Commit Graph

603 Commits

Author SHA1 Message Date
Uday Bondhugula cea9f28a2c Fix dominates() for block's.
- dominates() for blocks was assuming that there was only a single block at the
  top level whenever there was a hierarchy of blocks (as in the case of 'for'/'if'
  instructions).
- fix the comments as well

PiperOrigin-RevId: 227319738
2019-03-29 14:48:28 -07:00
Chris Lattner ae618428f6 Greatly simplify the ConvertToCFG pass, converting it from a module pass to a
function pass, and eliminating the need to copy over code and do
interprocedural updates.  While here, also improve it to make fewer empty
blocks, and rename it to "LowerIfAndFor" since that is what it does.  This is
a net reduction of ~170 lines of code.

As drive-bys, change the splitBlock method to *not* insert an unconditional
branch, since that behavior is annoying for all clients.  Also improve the
AsmPrinter to not crash when a block is referenced that isn't linked into a
function.

PiperOrigin-RevId: 227308856
2019-03-29 14:48:13 -07:00
Uday Bondhugula 545f3ce430 Fix ASAN failure in memref-dataflow-opt
- memrefsToErase had duplicates inserted into it; switch to SmallPtrSet.

PiperOrigin-RevId: 227299306
2019-03-29 14:47:58 -07:00
Feng Liu dfee0a6e9b Make PrintOpStatsPass a module pass
PrintOpStatsPass is maintaining state (op stats ) across functions and doing
per-module work - it should be a module pass.

PiperOrigin-RevId: 227294151
2019-03-29 14:47:43 -07:00
Uday Bondhugula b9fe6be6d4 Introduce memref store to load forwarding - a simple memref dataflow analysis
- the load/store forwarding relies on memref dependence routines as well as
  SSA/dominance to identify the memref store instance uniquely supplying a value
  to a memref load, and replaces the result of that load with the value being
  stored. The memref is also deleted when possible if only stores remain.

- add methods for post dominance for MLFunction blocks.

- remove duplicated getLoopDepth/getNestingDepth - move getNestingDepth,
  getMemRefAccess, getNumCommonSurroundingLoops into Analysis/Utils (were
  earlier static)

- add a helper method in FlatAffineConstraints - isRangeOneToOne.

PiperOrigin-RevId: 227252907
2019-03-29 14:47:28 -07:00
Uday Bondhugula 6e3462d251 Fix b/122139732; update FlatAffineConstraints::isEmpty() to eliminate IDs in a
better order.

- update isEmpty() to eliminate IDs in a better order. Speed improvement for
  complex cases (for eg. high-d reshape's involving mod's/div's).
- minor efficiency update to projectOut (was earlier making an extra albeit
  benign call to gaussianEliminateIds) (NFC).
- move getBestIdToEliminate further up in the file (NFC).
- add the failing test case.
- add debug info to checkMemRefAccessDependence.

PiperOrigin-RevId: 227244634
2019-03-29 14:47:13 -07:00
Chris Lattner dffc589ad2 Extend InstVisitor and Walker to handle arbitrary CFG functions, expand the
Function::walk functionality into f->walkInsts/Ops which allows visiting all
instructions, not just ops.  Eliminate Function::getBody() and
Function::getReturn() helpers which crash in CFG functions, and were only kept
around as a bridge.

This is step 25/n towards merging instructions and statements.

PiperOrigin-RevId: 227243966
2019-03-29 14:46:58 -07:00
Jacques Pienaar bbe3f4d9f5 Switch rewriters for relu, relu6, placeholder_input, softmax to patterns.
Add a constant F32 attribute for use with softmax legalization.

PiperOrigin-RevId: 227241643
2019-03-29 14:46:44 -07:00
Chris Lattner 8ef2552df7 Have the asmprinter take advantage of the new capabilities of the asmparser, by
printing the entry block in a CFG function's argument line.  Since I'm touching
all of the testcases anyway, change the argument list from printing as
"%arg : type" to "%arg: type" which is more consistent with bb arguments.

