Skip to content
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

[DRAFT][s4xbf16] Add shmem swizzling heuristics for loading into LinearLayouts #23

Open
wants to merge 1 commit into
base: llvm-head
Choose a base branch
from
Open

[DRAFT][s4xbf16] Add shmem swizzling heuristics for loading into LinearLayouts #23

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
53 changes: 52 additions & 1 deletion lib/Dialect/TritonGPU/Transforms/Utility.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -996,6 +996,46 @@ StringRef getAMDArch(Operation *module) {
return ref.drop_front(4); // drop the "hip:"
}

// Rough utility for obtaining a SharedEnc for a LinearEncoding,
// as we've replaced DotOpEnc with Linear in some cases
// (specifically, fp4ToFp and similar unpack-upcast thru join)
std::optional<ttg::SwizzledSharedEncodingAttr>
getSharedForLinear(ttg::LinearEncodingAttr enc,
ArrayRef<unsigned int> globalOrder, ArrayRef<int64_t> shape,
unsigned elemBitWidth, ttg::CTALayoutAttr ctaLayout) {
auto ctx = enc.getContext();
auto ll = enc.getLinearLayout();
auto rank = shape.size();

if (rank != 2)
return std::nullopt;

auto order = enc.getOrder();
assert(globalOrder.size() == rank);
// TODO add memdesc_trans support for dot(trans(cvt(src) #linear) #dot_op)
if (order != globalOrder)
return std::nullopt;

auto innerDim = order[0];
auto outerDim = order[1];
auto contigPerWarp = enc.getContigPerWarp();

constexpr unsigned BANK_SIZE{128};
auto elemBytes = elemBitWidth / 8;

auto vec = contigPerWarp[innerDim];
auto rowSize = elemBytes * (unsigned)shape[innerDim];
auto perPhase = std::max(BANK_SIZE / rowSize, 1u);
auto maxPhase = std::max(contigPerWarp[outerDim] / perPhase, 1u);

// cp.async does not support transfer size < 4B
if (vec * elemBytes < 4 && perPhase < maxPhase)
return std::nullopt;

return ttg::SwizzledSharedEncodingAttr::get(ctx, vec, perPhase, maxPhase,
order, ctaLayout);
}

// If all the transitive uses of the given value have are used by a convert to
// the same dot operand encoding, return the shared encoding that needs to be
// used to be compatible with users' layouts. If there are incompatible shared
Expand Down Expand Up @@ -1023,7 +1063,8 @@ getSharedEncIfAllUsersAreDotEnc(Value val, bool &incompatible) {
if (!isa<ttg::LocalLoadOp, ttg::ConvertLayoutOp>(user))
return std::nullopt;
auto enc =
cast<triton::gpu::TensorOrMemDesc>(user->getResult(0).getType()).getEncoding();
cast<triton::gpu::TensorOrMemDesc>(user->getResult(0).getType())
.getEncoding();
if (isa<ttg::DotOperandEncodingAttr>(enc)) {
auto srcTy = cast<triton::gpu::TensorOrMemDesc>(val.getType());
auto CTALayout = ttg::getCTALayout(srcTy.getEncoding());
Expand All @@ -1039,6 +1080,16 @@ getSharedEncIfAllUsersAreDotEnc(Value val, bool &incompatible) {
val.getContext(), /*vec=*/1, /*perPhase=*/1, /*maxPhase=*/1,
ttg::getOrder(srcTy.getEncoding()),
ttg::getCTALayout(srcTy.getEncoding()));
} else if (auto linearEnc = dyn_cast<ttg::LinearEncodingAttr>(enc)) {
auto srcTy = cast<triton::gpu::TensorOrMemDesc>(val.getType());
auto ctaLayout = ttg::getCTALayout(srcTy.getEncoding());
auto order = ttg::getOrder(srcTy.getEncoding());
unsigned bitWidth = srcTy.getElementType().getIntOrFloatBitWidth();
auto attrOpt = getSharedForLinear(linearEnc, order, srcTy.getShape(),
bitWidth, ctaLayout);
if (!attrOpt)
return std::nullopt;
tempAttr = *attrOpt;
} else {
return std::nullopt;
}
Expand Down