This is the follow-up to D144199 and suggestion from D144045.
We make use of loop info explicit via InstCombine pass parameter
rather than semi-arbitrary via caching.
The only InstCombine transform that uses LoopInfo currently is a
GEP fold in visitGEPOfGEP(), so that shows up as a failure in the
dedicated test for the fold as well as several LoopVectorizer tests
that run extra passes.
I don't see any pass manager regression tests that actually check
for pass options, but this is intended to be NFC for the pass
pipeline behavior - we only try to use loop info where it would
have been used before via caching .
Differential Revision: https://reviews.llvm.org/D144274
This patch adds the predicate as additional operand to VPReplicateRecipe
during initial construction. The predicated recipes are later moved into
replicate regions. This simplifies constructions and some VPlan
transformations, like fixed-order recurrence handling.
It also improves codegen in some cases (e.g. for in-loop reductions),
because the recipes remain in the same block.
Reviewed By: Ayal
Differential Revision: https://reviews.llvm.org/D143865
Previously, while calculating register usage due to invariants, it was assumed that invariant would always be part of widening
instructions. This resulted in calculating vector register types for vectors which cant be legalized(check the newly added test for more details).
An invariant might not always need a vector register. For e.g., invariant might just be used for iteration check.
This patch checks if the invariant is part of any widening instruction and considers register usage accordingly. Fixes issue 60493
Differential Revision: https://reviews.llvm.org/D143422
Previously, while calculating register usage due to invariants, it was assumed that invariant would always be part of widening
instructions. This resulted in calculating vector register types for vectors which cant be legalized(check the newly added test for more details).
An invariant might not always need a vector register. For e.g., invariant might just be used for iteration check.
This patch checks if the invariant is part of any widening instruction and considers register usage accordingly. Fixes issue 60493
Differential Revision: https://reviews.llvm.org/D143422
Update a few tests to add users to loads to avoid them being optimized
out by future changes. In cases the unused loads didn't matter for the
test, remove them.
This artifact can appear from the vectorizer. (add X, -1) is the
backedge taken count. It gets zero extended and then 1 is added to
it to get the trip count.
There is usually a dominating branch that rules out X being zero.
Alive: https://alive2.llvm.org/ce/z/NsRDwX
IR is now always parsed in opaque pointer mode, unless
-opaque-pointers=0 is explicitly given. There is no automatic
detection of typed pointers anymore.
The -opaque-pointers=0 option is added to any remaining IR tests
that haven't been migrated yet.
Differential Revision: https://reviews.llvm.org/D141912
Instcombine prefers this canonical form (see getPreferredVectorIndex),
as does IRBuilder when passing the index as an integer so we may as
well use the prefered form from creation.
NOTE: All test changes are mechanical with nothing else expected
beyond a change of index type from i32 to i64.
Differential Revision: https://reviews.llvm.org/D140983
This patch adds metadata to disable runtime unrolling to the vectorized
loop. If runtime unrolling/interleaving is considered profitable, LV
will interleave the loop directly. There should be no need to perform
runtime unrolling at a later stage.
Note that we already add metadata to disable runtime unrolling to the
scalar loop after vectorization.
The additional unrolling unnecessarily increases code size and compile
time. In addition to that we have several bug reports of unncessary
runtime unrolling for vectorized loops, e.g. PR40961
Compile-time improvements:
NewPM-O3: -1.04%
NewPM-ReleaseThinLTO: -0.59%
NewPM-ReleaseLTO-g: -0.97%
https://llvm-compile-time-tracker.com/compare.php?from=ce1be13a868d0f8afa367975558c1a6175cce33a&to=78bc2e67f22e9e10e61cdb6cdac4bb857d95eb1b&stat=instructions:uFixes#40306.
Reviewed By: lebedev.ri, nikic
Differential Revision: https://reviews.llvm.org/D115261
Check lines for some of these tests were regenerated. The difference
is that with opaque pointers SCEVExpander always emits i8 GEPs,
making the address calculation explicit. This is a known problem
that will be solved long term by making all address calculations
explicit.
Code generation now uses the start VPValue of induction recipes.
This makes it possible to adjust the start value of the epilogue
vector loop to use the 'resume' value of the main vector loop.
Fixes#59459.
Reviewed By: Ayal
Differential Revision: https://reviews.llvm.org/D92132
A few more that I missed in commit 3528e63d89305907b3d6e.
There could be more duplicates remaining, since I've only focused
on exactly duplicated "RUN: opt" lines (ignoring multi line RUN
lines ending with '\').
StoredValues only has entries for members of the interleave group. If
there are gaps, then using the index i here will either access a wrong
entry or be out-of-bounds.
Instead use a dedicated index that only gets incremented for members of
the interleave group.
Fixes#59090.
All of our insert/extract ops work on 128-bit lanes.
For `Insert`, we need to extract affected 128-bit lane,
unless it's being fully overwritten (FIXME: do we need to be
careful about legalization-induced padding that we obviously don't demand?),
perform insertions, and then insert the 128-bit lane back.
