Simply dropping the createPollyIRBuilder() function here, because
it doesn't do much. Also directly initialize Expander in
ScopExpander instead of going through the copy-constructor.
../polly/lib/Transform/ScheduleOptimizer.cpp:812:54: warning: comparison of integers of different signs: 'isl_size' (aka 'int') and 'const unsigned int' [-Wsign-compare]
isl_schedule_node_band_n_member(Node.get()) >
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ^
Static chunked OpenMP scheduling has not been treated correctly.
This patch fixes the problem that threads would not process their
(work-)chunks as intended.
Differential Revision: https://reviews.llvm.org/D61081
The primary motivation is to fix an assertion failure in
isl_basic_map_alloc_equality:
isl_assert(ctx, room_for_con(bmap, 1), return -1);
Although the assertion does not occur anymore, I could not identify
which of ISL's commits fixed it.
Compared to the previous ISL version, Polly requires some changes for this update
* Since ISL commit
20d3574 "perform parameter alignment by modifying both arguments to function"
isl_*_gist_* and similar functions do not always align the paramter
list anymore. This caused the parameter lists in JScop files to
become out-of-sync. Since many regression tests use JScop files with
a fixed parameter list and order, we explicitly call align_params to
ensure a predictable parameter list.
* ISL changed some return types to isl_size, a typedef of (signed) int.
This caused some issues where the return type was unsigned int before:
- No overload for std::max(unsigned,isl_size)
- It cause additional 'mixed signed/unsigned comparison' warnings.
Since they do not break compilation, and sizes larger than 2^31
were never supported, I am going to fix it separately.
* With the change to isl_size, commit
57d547 "isl_*_list_size: return isl_size"
also changed the return value in case of an error from 0 to -1. This
caused undefined looping over isl_iterator since the 'end iterator'
got index -1, never reached from the 'begin iterator' with index 0.
* Some internal changes in ISL caused the number of operations to
increase when determining access ranges to determine aliasing
overlaps. In one test, this caused exceeding the default limit of
800000. The operations-limit was disabled for this test.
The idiom
for (auto i = n - n; i < n; i += 1)
was intended to automatically derive the type of i from n
(signed/unsigned int) and avoid the 'mixed signed/unsigned comparison'
warning. However, almost-always-auto was never used in the LLVM coding
style (although we used it in Polly for some time) and I did never
intended to use this idiom upstream.
PVS Studio may warns about this idiom as 'warning: both sides of
operator are equivalent [misc-redundant-expression]'.
Remove the use of auto and directly use unsigned.
Also see http://llvm.org/PR44768
Commit 777180a "[ADT] Make StringRef's std::string conversion operator explicit"
caused Polly's GPU code generator to not compile anymore. The rest of
Polly has already been fixed in commit
0257a9 "Fix polly build after StringRef change."
Previously, the enums didn't account for all the possible cases, which
could cause misleading results (particularly for a "switch" on
FunctionModRefBehavior).
Fixes regression in polly from recent patch to add writeonly to memset.
While I'm here, also fix a few dubious uses of the FMRB_* enum values.
Differential Revision: https://reviews.llvm.org/D73154
Summary:
This is a follow up on https://reviews.llvm.org/D71473#inline-647262.
There's a caveat here that `Align(1)` relies on the compiler understanding of `Log2_64` implementation to produce good code. One could use `Align()` as a replacement but I believe it is less clear that the alignment is one in that case.
Reviewers: xbolva00, courbet, bollu
Subscribers: arsenm, dylanmckay, sdardis, nemanjai, jvesely, nhaehnle, hiraditya, kbarton, jrtc27, atanasyan, jsji, Jim, kerbowa, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D73099
Scope of changes:
1) Moved RecordedAssumptions vector to ScopBuilder. RecordedAssumptions are used only for Scop constructions.
2) Moved definition of RecordedAssumptionsTy to ScopHelper. It is required both by ScopBuilder and SCEVAffinator.
3) Add new function recordAssumption to ScopHelper. One of its argument is a reference to RecordedAssumption vector. This function is used by ScopBuilder and SCEVAffinator.
4) All RecordedAssumptions are created by ScopBuilder. isl::pw_aff
objects for corresponding SCEVs are created inside ScopBuilder. Scop
functions do not record any assumptions. Scop can use isl::pw_aff
objects which were created by ScopBuilder.
5) Removed functions for handling RecordedAssumptions from Scop class.
6) Removed constness from getScopArrayInfo functions.
7) Replaced SCEVVisitor struct from SCEVAffinator with taylored version, which allow to pass pointer to RecordedAssumptions as function argument.
Differential Revision: https://reviews.llvm.org/D68056
- Update documentation now that the move to monorepo has been made
- Do not tie compiler extension testing to LLVM_BUILD_EXAMPLES
- No need to specify LLVM libraries for plugins
- Add NO_MODULE option to match Polly specific requirements (i.e. building the
module *and* linking it statically)
- Issue a warning when building the compiler extension with
LLVM_BYE_LINK_INTO_TOOLS=ON, as it modifies the behavior of clang, which only
makes sense for testing purpose.
Still mark llvm/test/Feature/load_extension.ll as XFAIL because of a
ManagedStatic dependency that's going to be fixed in a seperate commit.
Differential Revision: https://reviews.llvm.org/D72327
Configure CMake to setup source-groups for Polly. Source groups
describe how source files should be organized in IDEs. By default, all
headers are dumped into one folder under PollyCore and all source files
into another. On disk, these files are organized into folders, but this
isn't reflected in the IDE. This change uses CMake source groups to have
the IDE reflect the on disk layout. This will make it easier to visualize
the project structure for users of Visual Studio and XCode
Patch by Christopher Tetreault <ctetreau@quicinc.com>
Reviewed By: Meinersbur, grosser
Differential Revision: https://reviews.llvm.org/D72117
There's quite a lot of references to Polly in the LLVM CMake codebase. However
the registration pattern used by Polly could be useful to other external
projects: thanks to that mechanism it would be possible to develop LLVM
extension without touching the LLVM code base.
This patch has two effects:
1. Remove all code specific to Polly in the llvm/clang codebase, replaicing it
with a generic mechanism
2. Provide a generic mechanism to register compiler extensions.
A compiler extension is similar to a pass plugin, with the notable difference
that the compiler extension can be configured to be built dynamically (like
plugins) or statically (like regular passes).
As a result, people willing to add extra passes to clang/opt can do it using a
separate code repo, but still have their pass be linked in clang/opt as built-in
passes.
Differential Revision: https://reviews.llvm.org/D61446
Previously, the polly unit tests were stuck in a infinite loop.
There was an edge case in StringRef::count() introduced by 9f6b13e5cc, where an empty 'Str' would cause the function to never exit.
Also fixed usage in polly.
Adapt for 05da2fe521 "Sink all InitializePasses.h includes" which
forgot the GPGPU files (presumably because POLLY_ENABLE_GPGPU_CODEGEN
is OFF by default).
Avoids the need to include TargetMachine.h from various places just for
an enum. Various other enums live here, such as the optimization level,
TLS model, etc. Data suggests that this change probably doesn't matter,
but it seems nice to have anyway.
This file lists every pass in LLVM, and is included by Pass.h, which is
very popular. Every time we add, remove, or rename a pass in LLVM, it
caused lots of recompilation.
I found this fact by looking at this table, which is sorted by the
number of times a file was changed over the last 100,000 git commits
multiplied by the number of object files that depend on it in the
current checkout:
recompiles touches affected_files header
342380 95 3604 llvm/include/llvm/ADT/STLExtras.h
314730 234 1345 llvm/include/llvm/InitializePasses.h
307036 118 2602 llvm/include/llvm/ADT/APInt.h
213049 59 3611 llvm/include/llvm/Support/MathExtras.h
170422 47 3626 llvm/include/llvm/Support/Compiler.h
162225 45 3605 llvm/include/llvm/ADT/Optional.h
158319 63 2513 llvm/include/llvm/ADT/Triple.h
140322 39 3598 llvm/include/llvm/ADT/StringRef.h
137647 59 2333 llvm/include/llvm/Support/Error.h
131619 73 1803 llvm/include/llvm/Support/FileSystem.h
Before this change, touching InitializePasses.h would cause 1345 files
to recompile. After this change, touching it only causes 550 compiles in
an incremental rebuild.
Reviewers: bkramer, asbirlea, bollu, jdoerfert
Differential Revision: https://reviews.llvm.org/D70211
Root cause is VectorBlockGenerator::copyStmt iterates all instructions
in basic block, however some load instructions may be not unnecessary
thus removed by simplification. As a result, these load instructions
don't have a corresponding array.
Looking at BlockGenerator::copyBB, it only iterates instructions list
of ScopStmt. Given it must be a block type scop in case of
vectorization, I think we should do the same in
VectorBlockGenerator::copyStmt.
