-
Notifications
You must be signed in to change notification settings - Fork 15.2k
[VPlan] Add exit phi operands during initial construction (NFC). #136455
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.
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
Changes from 1 commit
e931309
a854ca7
7052337
26d7348
d0081f7
ea95862
a900c63
5c37c32
412b278
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -9392,11 +9392,7 @@ collectUsersInExitBlocks(Loop *OrigLoop, VPRecipeBuilder &Builder, | |
| continue; | ||
|
Collaborator
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Independent: is this early continue needed: given that unreachable exit blocks have been emptied of their all their recipes - including phi ones?
Contributor
Author
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Removed in e268f71, thanks
Collaborator
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Independent: loop-unswitch this condition rather than check it for all phi's?
Contributor
Author
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Will do, thanks |
||
| } | ||
|
|
||
|
Collaborator
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Above "and add the exiting VPValue as operand" is now obsolete. Worth noting that users of multiple (early) exits are excluded?
Contributor
Author
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Done, aslo dropped .. modeled in VPlan, as all are now modeled in VPlan. |
||
| PHINode &ExitPhi = ExitIRI->getIRPhi(); | ||
| BasicBlock *ExitingBB = OrigLoop->getLoopLatch(); | ||
| Value *IncomingValue = ExitPhi.getIncomingValueForBlock(ExitingBB); | ||
| VPValue *V = Builder.getVPValueOrAddLiveIn(IncomingValue); | ||
| ExitIRI->addOperand(V); | ||
| VPValue *V = ExitIRI->getOperand(0); | ||
|
Collaborator
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Worth asserting that ExitIRI has a single operand? Expected to match its single middle-block predecessor.
Contributor
Author
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Done, thanks |
||
| if (V->isLiveIn()) | ||
| continue; | ||
| assert(V->getDefiningRecipe()->getParent()->getEnclosingLoopRegion() && | ||
|
|
||
| Original file line number | Diff line number | Diff line change | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
|
|
@@ -352,6 +352,19 @@ std::unique_ptr<VPlan> PlainCFGBuilder::buildPlainCFG( | |||||||||
| Plan->getEntry()->setOneSuccessor(getOrCreateVPBB(TheLoop->getHeader())); | ||||||||||
| Plan->getEntry()->setPlan(&*Plan); | ||||||||||
|
|
||||||||||
| // Add incoming operands for the VPIRInstructions wrapping the exit phis. | ||||||||||
|
||||||||||
| // Add incoming operands for the VPIRInstructions wrapping the exit phis. | |
| // Fix VPlan loop-closed-ssa exit phi's by add incoming operands to the VPIRInstructions wrapping them. |
more consistent with above comment.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Done thanks
Outdated
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
nit: can R iterate over EB->phis()?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Done thanks
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
assert PhiR is still w/o any operand?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Done thanks
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Order of operands set here corresponds to the order of predecessors of underlying IRBB, before EB VPBB has predecessors. This inconsistency requires attention later, when these predecessors are added, possibly in a different order. May be worth leaving a note.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Done, thanks
Outdated
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
| // The exit blocks are dead, remove any recipes to make sure no users remain | |
| // that may pessimize transforms. | |
| // The exit blocks are unreachable, remove their recipes to make sure no users remain | |
| // that may pessimize transforms. |
Have dce take care of this?
The blocks themselves are collected lazily when the plan is destroyed, hence we don't simply erase the exit blocks here. But should ExitBlocks be emptied?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Doing this elsewhere is a bit tricky at the moment; the phis themselves cannot be removed, as they are part of the wrapped IR leave nodes.
We could empty them here, but some transforms may later have to add them back in the future, so might be better to keep as-is.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The case
// 2) If we require a scalar epilogue, there is no conditional branch as
// we unconditionally branch to the scalar preheader. Do nothing.
is handle by early return above. Better place the explanation earlier, and replace "Do nothing" with "Empty the unreachable exit blocks of their recipes". The connection from scalar loop to exit blocks is (currently) outside of VPlan's scope.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Done, thanks. Also changed 2->1 as this is the first handled case.
| Original file line number | Diff line number | Diff line change | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
|
@@ -1139,7 +1139,7 @@ InstructionCost VPIRInstruction::computeCost(ElementCount VF, | |||||||||||||
| void VPIRInstruction::extractLastLaneOfOperand(VPBuilder &Builder) { | ||||||||||||||
| assert(isa<PHINode>(getInstruction()) && | ||||||||||||||
|
Collaborator
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Independent: should this be implemented for VPIRPhi instead?
