* Extract "value" functionality of `FlatAffineConstraints` into a new derived `FlatAffineValueConstraints` class. Current users of `FlatAffineConstraints` can use `FlatAffineValueConstraints` without additional code changes, thus NFC.
* `FlatAffineConstraints` no longer associates dimensions with SSA Values. All functionality that requires this, is moved to `FlatAffineValueConstraints`.
* `FlatAffineConstraints` no longer makes assumptions about where Values associated with dimensions are coming from.
Differential Revision: https://reviews.llvm.org/D107725
Changes include the following:
1. Single iteration reduction loops being sibling fused at innermost insertion level
are skipped from being considered as sequential loops.
Otherwise, the slice bounds of these loops is reset.
2. Promote loops that are skipped in previous step into outer loops.
3. Two utility function - buildSliceTripCountMap, getSliceIterationCount - are moved from
mlir/lib/Transforms/Utils/LoopFusionUtils.cpp to mlir/lib/Analysis/Utils.cpp
Reviewed By: bondhugula, vinayaka-polymage
Differential Revision: https://reviews.llvm.org/D104249
Fix FlatAffineConstraints::getConstantBoundOnDimSize to ensure that
returned bounds on dim size are always non-negative regardless of the
constraints on that dimension. Add an assertion at the user.
Differential Revision: https://reviews.llvm.org/D105171
Affine scalar replacement (and other affine passes, though not fixed here) don't properly handle operations with nested regions. This patch fixes the pass and two affine utilities to function properly given a non-affine internal region
This patch prevents the pass from throwing an internal compiler error when running on the added test case.
Differential Revision: https://reviews.llvm.org/D105058
Introduce a basic support for parallelizing affine loops with reductions
expressed using iteration arguments. Affine parallelism detector now has a flag
to assume such reductions are parallel. The transformation handles a subset of
parallel reductions that are can be expressed using affine.parallel:
integer/float addition and multiplication. This requires to detect the
reduction operation since affine.parallel only supports a fixed set of
reduction operators.
Reviewed By: chelini, kumasento, bondhugula
Differential Revision: https://reviews.llvm.org/D101171
Fixes a bug in affine fusion pipeline where an incorrect slice is computed.
After the slice computation is done, original domain of the the source is
compared with the new domain that will result if the fusion succeeds. If the
new domain must be a subset of the original domain for the slice to be
valid. If the slice computed is incorrect, fusion based on such a slice is
avoided.
Relevant test cases are added/edited.
Fixes https://bugs.llvm.org/show_bug.cgi?id=49203
Differential Revision: https://reviews.llvm.org/D98239
Fix 'isLoopParallel' utility so that 'iter_args' is taken into account
and loops with loop-carried dependences are not classified as parallel.
Reviewed By: tungld, vinayaka-polymage
Differential Revision: https://reviews.llvm.org/D97347
This patch adds support for producer-consumer fusion scenarios with
multiple producer stores to the AffineLoopFusion pass. The patch
introduces some changes to the producer-consumer algorithm, including:
* For a given consumer loop, producer-consumer fusion iterates over its
producer candidates until a fixed point is reached.
* Producer candidates are gathered beforehand for each iteration of the
consumer loop and visited in reverse program order (not strictly guaranteed)
to maximize the number of loops fused per iteration.
In general, these changes were needed to simplify the multi-store producer
support and remove some of the workarounds that were introduced in the past
to support more fusion cases under the single-store producer limitation.
This patch also preserves the existing functionality of AffineLoopFusion with
one minor change in behavior. Producer-consumer fusion didn't fuse scenarios
with escaping memrefs and multiple outgoing edges (from a single store).
Multi-store producer scenarios will usually (always?) have multiple outgoing
edges so we couldn't fuse any with escaping memrefs, which would greatly limit
the applicability of this new feature. Therefore, the patch enables fusion for
these scenarios. Please, see modified tests for specific details.
Reviewed By: andydavis1, bondhugula
Differential Revision: https://reviews.llvm.org/D92876
This patch adds support for producer-consumer fusion scenarios with
multiple producer stores to the AffineLoopFusion pass. The patch
introduces some changes to the producer-consumer algorithm, including:
* For a given consumer loop, producer-consumer fusion iterates over its
producer candidates until a fixed point is reached.
* Producer candidates are gathered beforehand for each iteration of the
consumer loop and visited in reverse program order (not strictly guaranteed)
to maximize the number of loops fused per iteration.
