Skip to content

[NFC][SIWholeQuadMode] Perform less lookups #124927

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
merged 1 commit into from
Jan 29, 2025
Merged

[NFC][SIWholeQuadMode] Perform less lookups #124927

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
33 changes: 12 additions & 21 deletions llvm/lib/Target/AMDGPU/SIWholeQuadMode.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -218,8 +218,8 @@ class SIWholeQuadMode : public MachineFunctionPass {
bool IsWQM);
MachineInstr *lowerKillF32(MachineBasicBlock &MBB, MachineInstr &MI);

void lowerBlock(MachineBasicBlock &MBB);
void processBlock(MachineBasicBlock &MBB, bool IsEntry);
void lowerBlock(MachineBasicBlock &MBB, BlockInfo &BI);
void processBlock(MachineBasicBlock &MBB, BlockInfo &BI, bool IsEntry);

bool lowerLiveMaskQueries();
bool lowerCopyInstrs();
Expand Down Expand Up @@ -1035,12 +1035,7 @@ MachineInstr *SIWholeQuadMode::lowerKillI1(MachineBasicBlock &MBB,
// Replace (or supplement) instructions accessing live mask.
// This can only happen once all the live mask registers have been created
// and the execute state (WQM/StrictWWM/Exact) of instructions is known.
void SIWholeQuadMode::lowerBlock(MachineBasicBlock &MBB) {
auto *BII = Blocks.find(&MBB);
if (BII == Blocks.end())
return;

const BlockInfo &BI = BII->second;
void SIWholeQuadMode::lowerBlock(MachineBasicBlock &MBB, BlockInfo &BI) {
if (!BI.NeedsLowering)
return;

Expand All @@ -1052,8 +1047,9 @@ void SIWholeQuadMode::lowerBlock(MachineBasicBlock &MBB) {

for (MachineInstr &MI : llvm::make_early_inc_range(
llvm::make_range(MBB.getFirstNonPHI(), MBB.end()))) {
if (StateTransition.count(&MI))
State = StateTransition[&MI];
auto MIState = StateTransition.find(&MI);
if (MIState != StateTransition.end())
State = MIState->second;

MachineInstr *SplitPoint = nullptr;
switch (MI.getOpcode()) {
Expand Down Expand Up @@ -1260,13 +1256,8 @@ void SIWholeQuadMode::fromStrictMode(MachineBasicBlock &MBB,
StateTransition[MI] = NonStrictState;
}

void SIWholeQuadMode::processBlock(MachineBasicBlock &MBB, bool IsEntry) {
auto *BII = Blocks.find(&MBB);
if (BII == Blocks.end())
return;

BlockInfo &BI = BII->second;

void SIWholeQuadMode::processBlock(MachineBasicBlock &MBB, BlockInfo &BI,
bool IsEntry) {
// This is a non-entry block that is WQM throughout, so no need to do
// anything.
if (!IsEntry && BI.Needs == StateWQM && BI.OutNeeds != StateExact) {
Expand Down Expand Up @@ -1809,11 +1800,11 @@ bool SIWholeQuadMode::runOnMachineFunction(MachineFunction &MF) {
if (GlobalFlags & StateWQM)
Blocks[&Entry].InNeeds |= StateWQM;
// Wave mode switching requires full lowering pass.
for (auto BII : Blocks)
processBlock(*BII.first, BII.first == &Entry);
for (auto &BII : Blocks)
processBlock(*BII.first, BII.second, BII.first == &Entry);
// Lowering blocks causes block splitting so perform as a second pass.
for (auto BII : Blocks)
lowerBlock(*BII.first);
for (auto &BII : Blocks)
lowerBlock(*BII.first, BII.second);
Changed = true;
}

Expand Down