Patch by bin.narwal <bin.narwal@gmail.com>
Differential Revision: https://reviews.llvm.org/D70076
ScopBuilder::buildEqivClassBlockStmts creates ScopStmts for instruction
groups in basic block and inserts these ScopStmts into Scop::StmtMap,
however, as described in llvm.org/PR38358, comment #5, StmtScops are
inserted into vector ScopStmt[BB] in wrong order. As a result,
ScopBuilder::buildSchedule creates wrong order sequence node.
Looking closer to code, it's clear there is no equivalent classes with
interleaving isOrderedInstruction(memory access) instructions after
joinOrderedInstructions. Afterwards, ScopStmts need to be created and
inserted in the original order of memory access instructions, however,
at the moment ScopStmts are inserted in the order of leader instructions
which are probably not memory access instructions.
The fix is simple with a standalone loop scanning
isOrderedInstruction(memory access) instructions in basic block and
inserting elements into LeaderToInstList one by one. The patch also
removes double reversing operations which are now unnecessary.
New test preserve-equiv-class-order-in-basic_block.ll is also added.
Differential Revision: https://reviews.llvm.org/D68941
llvm-svn: 375192
Since the removal of extensions nodes from schedule trees in r362257 it
is possible to emit parallel code for SCoPs containing
matrix-multiplications. However, the code looking for references used in
outlined statement was not prepared to handle CopyStmts introduced by
the matrix-matrix multiplication detection.
In this case, CopyStmts do not introduce references in addition to the
ones captured by MemoryAccesses, i.e. we change the assertion to accept
CopyStmts and add a regression test for this case.
This fixes llvm.org/PR43164
llvm-svn: 372188
lib/Transform/ScheduleOptimizer.cpp fails to compile on Solaris, both on the 9.x
branch (first noticed when running test-release.sh without -no-polly) and on trunk:
/var/llvm/llvm-9.0.0-rc4/rc4/llvm.src/tools/polly/lib/Transform/ScheduleOptimizer.cpp: In function ‘MicroKernelParamsTy getMicroKernelParams(const llvm::TargetTransformInfo*, polly::MatMulInfoTy)’:
/var/llvm/llvm-9.0.0-rc4/rc4/llvm.src/tools/polly/lib/Transform/ScheduleOptimizer.cpp:914:62: error: call of overloaded ‘sqrt(long unsigned int)’ is ambiguous
914 | ceil(sqrt(Nvec * LatencyVectorFma * ThroughputVectorFma) / Nvec) * Nvec;
| ^
In file included from /usr/gcc/9/lib/gcc/x86_64-pc-solaris2.11/9.1.0/include-fixed/math.h:24,
from /usr/gcc/9/include/c++/9.1.0/cmath:45,
from /var/llvm/llvm-9.0.0-rc4/rc4/llvm.src/include/llvm-c/DataTypes.h:28,
from /var/llvm/llvm-9.0.0-rc4/rc4/llvm.src/include/llvm/Support/DataTypes.h:16,
from /var/llvm/llvm-9.0.0-rc4/rc4/llvm.src/include/llvm/ADT/Hashing.h:47,
from /var/llvm/llvm-9.0.0-rc4/rc4/llvm.src/include/llvm/ADT/ArrayRef.h:12,
from /var/llvm/llvm-9.0.0-rc4/rc4/llvm.src/tools/polly/include/polly/ScheduleOptimizer.h:12,
from /var/llvm/llvm-9.0.0-rc4/rc4/llvm.src/tools/polly/lib/Transform/ScheduleOptimizer.cpp:48:
/usr/gcc/9/lib/gcc/x86_64-pc-solaris2.11/9.1.0/include-fixed/iso/math_iso.h:220:21: note: candidate: ‘long double std::sqrt(long double)’
220 | inline long double sqrt(long double __X) { return __sqrtl(__X); }
| ^~~~
/usr/gcc/9/lib/gcc/x86_64-pc-solaris2.11/9.1.0/include-fixed/iso/math_iso.h:186:15:
note: candidate: ‘float std::sqrt(float)’
186 | inline float sqrt(float __X) { return __sqrtf(__X); }
| ^~~~
/usr/gcc/9/lib/gcc/x86_64-pc-solaris2.11/9.1.0/include-fixed/iso/math_iso.h:74:15:
note: candidate: ‘double std::sqrt(double)’
74 | extern double sqrt __P((double));
| ^~~~
/var/llvm/llvm-9.0.0-rc4/rc4/llvm.src/tools/polly/lib/Transform/ScheduleOptimizer.cpp:915:67:
error: call of overloaded ‘ceil(long unsigned int)’ is ambiguous
915 | int Mr = ceil(Nvec * LatencyVectorFma * ThroughputVectorFma / Nr);
| ^
In file included from /usr/gcc/9/lib/gcc/x86_64-pc-solaris2.11/9.1.0/include-fixed/math.h:24,
from /usr/gcc/9/include/c++/9.1.0/cmath:45,
from /var/llvm/llvm-9.0.0-rc4/rc4/llvm.src/include/llvm-c/DataTypes.h:28,
from /var/llvm/llvm-9.0.0-rc4/rc4/llvm.src/include/llvm/Support/DataTypes.h:16,
from /var/llvm/llvm-9.0.0-rc4/rc4/llvm.src/include/llvm/ADT/Hashing.h:47,
from /var/llvm/llvm-9.0.0-rc4/rc4/llvm.src/include/llvm/ADT/ArrayRef.h:12,
from /var/llvm/llvm-9.0.0-rc4/rc4/llvm.src/tools/polly/include/polly/ScheduleOptimizer.h:12,
from /var/llvm/llvm-9.0.0-rc4/rc4/llvm.src/tools/polly/lib/Transform/ScheduleOptimizer.cpp:48:
/usr/gcc/9/lib/gcc/x86_64-pc-solaris2.11/9.1.0/include-fixed/iso/math_iso.h:196:21: note: candidate: ‘long double std::ceil(long double)’
196 | inline long double ceil(long double __X) { return __ceill(__X); }
| ^~~~
/usr/gcc/9/lib/gcc/x86_64-pc-solaris2.11/9.1.0/include-fixed/iso/math_iso.h:160:15:
note: candidate: ‘float std::ceil(float)’
160 | inline float ceil(float __X) { return __ceilf(__X); }
| ^~~~
/usr/gcc/9/lib/gcc/x86_64-pc-solaris2.11/9.1.0/include-fixed/iso/math_iso.h:76:15:
note: candidate: ‘double std::ceil(double)’
76 | extern double ceil __P((double));
| ^~~~
Fixed by adding casts to disambiguate, checked that it now compiles on both
amd64-pc-solaris2.11 and sparcv9-sun-solaris2.11 and on x86_64-pc-linux-gnu.
Differential Revision: https://reviews.llvm.org/D67442
llvm-svn: 371825
Function joinOrderedInstructions merges instructions when a leader is encountered twice.
It also notices that leaders in SeenLeaders may lose their leadership in previous merging,
and tries to handle the case using following code:
Instruction *PrevLeader = UnionFind.getLeaderValue(SeenLeaders.back());
However, this is wrong because it always gets leader for the last element of SeenLeaders,
and I believe it's wrong even we get leader for Prev here. As a result, Statements in cases
like the one in patch aren't merged as expected. After investigation, I believe it's
unnecessary to get leader instruction at all. This is based on fact: Although leaders in
SeenLeaders could lose leadership, they only lose to others in SeenLeaders, in other words,
one existing leader will be chosen as new leader of merged equivalent statements. We can
take advantage of this and simply check if current leader equals to Prev and break merging
if it does.
The patch also adds a new test.
Patch by bin.narwal <bin.narwal@gmail.com>
Differential Revision: https://reviews.llvm.org/D67007
llvm-svn: 371801
When reading code of Dependences::calculateDependences, I noticed that
WAR is computed specifically by buildWAR. Given ISL now
supports "kills" in approximate dataflow analysis, this patch takes
advantage of it.
This patch also cleans up a couple lines redundant codes.
Patch by bin.narwal <bin.narwal@gmail.com>
Differential Revision: https://reviews.llvm.org/D66741
llvm-svn: 370396
The while loop iterating parent loop in ScopBuilder::buildDomains is
unnecessary because either L or LD are later unused, this is a simple
patch removing it.
Patch by bin.narwal <bin.narwal@gmail.com>
Differential Revision: https://reviews.llvm.org/D66698
llvm-svn: 370368
When reading code in ScopBuilder::buildEqivClassBlockStmts, I think the
main statement flag computation can be simplified, here is the patch.
It's based on two simple facts that:
1. Instruction won't be removed once it's inserted into UnionFind.
2. Main statement must be set if there is non-trivial statement besides the last one.
The patch also saves std::find call.