In general, these changes were needed to simplify the multi-store producer
support and remove some of the workarounds that were introduced in the past
to support more fusion cases under the single-store producer limitation.
This patch also preserves the existing functionality of AffineLoopFusion with
one minor change in behavior. Producer-consumer fusion didn't fuse scenarios
with escaping memrefs and multiple outgoing edges (from a single store).
Multi-store producer scenarios will usually (always?) have multiple outgoing
edges so we couldn't fuse any with escaping memrefs, which would greatly limit
the applicability of this new feature. Therefore, the patch enables fusion for
these scenarios. Please, see modified tests for specific details.
Reviewed By: andydavis1, bondhugula
Differential Revision: https://reviews.llvm.org/D92876
Given that OpState already implicit converts to Operator*, this seems reasonable.
The alternative would be to add more functions to OpState which forward to Operation.
Reviewed By: rriddle, ftynse
Differential Revision: https://reviews.llvm.org/D92266
Refactoring/clean-up step needed to add support for producer-consumer fusion
with multi-store producer loops and, in general, to implement more general
loop fusion strategies in Affine. It introduces the following changes:
- AffineLoopFusion pass now uses loop fusion utilities more broadly to compute
fusion legality (canFuseLoops utility) and perform the fusion transformation
(fuseLoops utility).
- Loop fusion utilities have been extended to deal with AffineLoopFusion
requirements and assumptions while preserving both loop fusion utilities and
AffineLoopFusion current functionality within a unified implementation.
'FusionStrategy' has been introduced for this purpose and, in the future, it
will allow us to have a single loop fusion core implementation that will produce
different fusion outputs depending on the strategy used.
- Improve separation of concerns for legality and profitability analysis:
'isFusionProfitable' no longer filters out illegal scenarios that 'canFuse'
didn't detect, or the other way around. 'canFuse' now takes loop dependences
into account to determine the fusion loop depth (producer-consumer fusion only).
- As a result, maximal fusion now doesn't require any profitability analysis.
- Slices are now computed only once and reused across the legality, profitability
and fusion transformation steps (producer-consumer).
- Refactor some utilities and remove redundant copies of them.
This patch is NFCI and should preserve the existing functionality of both the
AffineLoopFusion pass and the affine fusion utilities.
Reviewed By: andydavis1, bondhugula
Differential Revision: https://reviews.llvm.org/D90798
This diff attempts to resolve the TODO in `getOpIndexSet` (formerly
known as `getInstIndexSet`), which states "Add support to handle IfInsts
surronding `op`".
Major changes in this diff:
1. Overload `getIndexSet`. The overloaded version considers both
`AffineForOp` and `AffineIfOp`.
2. The `getInstIndexSet` is updated accordingly: its name is changed to
`getOpIndexSet` and its implementation is based on a new API `getIVs`
instead of `getLoopIVs`.
3. Add `addAffineIfOpDomain` to `FlatAffineConstraints`, which extracts
new constraints from the integer set of `AffineIfOp` and merges it to
the current constraint system.
4. Update how a `Value` is determined as dim or symbol for
`ValuePositionMap` in `buildDimAndSymbolPositionMaps`.
Differential Revision: https://reviews.llvm.org/D84698
Fix memref region compute for 0-d memref accesses in certain cases (when
there are loops surrounding such 0-d accesses).
Differential Revision: https://reviews.llvm.org/D81792
This patch introduces interfaces for read and write ops with affine
restrictions. I used `read`/`write` intead of `load`/`store` for the
interfaces so that they can also be implemented by dma ops.
For now, they are only implemented by affine.load, affine.store,
affine.vector_load and affine.vector_store.
For testing purposes, this patch also migrates affine loop fusion and
required analysis to use the new interfaces. No other changes are made
beyond that.
Co-authored-by: Alex Zinenko <zinenko@google.com>
Reviewed By: bondhugula, ftynse
Differential Revision: https://reviews.llvm.org/D79829
Summary:
This makes a common pattern of
`dyn_cast_or_null<OpTy>(v.getDefiningOp())` more concise.
Differential Revision: https://reviews.llvm.org/D79681
Minor fixes and cleanup for ShapedType accessors, use
ShapedType::kDynamicSize, add ShapedType::isDynamicDim.
