/// order to merge values that are set under such a branch and feed their uses.
|
|
|
|
/// The phi nodes can be scalar or vector depending on the users of the value.
|
|
|
|
/// This recipe works in concert with VPBranchOnMaskRecipe.
|
|
|
|
class VPPredInstPHIRecipe : public VPRecipeBase {
|
|
|
|
Instruction *PredInst;
|
|
|
|
|
|
|
|
public:
|
|
|
|
/// Construct a VPPredInstPHIRecipe given \p PredInst whose value needs a phi
|
|
|
|
/// nodes after merging back from a Branch-on-Mask.
|
|
|
|
VPPredInstPHIRecipe(Instruction *PredInst)
|
|
|
|
: VPRecipeBase(VPPredInstPHISC), PredInst(PredInst) {}
|
|
|
|
~VPPredInstPHIRecipe() override = default;
|
|
|
|
|
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPRecipeBase *V) {
|
|
|
|
return V->getVPRecipeID() == VPRecipeBase::VPPredInstPHISC;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Generates phi nodes for live-outs as needed to retain SSA form.
|
|
|
|
void execute(VPTransformState &State) override;
|
|
|
|
|
|
|
|
/// Print the recipe.
|
2020-03-05 22:55:08 +08:00
|
|
|
void print(raw_ostream &O, const Twine &Indent,
|
|
|
|
VPSlotTracker &SlotTracker) const override;
|
2017-12-16 09:12:50 +08:00
|
|
|
};
|
|
|
|
|
|
|
|
/// A Recipe for widening load/store operations.
|
2020-03-19 03:58:13 +08:00
|
|
|
/// The recipe uses the following VPValues:
|
|
|
|
/// - For load: Address, optional mask
|
|
|
|
/// - For store: Address, stored value, optional mask
|
2017-12-16 09:12:50 +08:00
|
|
|
/// TODO: We currently execute only per-part unless a specific instance is
|
|
|
|
/// provided.
|
|
|
|
class VPWidenMemoryInstructionRecipe : public VPRecipeBase {
|
|
|
|
Instruction &Instr;
|
2019-11-28 19:54:38 +08:00
|
|
|
VPUser User;
|
2017-12-16 09:12:50 +08:00
|
|
|
|
2020-03-19 03:58:13 +08:00
|
|
|
void setMask(VPValue *Mask) {
|
|
|
|
if (!Mask)
|
|
|
|
return;
|
|
|
|
User.addOperand(Mask);
|
|
|
|
}
|
|
|
|
|
|
|
|
bool isMasked() const {
|
|
|
|
return (isa<LoadInst>(Instr) && User.getNumOperands() == 2) ||
|
|
|
|
(isa<StoreInst>(Instr) && User.getNumOperands() == 3);
|
|
|
|
}
|
|
|
|
|
2017-12-16 09:12:50 +08:00
|
|
|
public:
|
2020-03-19 03:58:13 +08:00
|
|
|
VPWidenMemoryInstructionRecipe(LoadInst &Load, VPValue *Addr, VPValue *Mask)
|
|
|
|
: VPRecipeBase(VPWidenMemoryInstructionSC), Instr(Load), User({Addr}) {
|
|
|
|
setMask(Mask);
|
|
|
|
}
|
|
|
|
|
|
|
|
VPWidenMemoryInstructionRecipe(StoreInst &Store, VPValue *Addr,
|
|
|
|
VPValue *StoredValue, VPValue *Mask)
|
|
|
|
: VPRecipeBase(VPWidenMemoryInstructionSC), Instr(Store),
|
|
|
|
User({Addr, StoredValue}) {
|
|
|
|
setMask(Mask);
|
2017-12-16 09:12:50 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPRecipeBase *V) {
|
|
|
|
return V->getVPRecipeID() == VPRecipeBase::VPWidenMemoryInstructionSC;
|
|
|
|
}
|
|
|
|
|
2019-11-28 19:54:38 +08:00
|
|
|
/// Return the address accessed by this recipe.
|
|
|
|
VPValue *getAddr() const {
|
|
|
|
return User.getOperand(0); // Address is the 1st, mandatory operand.
|
|
|
|
}
|
|
|
|
|
2019-10-07 22:24:33 +08:00
|
|
|
/// Return the mask used by this recipe. Note that a full mask is represented
|
|
|
|
/// by a nullptr.
|
2019-12-29 01:59:31 +08:00
|
|
|
VPValue *getMask() const {
|
2020-03-19 03:58:13 +08:00
|
|
|
// Mask is optional and therefore the last operand.
|
|
|
|
return isMasked() ? User.getOperand(User.getNumOperands() - 1) : nullptr;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Return the address accessed by this recipe.
|
|
|
|
VPValue *getStoredValue() const {
|
|
|
|
assert(isa<StoreInst>(Instr) &&
|
|
|
|
"Stored value only available for store instructions");
|
|
|
|
return User.getOperand(1); // Stored value is the 2nd, mandatory operand.
|
2019-10-07 22:24:33 +08:00
|
|
|
}
|
|
|
|
|
2017-12-16 09:12:50 +08:00
|
|
|
/// Generate the wide load/store.
|
|
|
|
void execute(VPTransformState &State) override;
|
|
|
|
|
|
|
|
/// Print the recipe.
|
2020-03-05 22:55:08 +08:00
|
|
|
void print(raw_ostream &O, const Twine &Indent,
|
|
|
|
VPSlotTracker &SlotTracker) const override;
|
2017-12-16 09:12:50 +08:00
|
|
|
};
|
|
|
|
|
2020-04-07 06:53:59 +08:00
|
|
|
/// A Recipe for widening the canonical induction variable of the vector loop.
|
|
|
|
class VPWidenCanonicalIVRecipe : public VPRecipeBase {
|
|
|
|
/// A VPValue representing the canonical vector IV.
|
|
|
|
VPValue Val;
|
|
|
|
|
|
|
|
public:
|
|
|
|
VPWidenCanonicalIVRecipe() : VPRecipeBase(VPWidenCanonicalIVSC) {}
|
|
|
|
~VPWidenCanonicalIVRecipe() override = default;
|
|
|
|
|
|
|
|
/// Return the VPValue representing the canonical vector induction variable of
|
|
|
|
/// the vector loop.
|
|
|
|
const VPValue *getVPValue() const { return &Val; }
|
|
|
|
VPValue *getVPValue() { return &Val; }
|
|
|
|
|
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPRecipeBase *V) {
|
|
|
|
return V->getVPRecipeID() == VPRecipeBase::VPWidenCanonicalIVSC;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Generate a canonical vector induction variable of the vector loop, with
|
|
|
|
/// start = {<Part*VF, Part*VF+1, ..., Part*VF+VF-1> for 0 <= Part < UF}, and
|
|
|
|
/// step = <VF*UF, VF*UF, ..., VF*UF>.
|
|
|
|
void execute(VPTransformState &State) override;
|
|
|
|
|
|
|
|
/// Print the recipe.
|
|
|
|
void print(raw_ostream &O, const Twine &Indent,
|
|
|
|
VPSlotTracker &SlotTracker) const override;
|
|
|
|
};
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
/// VPBasicBlock serves as the leaf of the Hierarchical Control-Flow Graph. It
|
|
|
|
/// holds a sequence of zero or more VPRecipe's each representing a sequence of
|
|
|
|
/// output IR instructions.
|
|
|
|
class VPBasicBlock : public VPBlockBase {
|
|
|
|
public:
|
2017-10-18 05:27:42 +08:00
|
|
|
using RecipeListTy = iplist<VPRecipeBase>;
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
private:
|
|
|
|
/// The VPRecipes held in the order of output instructions to generate.