In addition to being more consistent, this is a much nicer look for cfg functions.

PiperOrigin-RevId: 227240069
2019-03-29 14:46:29 -07:00
Chris Lattner aaa1d77e96 Clean up and improve the parser handling of basic block labels, now that we
have a designator.  This improves diagnostics and merges handling between CFG
and ML functions more.  This also eliminates hard coded parser knowledge of
terminator keywords, allowing dialects to define their own terminators.

PiperOrigin-RevId: 227239398
2019-03-29 14:46:13 -07:00
Chris Lattner 37579ae8c4 Introduce ^ as a basic block sigil, eliminating an ambiguity on the MLIR
syntax.

PiperOrigin-RevId: 227234174
2019-03-29 14:45:59 -07:00
Chris Lattner 56e2a6cc3b Merge the verifier logic for all functions into a unified framework, this
requires enhancing DominanceInfo to handle the structure of an ML function,
which is required anyway.  Along the way, this also fixes a const correctness
problem with Instruction::getBlock().

This is step 24/n towards merging instructions and statements.

PiperOrigin-RevId: 227228900
2019-03-29 14:45:43 -07:00
Chris Lattner 4a96a11d6d Enhance parsing of CFG and Ext functions to optionally allow named arguments in
the function signature, giving them common functionality to ml functions.  This
is a strictly additive patch that adds new capability without changing behavior
in a significant way (other than a few diagnostic cleanups).  A subsequent
patch will change the printer to use this behavior, which will require updating
a ton of testcases.  :)

This exposes the fact that we need to make a grammar change for block
arguments, as is tracked by b/122119779

This is step 23/n towards merging instructions and statements, and one of the
first steps towards eliminating the "cfg vs ml" distinction at a syntax and
semantic level.

PiperOrigin-RevId: 227228342
2019-03-29 14:45:28 -07:00
Jacques Pienaar 554848d617 Match multiple pattern nodes as input to rewrite.
* Allow multi input node patterns in the rewrite;
* Use number of nodes matched as benefit;
* Rewrite relu(add(...)) matching using the new pattern;

To allow for undefined ops, do string compare - will address soon!

PiperOrigin-RevId: 227225425
2019-03-29 14:45:14 -07:00
Chris Lattner 5b9c3f7cdb Tidy up references to "basic blocks" that should refer to blocks now. NFC.
PiperOrigin-RevId: 227196077
2019-03-29 14:44:59 -07:00
Chris Lattner be9ee4a98e Merge parser logic for CFG and ML functions, shrinking the code
by ~80 lines.  This causes a slight change to diagnostics, but
is otherwise behavior preserving.

This is step 22/n towards merging instructions and statements, MFC.

PiperOrigin-RevId: 227187857
2019-03-29 14:44:44 -07:00
Chris Lattner 456ad6a8e0 Standardize naming of statements -> instructions, revisting the code base to be
consistent and moving the using declarations over.  Hopefully this is the last
truly massive patch in this refactoring.

This is step 21/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 227178245
2019-03-29 14:44:30 -07:00
Uday Bondhugula b1d9cc4d1e Extend/complete dependence tester to utilize local var info.
- extend/complete dependence tester to utilize local var info while adding
  access function equality constraints; one more step closer to get slicing
  based fusion working in the general case of affine_apply's involving mod's/div's.
- update test case to reflect more accurate dependence information; remove
  inaccurate comment on test case mod_deps.
- fix a minor "bug" in equality addition in addMemRefAccessConstraints (doesn't
  affect correctness, but the fixed version is more intuitive).
- some more surrounding code clean up
- move simplifyAffineExpr out of anonymous AffineExprFlattener class - the
  latter has state, and the former should reside outside.

PiperOrigin-RevId: 227175600
2019-03-29 14:44:14 -07:00
Chris Lattner 315a466aed Rename BasicBlock and StmtBlock to Block, and make a pass cleaning it up. I did not make an effort to rename all of the 'bb' names in the codebase, since they are still correct and any specific missed once can be fixed up on demand.
The last major renaming is Statement -> Instruction, which is why Statement and
Stmt still appears in various places.