Patch by bin.narwal <bin.narwal@gmail.com>
Differential Revision: https://reviews.llvm.org/D66477
llvm-svn: 369972
Now that we've moved to C++14, we no longer need the llvm::make_unique
implementation from STLExtras.h. This patch is a mechanical replacement
of (hopefully) all the llvm::make_unique instances across the monorepo.
Differential revision: https://reviews.llvm.org/D66259
llvm-svn: 368935
Scope of changes:
1) Moved buildDomains function to ScopBuilder class.
2) Moved buildDomainsWithBranchConstraints function to ScopBuilder class.
3) Moved propagateDomainConstraints to ScopBuilder class.
4) Moved propagateDomainConstraintsToRegionExit to ScopBuilder class.
5) Moved propagateInvalidStmtDomains to ScopBuilder class.
6) Moved getPredecessorDomainConstraints function to ScopBuilder class.
7) Moved addLoopBoundsToHeaderDomain function to ScopBuilder class.
8) Moved getPwAff function to ScopBuilder class.
9) Moved buildConditionSets functions to ScopBuilder class.
10) Added updateMaxLoopDepth, notifyErrorBlock, getOrInitEmptyDomain, isDomainDefined, setDomain functions to Scop class. They are used by ScopBuilder.
11) Moved helper functions: getRegionNodeBasicBlock, getRegionNodeSuccessor, containsErrorBlock, createNextIterationMap, collectBoundedParts, partitionSetParts, buildConditionSet to ScopBuilder.cpp file.
Differential Revision: https://reviews.llvm.org/D65729
llvm-svn: 368100
Scope of changes:
1) Moved addUserAssumptions function to ScopBuilder class.
2) Moved buildConditionSets functions to polly namespace.
3) Moved getRepresentingInvariantLoadSCEV to public section of the Scop class
Differential Revision: https://reviews.llvm.org/D65241
llvm-svn: 368089
Scope of changes:
1. Moved buildSchedule functions to ScopBuilder.
2. Moved combineInSequence function to ScopBuilder.
3. Moved mapToDimension function to ScopBuilder.
4. Moved LoopStackTy to ScopBuilder.
5. Moved getLoopSurroundingScop to ScopHelper.
6. Moved getNumBlocksInLoop to ScopHelper.
7. Moved getNumBlocksInRegionNode to ScopHelper.
8. Moved getRegionNodeLoop to ScopHelper.
Differential Revision: https://reviews.llvm.org/D64223
llvm-svn: 366377
Scope of changes:
1) Moved finalizeAccesses to ScopBuilder
2) Moved updateAccessDimensionality to ScopBuilder
3) Moved foldSizeConstantsToRight to ScopBuilder
4) Moved foldSizeConstantsToRight to ScopBuilder
5) Moved assumeNoOutOfBounds to ScopBuilder
6) Moved markFortranArrays to ScopBuilder
7) Added iterator range for AccessFunctions vector.
Differential Revision: https://reviews.llvm.org/D63794
llvm-svn: 366374
Scope of changes:
1) Moved addUserContext to ScopBuilder.
2) Moved command line option UserContextStr to ScopBuilder.
Differential Revision: https://reviews.llvm.org/D63740
llvm-svn: 366266
Scope of changes:
1) Moved buildAliasChecks to ScopBuilder.
2) Moved buildAliasGroup to ScopBuilder.
3) Moved buildAliasGroups to ScopBuilder.
4) Moved buildAliasGroupsForAccesses to ScopBuilder.
5) Moved splitAliasGroupsByDomain to ScopBuilder.
6) Moved addNonEmptyDomainConstraints to ScopBuilder.
7) Moved buildMinMaxAccess to ScopBuilder.
8) Moved calculateMinMaxAccess to ScopBuilder.
9) Moved getAccessDomain to ScopBuilder.
10) Moved command line options used only by buildAliasChecks functions to ScopBuilder.
11) Refactored buildAliasGroup function. Added addAliasGroup function to Scop class for pushing back calculated min/max accesses.
12) Added function incrementNumberOfAliasingAssumptions which increments number of statistic variable AssumptionsAliasing. AssumptionsAliasing variable is defined by STATISTIC macro inside ScopInfo.cpp and it is also used by function trackAssumption from Scop class.
13) Added reference to OptimizationRemarkEmitter to ScopBuilder class.
14) Moved calculateMinMaxAccess function to ScopBuilder class.
Differential Revision: https://reviews.llvm.org/D63693
llvm-svn: 366262
Scope of changes:
1) Moved addRecordedAssumptions to ScopBuilder.
2) Moved Assumption struct outside Scop class.
3) Refactored addRecordedAssumptions function. Replaced while loop by
for range loop.
4) Added function to clear processed Assumptions.
Differential Revision: https://reviews.llvm.org/D63572
llvm-svn: 366260
The Loads.h API changed so that a Type parameter is now mandatory in
preparation for pointer types being opaque. Unfortunately I don't build
polly routinely and it still had some uses. This just provides the
(obvious) load type in each case.
llvm-svn: 365470
FunctionAnalysisManagerModuleProxy started to be used by the
AlwaysInlinerPass in r363287 and therefore had to be registered in the
New PassManager.
Should fix the regression tests
Polly :: ScopInliner/invariant-load-func.ll
Polly :: ScopInliner/simple-inline-loop.ll
llvm-svn: 363572
Moved addInvariantLoads and functions listed below to ScopBuilder:
isAParameter
canAlwaysBeHoisted
These functions were referenced only by getNonHoistableCtx.
Moved CLI parameter PollyAllowDereferenceOfAllFunctionParams to
ScopBuilder.
Added iterator range through InvariantEquivClasses.
Patch by Dominik Adamski <adamski.dominik@gmail.com>
Differential Revision: https://reviews.llvm.org/D63172
llvm-svn: 363216
This review is based on review: https://reviews.llvm.org/D62925 . It is
part of moving hoistInvariantLoads function and all functions referenced
only by hoistInvariantLoads to ScopBuilder.
Moved getNonHoistableCtx and functions listed below to ScopBuilder:
isRequiredInvariantLoad
hasNonHoistableBasePtrInScop
isAccessRangeTooComplex
These functions were referenced only by getNonHoistableCtx.
MaxDimensionsInAccessRange and MaxDisjunctsInDomain constant is marked
as extern and it is added to polly namespace. It is used by Scop and
ScopBuilder classes.
MaxDimensionsInAccessRange constant moved to ScopBuilder. It is not used
outside ScopBuilder.
Patch by Dominik Adamski <adamski.dominik@gmail.com>
Differential Revision: https://reviews.llvm.org/D63066
llvm-svn: 363214
Refactor Scop and ScopBuilder class:
1. Move hoistInvariantLoads function from Scop to ScopBuilder class.
2. Private functions (addInvariantLoads, getNonHoistableCtx) are moved
to public section of Scop class. hoistInvariantLoads function
references these functions. These functions will be moved to
ScopBuilder as well in the next steps.
Patch by Dominik Adamski <adamski.dominik@gmail.com>
Differential Revision: https://reviews.llvm.org/D62925
llvm-svn: 363121
Refactor Scop and ScopBuilder class. Move canonicalizeDynamicsBasePtrs
and corresponding static functions from ScopInfo.cpp to ScopBuilder.cpp
Patch by Dominik Adamski <adamski.dominik@gmail.com>
Differential Revision: https://reviews.llvm.org/D62781
llvm-svn: 362554
Extension nodes make schedule trees are less flexible: Many operations,
such as rescheduling, do not work on such schedule trees with extension.
As such, some functionality such as determining parallel loops in isl's
AST are disabled.
Currently, only the pattern-matching generalized matrix-matrix
multiplication optimization adds extension nodes (to add copy-in
statements).
This patch removes all extension nodes as the last step of the schedule
optimization by hoisting the extension node's added domain up to the
root domain node. All following passes can assume that schedule trees
work without restrictions, including the parallelism test. Mark the
outermost loop of the optimized matrix-matrix multiplication as parallel
such that -polly-parallel is able to parallelize that loop.
Differential Revision: https://reviews.llvm.org/D58202
llvm-svn: 362257
Refactor Scop and ScopBuilder class. Move
buildInvariantEquivalenceClasses function from Scop class to ScopBuilder
class.
Patch by: Dominik Adamski <adamski.dominik@gmail.com>
Differential Revision: https://reviews.llvm.org/D62351
llvm-svn: 361902
isl_map_from_union_map cannot determine the map's space if the union_map
is empty. polly::singleton was designed for this case. We pass the
expected map space to avoid crashing in isl_map_from_union_map.
This fixes an issue found by the aosp buildbot. Thanks to Eli Friedman
for the reproducer.
llvm-svn: 361290
At the end of a region statement, the PHINode must be generated
while the current IRBuilder's block is the region's exit node. For
obvious reasons: The PHINode references the region's exiting block.
A partial write would insert new control flow, i.e. insert new basic
blocks between the exiting blocks and the current block.