|
|
|
|
RecipeListTy Recipes;
|
|
|
|
|
|
|
|
public:
|
2017-10-18 05:27:42 +08:00
|
|
|
VPBasicBlock(const Twine &Name = "", VPRecipeBase *Recipe = nullptr)
|
|
|
|
: VPBlockBase(VPBasicBlockSC, Name.str()) {
|
|
|
|
if (Recipe)
|
|
|
|
appendRecipe(Recipe);
|
|
|
|
}
|
|
|
|
|
|
|
|
~VPBasicBlock() override { Recipes.clear(); }
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
/// Instruction iterators...
|
2017-10-18 05:27:42 +08:00
|
|
|
using iterator = RecipeListTy::iterator;
|
|
|
|
using const_iterator = RecipeListTy::const_iterator;
|
|
|
|
using reverse_iterator = RecipeListTy::reverse_iterator;
|
|
|
|
using const_reverse_iterator = RecipeListTy::const_reverse_iterator;
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
//===--------------------------------------------------------------------===//
|
|
|
|
/// Recipe iterator methods
|
|
|
|
///
|
|
|
|
inline iterator begin() { return Recipes.begin(); }
|
|
|
|
inline const_iterator begin() const { return Recipes.begin(); }
|
|
|
|
inline iterator end() { return Recipes.end(); }
|
|
|
|
inline const_iterator end() const { return Recipes.end(); }
|
|
|
|
|
|
|
|
inline reverse_iterator rbegin() { return Recipes.rbegin(); }
|
|
|
|
inline const_reverse_iterator rbegin() const { return Recipes.rbegin(); }
|
|
|
|
inline reverse_iterator rend() { return Recipes.rend(); }
|
|
|
|
inline const_reverse_iterator rend() const { return Recipes.rend(); }
|
|
|
|
|
|
|
|
inline size_t size() const { return Recipes.size(); }
|
|
|
|
inline bool empty() const { return Recipes.empty(); }
|
|
|
|
inline const VPRecipeBase &front() const { return Recipes.front(); }
|
|
|
|
inline VPRecipeBase &front() { return Recipes.front(); }
|
|
|
|
inline const VPRecipeBase &back() const { return Recipes.back(); }
|
|
|
|
inline VPRecipeBase &back() { return Recipes.back(); }
|
|
|
|
|
2018-06-18 19:34:17 +08:00
|
|
|
/// Returns a reference to the list of recipes.
|
|
|
|
RecipeListTy &getRecipeList() { return Recipes; }
|
|
|
|
|
2018-05-01 23:54:18 +08:00
|
|
|
/// Returns a pointer to a member of the recipe list.
|
2017-08-27 20:55:46 +08:00
|
|
|
static RecipeListTy VPBasicBlock::*getSublistAccess(VPRecipeBase *) {
|
|
|
|
return &VPBasicBlock::Recipes;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPBlockBase *V) {
|
|
|
|
return V->getVPBlockID() == VPBlockBase::VPBasicBlockSC;
|
|
|
|
}
|
|
|
|
|
2017-11-20 20:01:47 +08:00
|
|
|
void insert(VPRecipeBase *Recipe, iterator InsertPt) {
|
2017-08-27 20:55:46 +08:00
|
|
|
assert(Recipe && "No recipe to append.");
|
|
|
|
assert(!Recipe->Parent && "Recipe already in VPlan");
|
|
|
|
Recipe->Parent = this;
|
2017-11-20 20:01:47 +08:00
|
|
|
Recipes.insert(InsertPt, Recipe);
|
2017-08-27 20:55:46 +08:00
|
|
|
}
|
|
|
|
|
2017-11-20 20:01:47 +08:00
|
|
|
/// Augment the existing recipes of a VPBasicBlock with an additional
|
|
|
|
/// \p Recipe as the last recipe.
|
|
|
|
void appendRecipe(VPRecipeBase *Recipe) { insert(Recipe, end()); }
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
/// The method which generates the output IR instructions that correspond to
|
|
|
|
/// this VPBasicBlock, thereby "executing" the VPlan.
|
|
|
|
void execute(struct VPTransformState *State) override;
|
|
|
|
|
|
|
|
private:
|
|
|
|
/// Create an IR BasicBlock to hold the output instructions generated by this
|
|
|
|
/// VPBasicBlock, and return it. Update the CFGState accordingly.
|
|
|
|
BasicBlock *createEmptyBasicBlock(VPTransformState::CFGState &CFG);
|
|
|
|
};
|
|
|
|
|
|
|
|
/// VPRegionBlock represents a collection of VPBasicBlocks and VPRegionBlocks
|
|
|
|
/// which form a Single-Entry-Single-Exit subgraph of the output IR CFG.
|
|
|
|
/// A VPRegionBlock may indicate that its contents are to be replicated several
|
|
|
|
/// times. This is designed to support predicated scalarization, in which a
|
|
|
|
/// scalar if-then code structure needs to be generated VF * UF times. Having
|
|
|
|
/// this replication indicator helps to keep a single model for multiple
|
|
|
|
/// candidate VF's. The actual replication takes place only once the desired VF
|
|
|
|
/// and UF have been determined.
|
|
|
|
class VPRegionBlock : public VPBlockBase {
|
|
|
|
/// Hold the Single Entry of the SESE region modelled by the VPRegionBlock.
|
|
|
|
VPBlockBase *Entry;
|
|
|
|
|
|
|
|
/// Hold the Single Exit of the SESE region modelled by the VPRegionBlock.
|
|
|
|
VPBlockBase *Exit;
|
|
|
|
|
|
|
|
/// An indicator whether this region is to generate multiple replicated
|
|
|
|
/// instances of output IR corresponding to its VPBlockBases.
|
|
|
|
bool IsReplicator;
|
|
|
|
|
|
|
|
public:
|
|
|
|
VPRegionBlock(VPBlockBase *Entry, VPBlockBase *Exit,
|
|
|
|
const std::string &Name = "", bool IsReplicator = false)
|
|
|
|
: VPBlockBase(VPRegionBlockSC, Name), Entry(Entry), Exit(Exit),
|
|
|
|
IsReplicator(IsReplicator) {
|
|
|
|
assert(Entry->getPredecessors().empty() && "Entry block has predecessors.");
|
|
|
|
assert(Exit->getSuccessors().empty() && "Exit block has successors.");
|
|
|
|
Entry->setParent(this);
|
|
|
|
Exit->setParent(this);
|
|
|
|
}
|
2018-05-22 02:14:23 +08:00
|
|
|
VPRegionBlock(const std::string &Name = "", bool IsReplicator = false)
|
|
|
|
: VPBlockBase(VPRegionBlockSC, Name), Entry(nullptr), Exit(nullptr),
|
|
|
|
IsReplicator(IsReplicator) {}
|
2017-08-27 20:55:46 +08:00
|
|
|
|
2017-10-18 05:27:42 +08:00
|
|
|
~VPRegionBlock() override {
|
2017-08-27 20:55:46 +08:00
|
|
|
if (Entry)
|
|
|
|
deleteCFG(Entry);
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Method to support type inquiry through isa, cast, and dyn_cast.
|
|
|
|
static inline bool classof(const VPBlockBase *V) {
|
|
|
|
return V->getVPBlockID() == VPBlockBase::VPRegionBlockSC;
|
|
|
|
}
|
|
|
|
|
|
|
|
const VPBlockBase *getEntry() const { return Entry; }
|
|
|
|
VPBlockBase *getEntry() { return Entry; }
|
|
|
|
|
2018-05-22 02:14:23 +08:00
|
|
|
/// Set \p EntryBlock as the entry VPBlockBase of this VPRegionBlock. \p
|
|
|
|
/// EntryBlock must have no predecessors.