Contributor
Author
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Will update separately, thanks |
||||||||||||||
| "can only add exiting operands to phi nodes"); | ||||||||||||||
|
||||||||||||||
| "can only add exiting operands to phi nodes"); | |
| "can only update exiting operands to phi nodes"); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Doen thanks.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
And yet we continue to extract the first operand, only? extractLastLaneOf[First]Operand()
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yep, updated the name, thanks
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
| VPValue *Exiting = getOperand(0); | |
| if (!Exiting->isLiveIn()) { | |
| if (Exiting->isLiveIn()) | |
| return; | |
| VPValue *Exiting = getOperand(0); |
?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Will adjust separately, thanks
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
done in 71f2c1e
| Original file line number | Diff line number | Diff line change | ||||||
|---|---|---|---|---|---|---|---|---|
|
|
@@ -2501,35 +2501,41 @@ void VPlanTransforms::handleUncountableEarlyExit( | |||||||
| if (!ExitIRI) | ||||||||
|
||||||||
| break; | ||||||||
|
|
||||||||
| PHINode &ExitPhi = ExitIRI->getIRPhi(); | ||||||||
| VPValue *IncomingFromEarlyExit = RecipeBuilder.getVPValueOrAddLiveIn( | ||||||||
| ExitPhi.getIncomingValueForBlock(UncountableExitingBlock)); | ||||||||
|
|
||||||||
| unsigned EarlyExitIdx = 0; | ||||||||
|
||||||||
| unsigned EarlyExitIdx = 0; | |
| // By default, assume early exit operand is first, e.g., when the two exit blocks are distinct - VPEarlyExitBlock has a single predecessor. | |
| unsigned EarlyExitIdx = 0; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Added, thanks
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hmm, perhaps its better instead to set EarlyExitIdx to be the last operand, consistently?
If VPEarlyExitBlock has two predecessors, they are already ordered such that early exit is second, by construction. But its underlying IRBB may have its predecessors ordered the other way around, and it is this order which corresponds to the order of operands of VPEarlyExitBlock's phi recipes. Therefore, if early exit is the first predecessor of the underlying IRBB, we swap the operands of phi recipes, bringing them to match VPEarlyExitBlock's predecessor order with early exit being last (second).
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Independent: better ask instead if VPEarlyExitBlock has two predecessors? One (the last) was added above - VectorEarlyExitVPBB. This should match the case where OrigLoop has a unique exit block - which would be aka VPEarlyExitBlock, or rather its underlying EarlyExitIRBB.
| if (OrigLoop->getUniqueExitBlock()) { | |
| if (VPEarlyExitBlock->getNumPredecessors() == 2) { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Will check separately, thanks
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ok. It's more consistent with following explanation.
Outdated
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
| // After the transform, the first incoming value is coming from the | |
| // The incoming values currently correspond to the original IR predecessors. | |
| // After the transform, the first incoming value should correspond to the |
?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Added, thanks
Outdated
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
| // orignial loop latch, while the second operand is from the early exit. | |
| // original loop latch, while the second operand to the early exit. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Fixed thanks
Outdated
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
| // Sawp the phi operands, if the first predecessor in the original IR is | |
| // Swap the phi operands if the first predecessor in the original IR is |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Fixed, thanks
Outdated
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is this clearer?
| if (*pred_begin(VPEarlyExitBlock->getIRBasicBlock()) != | |
| OrigLoop->getLoopLatch()) | |
| if (*pred_begin(EarlyExitIRBB) == UncountableExitingBlock) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Done thanks
Outdated
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
| // If there's a unique exit block, VPEarlyExitBlock has 2 predecessors |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Done thanks
Outdated
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Independent: worth indicating that early.exit and latch.exit may be the same block, in https://llvm.org/docs/Vectorizers.html#early-exit-vectorization
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Will do separately, thanks
Outdated
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We're no longer adding this operand here, but only updating it with an extract.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Done thanks
Outdated
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Independent: the extract from the latch created for (first) operand uses extractLastLaneOfOperand(), the extract from early exit replaced below explicitly. Can this be done more consistently?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Not sure, we could as follow-up add the extracts from last lane for all exit phis during construction, then have the extract from the early exit updated here?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Independent: clamping range inside VPlanTransform? Limiting the range to scalar VF - is another VPlan constructed for the vector (sub)range?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, this is done early on when we are clamping the loop range for other reasons as well.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
(Independent) Still puzzled about clamping the range at this stage, when all VPlans were already created following range clampings. Rather than say asserting that the range contains either scalar or vector VF's but not both, and introduce extracts if it's the latter. Extracts added above for latch exit need not check vector VF's?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
the transform isn't run when all VPlans have been created, but in tryToBuildVPlanWithVPRecipes where we the range is also clamped in other places ( https://github.com/llvm/llvm-project/blob/main/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp#L9758)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ahh, very well, thanks.
(Independent) May be good to indicate somehow transforms that may clamp the range - that belong to tryToBuildVPlanWithVPRecipes stage, and prevent them from operating afterwards, perhaps by disabling range clamping afterwards?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The range should only be available in tryToBuildVPlanWithVPRecipes and not subsequent optimizations; we could seaprate things even clearer by first building all VPlans (i.e. only do tryToBuildVPlanWithVPRecipes) and then optimize all VPlans separatel
Outdated
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can continue to (re)set IncomingFromEarlyExit here, followed by resetting the operand (instead of adding it):
ExitIRI->setOperand(EarlyExitIdx, IncomingFromEarlyExit);
?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Done ,thanks
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Independent: can R iterate over ExitVPBB->phis() above?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Updated in 8c83355, thanks