shylie/llvm-project
1
0
Fork 0
Code Issues Pull Requests Actions 6 Packages Projects Releases Wiki Activity

[mlir][linalg] Reuploading: Apply shortened printing/parsing form to linalg.reduce.

Browse Source 
Differential Revision: https://reviews.llvm.org/D141259
This commit is contained in:
Aliia Khasanova 2023-01-09 08:25:03 +01:00
parent 3602d852a5
commit ef545ef62a
4 changed files with 61 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
mlir/include/mlir/Dialect/Linalg/IR/LinalgStructuredOps.td
View File

@ -255,6 +255,16 @@ def MapOp : LinalgStructuredBase_Op<"map", [
linalg.yield %0: f32
}
```
Shortened print form is available. Applies to simple maps with one
non-yield operation inside the body.
The example above will be printed as:
```
%add = linalg.map { arith.addf }
ins(%lhs, %rhs : tensor<64xf32>, tensor<64xf32>)
outs(%init: tensor<64xf32>)
```
}];
let arguments = (ins
@ -329,10 +339,22 @@ def ReduceOp : LinalgStructuredBase_Op<"reduce", [
outs(%init:tensor<16x64xf32>)
dimensions = [1]
(%in: f32, %out: f32) {
%0 = arith.addf %in, %out: f32
%0 = arith.addf %out, %in: f32
linalg.yield %0: f32
}
```
Shortened print form is available. Applies to simple (not variadic) reduces
with one non-yield operation inside the body. Applies only if the operation
takes `%out` as the first argument.
The example above will be printed as:
```
%reduce = linalg.reduce { arith.addf }
ins(%input:tensor<16x32x64xf32>)
outs(%init:tensor<16x64xf32>)
dimensions = [1]
```
}];
let arguments = (ins

43
mlir/lib/Dialect/Linalg/IR/LinalgOps.cpp
View File

@ -1046,7 +1046,8 @@ void MapOp::build(
static void addBodyWithPayloadOp(OpAsmParser &parser, OperationState &result,
const OperationName &payloadOpName,
const NamedAttrList &payloadOpAttrs,
ArrayRef<Value> operands) {
ArrayRef<Value> operands,
bool initFirst = false) {
OpBuilder b(parser.getContext());
Region *body = result.addRegion();
Block &block = body->emplaceBlock();
@ -1056,14 +1057,24 @@ static void addBodyWithPayloadOp(OpAsmParser &parser, OperationState &result,
block.addArgument(operand.getType().cast<ShapedType>().getElementType(),
b.getUnknownLoc());
}
SmallVector<Value> payloadOpOperands;
// If initFirst flag is enabled, we consider init as the first position of
// payload operands.
if (initFirst) {
payloadOpOperands.push_back(block.getArguments().back());
for (const auto &arg : block.getArguments().drop_back())
payloadOpOperands.push_back(arg);
} else {
payloadOpOperands = {block.getArguments().begin(),
block.getArguments().end()};
}
Operation *payloadOp = b.create(
result.location, b.getStringAttr(payloadOpName.getStringRef()),
block.getArguments(),
payloadOpOperands,
TypeRange{
result.operands.back().getType().cast<ShapedType>().getElementType()},
payloadOpAttrs);
b.create<YieldOp>(result.location, payloadOp->getResults());
}
@ -1102,7 +1113,9 @@ ParseResult MapOp::parse(OpAsmParser &parser, OperationState &result) {
// Retrieve the operation from the body, if it is the only one (except
// yield) and if it gets the same amount of arguments as the body does.
static Operation *findPayloadOp(Block *body) {
// If initFirst flag is enabled, we check that init takes the first position in
// operands of payload.
static Operation *findPayloadOp(Block *body, bool initFirst = false) {
if (body->getOperations().size() != 2)
return nullptr;
Operation &payload = body->getOperations().front();
@ -1111,10 +1124,22 @@ static Operation *findPayloadOp(Block *body) {
if (payload.getNumOperands() == 0 ||
payload.getNumOperands() != body->getNumArguments())
return nullptr;
for (const auto &[bbArg, operand] :
llvm::zip(payload.getOperands(), body->getArguments())) {
if (bbArg != operand)
if (initFirst) {
// check init
if (payload.getOperands().back() != body->getArgument(0))
return nullptr;
// check rest
for (const auto &[operand, bbArg] :
llvm::zip(payload.getOperands(), body->getArguments().drop_front())) {
if (bbArg != operand)
return nullptr;
}
} else {
for (const auto &[operand, bbArg] :
llvm::zip(payload.getOperands(), body->getArguments())) {
if (bbArg != operand)
return nullptr;
}
}
return &payload;
}
@ -1313,7 +1338,7 @@ ParseResult ReduceOp::parse(OpAsmParser &parser, OperationState &result) {
if (payloadOpName.has_value()) {
addBodyWithPayloadOp(parser, result, payloadOpName.value(), payloadOpAttrs,
makeArrayRef(result.operands));
makeArrayRef(result.operands), /*initFirst=*/true);
} else {
SmallVector<OpAsmParser::Argument> regionArgs;
if (parser.parseArgumentList(regionArgs, OpAsmParser::Delimiter::Paren,
@ -1336,7 +1361,7 @@ static void printDenseI64ArrayAttr(OpAsmPrinter &p, StringRef attributeName,
void ReduceOp::print(OpAsmPrinter &p) {
Block *mapper = getBody();
Operation *payloadOp = findPayloadOp(mapper);
Operation *payloadOp = findPayloadOp(mapper, /*initFirst=*/true);
if (payloadOp) {
printShortForm(p, payloadOp);
}

2
mlir/test/Dialect/Linalg/one-shot-bufferize.mlir
View File

@ -363,7 +363,7 @@ func.func @reduce(%input: tensor<16x32x64xf32>,
outs(%init:tensor<16x64xf32>)
dimensions = [1]
(%in: f32, %out: f32) {
%0 = arith.addf %in, %out: f32
%0 = arith.addf %out, %in: f32
linalg.yield %0: f32
}
func.return %reduce : tensor<16x64xf32>

6
mlir/test/Dialect/Linalg/roundtrip.mlir
View File

@ -414,7 +414,7 @@ func.func @reduce(%input: tensor<16x32x64xf32>,
outs(%init:tensor<16x64xf32>)
dimensions = [1]
(%in: f32, %out: f32) {
%0 = arith.addf %in, %out: f32
%0 = arith.addf %out, %in: f32
linalg.yield %0: f32
}
func.return %reduce : tensor<16x64xf32>
@ -433,7 +433,7 @@ func.func @reduce_memref(%input: memref<16x32x64xf32>,
outs(%init:memref<16x64xf32>)
dimensions = [1]
(%in: f32, %out: f32) {
%0 = arith.addf %in, %out: f32
%0 = arith.addf %out, %in: f32
linalg.yield %0: f32
}
func.return
@ -587,7 +587,7 @@ func.func @reduce_arith_with_attr(%input: tensor<16x32x64xf32>,
outs(%init:tensor<16x64xf32>)
dimensions = [1]
(%in: f32, %out: f32) {
%0 = arith.addf %in, %out fastmath<fast> : f32
%0 = arith.addf %out, %in fastmath<fast> : f32
linalg.yield %0: f32
}
func.return %reduce : tensor<16x64xf32>