|
|
|
|
void setEntry(VPBlockBase *EntryBlock) {
|
|
|
|
assert(EntryBlock->getPredecessors().empty() &&
|
|
|
|
"Entry block cannot have predecessors.");
|
|
|
|
Entry = EntryBlock;
|
|
|
|
EntryBlock->setParent(this);
|
|
|
|
}
|
|
|
|
|
2018-07-31 05:33:31 +08:00
|
|
|
// FIXME: DominatorTreeBase is doing 'A->getParent()->front()'. 'front' is a
|
|
|
|
// specific interface of llvm::Function, instead of using
|
|
|
|
// GraphTraints::getEntryNode. We should add a new template parameter to
|
|
|
|
// DominatorTreeBase representing the Graph type.
|
|
|
|
VPBlockBase &front() const { return *Entry; }
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
const VPBlockBase *getExit() const { return Exit; }
|
|
|
|
VPBlockBase *getExit() { return Exit; }
|
|
|
|
|
2018-05-22 02:14:23 +08:00
|
|
|
/// Set \p ExitBlock as the exit VPBlockBase of this VPRegionBlock. \p
|
|
|
|
/// ExitBlock must have no successors.
|
|
|
|
void setExit(VPBlockBase *ExitBlock) {
|
|
|
|
assert(ExitBlock->getSuccessors().empty() &&
|
|
|
|
"Exit block cannot have successors.");
|
|
|
|
Exit = ExitBlock;
|
|
|
|
ExitBlock->setParent(this);
|
|
|
|
}
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
/// An indicator whether this region is to generate multiple replicated
|
|
|
|
/// instances of output IR corresponding to its VPBlockBases.
|
|
|
|
bool isReplicator() const { return IsReplicator; }
|
|
|
|
|
|
|
|
/// The method which generates the output IR instructions that correspond to
|
|
|
|
/// this VPRegionBlock, thereby "executing" the VPlan.
|
|
|
|
void execute(struct VPTransformState *State) override;
|
|
|
|
};
|
|
|
|
|
2019-12-03 02:21:07 +08:00
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
// GraphTraits specializations for VPlan Hierarchical Control-Flow Graphs //
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
|
|
|
// The following set of template specializations implement GraphTraits to treat
|
|
|
|
// any VPBlockBase as a node in a graph of VPBlockBases. It's important to note
|
|
|
|
// that VPBlockBase traits don't recurse into VPRegioBlocks, i.e., if the
|
|
|
|
// VPBlockBase is a VPRegionBlock, this specialization provides access to its
|
|
|
|
// successors/predecessors but not to the blocks inside the region.
|
|
|
|
|
|
|
|
template <> struct GraphTraits<VPBlockBase *> {
|
|
|
|
using NodeRef = VPBlockBase *;
|
|
|
|
using ChildIteratorType = SmallVectorImpl<VPBlockBase *>::iterator;
|
|
|
|
|
|
|
|
static NodeRef getEntryNode(NodeRef N) { return N; }
|
|
|
|
|
|
|
|
static inline ChildIteratorType child_begin(NodeRef N) {
|
|
|
|
return N->getSuccessors().begin();
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline ChildIteratorType child_end(NodeRef N) {
|
|
|
|
return N->getSuccessors().end();
|
|
|
|
}
|
|
|
|
};
|
|
|
|
|
|
|
|
template <> struct GraphTraits<const VPBlockBase *> {
|
|
|
|
using NodeRef = const VPBlockBase *;
|
|
|
|
using ChildIteratorType = SmallVectorImpl<VPBlockBase *>::const_iterator;
|
|
|
|
|
|
|
|
static NodeRef getEntryNode(NodeRef N) { return N; }
|
|
|
|
|
|
|
|
static inline ChildIteratorType child_begin(NodeRef N) {
|
|
|
|
return N->getSuccessors().begin();
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline ChildIteratorType child_end(NodeRef N) {
|
|
|
|
return N->getSuccessors().end();
|
|
|
|
}
|
|
|
|
};
|
|
|
|
|
|
|
|
// Inverse order specialization for VPBasicBlocks. Predecessors are used instead
|
|
|
|
// of successors for the inverse traversal.
|
|
|
|
template <> struct GraphTraits<Inverse<VPBlockBase *>> {
|
|
|
|
using NodeRef = VPBlockBase *;
|
|
|
|
using ChildIteratorType = SmallVectorImpl<VPBlockBase *>::iterator;
|
|
|
|
|
|
|
|
static NodeRef getEntryNode(Inverse<NodeRef> B) { return B.Graph; }
|
|
|
|
|
|
|
|
static inline ChildIteratorType child_begin(NodeRef N) {
|
|
|
|
return N->getPredecessors().begin();
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline ChildIteratorType child_end(NodeRef N) {
|
|
|
|
return N->getPredecessors().end();
|
|
|
|
}
|
|
|
|
};
|
|
|
|
|
|
|
|
// The following set of template specializations implement GraphTraits to
|
|
|
|
// treat VPRegionBlock as a graph and recurse inside its nodes. It's important
|
|
|
|
// to note that the blocks inside the VPRegionBlock are treated as VPBlockBases
|
|
|
|
// (i.e., no dyn_cast is performed, VPBlockBases specialization is used), so
|
|
|
|
// there won't be automatic recursion into other VPBlockBases that turn to be
|
|
|
|
// VPRegionBlocks.
|
|
|
|
|
|
|
|
template <>
|
|
|
|
struct GraphTraits<VPRegionBlock *> : public GraphTraits<VPBlockBase *> {
|
|
|
|
using GraphRef = VPRegionBlock *;
|
|
|
|
using nodes_iterator = df_iterator<NodeRef>;
|
|
|
|
|
|
|
|
static NodeRef getEntryNode(GraphRef N) { return N->getEntry(); }
|
|
|
|
|
|
|
|
static nodes_iterator nodes_begin(GraphRef N) {
|
|
|
|
return nodes_iterator::begin(N->getEntry());
|
|
|
|
}
|
|
|
|
|
|
|
|
static nodes_iterator nodes_end(GraphRef N) {
|
|
|
|
// df_iterator::end() returns an empty iterator so the node used doesn't
|
|
|
|
// matter.
|
|
|
|
return nodes_iterator::end(N);
|
|
|
|
}
|
|
|
|
};
|
|
|
|
|
|
|
|
template <>
|
|
|
|
struct GraphTraits<const VPRegionBlock *>
|
|
|
|
: public GraphTraits<const VPBlockBase *> {
|
|
|
|
using GraphRef = const VPRegionBlock *;
|
|
|
|
using nodes_iterator = df_iterator<NodeRef>;
|
|
|
|
|
|
|
|
static NodeRef getEntryNode(GraphRef N) { return N->getEntry(); }
|
|
|
|
|
|
|
|
static nodes_iterator nodes_begin(GraphRef N) {
|
|
|
|
return nodes_iterator::begin(N->getEntry());
|
|
|
|
}
|
|
|
|
|
|
|
|
static nodes_iterator nodes_end(GraphRef N) {
|
|
|
|
// df_iterator::end() returns an empty iterator so the node used doesn't
|
|
|
|
// matter.
|
|
|
|
return nodes_iterator::end(N);
|
|
|
|
}
|
|
|
|
};
|
|
|
|
|
|
|
|
template <>
|
|
|
|
struct GraphTraits<Inverse<VPRegionBlock *>>
|
|
|
|
: public GraphTraits<Inverse<VPBlockBase *>> {
|
|
|
|
using GraphRef = VPRegionBlock *;
|
|
|
|
using nodes_iterator = df_iterator<NodeRef>;
|
|
|
|
|
|
|
|
static NodeRef getEntryNode(Inverse<GraphRef> N) {
|
|
|
|
return N.Graph->getExit();
|
|
|
|
}
|
|
|
|
|
|
|
|
static nodes_iterator nodes_begin(GraphRef N) {
|
|
|
|
return nodes_iterator::begin(N->getExit());
|
|
|
|
}
|
|
|
|
|
|
|
|
static nodes_iterator nodes_end(GraphRef N) {
|
|
|
|
// df_iterator::end() returns an empty iterator so the node used doesn't
|
|
|
|
// matter.
