We previously had a different interpretation of unroll transformation
attributes than how LoopUnroll interpreted it. In particular,
llvm.loop.unroll.enable was needed explicitly to enable it and disabling
metadata was ignored.
Additionally, it required that either full unrolling or an unroll factor
to be specified or fail otherwise. An unroll factor is still required,
but the transformation is ignored with the hope that LoopUnroll is going
to apply the unrolling, since Polly currently does not implement an
heuristic.
Fixes llvm.org/PR50109
We enumerated the cross product Domain x Scatter, but sorted only be the
scatter key. In case there are are multiple statement instances per
scatter value, the order between statement instances of the same loop
iteration was undefined.
Propertly enumerate and sort only by the scatter value, and group the
domains using the scatter dimension again.
Thanks to Leonard Chan for the report.
Make Polly look for unrolling metadata (https://llvm.org/docs/TransformMetadata.html#loop-unrolling) that is usually only interpreted by the LoopUnroll pass and apply it to the SCoP's schedule.
While not that useful by itself (there already is an unroll pass), it introduces mechanism to apply arbitrary loop transformation directives in arbitrary order to the schedule. Transformations are applied until no more directives are found. Since ISL's rescheduling would discard the manual transformations and it is assumed that when the user specifies the sequence of transformations, they do not want any other transformations to apply. Applying user-directed transformations can be controlled using the `-polly-pragma-based-opts` switch and is enabled by default.
This does not influence the SCoP detection heuristic. As a consequence, loop that do not fulfill SCoP requirements or the initial profitability heuristic will be ignored. `-polly-process-unprofitable` can be used to disable the latter.
Other than manually editing the IR, there is currently no way for the user to add loop transformations in an order other than the order in the default pipeline, or transformations other than the one supported by clang's LoopHint. See the `unroll_double.ll` test as example that clang currently is unable to emit. My own extension of `#pragma clang loop` allowing an arbitrary order and additional transformations is available here: https://github.com/meinersbur/llvm-project/tree/pragma-clang-loop. An effort to upstream this functionality as `#pragma clang transform` (because `#pragma clang loop` has an implicit transformation order defined by the loop pipeline) is D69088.
Additional transformations from my downstream pragma-clang-loop branch are tiling, interchange, reversal, unroll-and-jam, thread-parallelization and array packing. Unroll was chosen because it uses already-defined metadata and does not require correctness checks.
Reviewed By: sebastiankreutzer
Differential Revision: https://reviews.llvm.org/D97977
This reverts commit 329aeb5db4,
and relands commit 61f006ac65.
This is a continuation of D89456.
As it was suggested there, now that SCEV models `PtrToInt`,
we can try to improve SCEV's pointer handling.
In particular, i believe, i will need this in the future
to further fix `SCEVAddExpr`operation type handling.
This removes special handling of `ConstantPointerNull`
from `ScalarEvolution::createSCEV()`, and add constant folding
into `ScalarEvolution::getPtrToIntExpr()`.
This way, `null` constants stay as such in SCEV's,
but gracefully become zero integers when asked.
Reviewed By: Meinersbur
Differential Revision: https://reviews.llvm.org/D98147
This is a continuation of D89456.
As it was suggested there, now that SCEV models `PtrToInt`,
we can try to improve SCEV's pointer handling.
In particular, i believe, i will need this in the future
to further fix `SCEVAddExpr`operation type handling.
This removes special handling of `ConstantPointerNull`
from `ScalarEvolution::createSCEV()`, and add constant folding
into `ScalarEvolution::getPtrToIntExpr()`.
This way, `null` constants stay as such in SCEV's,
but gracefully become zero integers when asked.
Reviewed By: Meinersbur
Differential Revision: https://reviews.llvm.org/D98147
Emit llvm.loop.parallel_accesses metadata instead of
llvm.mem.parallel_loop_access. The latter is deprecated because it
assumes that LoopIDs are persistent, which they are not.
We also emit parallel access metadata for all surrounding parallel
loops, not just the innermost parallel.
DetectionContext objects are stored as values in a DenseMap. When the
DenseMap reaches its maximum load factor, it is resized and all its
objects moved to a new memory allocation. Unfortunately Scop object have
a reference to its DetectionContext. When the DenseMap resizes, all the
DetectionContexts reference now point to invalid memory, even if caused
by an unrelated DetectionContext.