But hold on. If we are operating on an 256-bit legal vector,
and thus have two 128-bit subvectors, and are fully overwriting them both,
we don't actually need to insert *both* subvectors,
only the second one, into the implicitly-widened first one.
Also, `Insert` wasn't actually querying the costs,
but just assuming them to be `1`.
`getShuffleCost(TTI::SK_ExtractSubvector)` notes:
```
// Note that in general, the insertion starting at the beginning of a vector
// isn't free, because we need to preserve the rest of the wide vector.
```
... so as far as i can tell, we didn't account for that.
I was hoping this would allow vectorization at a higher VF at one case i looked at,
but the subvector insertion cost is still dis-advising that.
The change for `Extract` is NFC, and is for consistency only,
i wanted to get rid of of that weird explicit discounting of insertion of 0'th element,
since the general code should already deal with that.
Reviewed By: RKSimon
Differential Revision: https://reviews.llvm.org/D137913
This extends the previously added uniform store case to handle stores of loop varying values to a loop invariant address. Note that the placement of this code only allows unpredicated stores; this is important for correctness. (That is "IsPredicated" is always false at this point in the function.)
This patch does not include scalable types. The diff felt "large enough" as it were; I'll handle that in a separate patch. (It requires some changes to cost modeling.)
Differential Revision: https://reviews.llvm.org/D133580
This mainly just adds costs for the targets where we have actual funnelshift/rotate instructions (VBMI2/XOP etc.) - the cases where we expand still need addressing, although for many the default shift+or expansion, especially for uniform cases, isn't that bad.
This was achieved with the 'cost-tables vs llvm-mca' script D103695
Epilogue vectorization uses isScalarAfterVectorization to check if
widened versions for inductions need to be generated and bails out in
those cases.
At the moment, there are scenarios where isScalarAfterVectorization
returns true but VPWidenPointerInduction::onlyScalarsGenerated would
return false, causing widening.
This can lead to widened phis with incorrect start values being created
in the epilogue vector body.
This patch addresses the issue by storing the cost-model decision in
VPWidenPointerInductionRecipe and restoring the behavior before 151c144.
This effectively reverts 151c144, but the long-term fix is to properly
support widened inductions during epilogue vectorization
Fixes#57712.
This is the first patch in a series intended for removing flag
-enable-new-pm=0 from lit tests. This is part of a bigger
effort of completely removing legacy code related to legacy
pass manager in favor of currently default new pass manager.
In this patch flag has been removed only from tests where no significant
change has been required because checks has been duplicated for
both PMs.
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D134150
This was originally part of D133788. There are no visible
regressions. All of the diffs show a large unsigned constant
becoming a small negative constant. This should be better
for analysis (and slightly less compile-time) and codegen.
Vector shift by const uniform is the cheapest shift instruction we have, non-const uniform have a marginally higher cost - some targets 'splat' the amount internally to use the shift-per-element instruction, others see a higher cost for the explicit zeroing of the upper bits for the (64-bit) shift amount.
This was achieved with an updated version of the 'cost-tables vs llvm-mca' script D103695 (I'll update the patch soon for reference)
VPReplicateRecipe::isUniform actually means uniform-per-parts, hence a
scalar instruction is generated per-part.
This is a potential alternative D132892. For now the current patch only
catches cases where the address is trivially invariant (defined outside
VPlan), while D132892 catches any address that is considered invariant
by SCEV AFAICT.
It should be possible to hoist fully invariant recipes feeding loads out
of the vector loop region as well, but in practice LICM should do that
already.
This version of the patch artificially limits this to loads to make it
easier to compare, but this restriction should be easily liftable.
Reviewed By: reames
Differential Revision: https://reviews.llvm.org/D133019
Don't demand low order bits from the LHS of an Add if:
- they are not demanded in the result, and
- they are known to be zero in the RHS, so they can't possibly
overflow and affect higher bit positions
This is intended to avoid a regression from a future patch to change
the order of canonicalization of ADD and AND.
Differential Revision: https://reviews.llvm.org/D130075
This extends the handling of uniform memory operations to handle the case where a store is storing a loop invariant value. Unlike the general case of a store to an invariant address where we must use the last active lane, in this case we can use any lane since all lanes must produce the same result.
For context, the basic structure of the existing code and how the change fits in:
* First, we select a widening strategy. (The result is irrelevant for this patch.)
* Then we determine if a computation is uniform within all lanes of VF. (Note this is the uniform-per-part definition, not LAI's uniform across all unrolled iterations definition.)
* If it is, we overrule the widening strategy, and unconditionally scalarize.
* VPReplicationRecipe - which is what actually does the scalarization - knows how to handle unform-per-part values including for scalable vectors. However, we do need to know that the expression is safe to execute without predication - e.g. the uniform mem op was unconditional in the original loop. (This part was split off and already landed.)
An obvious question is why not simply implement the generic case? The answer is that I'm going to, but doing so without a canonicalization towards uniform causes regressions due to bad interaction with scalarization/uniformity of values feeding the uniform mem-op. This patch is needed to avoid those regressions.
Differential Revision: https://reviews.llvm.org/D130364