Differential Revision: https://reviews.llvm.org/D77710
Support to recognize and deal with aligned_alloc was recently added to
LLVM's TLI/MemoryBuiltins and its various optimization passes. This
revision adds support for generation of aligned_alloc's when lowering
AllocOp from std to LLVM. Setting 'use-aligned_alloc=1' will lead to
aligned_alloc being used for all heap allocations. An alignment and size
that works with the constraints of aligned_alloc is chosen.
Using aligned_alloc is preferable to "using malloc and adjusting the
allocated pointer to align for indexing" because the pointer access
arithmetic done for the latter only makes it harder for LLVM passes to
deal with for analysis, optimization, attribute deduction, and rewrites.
Differential Revision: https://reviews.llvm.org/D77528
Fix point-wise copy generation to work with bounds that have max/min.
Change structure of copy loop nest to use absolute loop indices and
subtracting base from the indexes of the fast buffers. Update supporting
utilities: Fix FlatAffineConstraints::getLowerAndUpperBound to look at
equalities as well and for a missing division. Update unionBoundingBox
to not discard common constraints (leads to a tighter system). Update
MemRefRegion::getConstantBoundingSizeAndShape to add memref dimension
constraints. Run removeTrivialRedundancy at the end of
MemRefRegion::compute. Run single iteration loop promotion and
load/store canonicalization after affine data copy (in its test pass as
well).
Differential Revision: https://reviews.llvm.org/D77320
This patch introduces a utility to separate full tiles from partial
tiles when tiling affine loop nests where trip counts are unknown or
where tile sizes don't divide trip counts. A conditional guard is
generated to separate out the full tile (with constant trip count loops)
into the then block of an 'affine.if' and the partial tile to the else
block. The separation allows the 'then' block (which has constant trip
count loops) to be optimized better subsequently: for eg. for
unroll-and-jam, register tiling, vectorization without leading to
cleanup code, or to offload to accelerators. Among techniques from the
literature, the if/else based separation leads to the most compact
cleanup code for multi-dimensional cases (because a single version is
used to model all partial tiles).
INPUT
affine.for %i0 = 0 to %M {
affine.for %i1 = 0 to %N {
"foo"() : () -> ()
}
}
OUTPUT AFTER TILING W/O SEPARATION
map0 = affine_map<(d0) -> (d0)>
map1 = affine_map<(d0)[s0] -> (d0 + 32, s0)>
affine.for %arg2 = 0 to %M step 32 {
affine.for %arg3 = 0 to %N step 32 {
affine.for %arg4 = #map0(%arg2) to min #map1(%arg2)[%M] {
affine.for %arg5 = #map0(%arg3) to min #map1(%arg3)[%N] {
"foo"() : () -> ()
}
}
}
}
OUTPUT AFTER TILING WITH SEPARATION
map0 = affine_map<(d0) -> (d0)>
map1 = affine_map<(d0) -> (d0 + 32)>
map2 = affine_map<(d0)[s0] -> (d0 + 32, s0)>
#set0 = affine_set<(d0, d1)[s0, s1] : (-d0 + s0 - 32 >= 0, -d1 + s1 - 32 >= 0)>
affine.for %arg2 = 0 to %M step 32 {
affine.for %arg3 = 0 to %N step 32 {
affine.if #set0(%arg2, %arg3)[%M, %N] {
// Full tile.
affine.for %arg4 = #map0(%arg2) to #map1(%arg2) {
affine.for %arg5 = #map0(%arg3) to #map1(%arg3) {
"foo"() : () -> ()
}
}
} else {
// Partial tile.
affine.for %arg4 = #map0(%arg2) to min #map2(%arg2)[%M] {
affine.for %arg5 = #map0(%arg3) to min #map2(%arg3)[%N] {
"foo"() : () -> ()
}
}
}
}
}
The separation is tested via a cmd line flag on the loop tiling pass.
The utility itself allows one to pass in any band of contiguously nested
loops, and can be used by other transforms/utilities. The current
implementation works for hyperrectangular loop nests.
Signed-off-by: Uday Bondhugula <uday@polymagelabs.com>
Differential Revision: https://reviews.llvm.org/D76700
- add method to get back an integer set from flat affine constraints;
this allows a round trip
- use this to complete the simplification of integer sets in
-simplify-affine-structures
- update FlatAffineConstraints::removeTrivialRedundancy to also do GCD
tightening and normalize by GCD (while still keeping it linear time).
Signed-off-by: Uday Bondhugula <uday@polymagelabs.com>
Summary:
Change AffineOps Dialect structure to better group both IR and Tranforms. This included extracting transforms directly related to AffineOps. Also move AffineOps to Affine.