We fix this by generating the PHI nodes (region exit values) before
generating any MemoryAccess's stores.
This should fix the AOSP buildbot.
Reported-by: Eli Friedman <efriedma@quicinc.com>
llvm-svn: 361204
Fix scan-analyzer issue:
Value stored to 'WARMemAccesses' during its initialization is never read
Patch by Dominik Adamski <adamski.dominik@gmail.com>
Signed-off-by: Dominik Adamski <adamski.dominik@gmail.com>
llvm-svn: 361196
In certain cases, it's possible for delinearization to decide one of the
array dimensions should be some function of an induction variable inside
the scop. Make sure if this happens, we refuse to use those dimensions
for delinearization.
Usually, we end up rejecting the scop before it actually crashes, but it
looks like it's possible to slip past other checks in certain cases
involving smax expressions.
Fixes a crash that started showing up this week on the polly AOSP
builder. As far as I can tell, this is a longstanding issue, though;
it was just exposed by better SCEV analysis of smin expressions.
Differential Revision: https://reviews.llvm.org/D61807
llvm-svn: 360708
PHI nodes (reads) could point to multiple instances of predecessor
blocks (PHI writes) when in an invalid context. Fix by removing PHI
instances that are in an invalid or ouside assumed context.
This fixes llvm.org/PR41656.
llvm-svn: 360454
This removes unused includes (and forward declarations) as
suggested by include-what-you-use. If a transitive include of a removed
include is required to compile a file, I added the required header (or
forward declaration if suggested by include-what-you-use).
This should reduce compilation time and reduce the number of iterative
recompilations when a header was changed.
llvm-svn: 357209
Split out from D59749. The current implementation of isWrappedSet()
doesn't do what it says on the tin, and treats ranges like
[X, Max] as wrapping, because they are represented as [X, 0) when
using half-inclusive ranges. This also makes it inconsistent with
the semantics of isSignWrappedSet().
This patch renames isWrappedSet() to isUpperWrapped(), in preparation
for the introduction of a new isWrappedSet() method with corrected
behavior.
llvm-svn: 357107
Compiling with -polly-target=hybrid was causing Polly to occur two times
in the pipeline. The reason was how the ManagedMemoryRewritePass was
registered in the pass manager. ManagedMemoryRewritePass being a
ModulePass was forcing all previous passes to get recomputed. This
commit avoids Polly to appear two times in the pipeline registering the
ManagedMemoryRewritePass later in the pass manager.
Patch by Lorenzo Chelini <l.chelini@icloud.com>
Differential Revision: https://reviews.llvm.org/D59263
llvm-svn: 356965
The ParallelLoopGenerator class is changed such that GNU OpenMP specific
code was removed, allowing to use it as super class in a
template-pattern. Therefore, the code has been reorganized and one may
not use the ParallelLoopGenerator directly anymore, instead specific
implementations have to be provided. These implementations contain the
library-specific code. As such, the "GOMP" (code completely taken from
the existing backend) and "KMP" variant were created.
For "check-polly" all tests that involved "GOMP": equivalents were added
that test the new functionalities, like static scheduling and different
chunk sizes. "docs/UsingPollyWithClang.rst" shows how the alternative
backend may be used.
Patch by Michael Halkenhäuser <michaelhalk@web.de>
Differential Revision: https://reviews.llvm.org/D59100
llvm-svn: 356434
to reflect the new license.
We understand that people may be surprised that we're moving the header
entirely to discuss the new license. We checked this carefully with the
Foundation's lawyer and we believe this is the correct approach.
Essentially, all code in the project is now made available by the LLVM
project under our new license, so you will see that the license headers
include that license only. Some of our contributors have contributed
code under our old license, and accordingly, we have retained a copy of
our old license notice in the top-level files in each project and
repository.
llvm-svn: 351636
This patch should not introduce any behavior changes. It consists of
mostly one of two changes:
1. Replacing fall through comments with the LLVM_FALLTHROUGH macro
2. Inserting 'break' before falling through into a case block consisting
of only 'break'.
We were already using this warning with GCC, but its warning behaves
slightly differently. In this patch, the following differences are
relevant:
1. GCC recognizes comments that say "fall through" as annotations, clang
doesn't
2. GCC doesn't warn on "case N: foo(); default: break;", clang does
3. GCC doesn't warn when the case contains a switch, but falls through
the outer case.
I will enable the warning separately in a follow-up patch so that it can
be cleanly reverted if necessary.
Reviewers: alexfh, rsmith, lattner, rtrieu, EricWF, bollu
Differential Revision: https://reviews.llvm.org/D53950
llvm-svn: 345882
This removes the primary remaining API producing `TerminatorInst` which
will reduce the rate at which code is introduced trying to use it and
generally make it much easier to remove the remaining APIs across the
codebase.
Also clean up some of the stragglers that the previous mechanical update
of variables missed.
Users of LLVM and out-of-tree code generally will need to update any
explicit variable types to handle this. Replacing `TerminatorInst` with
`Instruction` (or `auto`) almost always works. Most of these edits were
made in prior commits using the perl one-liner:
```
perl -i -ple 's/TerminatorInst(\b.* = .*getTerminator\(\))/Instruction\1/g'
```
This also my break some rare use cases where people overload for both
`Instruction` and `TerminatorInst`, but these should be easily fixed by
removing the `TerminatorInst` overload.
llvm-svn: 344504
IslAst could mark two nested outer loops as "OutermostParallel". It
caused that the code generator tried to OpenMP-parallelize both loops,
which it is not prepared loop.
It was because the recursive AST build algorithm managed a flag
"InParallelFor" to ensure that no nested loop is also marked as
"OutermostParallel". Unfortunatetly the same flag was used by nodes
marked as SIMD, and reset to false after the SIMD node. Since loops can
be marked as SIMD inside "OutermostParallel" loops, the recursive
algorithm again tried to mark loops as "OutermostParellel" although
still nested inside another "OutermostParallel" loop.
The fix exposed another bug: The function "astScheduleDimIsParallel" was
only called when a loop was potentially "OutermostParallel" or
"InnermostParallel", but as a side-effect also determines the minimum
dependence distance. Hence, changing when we need to know whether a loop
is "OutermostParallel" also changed which loop was annotated with
"#pragma minimal dependence distance".
Moreover, some complex condition linked with "InParallelFor" determined
whether a loop should be an "InnermostParallel" loop. It missed some
situations where it would not use mark as such although being inside an
SIMD mark node, and therefore not be annotated using "#pragma simd".
The changes in particular:
1. Split the "InParallelFor" flag into an "InParallelFor" and an
"InSIMD" flag.
2. Unconditionally call "astScheduleDimIsParallel" for its side-effects
and store the result in "InParallel" for later use.
3. Simplify the condition when a loop is "InnermostParallel".
Fixes llvm.org/PR33153 and llvm.org/PR38073.
llvm-svn: 343212
The general-purpose add() now sometimes adds unexpected loop-variant
pointers to the AliasSetTracker, so certain loops would be rejected with
-polly-allow-modref-calls. Use addUnknown() instead, which has the old
behavior.
I'm not completely convinced the resulting behavior is actually
correct: ScopDetection::isValidAccess seems to mostly ignore
"unknown" instructions in the AliasSetTracker. But it's not any worse
than what was happening before.
Committing without pre-commit review to unbreak the buildbots; the
following tests were failing:
test/ScopInfo/mod_ref_access_pointee_arguments.ll
test/ScopInfo/mod_ref_read_pointee_arguments.ll
test/ScopInfo/multidim_2d_with_modref_call_2.ll
llvm-svn: 342010
Summary:
Update all rdtscp callsites in PerfMonitor so that they conform with the signature changes introduced in r341698.
Reviewers: grosser, bollu
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D51928
llvm-svn: 341946
This is a bit awkward in a handful of places where we didn't even have
an instruction and now we have to see if we can build one. But on the
whole, this seems like a win and at worst a reasonable cost for removing
`TerminatorInst`.
All of this is part of the removal of `TerminatorInst` from the
`Instruction` type hierarchy.
llvm-svn: 340701
The method AliasSetTracker::getAliasSetForPointer was removed and replaced by AliasSetTracker::getAliasSetFor for the restructuring in r339930.
Since Polly uses AliasSetTracker::getAliasSetForPointer, a temporary fix has been committed in r339937 with a comment:
Can someone from polly please migrate usage and then delete the wrapper?
This commit is doing exactly that.
llvm-svn: 340072
We upstreamed the export of isl_val_2exp, to the official cpp bindings.
In this process, we concluded that pow2 is a better and more widely used
name for this functionality. Hence, both the official isl-cpp bindings
and our derived variant use now the term pow2.
llvm-svn: 339312
Latest isl versions require clang to build a 'dist' package. Make sure
we actually ask for it.