Even worse, NewPM's ScopPassManager called isMaxRegionInScop with the
Verify=true parameter before each pass. This caused the old
DetectionContext to be removed an a new on created and re-verified.
Of course, the Scop object was already created pointing to the old
DetectionContext. Because the new DetectionContext would
usually be stored at the same position in the DenseMap, the reference
would usually reference the new DetectionContext of the same Region.
Usually.
If not, the old position still points to memory in the DenseMap
allocation (unless also a resizing occurs) such that tools like Valgrind
and AddressSanitizer would not be able to diagnose this.
Instead of storing the DetectionContext inside the DenseMap, use a
std::unique_ptr to a DetectionContext allocation, i.e. it will not move
around anymore. This also allows use to remove the very strange
DetectionContext(const DetectionContext &&)
copy/move(?) constructor. DetectionContext objects now are neither
copied nor moved.
As a result, every re-verification of a DetectionContext will use a new
allocation. Therefore, once a Scop object has been created using a
DetectionContext, it must not be re-verified (the Scop data structure
requires its underlying Region to not change before code generation
anyway). The NewPM may call isMaxRegionInScop only with
Validate=false parameter.
In addition to that regression tests should not test the intire pass
pipeline (unless they are testing the pipeline itself), the Polly-ACC
currently does not support the new pass manager. If enabled by default,
such tests will therefore fail.
Use the -polly-gpu-runtime and -polly-gpu-arch options also as default
values for the PPCGCodeGeneration pass. This requires to move the option
to be moved from the pipeline-building Register passes to the
PPCGCodeGeneration implementation.
Fixes the spir-typesize.ll buildbot fail.
ZoneAlgorithms's computePHI relies on being provided with consistent a
schedule to compute the statement prodecessors of a statement containing
PHINodes. Otherwise unexpected results such as PHI nodes with multiple
predecessors can occur which would result in problems in the
algorithms expecting consistent data.
In the added test case, statement instances are scrubbed from the
SCoP their execution would result in undefined behavior (Due to a nsw
overflow). As already being undefined behavior in LLVM-IR, neither
AssumedContext nor InvalidContext are updated, giving computePHI no
means to avoid these cases.
Intoduce a new SCoP property, the DefinedBehaviorContext, that among
the runtime-checked conditions, also tracks the assumptions not needing
a runtime check, in particular those affecting the assumed control flow.
This replaces the manual combination of the 3 other contexts that was
already done in computePHI and setNewAccessRelation. Currently, the only
additional assumption is that loop induction variables will nsw flag for
not wrap, but potentially more can be added. Use in
hasFeasibleRuntimeContext, isl::ast_build and gisting are other
potential uses.
To limit computational complexity, the DefinedBehaviorContext is not
availabe if it grows too large (atm hardcoded to 8 disjuncts).
Possible other fixes include bailing out in computePHI when
inconsistencies are detected, choose an arbitrary value for inconsistent
cases (since it is undefined behavior anyways), or make the code
receiving the result from ComputePHI handle inconsistent data. All of
them reduce the quality of implementation having to bail out more often
and disabling the ability to assert on actually wrong results.
This fixes llvm.org/PR48783.
In preparation for turning on opt's -enable-new-pm by default, this pins
uses of passes via the legacy "opt -passname" with pass names beginning
with "polly-" and "polyhedral-info" to the legacy PM. Many of these
tests use -analyze, which isn't supported in the new PM.
(This doesn't affect uses of "opt -passes=passname").
rL240766 accidentally removed `-polly-prepare` in
phi_not_grouped_at_top.ll, and it also doesn't use the output of
-analyze.
Reviewed By: Meinersbur
Differential Revision: https://reviews.llvm.org/D94266
This fixes llvm.org/PR48554
Some test cases had to be updated because the hash function for
union_maps have been changed which affects the output order.
This patch updates IRBuilder to create insertelement/shufflevector using poison as a placeholder.
Reviewed By: nikic
Differential Revision: https://reviews.llvm.org/D93793
MemoryAccess::setNewAccessRelation() in assert-builds checks whether the
access relation for a READ has a memory location for every instance of
the domain. Otherwise, we would not have value to load from. That check
already considered that instances outside the Scop's context do not
matter since they are never executed (or would be undefined behavior).
In this patch also take instances of the InvalidContext into account,
as these can also be assumed to never occur. InvalidContext was
introduced to avoid the computational complexity of subtracting
restrictions from the AssumedContext. However, this additional check in
setNewAccessRelation is only done in assert-builds.