|
|
|
|
return nodes_iterator::end(N);
|
|
|
|
}
|
|
|
|
};
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
/// VPlan models a candidate for vectorization, encoding various decisions take
|
|
|
|
/// to produce efficient output IR, including which branches, basic-blocks and
|
|
|
|
/// output IR instructions to generate, and their cost. VPlan holds a
|
|
|
|
/// Hierarchical-CFG of VPBasicBlocks and VPRegionBlocks rooted at an Entry
|
|
|
|
/// VPBlock.
|
|
|
|
class VPlan {
|
2017-11-20 20:01:47 +08:00
|
|
|
friend class VPlanPrinter;
|
2020-03-05 22:55:08 +08:00
|
|
|
friend class VPSlotTracker;
|
2017-11-20 20:01:47 +08:00
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
/// Hold the single entry to the Hierarchical CFG of the VPlan.
|
|
|
|
VPBlockBase *Entry;
|
|
|
|
|
|
|
|
/// Holds the VFs applicable to this VPlan.
|
[llvm][LV] Replace `unsigned VF` with `ElementCount VF` [NFCI]
Changes:
* Change `ToVectorTy` to deal directly with `ElementCount` instances.
* `VF == 1` replaced with `VF.isScalar()`.
* `VF > 1` and `VF >=2` replaced with `VF.isVector()`.
* `VF <=1` is replaced with `VF.isZero() || VF.isScalar()`.
* Replaced the uses of `llvm::SmallSet<ElementCount, ...>` with
`llvm::SmallSetVector<ElementCount, ...>`. This avoids the need of an
ordering function for the `ElementCount` class.
* Bits and pieces around printing the `ElementCount` to string streams.
To guarantee that this change is a NFC, `VF.Min` and asserts are used
in the following places:
1. When it doesn't make sense to deal with the scalable property, for
example:
a. When computing unrolling factors.
b. When shuffle masks are built for fixed width vector types
In this cases, an
assert(!VF.Scalable && "<mgs>") has been added to make sure we don't
enter coepaths that don't make sense for scalable vectors.
2. When there is a conscious decision to use `FixedVectorType`. These
uses of `FixedVectorType` will likely be removed in favour of
`VectorType` once the vectorizer is generic enough to deal with both
fixed vector types and scalable vector types.
3. When dealing with building constants out of the value of VF, for
example when computing the vectorization `step`, or building vectors
of indices. These operation _make sense_ for scalable vectors too,
but changing the code in these places to be generic and make it work
for scalable vectors is to be submitted in a separate patch, as it is
a functional change.
4. When building the potential VFs in VPlan. Making the VPlan generic
enough to handle scalable vectorization factors is a functional change
that needs a separate patch. See for example `void
LoopVectorizationPlanner::buildVPlans(unsigned MinVF, unsigned
MaxVF)`.
5. The class `IntrinsicCostAttribute`: this class still uses `unsigned
VF` as updating the field to use `ElementCount` woudl require changes
that could result in changing the behavior of the compiler. Will be done
in a separate patch.
7. When dealing with user input for forcing the vectorization
factor. In this case, adding support for scalable vectorization is a
functional change that migh require changes at command line.
Note that in some places the idiom
```
unsigned VF = ...
auto VTy = FixedVectorType::get(ScalarTy, VF)
```
has been replaced with
```
ElementCount VF = ...
assert(!VF.Scalable && ...);
auto VTy = VectorType::get(ScalarTy, VF)
```
The assertion guarantees that the new code is (at least in debug mode)
functionally equivalent to the old version. Notice that this change had been
possible because none of the methods that are specific to `FixedVectorType`
were used after the instantiation of `VTy`.
Reviewed By: rengolin, ctetreau
Differential Revision: https://reviews.llvm.org/D85794
2020-08-08 06:03:24 +08:00
|
|
|
SmallSetVector<ElementCount, 2> VFs;
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
/// Holds the name of the VPlan, for printing.
|
|
|
|
std::string Name;
|
|
|
|
|
2018-05-22 02:14:23 +08:00
|
|
|
/// Holds all the external definitions created for this VPlan.
|
|
|
|
// TODO: Introduce a specific representation for external definitions in
|
|
|
|
// VPlan. External definitions must be immutable and hold a pointer to its
|
|
|
|
// underlying IR that will be used to implement its structural comparison
|
|
|
|
// (operators '==' and '<').
|
2018-06-09 13:04:20 +08:00
|
|
|
SmallPtrSet<VPValue *, 16> VPExternalDefs;
|
2018-05-22 02:14:23 +08:00
|
|
|
|
2018-10-18 23:03:15 +08:00
|
|
|
/// Represents the backedge taken count of the original loop, for folding
|
|
|
|
/// the tail.
|
|
|
|
VPValue *BackedgeTakenCount = nullptr;
|
|
|
|
|
2017-11-20 20:01:47 +08:00
|
|
|
/// Holds a mapping between Values and their corresponding VPValue inside
|
|
|
|
/// VPlan.
|
|
|
|
Value2VPValueTy Value2VPValue;
|
|
|
|
|
2018-07-31 09:57:29 +08:00
|
|
|
/// Holds the VPLoopInfo analysis for this VPlan.
|
|
|
|
VPLoopInfo VPLInfo;
|
|
|
|
|
2018-09-14 10:02:57 +08:00
|
|
|
/// Holds the condition bit values built during VPInstruction to VPRecipe transformation.
|
|
|
|
SmallVector<VPValue *, 4> VPCBVs;
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
public:
|
2020-03-05 22:55:08 +08:00
|
|
|
VPlan(VPBlockBase *Entry = nullptr) : Entry(Entry) {
|
|
|
|
if (Entry)
|
|
|
|
Entry->setPlan(this);
|
|
|
|
}
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
~VPlan() {
|
|
|
|
if (Entry)
|
|
|
|
VPBlockBase::deleteCFG(Entry);
|
2017-11-20 20:01:47 +08:00
|
|
|
for (auto &MapEntry : Value2VPValue)
|
2020-04-25 08:44:38 +08:00
|
|
|
delete MapEntry.second;
|
2018-10-18 23:03:15 +08:00
|
|
|
if (BackedgeTakenCount)
|
2020-04-25 08:44:38 +08:00
|
|
|
delete BackedgeTakenCount;
|
2018-05-22 02:14:23 +08:00
|
|
|
for (VPValue *Def : VPExternalDefs)
|
|
|
|
delete Def;
|
2018-09-14 10:02:57 +08:00
|
|
|
for (VPValue *CBV : VPCBVs)
|
|
|
|
delete CBV;
|
2017-08-27 20:55:46 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/// Generate the IR code for this VPlan.
|
|
|
|
void execute(struct VPTransformState *State);
|
|
|
|
|
|
|
|
VPBlockBase *getEntry() { return Entry; }
|
|
|
|
const VPBlockBase *getEntry() const { return Entry; }
|
|
|
|
|
2020-03-03 18:28:53 +08:00
|
|
|
VPBlockBase *setEntry(VPBlockBase *Block) {
|
|
|
|
Entry = Block;
|
|
|
|
Block->setPlan(this);
|
|
|
|
return Entry;
|
|
|
|
}
|
2017-08-27 20:55:46 +08:00
|
|
|
|
2018-10-18 23:03:15 +08:00
|
|
|
/// The backedge taken count of the original loop.
|
|
|
|
VPValue *getOrCreateBackedgeTakenCount() {
|
|
|
|
if (!BackedgeTakenCount)
|
|
|
|
BackedgeTakenCount = new VPValue();
|
|
|
|
return BackedgeTakenCount;
|
|
|
|
}
|
|
|
|
|
[llvm][LV] Replace `unsigned VF` with `ElementCount VF` [NFCI]
Changes:
* Change `ToVectorTy` to deal directly with `ElementCount` instances.