While being there, also make sure we build isl on all cores.
llvm-svn: 339094
Silence the warning
warning: ignoring return value of function declared with 'warn_unused_result' attribute [-Wunused-result]
JSONExporter is a developer tool, there is no mechanism for error
handling. Print the parser error and abort with a fatal error.
llvm-svn: 338659
The domain generation used nullptr to mark the domain of an error block
as never-executed. Later, nullptr domains are recreated with a
zero-tuple domain that then mismatches with the expected domain the
error block within the loop.
Instead of using nullptr, assign an empty domain which preserves the
expected space. Remove empty domains during SCoP simplification.
Fixes llvm.org/PR38218.
llvm-svn: 338646
The latest version of the isl C++ bindings does not export the 'set'
method yet. Fall back to the C interface until this method can be
exported.
llvm-svn: 338512
These namespace qualifiers are not needed. Dropping them brings us
closer to the official isl namespace qualifiers.
While the delta of this change set is large, it only mechanically drops
the 'isl::' prefixes.
llvm-svn: 338505
The main difference in this change is that isl_stat is now always
checked by default. As we elminiated most used of isl_stat, thanks to
Philip Pfaffe's implementation of foreach, only a small set of changes
is needed.
This change does not include the following recent changes to isl's C++
bindings:
- stricter error handling for isl_bool
- dropping of the isl::namespace qualifiers
The former requires a larger patch in Polly and consequently should go
through a patch-review. The latter will be applied in the next commit to
keep this commit free from noise.
We also still apply a couple of other changes on top of the official isl
bindings. This delta is expected to shrink over time.
llvm-svn: 338504
This time we replace for loops where the return isl::stat::error has
been used to carry status information.
There are still two uses of foreach remaining as we do not have a
corresponding for implementation for pw_aff functions.
llvm-svn: 337239
After Philip added support for range-based for loops to our C++
bindings, we now convert another bunch of foreach calls to range-for loops.
This improves general readability of the code.
llvm-svn: 337201
This is a maintenance update. Besides many minor changes it ships two
functions "isl_*_list_size" and "isl_*_list_get_at" which will allow us
to simplify the iterator implementation in Polly.
llvm-svn: 336425
Summary: This patch updates the isl interface used in `foldSizeConstantsToRight()` to the new C++ interface.
Reviewers: chelini, grosser, philip.pfaffe, Meinersbur
Reviewed By: grosser
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D48965
llvm-svn: 336362
ScalarEvolution::getSCEV dereferences its argument, s.t. passing nullptr
leads to undefined behaviour.
Check for nullptr before calling it instead of checking its argument
afterwards.
llvm-svn: 336350
Summary:
It appears that llvm uses unbuffered C++ streams. So, we should not
mix C and C++ stream operations, because that will give us mixed
up output.
Reviewers: efriedma, jdoerfert, Meinersbur, gareevroman, sebpop, zinob, huihuiz, pollydev, grosser, singam-sanjay, philip.pfaffe
Reviewed By: philip.pfaffe
Subscribers: nemanjai, kbarton
Differential Revision: https://reviews.llvm.org/D40126
llvm-svn: 336288
This change has no impact on upstream Polly directly, but reduces output
noise for some internal isl versions we are testing. In general, storing
simpler and more canonical output is a good idea. Hence, it seems useful
to upstream this change.
llvm-svn: 336281
Summary:
This patch adds <isl_space.h> to gpu_tree.c. This prevents a segfault
when allocating a new isl_space in the function create_sync_domain(), as
the compiler now knows that the return type is a pointer instead of
assuming the function returns an int.
This has been updated in upstream PPCG, so we should bump up our PPCG
version.
Initially discovered by Philip Pfaffe in Polly.
Reviewers: grosser, bollu, philip.pfaffe
Reviewed By: bollu
Subscribers: nemanjai, kbarton, llvm-commits
Differential Revision: https://reviews.llvm.org/D48883
Contributed-by: Alain Denzler <alaindenzler@gmail.com>
llvm-svn: 336251
Summary:
This patch changes the return types for ocl_get_* functions during SPIR code generation. Because these functions return size_t types, the return type needs to be changed to the actual size of size_t on the device.
Based on work by Michal Babej and Pekka Jääskeläinen
Patch by: Alain Denzler
Reviewers: grosser, philip.pfaffe, bollu
Reviewed By: grosser, philip.pfaffe
Subscribers: nemanjai, kbarton, llvm-commits
Differential Revision: https://reviews.llvm.org/D48774
llvm-svn: 336080
Summary: This patch aims to provide support for detecting load patterns which are collectively invariant but right now `isHoistableLoad()` is checking each load instruction individually which cannot detect the load pattern as a whole.
Patch by: Sahil Girish Yerawar
Reviewers: bollu, philip.pfaffe, Meinersbur
Reviewed By: philip.pfaffe, Meinersbur
Differential Revision: https://reviews.llvm.org/D48026
llvm-svn: 335949
The number of SCEV expressions is usually linear in the number of IR
instructions being modeled. However, a naive SCEV visitor is not. For
an expression like x*x, "x" will be visited twice. If x is itself an
expression like x*x, that will be visited twice, etc, and the overall
runtime is O(2^N) in the number of SCEV expressions.
To prevent this from happening, add a cache, so we only visit each SCEV
expression once.
Not sure this is the best solution. Maybe we can instead check whether
the SCEV is scop-invariant (in which case we never need to map the
value). But we don't have a utility for that at the moment.
Differential Revision: https://reviews.llvm.org/D47087
llvm-svn: 335783
Move the optimized getDefToTarget() from ForwardOpTree to ZoneAlgo such
that it can be used by makeValInst.
This reduces the compile time of GrTestUtils of the aosp buildbot from
2m46s to 21s, which should fix the timeout issue.
Differential Revision: https://reviews.llvm.org/D48579
llvm-svn: 335606
Summary:
The function is currently awfully complicated. Drop the IILE and use
StringRef over std::string.
Reviewers: Meinersbur, grosser, bollu
Reviewed By: Meinersbur
Subscribers: nemanjai, kbarton, bollu, llvm-commits, pollydev
Differential Revision: https://reviews.llvm.org/D48070
llvm-svn: 334695
First build the surrounding loops and then build up the polyhedral
structures. Before r326664 we had to mix these updates, clean this
up to improve readability (slightly).
llvm-svn: 334412
In case the schedule has not changed and the operand tree root uses a
value defined in an ancestor loop, the def-to-target mapping is trivial.
For instance, the SCoP
for (int i < 0; i < N; i+=1) {
DefStmt:
D = ...;
for (int j < 0; j < N; j+=1) {
TargetStmt:
use(D);
}
}
has DefStmt-to-TargetStmt mapping of
{ DefStmt[i] -> TargetStmt[i,j] }
This should apply on the majority of def-to-target mappings.
This patch detects this case and directly constructs the expected
mapping. It assumes that the mapping never crosses the loop header
DefStmt is in, which ForwardOpTree does not support at the moment
anyway.
Differential Revision: https://reviews.llvm.org/D47752
llvm-svn: 334134
The aosp-O3-polly-before-vectorizer-unprofitable buildbot currently
fails in ZoneAlgorithm::isNormalized, presumably because an
out-of-quota happens in that function.
Modify ZoneAlgorithm::isNormalized to return an isl::boolean such
it can report an error.
In the failing case, it was called in an assertion in ForwardOpTree.
Allow to pass the assertion in an out-of-quota event, a condition that
is later checked before forwarding an operand tree.
llvm-svn: 333709
This update adds new list types and adds functions which convert isl
data types to lists of their contained elements:
isl::set::get_basic_set_list
isl::union_set::get_set_list
isl::map::get_basic_map_list
isl::union_map::get_map_list
llvm-svn: 333688
Besides other changes, this update introduces functions to translate a
maps and sets into lists of their elements. These lists are useful as
we can define iterators for lists, which allow us to replace many uses
of foreach.
llvm-svn: 333621
When forwarding a LoadInst to another statement, a map that translates
their domain is needed. Before this patch, is was computed by appending
the def-to-use map to the def-to-target of the operand tree's target.
This patch lets the new method getDefToTarget do this. This is
computationally less expensive due to:
* Caching of the result such that it can be used for multiple operands
tree to the same target.
* The map is only computed when there is a LoadInst that needs it.
* It is only computed for the statement requiring the translator map,
instead of having an intermediate result for every edge in the
operand tree.
The downside is that this scheme cannot handle forwarding from a
previous loop iteration (which would require the entire path from
statement to target). Since ForwardOpTree currently does not support
forwarding across loop iterations (SCEV expressions would need to be
transformed), this was not needed anyway.
Differential Revision: https://reviews.llvm.org/D47385
llvm-svn: 333426
Summary: This patch updates `Scop::addUserContext()` function to the new C++ interface and replaces the `auto` keyword with explicit type wherever used in this function.