The assertion case with an InvalidContext may occur with DeLICM on a
conditionally infinite loops, as it is the case in the following code:
for (int i = 0; i < n; i+=b)
vreg = ...;
*Dest = vreg;
The loop is infinite when b=0, and [b] -> { : b = 0 } is part of the
InvalidContext. When DeLICM tries to map the memory for %vreg to *Dest,
there is no store instance that uses the value of vreg when b = 0, hence
no location to map it to. However, the case is irrelevant since Polly's
runtime condition check ensures that this is never case.
Fixes llvm.org/PR48445
ScalarEvolution::getSCEV cannot be used during codegen. ScalarEvolution
assumes a stable IR and control flow which is under construction during
Polly's CodeGen. In particular, it uses DominatorTree for compute the
backedge taken count. However the DominatorTree is not updated during
codegen.
In this case, SCEV was used to determine the base pointer of an array
access. Replace it by our own function. Polly generates only GEP and
BitCasts for array acceses, i.e. it is sufficient to handle these to to
find the base pointer.
Fixes llvm.org/PR48422
Operand tree forwarding can cause the change of an access kind; in
particular change from a scalar kind to an array kind if the scalar
dependency is not necessary. Such an access cannot and doesn't need to
be forwarded anymore.
Fixes llvm.org/PR48034
Print to dbgs() any taken action.
Also, read-only scalars do not require any action unless
-polly-analyze-read-only-scalars=true is used. Better refect this by
using ForwardingAction::triviallyForwardable and thus not bumping the
statistics.
ScopBuilder distributes independent instructions between statements.
Only modeled (e.g. not synthesizable) instructions are represented.
To compute independence, non-modeled instructions were used in some
parts of determining instruction independence, which could lead to the
re-introduction of non-model instructions.
In particular, required invariant loads could be added to instruction
list, which then led to redundant MemoryAccesses for such a load.
This fixes llvm.org/PR48059.
If we've got an SCEVPtrToIntExpr(op), where op is not an SCEVUnknown,
we want to sink the SCEVPtrToIntExpr into an operand,
so that the operation is performed on integers,
and eventually we end up with just an `SCEVPtrToIntExpr(SCEVUnknown)`.
Reviewed By: mkazantsev
Differential Revision: https://reviews.llvm.org/D89692
And use it to model LLVM IR's `ptrtoint` cast.
This is essentially an alternative to D88806, but with no chance for
all the problems it caused due to having the cast as implicit there.
(see rG7ee6c402474a2f5fd21c403e7529f97f6362fdb3)
As we've established by now, there are at least two reasons why we want this:
* It will allow SCEV to actually model the `ptrtoint` casts
and their operands, instead of treating them as `SCEVUnknown`
* It should help with initial problem of PR46786 - this should eventually allow us
to not loose pointer-ness of an expression in more cases
As discussed in [[ https://bugs.llvm.org/show_bug.cgi?id=46786 | PR46786 ]], in principle,
we could just extend `SCEVUnknown` with a `is ptrtoint` cast, because `ScalarEvolution::getPtrToIntExpr()`
should sink the cast as far down into the expression as possible,
so in the end we should always end up with `SCEVPtrToIntExpr` of `SCEVUnknown`.
But i think that it isn't the best solution, because it doesn't really matter
from memory consumption side - there probably won't be *that* many `SCEVPtrToIntExpr`s
for it to matter, and it allows for much better discoverability.
Reviewed By: mkazantsev
Differential Revision: https://reviews.llvm.org/D89456
Recursively traversing the operand tree leads to an exponential blowup
if instructions are used multiple times due to every path leading to an
additional copy of the instructions after forwarding. This problem was
marked as a TODO in the code and was reported as a bug in llvm.org/PR47340.
Fix by caching already visited instructions and returning the cached
version when already visited. Instead of calling forwardTree() twice,
return a ForwardingAction structure that contains a lambda which will
carry-out the forwarding when requested. The lambdas are executed in
reverse-postorder to mimic the previous recursive calls unless there
is a reuse.
Fixes llvm.org/PR47340
Polly incorrectly dropped the address space specified for a load instruction when it vectorized the code.
Reviewed By: Meinersbur
Differential Revision: https://reviews.llvm.org/D88907
While we haven't encountered an earth-shattering problem with this yet,
by now it is pretty evident that trying to model the ptr->int cast
implicitly leads to having to update every single place that assumed
no such cast could be needed. That is of course the wrong approach.