* `VF == 1` replaced with `VF.isScalar()`.
* `VF > 1` and `VF >=2` replaced with `VF.isVector()`.
* `VF <=1` is replaced with `VF.isZero() || VF.isScalar()`.
* Replaced the uses of `llvm::SmallSet<ElementCount, ...>` with
`llvm::SmallSetVector<ElementCount, ...>`. This avoids the need of an
ordering function for the `ElementCount` class.
* Bits and pieces around printing the `ElementCount` to string streams.
To guarantee that this change is a NFC, `VF.Min` and asserts are used
in the following places:
1. When it doesn't make sense to deal with the scalable property, for
example:
a. When computing unrolling factors.
b. When shuffle masks are built for fixed width vector types
In this cases, an
assert(!VF.Scalable && "<mgs>") has been added to make sure we don't
enter coepaths that don't make sense for scalable vectors.
2. When there is a conscious decision to use `FixedVectorType`. These
uses of `FixedVectorType` will likely be removed in favour of
`VectorType` once the vectorizer is generic enough to deal with both
fixed vector types and scalable vector types.
3. When dealing with building constants out of the value of VF, for
example when computing the vectorization `step`, or building vectors
of indices. These operation _make sense_ for scalable vectors too,
but changing the code in these places to be generic and make it work
for scalable vectors is to be submitted in a separate patch, as it is
a functional change.
4. When building the potential VFs in VPlan. Making the VPlan generic
enough to handle scalable vectorization factors is a functional change
that needs a separate patch. See for example `void
LoopVectorizationPlanner::buildVPlans(unsigned MinVF, unsigned
MaxVF)`.
5. The class `IntrinsicCostAttribute`: this class still uses `unsigned
VF` as updating the field to use `ElementCount` woudl require changes
that could result in changing the behavior of the compiler. Will be done
in a separate patch.
7. When dealing with user input for forcing the vectorization
factor. In this case, adding support for scalable vectorization is a
functional change that migh require changes at command line.
Note that in some places the idiom
```
unsigned VF = ...
auto VTy = FixedVectorType::get(ScalarTy, VF)
```
has been replaced with
```
ElementCount VF = ...
assert(!VF.Scalable && ...);
auto VTy = VectorType::get(ScalarTy, VF)
```
The assertion guarantees that the new code is (at least in debug mode)
functionally equivalent to the old version. Notice that this change had been
possible because none of the methods that are specific to `FixedVectorType`
were used after the instantiation of `VTy`.
Reviewed By: rengolin, ctetreau
Differential Revision: https://reviews.llvm.org/D85794
2020-08-08 06:03:24 +08:00
|
|
|
void addVF(ElementCount VF) { VFs.insert(VF); }
|
2017-08-27 20:55:46 +08:00
|
|
|
|
[llvm][LV] Replace `unsigned VF` with `ElementCount VF` [NFCI]
Changes:
* Change `ToVectorTy` to deal directly with `ElementCount` instances.
* `VF == 1` replaced with `VF.isScalar()`.
* `VF > 1` and `VF >=2` replaced with `VF.isVector()`.
* `VF <=1` is replaced with `VF.isZero() || VF.isScalar()`.
* Replaced the uses of `llvm::SmallSet<ElementCount, ...>` with
`llvm::SmallSetVector<ElementCount, ...>`. This avoids the need of an
ordering function for the `ElementCount` class.
* Bits and pieces around printing the `ElementCount` to string streams.
To guarantee that this change is a NFC, `VF.Min` and asserts are used
in the following places:
1. When it doesn't make sense to deal with the scalable property, for
example:
a. When computing unrolling factors.
b. When shuffle masks are built for fixed width vector types
In this cases, an
assert(!VF.Scalable && "<mgs>") has been added to make sure we don't
enter coepaths that don't make sense for scalable vectors.
2. When there is a conscious decision to use `FixedVectorType`. These
uses of `FixedVectorType` will likely be removed in favour of
`VectorType` once the vectorizer is generic enough to deal with both
fixed vector types and scalable vector types.
3. When dealing with building constants out of the value of VF, for
example when computing the vectorization `step`, or building vectors
of indices. These operation _make sense_ for scalable vectors too,
but changing the code in these places to be generic and make it work
for scalable vectors is to be submitted in a separate patch, as it is
a functional change.
4. When building the potential VFs in VPlan. Making the VPlan generic
enough to handle scalable vectorization factors is a functional change
that needs a separate patch. See for example `void
LoopVectorizationPlanner::buildVPlans(unsigned MinVF, unsigned
MaxVF)`.
5. The class `IntrinsicCostAttribute`: this class still uses `unsigned
VF` as updating the field to use `ElementCount` woudl require changes
that could result in changing the behavior of the compiler. Will be done
in a separate patch.
7. When dealing with user input for forcing the vectorization
factor. In this case, adding support for scalable vectorization is a
functional change that migh require changes at command line.
Note that in some places the idiom
```
unsigned VF = ...
auto VTy = FixedVectorType::get(ScalarTy, VF)
```
has been replaced with
```
ElementCount VF = ...
assert(!VF.Scalable && ...);
auto VTy = VectorType::get(ScalarTy, VF)
```
The assertion guarantees that the new code is (at least in debug mode)
functionally equivalent to the old version. Notice that this change had been
possible because none of the methods that are specific to `FixedVectorType`
were used after the instantiation of `VTy`.
Reviewed By: rengolin, ctetreau
Differential Revision: https://reviews.llvm.org/D85794
2020-08-08 06:03:24 +08:00
|
|
|
bool hasVF(ElementCount VF) { return VFs.count(VF); }
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
const std::string &getName() const { return Name; }
|
|
|
|
|
|
|
|
void setName(const Twine &newName) { Name = newName.str(); }
|
|
|
|
|
2018-05-22 02:14:23 +08:00
|
|
|
/// Add \p VPVal to the pool of external definitions if it's not already
|
|
|
|
/// in the pool.
|
|
|
|
void addExternalDef(VPValue *VPVal) {
|
|
|
|
VPExternalDefs.insert(VPVal);
|
|
|
|
}
|
|
|
|
|
2018-09-14 10:02:57 +08:00
|
|
|
/// Add \p CBV to the vector of condition bit values.
|
|
|
|
void addCBV(VPValue *CBV) {
|
|
|
|
VPCBVs.push_back(CBV);
|
|
|
|
}
|
|
|
|
|
2017-11-20 20:01:47 +08:00
|
|
|
void addVPValue(Value *V) {
|
|
|
|
assert(V && "Trying to add a null Value to VPlan");
|
|
|
|
assert(!Value2VPValue.count(V) && "Value already exists in VPlan");
|
2020-03-19 05:23:12 +08:00
|
|
|
Value2VPValue[V] = new VPValue(V);
|
2017-11-20 20:01:47 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
VPValue *getVPValue(Value *V) {
|
|
|
|
assert(V && "Trying to get the VPValue of a null Value");
|
|
|
|
assert(Value2VPValue.count(V) && "Value does not exist in VPlan");
|
|
|
|
return Value2VPValue[V];
|
|
|
|
}
|
|
|
|
|
2019-11-28 19:54:38 +08:00
|
|
|
VPValue *getOrAddVPValue(Value *V) {
|
|
|
|
assert(V && "Trying to get or add the VPValue of a null Value");
|
|
|
|
if (!Value2VPValue.count(V))
|
|
|
|
addVPValue(V);
|
|
|
|
return getVPValue(V);
|
|
|
|
}
|
|
|
|
|
2018-07-31 09:57:29 +08:00
|
|
|
/// Return the VPLoopInfo analysis for this VPlan.