Reviewers: grosser, bollu, philip.pfaffe, chelini, Meinersbur
Reviewed By: grosser
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D47438
llvm-svn: 333366
Besides normal updates this change also contains a bug-fix to in
isl_coalesce which broke the AOSP buildbot. Thanks to Michael Kruse for
reporting this bug and Sven Verdoolage for fixing this bug.
llvm-svn: 333118
An assertion was not prepared to be passed a nullptr because the
out-of-quota limit was exceeded. Bail-out before the assertion
since the assertion does not apply on out-of-quote.
This fixes llvm.org/PR37477.
llvm-svn: 332488
Summary: This patch aims to remove the usage of old C-styled isl functions (in this case `isl_set_n_basic_set()`) in favor of new C++ isl interface based methods in `ScopInfo.cpp`.
Patch by Sahil Yerawar
Differential Revision: https://reviews.llvm.org/D46935
llvm-svn: 332471
Summary: Creating the Scop name is expensive, because creating the
Region name it's derived from is expensive. So create the name lazily,
because getName() is actually called rarely.
This is a reiteration of r328666, which introduced a use-after-free and
got reverted in r331363.
Differential Revision: https://reviews.llvm.org/D46868
llvm-svn: 332359
The DEBUG() macro is very generic so it might clash with other projects.
The renaming was done as follows:
- git grep -l 'DEBUG' | xargs sed -i 's/\bDEBUG\s\?(/LLVM_DEBUG(/g'
- git diff -U0 master | ../clang/tools/clang-format/clang-format-diff.py -i -p1 -style LLVM
Differential Revision: https://reviews.llvm.org/D44978
llvm-svn: 332352
nullptr is not a valid affine expression, and none of the callers check
for null, so we eventually hit an isl error and crash.
Instead, invalidate the scop and return a constant zero.
Differential Revision: https://reviews.llvm.org/D46445
llvm-svn: 332309
The condition was introduced in r267142 to mitigate a long compile-time
case. In r306087, a max-computation limit was introduced that should
handle the same case while leaving the max disjuncts heuristic it
should have replaced intact.
Today, the max disjuncts bail-out causes problems in that it prematurely
stops SCoPs from being detected, e.g. in SPEC's lbm. This would hit less
like if isl_set_coalesce would be called after isl_set_remove_divs
(which makes more basic_set likely to be coalescable) instead of before.
This patch tries to remove the premature max-disjuncts bail-out
condition by using simple_hull() to reduce the computational overhead,
instead of directly invalidating that SCoP.
Differential Revision: https://reviews.llvm.org/D45066
Contributed-by: Sahil Girish Yerawar <cs15btech11044@iith.ac.in>
llvm-svn: 331891
This reverts commit 0f9dc03765dc301fff7a52e2a0e1dd3e5f3130c5, r328666.
The change introduced a use-after-free, caused by the temporary name string
being destroyed after converting it to a StringRef.
llvm-svn: 331363
These functions have been legacy leftovers which we used before the
official C++ bindings existed. As all uses of these legacy functions
have been removed, this polly-specific extension can also be dropped.
llvm-svn: 331130
Add the options -polly-codegen-trace-stmts and
-polly-codegen-trace-scalars. When enabled, adds a call to the
beginning of every generated statement that prints the executed
statement instance. With -polly-codegen-trace-scalars, it also prints
the value of all scalars that are used in the statement, and PHIs
defined in the beginning of the statement.
Differential Revision: https://reviews.llvm.org/D45743
llvm-svn: 330864
The current statement domain derivation algorithm does not (always)
consider that different exit blocks of a loop can have different
conditions to be reached.
From the code
for (int i = n; ; i-=2) {
if (i <= 0) goto even;
if (i <= 1) goto odd;
A[i] = i;
}
even:
A[0] = 42;
return;
odd:
A[1] = 21;
return;
Polly currently derives the following domains:
Stmt_even_critedge
Domain :=
[n] -> { Stmt_even_critedge[] };
Stmt_odd
Domain :=
[n] -> { Stmt_odd[] : (1 + n) mod 2 = 0 and n > 0 };
while the domain for the odd case is correct, Stmt_even is assumed to be
executed unconditionally, which is obviously wrong. While projecting out
the loop dimension in `adjustDomainDimensions`, it does not consider
that there are other exit condition that have matched before.
I don't know a how to fix this without changing a lot of code. Therefore
This patch rejects loops with multiple exist blocks to fix the
miscompile of test-suite's uuencode.
The odd condition is transformed by LLVM to
%cmp1 = icmp eq i64 %indvars.iv, 1
such that the project_out in adjustDomainDimensions() indeed only
matches for odd n (using this condition only, we'd have an infinite loop
otherwise).
The even condition manifests as
%cmp = icmp slt i64 %indvars.iv, 3
Because buildDomainsWithBranchConstraints() does not consider other exit
conditions, it has to assume that the induction variable will eventually
be lower than 3 and taking this exit.
IMHO we need to reuse the algorithm that determines the number of
iterations (addLoopBoundsToHeaderDomain) to determine which exit
condition applies first. It has to happen in
buildDomainsWithBranchConstraints() because the result will need to
propagate to successor BBs. Currently addLoopBoundsToHeaderDomain() just
look for union of all backedge conditions (which means leaving not the
loop here). The patch in llvm.org/PR35465 changes it to look for exit
conditions instead. This is required because there might be other exit
conditions that do not alternatively go back to the loop header.
Differential Revision: https://reviews.llvm.org/D45649
llvm-svn: 330858
Before this patch, ISL_ASSERT only printed an error message to stderr.
This can be easily missed if the program continues or just fails later.
To fail-early and help error diagnostics (e.g. using bugpoint), call
abort() when an assertion does not hold.
I seem to just have forgotten to add this abort() when I originally
proposed the ISL_ASSERT macro.
Suggested-By: Eli Friedman <efriedma@codeaurora.org>
Differential Revision: https://reviews.llvm.org/D45171
llvm-svn: 330467
Add the switch -polly-debug-func to define the name of a debug
function. This function is ignored for any validity check.
Its purpose is to allow to observe a value after transformation by a
SCoP, and to follow which statements are executed in which order. For
instance, consider the following code:
static void dbg_printf(int sum, int i) {
fprintf(stderr, "The value of sum is %d, i=%d\n", sum, i);
fflush(stderr);
}
void func(int n) {
int sum = 0;
for (int i = 0; i < 16; i+=1) {
sum += i;
dbg_printf(sum, i);
}
}
Executing this after Polly's codegen with -polly-debug-func=dbg_printf
reveals the new execution order and the assumed values at that point of
execution.
Differential Revision: https://reviews.llvm.org/D45728
llvm-svn: 330466
In r330292 this assert was turned incorrectly into an unreachable, but
the correct behavior (thanks Michael) is to assert for anything that is
not 64 bit, but falltrough for 64 bit. I document this in the source
code.
llvm-svn: 330309
Summary:
As of rL329273, LLVM has a mechanism to load new-pm plugins in opt. Use
this API in Polly.
Reviewers: grosser, Meinersbur, bollu
Reviewed By: grosser, Meinersbur
Subscribers: lksbhm, bollu, pollydev, llvm-commits
Differential Revision: https://reviews.llvm.org/D45484
llvm-svn: 330181
A check in assert-builds was meant to verify that a load provides a
value in all statement instances (i.e. its domain). The domain is
commonly gist'ed within the parameter context to contain fewer
constraints. However, statement instances outside the context are
no valid executions, hence the value provided can be undefined.
Refine the check for valid loads to only needed to be defined within
the SCoP context.
In addition, the JSONImporter had to be changed to allow importing
access relations that are broader than the current access relation,
but still defined over all statement instances.
This should fix the compiler crash in test-suite's oggenc of the
-polly-process-unprofitable buildbot.
llvm-svn: 329655
Commit r329640 introduced the removal of all MemoryAccesses of a Scop.
It accidentally continued iterating over a vector whose iterators
have been invalidated by a MemoryAccess removal.
Make a copy of the MemoryAccesses to remove to iterate over while
removing them.
llvm-svn: 329653
Removing a statement left its MemoryAccesses in some lists and maps of
the SCoP. Which lists depends on at which phase of the SCoP
construction the statement is deleted. Follow-up passes could still see
the already deleted MemoryAccesses by iterating through these
lists/maps, resulting in an access violation.
When removing a ScopStmt, also remove all its MemoryAccesses by using
the same mechnism that removes a MemoryAccess.
llvm-svn: 329640
std::remove, despite its name, does not remove elements from a list, but
only moves them to the end of a list. Call erase() to shorten the
vector to the remaining elements.
Test case included in next commit.
llvm-svn: 329639
This patch removes the heuristic in
- Polly :: lib/Support/ScopHelper.cpp
The heuristic forces blocks that directly follow a loop header to not to be considered error blocks.