This is step 19/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 227163082
2019-03-29 14:43:58 -07:00
Jacques Pienaar 2a463c36b1 Add convenience wrapper for operator in tblgen
Add convenience wrapper to make it easier to iterate over attributes and operands of operator defined in TableGen file. Use this class in RewriterGen (not used in the op generator yet, will do shortly). Change the RewriterGen to pass the bound arguments explicitly, this is in preparation for multi-op matching.

PiperOrigin-RevId: 227156748
2019-03-29 14:43:43 -07:00
Chris Lattner 69f9f6e21c Merge ext/cfg/ml function printing logic in the AsmPrinter (shrinking it
by about 100 LOC), without changing any existing behavior.

This is step 20/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 227155000
2019-03-29 14:43:29 -07:00
Chris Lattner 69d9e990fa Eliminate the using decls for MLFunction and CFGFunction standardizing on
Function.

This is step 18/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 227139399
2019-03-29 14:43:13 -07:00
Jacques Pienaar f845bc4542 Fix incorrect names due to merging of tblgen tools.
PiperOrigin-RevId: 227131485
2019-03-29 14:42:57 -07:00
Chris Lattner d798f9bad5 Rename BBArgument -> BlockArgument, Op::getOperation -> Op::getInst(),
StmtResult -> InstResult, StmtOperand -> InstOperand, and remove the old names.

This is step 17/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 227121537
2019-03-29 14:42:40 -07:00
Chris Lattner 5187cfcf03 Merge Operation into OperationInst and standardize nomenclature around
OperationInst.  This is a big mechanical patch.

This is step 16/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 227093712
2019-03-29 14:42:23 -07:00
Uday Bondhugula 3b021d7f2e Update vim syntax file to highlight core ops
PiperOrigin-RevId: 227082502
2019-03-29 14:42:05 -07:00
Chris Lattner 471c976413 Rework inherentance hierarchy: Operation now derives from Statement, and
OperationInst derives from it.  This allows eliminating some forwarding
functions, other complex code handling multiple paths, and the 'isStatement'
bit tracked by Operation.

This is the last patch I think I can make before the big mechanical change
merging Operation into OperationInst, coming next.

This is step 15/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 227077411
2019-03-29 14:41:49 -07:00
Feng Liu 9b20a4ccdf add a method to get FloatAttr value as double
Sometimes we have to get the raw value of the FloatAttr to invoke APIs from
non-MLIR libraries (i.e. in the tpu_ops.inc and convert_tensor.cc files). Using
`FloatAttr::getValue().convertToFloat()` and
`FloatAttr::getValue().convertToDouble()` is not safe because interally they
checke the semantics of the APFloat in the attribute, and the semantics is not
always specified (the default value is f64 then convertToFloat will fail) or
inferred incorrectly (for example, using 1.0 instead of 1.f for IEEEFloat).
Calling these convert methods without knowing the semantics can usually crash
the compiler.

This new method converts the value of a FloatAttr to double even if it loses
precision. Currently this method can be used to read in f32 data from arrays.

PiperOrigin-RevId: 227076616
2019-03-29 14:41:34 -07:00
Chris Lattner bd24a131d3 Fix an ASAN detected bug introduced by cr/227067644. While MLFunctions always
have one block, CFG Functions don't necessarily have one (e.g. when they are
being first constructed by the parser).

PiperOrigin-RevId: 227075636
2019-03-29 14:41:19 -07:00
Chris Lattner 1b430f1d32 Delicately re-layer Operation, Statement, and OperationStmt, reworking
#includes so Statements.h includes Operation.h but nothing else does.  This is
in preparation to eliminate the Operation class and the complexity it brings
with it.  I split this patch off because it is just moving stuff around, the
next patch will be more complex.

This is step 14/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 227071777
2019-03-29 14:41:05 -07:00
Chris Lattner 4fbcd1ac52 Minor renamings: Trim the "Stmt" prefix off
StmtSuccessorIterator/StmtSuccessorIterator, and rename and move the
CFGFunctionViewGraph pass to ViewFunctionGraph.