|
|
|
|
VPLoopInfo &getVPLoopInfo() { return VPLInfo; }
|
|
|
|
const VPLoopInfo &getVPLoopInfo() const { return VPLInfo; }
|
|
|
|
|
2019-12-03 19:58:31 +08:00
|
|
|
/// Dump the plan to stderr (for debugging).
|
|
|
|
void dump() const;
|
|
|
|
|
2020-04-29 17:22:30 +08:00
|
|
|
/// Returns a range mapping the values the range \p Operands to their
|
|
|
|
/// corresponding VPValues.
|
|
|
|
iterator_range<mapped_iterator<Use *, std::function<VPValue *(Value *)>>>
|
|
|
|
mapToVPValues(User::op_range Operands) {
|
|
|
|
std::function<VPValue *(Value *)> Fn = [this](Value *Op) {
|
|
|
|
return getOrAddVPValue(Op);
|
|
|
|
};
|
|
|
|
return map_range(Operands, Fn);
|
|
|
|
}
|
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
private:
|
|
|
|
/// Add to the given dominator tree the header block and every new basic block
|
|
|
|
/// that was created between it and the latch block, inclusive.
|
2019-12-09 16:11:30 +08:00
|
|
|
static void updateDominatorTree(DominatorTree *DT, BasicBlock *LoopLatchBB,
|
2017-08-27 20:55:46 +08:00
|
|
|
BasicBlock *LoopPreHeaderBB,
|
2019-12-09 16:11:30 +08:00
|
|
|
BasicBlock *LoopExitBB);
|
2017-08-27 20:55:46 +08:00
|
|
|
};
|
|
|
|
|
|
|
|
/// VPlanPrinter prints a given VPlan to a given output stream. The printing is
|
|
|
|
/// indented and follows the dot format.
|
|
|
|
class VPlanPrinter {
|
2019-12-03 19:58:31 +08:00
|
|
|
friend inline raw_ostream &operator<<(raw_ostream &OS, const VPlan &Plan);
|
2017-08-27 20:55:46 +08:00
|
|
|
friend inline raw_ostream &operator<<(raw_ostream &OS,
|
|
|
|
const struct VPlanIngredient &I);
|
|
|
|
|
|
|
|
private:
|
|
|
|
raw_ostream &OS;
|
2019-12-03 19:58:31 +08:00
|
|
|
const VPlan &Plan;
|
2019-11-03 19:17:05 +08:00
|
|
|
unsigned Depth = 0;
|
2017-08-27 20:55:46 +08:00
|
|
|
unsigned TabWidth = 2;
|
|
|
|
std::string Indent;
|
|
|
|
unsigned BID = 0;
|
|
|
|
SmallDenseMap<const VPBlockBase *, unsigned> BlockID;
|
|
|
|
|
2020-03-05 22:55:08 +08:00
|
|
|
VPSlotTracker SlotTracker;
|
|
|
|
|
|
|
|
VPlanPrinter(raw_ostream &O, const VPlan &P)
|
|
|
|
: OS(O), Plan(P), SlotTracker(&P) {}
|
2017-10-18 05:27:42 +08:00
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
/// Handle indentation.
|
|
|
|
void bumpIndent(int b) { Indent = std::string((Depth += b) * TabWidth, ' '); }
|
|
|
|
|
|
|
|
/// Print a given \p Block of the Plan.
|
|
|
|
void dumpBlock(const VPBlockBase *Block);
|
|
|
|
|
|
|
|
/// Print the information related to the CFG edges going out of a given
|
|
|
|
/// \p Block, followed by printing the successor blocks themselves.
|
|
|
|
void dumpEdges(const VPBlockBase *Block);
|
|
|
|
|
|
|
|
/// Print a given \p BasicBlock, including its VPRecipes, followed by printing
|
|
|
|
/// its successor blocks.
|
|
|
|
void dumpBasicBlock(const VPBasicBlock *BasicBlock);
|
|
|
|
|
|
|
|
/// Print a given \p Region of the Plan.
|
|
|
|
void dumpRegion(const VPRegionBlock *Region);
|
|
|
|
|
|
|
|
unsigned getOrCreateBID(const VPBlockBase *Block) {
|
|
|
|
return BlockID.count(Block) ? BlockID[Block] : BlockID[Block] = BID++;
|
|
|
|
}
|
|
|
|
|
|
|
|
const Twine getOrCreateName(const VPBlockBase *Block);
|
|
|
|
|
|
|
|
const Twine getUID(const VPBlockBase *Block);
|
|
|
|
|
|
|
|
/// Print the information related to a CFG edge between two VPBlockBases.
|
|
|
|
void drawEdge(const VPBlockBase *From, const VPBlockBase *To, bool Hidden,
|
|
|
|
const Twine &Label);
|
|
|
|
|
|
|
|
void dump();
|
|
|
|
|
|
|
|
static void printAsIngredient(raw_ostream &O, Value *V);
|
|
|
|
};
|
|
|
|
|
|
|
|
struct VPlanIngredient {
|
|
|
|
Value *V;
|
2017-10-18 05:27:42 +08:00
|
|
|
|
2017-08-27 20:55:46 +08:00
|
|
|
VPlanIngredient(Value *V) : V(V) {}
|
|
|
|
};
|
|
|
|
|
|
|
|
inline raw_ostream &operator<<(raw_ostream &OS, const VPlanIngredient &I) {
|
|
|
|
VPlanPrinter::printAsIngredient(OS, I.V);
|
|
|
|
return OS;
|
|
|
|
}
|
|
|
|
|
2019-12-03 19:58:31 +08:00
|
|
|
inline raw_ostream &operator<<(raw_ostream &OS, const VPlan &Plan) {
|
2017-08-27 20:55:46 +08:00
|
|
|
VPlanPrinter Printer(OS, Plan);
|
|
|
|
Printer.dump();
|
|
|
|
return OS;
|
|
|
|
}
|
|
|
|
|
2018-05-22 02:14:23 +08:00
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
// VPlan Utilities
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
|
|
|
/// Class that provides utilities for VPBlockBases in VPlan.
|
|
|
|
class VPBlockUtils {
|
|
|
|
public:
|
|
|
|
VPBlockUtils() = delete;
|
|
|
|
|
|
|
|
/// Insert disconnected VPBlockBase \p NewBlock after \p BlockPtr. Add \p
|
2018-07-09 23:57:09 +08:00
|
|
|
/// NewBlock as successor of \p BlockPtr and \p BlockPtr as predecessor of \p
|
|
|
|
/// NewBlock, and propagate \p BlockPtr parent to \p NewBlock. If \p BlockPtr
|
|
|
|
/// has more than one successor, its conditional bit is propagated to \p
|
|
|
|
/// NewBlock. \p NewBlock must have neither successors nor predecessors.
|
2018-05-22 02:14:23 +08:00
|
|
|
static void insertBlockAfter(VPBlockBase *NewBlock, VPBlockBase *BlockPtr) {
|
|
|
|
assert(NewBlock->getSuccessors().empty() &&
|
|
|
|
"Can't insert new block with successors.");
|
|
|
|
// TODO: move successors from BlockPtr to NewBlock when this functionality
|
|
|
|
// is necessary. For now, setBlockSingleSuccessor will assert if BlockPtr
|
|
|
|
// already has successors.