It was introduced in r249611 with the following commit message:
> This replaces the support for user defined error functions by a
> heuristic that tries to determine if a call to a non-pure function
> should be considered "an error". If so the block is assumed not to be
> executed at runtime. While treating all non-pure function calls as
> errors will allow a lot more regions to be analyzed, it will also
> cause us to dismiss a lot again due to an infeasible runtime context.
> This patch tries to limit that effect. A non-pure function call is
> considered an error if it is executed only in conditionally with
> regards to a cheap but simple heuristic.
In the code below `CCK_Abort2()` would be considered as an error block, but not `CCK_Abort1()` due to this heuristic.
```
for (int i = 0; i < n; i+=1) {
if (ErrorCondition1)
CCK_Abort1(); // No __attribute__((noreturn))
if (ErrorCondition2)
CCK_Abort2(); // No __attribute__((noreturn))
}
```
This does not seem useful. Checking error conditions in the beginning of some work is quite common. It causes a switch default-case to be not considered an error block in SPEC's cactuBSSN. The comment justifying the heuristic mentions a "load", which does not seem to be applicable here. It has been proposed to remove the heuristic.
In addition, the patch fixes the following test cases:
- Polly :: ScopDetect/mod_ref_read_pointer.ll
- Polly :: ScopInfo/max-loop-depth.ll
- Polly :: ScopInfo/mod_ref_access_pointee_arguments.ll
- Polly :: ScopInfo/mod_ref_read_pointee_arguments.ll
- Polly :: ScopInfo/mod_ref_read_pointer.ll
- Polly :: ScopInfo/mod_ref_read_pointers.ll
The test cases failed after removing the heuristic.
Differential Revision: https://reviews.llvm.org/D45274
Contributed-by: Lorenzo Chelini <l.chelini@icloud.com>
llvm-svn: 329548
The warning is:
isl_union_map.c(2041): warning C4221: nonstandard extension used: 'filter_user': cannot be initialized using address of automatic variable 'data'
for the following code (and others)
struct isl_un_op_drop_user_data data = { &isl_set_is_wrapping };
struct isl_un_op_control control = {
.filter = &un_op_filter_drop_user,
.filter_user = &data,
.fn_map = &isl_set_wrapped_domain_map,
};
llvm-svn: 329328
Summary:
When checking the parallelism of a scheduling dimension, we first check if excluding reduction dependences the loop is parallel or not.
If the loop is not parallel, then we need to return the minimal dependence distance of all data dependences, including the previously subtracted reduction dependences.
Reviewers: grosser, Meinersbur, efriedma, eli.friedman, jdoerfert, bollu
Reviewed By: Meinersbur
Subscribers: llvm-commits, pollydev
Tags: #polly
Differential Revision: https://reviews.llvm.org/D45236
llvm-svn: 329214
This gets very expensive for basic blocks which don't have a name: it
calls printAsOperand, which numbers the entire module. We don't
normally need the name anyway, though; it's only used for debug dumps,
so don't compute it by default.
Differential Revision: https://reviews.llvm.org/D44946
llvm-svn: 328666
Summary:
r327219 added wrappers to std::sort which randomly shuffle the container before sorting.
This will help in uncovering non-determinism caused due to undefined sorting
order of objects having the same key.
To make use of that infrastructure we need to invoke llvm::sort instead of std::sort.
Reviewers: grosser, efriedma, jdoerfert, bollu, sebpop
Reviewed By: sebpop
Subscribers: sebpop, mehdi_amini, llvm-commits, pollydev
Tags: #polly
Differential Revision: https://reviews.llvm.org/D44361
llvm-svn: 327361
Piecewise affine expressions have directly corresponding mathematical
operators. Introduce these operators as overloads as this makes writing
code with isl::pw_aff expressions more directly readable.
We can now write:
A = B + C instead of A = B.add(C)
llvm-svn: 327216
isl does not guarantee that set dimension ids will be preserved, so using them
to carry information is not a good idea. Furthermore, the loop information can
be derived without problem from the statement itself. As this even requires
less code than propagating loop information on set dimension ids, starting from
this commit we just derive the loop information in collectSurroundingLoops
directly from the IR.
Interestingly this also results in a couple of isl sets to take a simpler
representation.
llvm-svn: 326664
During codegen, Polly attempts to clear all loops from ScalarEvolution
and LoopInfo, and it does so one block at a time. This causes undefined
behaviour, since this way a loop header might be removed from a loop
before the entire loop is erased, causing ScalarEvolution to run into an
error.
Instead, just delete the entire loop atomically. This fixes currently
failing testcases.
llvm-svn: 326643
Also un-revert (isl_pw_*_alloc: add missing check for compatible spaces, Wed Sep
6 12:18:04 2017 +0200).
This patch is a proposed fix to avoid asserts due to stricter space checking
within isl, which resulted in failures when converting a schedule tree to
a schedule map.
llvm-svn: 326073
As part of this cleanup a couple of unnecessary isl::manage(obj.copy()) pattern
are eliminated as well.
We checked for all potential cleanups by scanning for:
"grep -R isl::manage\( lib/ | grep copy"
llvm-svn: 325558
This update:
- Removes several deprecated functions (e.g., isl_band).
- Improves the pretty-printing of sets by detecting modulos and "false"
equalities.
- Minor improvements to coalescing and increased robustness of the isl
scheduler.
This update does not yet include isl commit isl-0.18-90-gd00cb45
(isl_pw_*_alloc: add missing check for compatible spaces, Wed Sep 6 12:18:04
2017 +0200), as this additional check is too tight and unfortunately causes
two test case failures in Polly. A patch has been submitted to isl and will be
included in the next isl update for Polly.
llvm-svn: 325557
Two or more PHIs mutually using each other directly or indirectly as
incoming value could cause that a PHI WRITE be added before the PHI READ
(i.e. it overwrites the current incoming value with the next incoming
value before it being read).
Fix by ensuring that the PHI WRITE and PHI READ are in the same statement.
This should fix the miscompile of SingleSource/Benchmark/Misc/whetstone
from the test-suite.
llvm-svn: 324934
Splitting basic blocks into multiple statements if there are now
additional scalar dependencies gives more freedom to the scheduler, but
more statements also means higher compile-time complexity. Switch to
finer statement granularity, the additional compile time should be
limited by the number of operations quota.
The regression tests are written for the -polly-stmt-granularity=bb
setting, therefore we add that flag to those tests that break with the
new default. Some of the tests only fail because the statements are
named differently due to a basic block resulting in multiple statements,
but which are removed during simplification of statements without
side-effects. Previous commits tried to reduce this effect, but it is
not completely avoidable.
Differential Revision: https://reviews.llvm.org/D42151
llvm-svn: 324169
Do not add a "_last" suffix to the statement name if there is no (other)
main statement for a basic block. In other words, it becomes the main
statement itself. This further reduces the statement naming difference
between -polly-stmt-granularity=bb and
-polly-stmt-granularity=scalar-indep.
llvm-svn: 324168
Theoretically, a PHI write can be added to any statement that represents
the incoming basic block. We previously always chose the last because
the incoming value's definition is guaranteed to be defined.
With this patch the PHI write is added to the statement that defines the
incoming value. It avoids the requirement for a scalar dependency between
the defining statement and the statement containing the write. As such the
logic for -polly-stmt-granularity=scalar-indep that ensures that there is
such scalar dependencies can be removed.
Differential Revision: https://reviews.llvm.org/D42147
llvm-svn: 323284
VirtualUse::create is only called for MemoryKind::Value, but its
consistency nonetheless checked in verifyUses(). PHI uses are always
inter-stmt dependencies, which was not considered by the constructor
method. The virtual and non-virtual execution paths were the same, such
that verifyUses did not encounter any inconsistencies.
llvm-svn: 323283
The goal is to have -polly-stmt-granularity=bb and
-polly-stmt-granularity=scalar-indep to have the same names if there is
just one statement per basic block.
This fixes a fluke when Polybench's jacobi-2d is optimized differently
depending on the -polly-stmt-granularity option, although both options
create the same SCoP, just with different statement names.
The new naming scheme is:
With -polly-use-llvm-names=0:
Stmt<BBIdx as decimal><Idx within BB as letter>
With -polly-use-llvm-names=1:
Stmt_BBName_<Idx within BB as letter>
The <Idx within BB> suffix is omitted for the main statement of a BB. The
main statement is either the one containing the first store or call
(those cannot be removed by the simplifyer), or if there is no such
instruction, the first. If after simplification there is just a single
statement left, it should be the main statement and have the same names as
with -polly-stmt-granularity=bb.
Differential Revision: https://reviews.llvm.org/D42136
llvm-svn: 322852
This will give control of the statement's name to the caller.
Required to give -polly-stmt-granularity=scalar-indep more control
over the name of the generated statement in a follow-up commit.
llvm-svn: 322851
isl_val_get_num_si crashes on overflow, so don't use it on arbitrary
integers.