This is step 13/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 227069438
2019-03-29 14:40:51 -07:00
Uday Bondhugula 294687ef59 Fix affine expr flattener bug introduced by cl/225452174.
- inconsistent local var constraint size when repeatedly using the same
  flattener for all expressions in a map.

PiperOrigin-RevId: 227067836
2019-03-29 14:40:37 -07:00
Chris Lattner 4c05f8cac6 Merge CFGFuncBuilder/MLFuncBuilder/FuncBuilder together into a single new
FuncBuilder class.  Also rename SSAValue.cpp to Value.cpp

This is step 12/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 227067644
2019-03-29 14:40:22 -07:00
Chris Lattner 3f190312f8 Merge SSAValue, CFGValue, and MLValue together into a single Value class, which
is the new base of the SSA value hierarchy.  This CL also standardizes all the
nomenclature and comments to use 'Value' where appropriate.  This also eliminates a large number of cast<MLValue>(x)'s, which is very soothing.

This is step 11/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 227064624
2019-03-29 14:40:06 -07:00
Chris Lattner 776b035646 Eliminate the Instruction, BasicBlock, CFGFunction, MLFunction, and ExtFunction classes, using the Statement/StmtBlock hierarchy and Function instead.
This *only* changes the internal data structures, it does not affect the user visible syntax or structure of MLIR code.  Function gets new "isCFG()" sorts of predicates as a transitional measure.

This patch is gross in a number of ways, largely in an effort to reduce the amount of mechanical churn in one go.  It introduces a bunch of using decls to keep the old names alive for now, and a bunch of stuff needs to be renamed.

This is step 10/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 227044402
2019-03-29 14:39:49 -07:00
Alex Zinenko a63f440601 LoopAnalysis: isContiguousAccess fail gracefully
Existing implementation of isContiguousAccess asserts that one of the
function arguments is within certain range, depending on another parameter.
However, the value of this argument may come from outside, in particular in the
loop vectorization pass it may come from command line arguments.  This leads
to 'mlir-opt' crashing on an assertion depending on flags.  Handle the error
gracefully by reporting error returning a negative result instead.  This
negative result prevents any further transformation by the vectorizer so the IR
remains valid.

PiperOrigin-RevId: 227029496
2019-03-29 14:39:34 -07:00
Jacques Pienaar 057984d05d Move print op stats pass to analysis.
Move PrintOpStatsPass out of tools and to other passes (moved to Analysis as it
doesn't modify the program but it is different than the other analysis passes
as it is only consumer at present is the user).

PiperOrigin-RevId: 227018996
2019-03-29 14:39:19 -07:00
Jacques Pienaar 150b1a859e Merge mlir-op-gen and mlir-rewriter-gen into mlir-tblgen.
Unify the two tools before factoring out helper classes.

PiperOrigin-RevId: 227015406
2019-03-29 14:39:05 -07:00
Chris Lattner abf72a8bb1 Rename findFunction from the ML side of the house to be named getFunction(),
making it more similar to the CFG side of things.  It is true that in a deeply
nested case that this is not a guaranteed O(1) time operation, and that 'get'
could lead compiler hackers to think this is cheap, but we need to merge these
and we can look into solutions for this in the future if it becomes a problem
in practice.

This is step 9/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 226983931
2019-03-29 14:38:49 -07:00
Mehdi Amini 4e5337601e Inline Instruction's operands as TrailingObjects
For performance/memory saving purpose, having the Instruction holding a
std::vector for the operands isn't a really good tradeoff.  The only reason for
this was to support adding/removing easily BasicBlock arguments to Terminator.
Since this isn't the most common operation, we instead force a pre-allocated
list of operands on Instructions at creation time.
PiperOrigin-RevId: 226981227
2019-03-29 14:38:34 -07:00
Chris Lattner 036f87b15f Rename CFGFunctionGraphTraits.h -> FunctionGraphTraits.h and add
graph specializations for doing CFG traversals of ML Functions, making the two
sorts of functions have the same capabilities.