Let's back this out, and re-attempt with some another approach,
possibly one originally suggested by Eli Friedman in
https://bugs.llvm.org/show_bug.cgi?id=46786#c20
which should hopefully spare us this pain and more.
This reverts commits 1fb6104293,
7324616660,
aaafe350bb,
e92a8e0c74.
I've kept&improved the tests though.
This relands commit 1c021c64ca which was
reverted in commit 17cec6a11a because
an assertion was being triggered, since `BuildConstantFromSCEV()`
wasn't updated to handle the case where the constant we want to truncate
is actually a pointer. I was unsuccessful in coming up with a test case
where we'd end there with constant zext/sext of a pointer,
so i didn't handle those cases there until there is a test case.
Original commit message:
While we indeed can't treat them as no-ops, i believe we can/should
do better than just modelling them as `unknown`. `inttoptr` story
is complicated, but for `ptrtoint`, it seems straight-forward
to model it just as a zext-or-trunc of unknown.
This may be important now that we track towards
making inttoptr/ptrtoint casts not no-op,
and towards preventing folding them into loads/etc
(see D88979/D88789/D88788)
Reviewed By: mkazantsev
Differential Revision: https://reviews.llvm.org/D88806
> While we indeed can't treat them as no-ops, i believe we can/should
> do better than just modelling them as `unknown`. `inttoptr` story
> is complicated, but for `ptrtoint`, it seems straight-forward
> to model it just as a zext-or-trunc of unknown.
>
> This may be important now that we track towards
> making inttoptr/ptrtoint casts not no-op,
> and towards preventing folding them into loads/etc
> (see D88979/D88789/D88788)
>
> Reviewed By: mkazantsev
>
> Differential Revision: https://reviews.llvm.org/D88806
It caused the following assert during Chromium builds:
llvm/lib/IR/Constants.cpp:1868:
static llvm::Constant *llvm::ConstantExpr::getTrunc(llvm::Constant *, llvm::Type *, bool):
Assertion `C->getType()->isIntOrIntVectorTy() && "Trunc operand must be integer"' failed.
See code review for a link to a reproducer.
This reverts commit 1c021c64ca.
While we indeed can't treat them as no-ops, i believe we can/should
do better than just modelling them as `unknown`. `inttoptr` story
is complicated, but for `ptrtoint`, it seems straight-forward
to model it just as a zext-or-trunc of unknown.
This may be important now that we track towards
making inttoptr/ptrtoint casts not no-op,
and towards preventing folding them into loads/etc
(see D88979/D88789/D88788)
Reviewed By: mkazantsev
Differential Revision: https://reviews.llvm.org/D88806
VirtualUse of type UseKind::Inter expects the definition of a
llvm::Value to be represented in another statement. In the bug report
that statement has been removed due to its domain being empty.
Scop::InstStmtMap for the llvm::Value's defintion still pointed to the
removed statement, which resulted in the use-after-free.
The defintion statement was removed by Simplify because it was
considered to not be reachable by other uses; trivially because it is
never executed due to its empty domain. However, no such thing happend
to the using statement using the value altough its domain is also empty.
Fix by always removing statements with empty domains in Simplify since
these are not properly analyzable. A UseKind::Inter should always have a
statement with its defintion due to LLVM's SSA form.
Scop::removeStmtNotInDomainMap() also removes statements with empty
domains but does so without considering the context as used by
Simplify's analyzes.
In another angle, InstStmtMap pointing to removed statements should not
happen either and ForwardOpTree would have bailed out if the llvm::Value
definition was not represented by a statement. This will be corrected in
a followup-commit.
This fixes llvm.org/PR47098
The test failed since commit
bc10888dc "DomTree: Make PostDomTree indifferent to block successors swap"
which is a re-commit of
c35585e20 "DomTree: Make PostDomTree immune to block successors swap"
The schedule of a fused loop has one isl_space per statement, such that
a conversion to a isl_map fails. However, the prevectorization is
interested in the schedule space only: Converting to the non-union
representation only after extracting the schedule range fixes the problem.
This fixes llvm.org/PR46578
The member LastSchedule was never set, such that printScop would always
print "n/a" instead of the last schedule.
To ensure that the isl_ctx lives as least as long as the stored
schedule, also store a shared_ptr.
Also set the schedule tree output style to ISL_YAML_STYLE_BLOCK to avoid
printing everything on a single line.
`opt -polly-opt-isl -analyze` will be used in the next commit.