Differential Revision: https://reviews.llvm.org/D76161
HasNoSideEffect can now be implemented using the MemoryEffectInterface, removing the need to check multiple things for the same information. This also removes an easy foot-gun for users as 'Operation::hasNoSideEffect' would ignore operations that dynamically, or recursively, have no side effects. This also leads to an immediate improvement in some of the existing users, such as DCE, now that they have access to more information.
Differential Revision: https://reviews.llvm.org/D76036
Summary: For example, DenseElementsAttr currently does not properly round-trip unsigned integer values.
Differential Revision: https://reviews.llvm.org/D75374
Summary:
NFC - Moved StandardOps/Ops.h to a StandardOps/IR dir to better match surrounding
directories. This is to match other dialects, and prepare for moving StandardOps
related transforms in out for Transforms and into StandardOps/Transforms.
Differential Revision: https://reviews.llvm.org/D74940
Thus far IntegerType has been signless: a value of IntegerType does
not have a sign intrinsically and it's up to the specific operation
to decide how to interpret those bits. For example, std.addi does
two's complement arithmetic, and std.divis/std.diviu treats the first
bit as a sign.
This design choice was made some time ago when we did't have lots
of dialects and dialects were more rigid. Today we have much more
extensible infrastructure and different dialect may want different
modelling over integer signedness. So while we can say we want
signless integers in the standard dialect, we cannot dictate for
others. Requiring each dialect to model the signedness semantics
with another set of custom types is duplicating the functionality
everywhere, considering the fundamental role integer types play.
This CL extends the IntegerType with a signedness semantics bit.
This gives each dialect an option to opt in signedness semantics
if that's what they want and helps code sharing. The parser is
modified to recognize `si[1-9][0-9]*` and `ui[1-9][0-9]*` as
signed and unsigned integer types, respectively, leaving the
original `i[1-9][0-9]*` to continue to mean no indication over
signedness semantics. All existing dialects are not affected (yet)
as this is a feature to opt in.
More discussions can be found at:
https://groups.google.com/a/tensorflow.org/d/msg/mlir/XmkV8HOPWpo/7O4X0Nb_AQAJ
Differential Revision: https://reviews.llvm.org/D72533
This is an initial step to refactoring the representation of OpResult as proposed in: https://groups.google.com/a/tensorflow.org/g/mlir/c/XXzzKhqqF_0/m/v6bKb08WCgAJ
This change will make it much simpler to incrementally transition all of the existing code to use value-typed semantics.
PiperOrigin-RevId: 286844725
This CL utilizies the more robust fusion feasibility analysis being built out in LoopFusionUtils, which will eventually be used to replace the current affine loop fusion pass.
PiperOrigin-RevId: 281112340
- fix store to load forwarding for a certain set of cases (where
forwarding shouldn't have happened); use AffineValueMap difference
based MemRefAccess equality checking; utility logic is also greatly
simplified
- add missing equality/inequality operators for AffineExpr ==/!= ints
- add == != operators on MemRefAccess
Closestensorflow/mlir#136
COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/136 from bondhugula:store-load-forwarding d79fd1add8bcfbd9fa71d841a6a9905340dcd792
PiperOrigin-RevId: 270457011
- add canonicalization pattern to compose maps into affine loads/stores;
templatize the pattern and reuse it for affine.apply as well
- rename getIndices -> getMapOperands() (getIndices is confusing since
these are no longer the indices themselves but operands to the map
whose results are the indices). This also makes the accessor uniform
across affine.apply/load/store. Change arg names on the affine
load/store builder to avoid confusion. Drop an unused confusing build
method on AffineStoreOp.
- update incomplete doc comment for canonicalizeMapAndOperands (this was
missed from a previous update).
Addresses issue tensorflow/mlir#121
Signed-off-by: Uday Bondhugula <uday@polymagelabs.com>
Closestensorflow/mlir#122
COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/122 from bondhugula:compose-load-store e71de1771e56a85c4282c10cb43f30cef0701c4f
PiperOrigin-RevId: 269619540
This is done by providing a walk callback that returns a WalkResult. This result is either `advance` or `interrupt`. `advance` means that the walk should continue, whereas `interrupt` signals that the walk should stop immediately. An example is shown below:
auto result = op->walk([](Operation *op) {
if (some_invariant)
return WalkResult::interrupt();
return WalkResult::advance();
});
if (result.wasInterrupted())
...;
PiperOrigin-RevId: 266436700
