Skip to content

[BACKEND] Relax patterns to move sink broadcast and hoist convert #2331

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ptillet merged 1 commit into from
Sep 18, 2023
Merged

[BACKEND] Relax patterns to move sink broadcast and hoist convert #2331

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 6 additions & 4 deletions lib/Dialect/Triton/Transforms/ReorderBroadcast.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -116,18 +116,20 @@ struct MoveBroadcastAfterElementwisePattern

auto operands = op->getOperands();
bool seenBroadcast = false;
Type srcType;
ArrayRef<int64_t> srcShape;
for (auto operand : operands) {
auto definingOp = operand.getDefiningOp();
if (!definingOp) {
return mlir::failure();
}

auto getSrcShape = [](triton::BroadcastOp b) {
return b.getSrc().getType().cast<RankedTensorType>().getShape();
};
if (auto broadcastOp = llvm::dyn_cast<triton::BroadcastOp>(definingOp)) {
if (!seenBroadcast) {
seenBroadcast = true;
srcType = broadcastOp.getSrc().getType();
} else if (srcType != broadcastOp.getSrc().getType()) {
srcShape = getSrcShape(broadcastOp);
} else if (srcShape != getSrcShape(broadcastOp)) {
// If the broadcast have different types we cannot re-order.
return mlir::failure();
}
Expand Down
16 changes: 11 additions & 5 deletions lib/Dialect/TritonGPU/Transforms/RemoveLayoutConversions.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -929,7 +929,7 @@ static void hoistConvertOnTopOfExtOrBroadcast(ConvertLayoutOp convertOp) {

auto isExtOrBroadcastOp = [](Operation *op) {
return isa<arith::ExtSIOp, arith::ExtUIOp, arith::ExtFOp,
triton::BroadcastOp>(op);
triton::BroadcastOp, triton::ExpandDimsOp>(op);
};
// 1. Take a backward slice of all the tensor dependencies.
SetVector<Value> slice;
Expand All @@ -950,8 +950,11 @@ static void hoistConvertOnTopOfExtOrBroadcast(ConvertLayoutOp convertOp) {
if (isExtOrBroadcastOp(op)) {
SetVector<Value> tempSlice;
DenseMap<Value, Attribute> tempLayout;
std::optional<Attribute> srcEncoding = inferSrcEncoding(op, layout[v]);
if (!srcEncoding)
return;
LogicalResult result = getRematerializableSlice(
op->getOperand(0), layout[v], tempSlice, tempLayout);
op->getOperand(0), *srcEncoding, tempSlice, tempLayout);
// If we can rematerialize the rest of the ext slice we can ignore this
// ext as it won't need a convert.
if (result.succeeded()) {
Expand All @@ -969,13 +972,16 @@ static void hoistConvertOnTopOfExtOrBroadcast(ConvertLayoutOp convertOp) {

if (extOrBroadcatOp == nullptr)
return;
std::optional<Attribute> srcEncoding =
inferSrcEncoding(extOrBroadcatOp, layout[extOrBroadcatOp->getResult(0)]);
if (!srcEncoding)
return;
// Move the convert before the ext op and rewrite the slice.
OpBuilder builder(extOrBroadcatOp);
auto tensorType =
extOrBroadcatOp->getOperand(0).getType().cast<RankedTensorType>();
auto newType =
RankedTensorType::get(tensorType.getShape(), tensorType.getElementType(),
layout[extOrBroadcatOp->getResult(0)]);
auto newType = RankedTensorType::get(
tensorType.getShape(), tensorType.getElementType(), *srcEncoding);
auto newConvertOp = builder.create<ConvertLayoutOp>(
convertOp.getLoc(), newType, extOrBroadcatOp->getOperand(0));
IRMapping mapping;
Expand Down
12 changes: 12 additions & 0 deletions test/Triton/reorder-broadcast.mlir
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -53,3 +53,15 @@ tt.func @test_broadcast_binary_op_pattern(%arg0: tensor<128x1xf32>, %arg1: tenso

tt.return %mul, %mul1 : tensor<128x128xf32>, tensor<128x128xf32>
}

// CHECK-LABEL: @test_broadcast_mix_type_op_pattern
tt.func @test_broadcast_mix_type_op_pattern(%arg0: tensor<128x1xf32>, %arg1: f32, %arg2: tensor<1x128xf32>, %arg3: tensor<128x1xi1>) -> (tensor<128x128xf32>) {
// CHECK: %[[sel:.*]] = arith.select %{{.*}}, %{{.*}}, %{{.*}} : tensor<128x1xi1>, tensor<128x1xf32>
// CHECK-NEXT: %{{.*}} = tt.broadcast %[[sel]] : (tensor<128x1xf32>) -> tensor<128x128xf32>
%broadcast0 = tt.broadcast %arg0 : (tensor<128x1xf32>) -> tensor<128x128xf32>
%broadcast1 = tt.splat %arg1 : (f32) -> tensor<128x128xf32>
%cond = tt.broadcast %arg3 : (tensor<128x1xi1>) -> tensor<128x128xi1>
%sel = arith.select %cond, %broadcast0, %broadcast1 : tensor<128x128xi1>, tensor<128x128xf32>

tt.return %sel : tensor<128x128xf32>
}
4 changes: 3 additions & 1 deletion test/TritonGPU/combine.mlir
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1189,10 +1189,12 @@ module attributes {"triton_gpu.num-warps" = 2 : i32} {

// CHECK-LABEL: reduce_cvt2
// Match the reduction
// CHECK-NOT: triton_gpu.convert_layout
// CHECK: tt.reduce
// CHECK-SAME: axis = 1
// CHECK: (tensor<1x256xf32, #blocked>) -> tensor<1xf32, #triton_gpu.slice<{dim = 1, parent = #blocked}>>
// CHECK: (tensor<1x256xf32, #{{.*}}>) -> tensor<1xf32, #triton_gpu.slice<{dim = 1, parent = #{{.*}}}>>
// CHECK: triton_gpu.convert_layout
// CHECK: tt.expand_dims
// CHECK-NOT: triton_gpu.convert_layout
// CHECK: tt.return
#blocked = #triton_gpu.blocked<{sizePerThread = [1, 1], threadsPerWarp = [1, 32], warpsPerCTA = [1, 4], order = [0, 1], CTAsPerCGA = [1, 1], CTASplitNum = [1, 1], CTAOrder = [1, 0]}>
Expand Down