This is step 8/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 226968502
2019-03-29 14:38:19 -07:00
Chris Lattner 3bd8ff6699 Eliminate the MLFuncArgument class representing arguments to MLFunctions: use the
BlockArgument arguments of the entry block instead.  This makes MLFunctions and
CFGFunctions work more similarly.

This is step 7/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 226966975
2019-03-29 14:38:04 -07:00
Chris Lattner 5ff0001dc7 Introduce a new StmtBlockList type to hold a list of StmtBlocks. Use it in
MLFunction, IfStmt, ForStmt even though they currently only contain exactly one
block in that list.

This is step 6/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 226960278
2019-03-29 14:37:49 -07:00
Feng Liu 63068da4d9 Support NameLoc and CallSiteLoc for mlir::Location
The NameLoc can be used to represent a variable, node or method. The
CallSiteLoc has two fields, one represents the concrete location and another
one represents the caller's location. Multiple CallSiteLocs can be chained as
a call stack.

For example, the following call stack
```
AAA
at file1:1
at file2:135
at file3:34
```

can be formed by call0:

```
auto name = NameLoc::get("AAA");
auto file1 = FileLineColLoc::get("file1", 1);
auto file2 = FileLineColLoc::get("file2", 135);
auto file3 = FileLineColLoc::get("file3", 34);
auto call2 = CallSiteLoc::get(file2, file3);
auto call1 = CallSiteLoc::get(file1, call2);
auto call0 = CallSiteLoc::get(name, call1);
```

PiperOrigin-RevId: 226941797
2019-03-29 14:37:34 -07:00
Alex Zinenko eb0f9f37af SuperVectorization: fix 'isa' assertion
Supervectorization uses null pointers to SSA values as a means of communicating
the failure to vectorize.  In operation vectorization, all operations producing
the values of operation arguments must be vectorized for the given operation to
be vectorized.  The existing check verified if any of the value "def"
statements was vectorized instead, sometimes leading to assertions inside `isa`
called on a null pointer.  Fix this to check that all "def" statements were
vectorized.

PiperOrigin-RevId: 226941552
2019-03-29 14:37:20 -07:00
Alex Zinenko 9403f80dd3 LLVM IR lowering: support SubIOp and SubFOp
The binary subtraction operations were not supported by the lowering because
they were not essential for the testing flow.  Add support for these
operations.

PiperOrigin-RevId: 226941463
2019-03-29 14:37:05 -07:00
Jacques Pienaar 58d50a6325 Rename convenience methods to make type explicit.
PiperOrigin-RevId: 226939383
2019-03-29 14:36:50 -07:00
Chris Lattner d613f5ab65 Refactor MLFunction to contain a StmtBlock for its body instead of inheriting
from it.  This is necessary progress to squaring away the parent relationship
that a StmtBlock has with its enclosing if/for/fn, and makes room for functions
to have more than one block in the future.  This also removes IfClause and ForStmtBody.

This is step 5/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 226936541
2019-03-29 14:36:35 -07:00
Chris Lattner 9a4060d3f5 Eliminate the ability to add operands to an instruction, used in a narrow case
for SSA values in terminators, but easily worked around.  At the same time,
move the StmtOperand list in a OperationStmt to the end of its trailing
objects list so we can *reduce* the number of operands, without affecting
offsets to the other stuff in the allocation.

This is important because we want OperationStmts to be consequtive, including
their operands - we don't want to use an std::vector of operands like
Instructions have.

This is patch 4/n towards merging instructions and statements, NFC.

PiperOrigin-RevId: 226865727
2019-03-29 14:36:20 -07:00
Chris Lattner eadaa1101c Implement StmtBlocks support for arguments and pred/succ iteration. This isn't
tested yet, but will when stuff starts switching over to it.  This is part 3/n of merging CFGFunctions and MLFunctions.

PiperOrigin-RevId: 226794787
2019-03-29 14:36:05 -07:00