|
|
|
|
BlockPtr->setOneSuccessor(NewBlock);
|
|
|
|
NewBlock->setPredecessors({BlockPtr});
|
|
|
|
NewBlock->setParent(BlockPtr->getParent());
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Insert disconnected VPBlockBases \p IfTrue and \p IfFalse after \p
|
|
|
|
/// BlockPtr. Add \p IfTrue and \p IfFalse as succesors of \p BlockPtr and \p
|
|
|
|
/// BlockPtr as predecessor of \p IfTrue and \p IfFalse. Propagate \p BlockPtr
|
2018-07-09 23:57:09 +08:00
|
|
|
/// parent to \p IfTrue and \p IfFalse. \p Condition is set as the successor
|
|
|
|
/// selector. \p BlockPtr must have no successors and \p IfTrue and \p IfFalse
|
|
|
|
/// must have neither successors nor predecessors.
|
2018-05-22 02:14:23 +08:00
|
|
|
static void insertTwoBlocksAfter(VPBlockBase *IfTrue, VPBlockBase *IfFalse,
|
2018-07-09 23:57:09 +08:00
|
|
|
VPValue *Condition, VPBlockBase *BlockPtr) {
|
2018-05-22 02:14:23 +08:00
|
|
|
assert(IfTrue->getSuccessors().empty() &&
|
|
|
|
"Can't insert IfTrue with successors.");
|
|
|
|
assert(IfFalse->getSuccessors().empty() &&
|
|
|
|
"Can't insert IfFalse with successors.");
|
2018-07-09 23:57:09 +08:00
|
|
|
BlockPtr->setTwoSuccessors(IfTrue, IfFalse, Condition);
|
2018-05-22 02:14:23 +08:00
|
|
|
IfTrue->setPredecessors({BlockPtr});
|
|
|
|
IfFalse->setPredecessors({BlockPtr});
|
|
|
|
IfTrue->setParent(BlockPtr->getParent());
|
|
|
|
IfFalse->setParent(BlockPtr->getParent());
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Connect VPBlockBases \p From and \p To bi-directionally. Append \p To to
|
|
|
|
/// the successors of \p From and \p From to the predecessors of \p To. Both
|
|
|
|
/// VPBlockBases must have the same parent, which can be null. Both
|
|
|
|
/// VPBlockBases can be already connected to other VPBlockBases.
|
|
|
|
static void connectBlocks(VPBlockBase *From, VPBlockBase *To) {
|
|
|
|
assert((From->getParent() == To->getParent()) &&
|
|
|
|
"Can't connect two block with different parents");
|
|
|
|
assert(From->getNumSuccessors() < 2 &&
|
|
|
|
"Blocks can't have more than two successors.");
|
|
|
|
From->appendSuccessor(To);
|
|
|
|
To->appendPredecessor(From);
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Disconnect VPBlockBases \p From and \p To bi-directionally. Remove \p To
|
|
|
|
/// from the successors of \p From and \p From from the predecessors of \p To.
|
|
|
|
static void disconnectBlocks(VPBlockBase *From, VPBlockBase *To) {
|
|
|
|
assert(To && "Successor to disconnect is null.");
|
|
|
|
From->removeSuccessor(To);
|
|
|
|
To->removePredecessor(From);
|
|
|
|
}
|
2019-01-24 06:43:12 +08:00
|
|
|
|
|
|
|
/// Returns true if the edge \p FromBlock -> \p ToBlock is a back-edge.
|
|
|
|
static bool isBackEdge(const VPBlockBase *FromBlock,
|
|
|
|
const VPBlockBase *ToBlock, const VPLoopInfo *VPLI) {
|
|
|
|
assert(FromBlock->getParent() == ToBlock->getParent() &&
|
|
|
|
FromBlock->getParent() && "Must be in same region");
|
|
|
|
const VPLoop *FromLoop = VPLI->getLoopFor(FromBlock);
|
|
|
|
const VPLoop *ToLoop = VPLI->getLoopFor(ToBlock);
|
|
|
|
if (!FromLoop || !ToLoop || FromLoop != ToLoop)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
// A back-edge is a branch from the loop latch to its header.
|
|
|
|
return ToLoop->isLoopLatch(FromBlock) && ToBlock == ToLoop->getHeader();
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Returns true if \p Block is a loop latch
|
|
|
|
static bool blockIsLoopLatch(const VPBlockBase *Block,
|
|
|
|
const VPLoopInfo *VPLInfo) {
|
|
|
|
if (const VPLoop *ParentVPL = VPLInfo->getLoopFor(Block))
|
|
|
|
return ParentVPL->isLoopLatch(Block);
|
|
|
|
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Count and return the number of succesors of \p PredBlock excluding any
|
|
|
|
/// backedges.
|
|
|
|
static unsigned countSuccessorsNoBE(VPBlockBase *PredBlock,
|
|
|
|
VPLoopInfo *VPLI) {
|
|
|
|
unsigned Count = 0;
|
|
|
|
for (VPBlockBase *SuccBlock : PredBlock->getSuccessors()) {
|
|
|
|
if (!VPBlockUtils::isBackEdge(PredBlock, SuccBlock, VPLI))
|
|
|
|
Count++;
|
|
|
|
}
|
|
|
|
return Count;
|
|
|
|
}
|
2018-05-22 02:14:23 +08:00
|
|
|
};
|
2018-06-09 01:30:45 +08:00
|
|
|
|
2018-11-13 23:58:18 +08:00
|
|
|
class VPInterleavedAccessInfo {
|
|
|
|
DenseMap<VPInstruction *, InterleaveGroup<VPInstruction> *>
|
|
|
|
InterleaveGroupMap;
|
|
|
|
|
|
|
|
/// Type for mapping of instruction based interleave groups to VPInstruction
|
|
|
|
/// interleave groups
|
|
|
|
using Old2NewTy = DenseMap<InterleaveGroup<Instruction> *,
|
|
|
|
InterleaveGroup<VPInstruction> *>;
|
|
|
|
|
|
|
|
/// Recursively \p Region and populate VPlan based interleave groups based on
|
|
|
|
/// \p IAI.
|
|
|
|
void visitRegion(VPRegionBlock *Region, Old2NewTy &Old2New,
|
|
|
|
InterleavedAccessInfo &IAI);
|
|
|
|
/// Recursively traverse \p Block and populate VPlan based interleave groups
|
|
|
|
/// based on \p IAI.
|
|
|
|
void visitBlock(VPBlockBase *Block, Old2NewTy &Old2New,
|
|
|
|
InterleavedAccessInfo &IAI);
|
|
|
|
|
|
|
|
public:
|
|
|
|
VPInterleavedAccessInfo(VPlan &Plan, InterleavedAccessInfo &IAI);
|
|
|
|
|
|
|
|
~VPInterleavedAccessInfo() {
|
|
|
|
SmallPtrSet<InterleaveGroup<VPInstruction> *, 4> DelSet;
|
|
|
|
// Avoid releasing a pointer twice.
|
|
|
|
for (auto &I : InterleaveGroupMap)
|
|
|
|
DelSet.insert(I.second);
|
|
|
|
for (auto *Ptr : DelSet)
|
|
|
|
delete Ptr;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Get the interleave group that \p Instr belongs to.
|
|
|
|
///
|
|
|
|
/// \returns nullptr if doesn't have such group.
|
|
|
|
InterleaveGroup<VPInstruction> *
|
|
|
|
getInterleaveGroup(VPInstruction *Instr) const {
|
|
|
|
if (InterleaveGroupMap.count(Instr))
|
|
|
|
return InterleaveGroupMap.find(Instr)->second;
|
|
|
|
return nullptr;
|
|
|
|
}
|
|
|
|
};
|
|
|
|
|
[VPlan, SLP] Add simple SLP analysis on top of VPlan.
This patch adds an initial implementation of the look-ahead SLP tree
construction described in 'Look-Ahead SLP: Auto-vectorization in the Presence
of Commutative Operations, CGO 2018 by Vasileios Porpodas, Rodrigo C. O. Rocha,
Luís F. W. Góes'.
It returns an SLP tree represented as VPInstructions, with combined
instructions represented as a single, wider VPInstruction.
This initial version does not support instructions with multiple
different users (either inside or outside the SLP tree) or
non-instruction operands; it won't generate any shuffles or
insertelement instructions.
It also just adds the analysis that builds an SLP tree rooted in a set
of stores. It does not include any cost modeling or memory legality
checks. The plan is to integrate it with VPlan based cost modeling, once
available and to only apply it to operations that can be widened.
A follow-up patch will add a support for replacing instructions in a
VPlan with their SLP counter parts.
Reviewers: Ayal, mssimpso, rengolin, mkuper, hfinkel, hsaito, dcaballe, vporpo, RKSimon, ABataev
Reviewed By: rengolin
Differential Revision: https://reviews.llvm.org/D4949
llvm-svn: 346857
2018-11-14 21:11:49 +08:00
|
|
|
/// Class that maps (parts of) an existing VPlan to trees of combined
|
|
|
|
/// VPInstructions.
|
|
|
|
class VPlanSlp {
|
|
|
|
enum class OpMode { Failed, Load, Opcode };
|
|
|
|
|
|
|
|
/// A DenseMapInfo implementation for using SmallVector<VPValue *, 4> as
|
|
|
|
/// DenseMap keys.
|
|
|
|
struct BundleDenseMapInfo {
|
|
|
|
static SmallVector<VPValue *, 4> getEmptyKey() {
|
|
|
|
return {reinterpret_cast<VPValue *>(-1)};
|
|
|
|
}
|
|
|
|
|
|
|
|
static SmallVector<VPValue *, 4> getTombstoneKey() {
|
|
|
|
return {reinterpret_cast<VPValue *>(-2)};
|
|
|
|
}
|
|
|
|
|
|
|
|
static unsigned getHashValue(const SmallVector<VPValue *, 4> &V) {
|
|
|
|
return static_cast<unsigned>(hash_combine_range(V.begin(), V.end()));
|
|
|
|
}
|
|
|
|
|
|
|
|
static bool isEqual(const SmallVector<VPValue *, 4> &LHS,
|
|
|
|
const SmallVector<VPValue *, 4> &RHS) {
|
|
|
|
return LHS == RHS;
|
|
|
|
}
|
|
|
|
};
|
|
|
|
|
|
|
|
/// Mapping of values in the original VPlan to a combined VPInstruction.
|
|
|
|
DenseMap<SmallVector<VPValue *, 4>, VPInstruction *, BundleDenseMapInfo>
|
|
|
|
BundleToCombined;
|
|
|
|
|
|
|
|
VPInterleavedAccessInfo &IAI;
|
|
|
|
|
|
|
|
/// Basic block to operate on. For now, only instructions in a single BB are
|
|
|
|
/// considered.
|
|
|
|
const VPBasicBlock &BB;
|
|
|
|
|
|
|
|
/// Indicates whether we managed to combine all visited instructions or not.
|
|
|
|
bool CompletelySLP = true;
|
|
|
|
|
|
|
|
/// Width of the widest combined bundle in bits.
|
|
|
|
unsigned WidestBundleBits = 0;
|
|
|
|
|
|
|
|
using MultiNodeOpTy =
|
|
|
|
typename std::pair<VPInstruction *, SmallVector<VPValue *, 4>>;
|
|
|
|
|
|
|
|
// Input operand bundles for the current multi node. Each multi node operand
|
|
|
|
// bundle contains values not matching the multi node's opcode. They will
|
|
|
|
// be reordered in reorderMultiNodeOps, once we completed building a
|
|
|
|
// multi node.
|
|
|
|
SmallVector<MultiNodeOpTy, 4> MultiNodeOps;
|
|
|
|
|
|
|
|
/// Indicates whether we are building a multi node currently.
|
|
|
|
bool MultiNodeActive = false;
|
|
|
|
|
|
|
|
/// Check if we can vectorize Operands together.
|
|
|
|
bool areVectorizable(ArrayRef<VPValue *> Operands) const;
|
|
|
|
|
|
|
|
/// Add combined instruction \p New for the bundle \p Operands.
|
|
|
|
void addCombined(ArrayRef<VPValue *> Operands, VPInstruction *New);
|
|
|
|
|
|
|
|
/// Indicate we hit a bundle we failed to combine. Returns nullptr for now.
|
|
|
|
VPInstruction *markFailed();
|
|
|
|
|
|
|
|
/// Reorder operands in the multi node to maximize sequential memory access
|
|
|
|
/// and commutative operations.
|
|
|
|
SmallVector<MultiNodeOpTy, 4> reorderMultiNodeOps();
|
|
|
|
|
|
|
|
/// Choose the best candidate to use for the lane after \p Last. The set of
|
|
|
|
/// candidates to choose from are values with an opcode matching \p Last's
|
|
|
|
/// or loads consecutive to \p Last.
|
|
|
|
std::pair<OpMode, VPValue *> getBest(OpMode Mode, VPValue *Last,
|
2018-11-14 23:58:40 +08:00
|
|
|
SmallPtrSetImpl<VPValue *> &Candidates,
|
[VPlan, SLP] Add simple SLP analysis on top of VPlan.
This patch adds an initial implementation of the look-ahead SLP tree
construction described in 'Look-Ahead SLP: Auto-vectorization in the Presence
of Commutative Operations, CGO 2018 by Vasileios Porpodas, Rodrigo C. O. Rocha,
Luís F. W. Góes'.
It returns an SLP tree represented as VPInstructions, with combined
instructions represented as a single, wider VPInstruction.
This initial version does not support instructions with multiple
different users (either inside or outside the SLP tree) or
non-instruction operands; it won't generate any shuffles or
insertelement instructions.
It also just adds the analysis that builds an SLP tree rooted in a set
of stores. It does not include any cost modeling or memory legality
checks. The plan is to integrate it with VPlan based cost modeling, once
available and to only apply it to operations that can be widened.
A follow-up patch will add a support for replacing instructions in a
VPlan with their SLP counter parts.
Reviewers: Ayal, mssimpso, rengolin, mkuper, hfinkel, hsaito, dcaballe, vporpo, RKSimon, ABataev
Reviewed By: rengolin
Differential Revision: https://reviews.llvm.org/D4949
llvm-svn: 346857
2018-11-14 21:11:49 +08:00
|
|
|
VPInterleavedAccessInfo &IAI);
|
|
|
|
|
|
|
|
/// Print bundle \p Values to dbgs().
|
|
|
|
void dumpBundle(ArrayRef<VPValue *> Values);
|
|
|
|
|
|
|
|
public:
|
|
|
|
VPlanSlp(VPInterleavedAccessInfo &IAI, VPBasicBlock &BB) : IAI(IAI), BB(BB) {}
|
|
|
|
|
|
|
|
~VPlanSlp() {
|
|
|
|
for (auto &KV : BundleToCombined)
|
|
|
|
delete KV.second;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Tries to build an SLP tree rooted at \p Operands and returns a
|
|
|
|
/// VPInstruction combining \p Operands, if they can be combined.
|
|
|
|
VPInstruction *buildGraph(ArrayRef<VPValue *> Operands);
|
|
|
|
|
|
|
|
/// Return the width of the widest combined bundle in bits.
|
|
|
|
unsigned getWidestBundleBits() const { return WidestBundleBits; }
|
|
|
|
|
|
|
|
/// Return true if all visited instruction can be combined.
|
|
|
|
bool isCompletelySLP() const { return CompletelySLP; }
|
|
|
|
};
|
2017-10-18 05:27:42 +08:00
|
|
|
} // end namespace llvm
|
2017-08-27 20:55:46 +08:00
|
|
|
|
|
|
|
#endif // LLVM_TRANSFORMS_VECTORIZE_VPLAN_H
|