Testcase only crashes on platforms where long is 32 bits because of the
signature of isl_val_get_num_si; not sure if it's possible to write a
testcase which crashes if long is 64 bits.
There are a few other places in polly which use isl_val_get_num_si;
they probably need to be fixed as well. I don't think polly uses any
of the other "long" isl APIs in an unsafe manner.
Differential Revision: https://reviews.llvm.org/D42129
llvm-svn: 322766
Print same or similar structure elements together. Previously, the
value could take more importance that the space structure if visited
first in the space nest tree.
Before:
{
Left[0] -> Right[i]: i >= 0;
Left[1] -> AnotherRight[i];
Left[2] -> Right[-1]
}
After:
{
Left[0] -> Right[i]: i >= 0;
Left[2] -> Right[-1];
Left[1] -> AnotherRight[i]
}
llvm-svn: 322581
CMake insists that for each target, one uses only the non-keyword
version of target_link_library
target_link_library(mytarget lib)
or the one with PUBLIC/PRIVATE/INTERFACE keyword:
target_link_library(mytarget PUBLIC lib)
Otherwise, CMake fails with the error message:
The keyword signature for target_link_libraries has already been used with
the target "mytarget". All uses of target_link_libraries with a target
must be either all-keyword or all-plain.
Change all occurances of target_link_library to the newer keyworded
version to avoid such errors. Some already have been changed in r319840,
but might not be sufficient for all build configurations to build
the doxygen manual.
Reported-by: Tanya Lattner <tanyalattner@llvm.org>
llvm-svn: 322376
Memory transfer instructions take two pointers. It is not defined to
which of those a noalias annotation applies. To ensure correctness,
do not add noalias annotations to memcpy/memmove instructions anymore.
The caused a miscompile with test-suite's MultiSource/Applications/obsequi.
Since r321138, the MemCpyOpt pass would remove memcpy/memmove calls if
known to copy uninitialized memory. In that case, it was initialized
by another memcpy, but the annotation for the target pointer said
it would not alias. The annotation was actually meant for the source
pointer, which was was an alloca and could not alias with the target
pointer.
llvm-svn: 321371
Summary:
This can be seen as a follow-up on my previous differential [D33411](https://reviews.llvm.org/D33411).
We received a bug report where this error was triggered. I have tried my best to recreate the issue in a minimal lit testcase which is also part of this differential.
I only handle return instructions as predecessors to a virtual TLR-exit right now. From inspecting the codebase, it seems `unreachable` instructions may also be of interest here. If requested, I can extend my patches to consider them as well. I would also apply this on `ScopHelper.cpp::isErrorBlock` (see D33411), of course.
Reviewers: philip.pfaffe, bollu
Reviewed By: bollu
Subscribers: Meinersbur, pollydev, llvm-commits
Tags: #polly
Differential Revision: https://reviews.llvm.org/D40492
llvm-svn: 319431
Isl does not allow generating isl_ast_expr from an isl_pw_aff that has an
empty domain (i.e. has no pieces). We already detected the case if the
isl_pw_aff comes with an empty domain.
isl_ast_build also considers the domain empty if it is disjoint with the
parameter context (e.g. parameters values that we exclude by runtime
versioning).
Intersect the access relation domain with the parameter context to
also detect such practically empty access domains. The effective
pointer used in the generated code is unimportand because it will never
be executed.
This fixes llvm.org/PR35362
llvm-svn: 318806
Summary:
Most changes are mechanical, but in one place I changed the program semantics
by fixing a likely bug:
In `Scop::hasFeasibleRuntimeContext()`, I'm now explicitely handling the
error-case. Before, when the call to `addNonEmptyDomainConstraints()`
returned a null set, this (probably) accidentally worked because
isl_bool_error converts to true. I'm checking for nullptr now.
Reviewers: grosser, Meinersbur, bollu
Reviewed By: Meinersbur
Subscribers: nemanjai, kbarton, pollydev, llvm-commits
Differential Revision: https://reviews.llvm.org/D39971
llvm-svn: 318632
Summary:
There is a potential use-after-free bug in Scop::buildSchedule(Region *,
LoopStackTy &, LoopInfo &). Before, we took a reference to LoopStack.back()
which is a use after free, since back is popped off further below. This didn't
crash before by pure chance, since LoopStack is actually a vector, and the
memory isn't freed upon pop. I turned this into an iterator-based algorithm.
Reviewers: grosser, bollu, Meinersbur
Reviewed By: Meinersbur
Subscribers: llvm-commits, pollydev
Differential Revision: https://reviews.llvm.org/D39979
llvm-svn: 318415
Put the analysis part of reloadKnownContent under an isl
max-operations quota scope, as has already been done for
forwardKnownLoad.
This should fix the aosp timeout of "GrTestUtils.cpp".
llvm-svn: 317495
Represent PHIs by their incoming values instead of an opaque value of
themselves. This allows ForwardOpTree to "look through" the PHIs and
forward the incoming values since forwardings PHIs is currently not
supported.
This is particularly useful to cope with PHIs inserted by GVN LoadPRE.
The incoming values all resolve to a load from a single array element
which then can be forwarded.
It should in theory also reduce spurious conflicts in value mapping
(DeLICM), but I have not yet found a profitable case yet, so it is
not included here.
To avoid transitive closure and potentially necessary overapproximations
of those, PHIs that may reference themselves are excluded from
normalization and keep their opaque self-representation.
Differential Revision: https://reviews.llvm.org/D39333
llvm-svn: 317008
ForwardOpTree may already transform a scalar access to an array
accesses. The access remains implicit (isOriginalScalarKind(), meaning
that the access is always executed at the begin/end of a statement), but
targets an array (isLatestArrayKind(), which is unrelated to whether the
execution is implicit/explicit).
Fix by properly using isOriginalXXX() to determine execution order.
This fixes the buildbots on MultiSource/Benchmarks/DOE-ProxyApps-C/miniGMG.
llvm-svn: 316995
When collecting base pointers that need to be made available in parallel
subfunctions, use the base pointer associated with the latest
ScopArrayInfo, instead of the original one.
llvm-svn: 316983
Summary:
When GPUNodeBuilder creates loops inside the kernel, it dispatches to
IslNodeBuilder. This however is surprisingly dangerous, since it accesses the
AST Node's user through the wrong type. This patch fixes this problem by
overriding createFor correctly.
This fixes PR35010.
Reviewers: grosser, bollu, Meinersbur
Reviewed By: Meinersbur
Subscribers: Meinersbur, nemanjai, pollydev, llvm-commits, kbarton
Differential Revision: https://reviews.llvm.org/D39364
llvm-svn: 316872
For scalar accesses, change the access target to an array element that
is known to contain the same value.
This may become an alternative to forwardKnownLoad which creates new
loads (and therefore closer to forwarding speculatives). Reloading does
not require the known value originating from a load, but can be a store
as well.
Differential Revision: https://reviews.llvm.org/D39325
llvm-svn: 316766
Previously we marked scalars based on the original access function. However,
when a scalar read access is redirected, the original definition
(or incoming values of a PHI) is not used anymore, and can be deleted
(unless referenced by use that has not been redirected).
llvm-svn: 316660
Add check and skip when the store used to determine the target accesses
multiple array elements. Only a single array location should for
mapping the scalar. Having multiple creates problems when deciding which
element to load from. While MemoryAccess::getAddressFunction() should
select just one of them, other problems arise in code that assumes
that there is just one target element per statement instance.
This fixes llvm.org/PR34989
This also reverts r313902 which fixed llvm.org/PR34485 also caused by
a non-functional target array element. This patch avoids the situation
to occur in the first place.
llvm-svn: 316432
The option splits BasicBlocks into minimal statements such that no
additional scalar dependencies are introduced.
The algorithm is based on a union-find structure, and unites sets if
putting them into separate statements would introduce a scalar
dependencies. As a consequence, instructions may be split into separate
statements such their relative order is different than the statements
they are in. This is accounted for instructions whose relative order
matters (e.g. memory accesses).
The algorithm is generic in that heuristic changes can be made
relatively easily. We might relax the order requirement for read-reads
or accesses to different base pointers. Forwardable instructions can be
made to not cause a join.
This implementation gives us a speed-up of 82% in SPEC 2006 456.hmmer
benchmark by allowing loop-distribution in a hot loop such that one of
the loops can be vectorized.
Differential Revision: https://reviews.llvm.org/D38403
llvm-svn: 314983
The option is introduced with only one possible value
-polly-stmt-granularity=bb which represents the current behaviour, which
is outlined into the new function buildSequentialBlockStmts().
More options will be added in future commits.
llvm-svn: 314900
We make sure that the final reload of an invariant scalar memory access uses the
same stack slot into which the invariant memory access was stored originally.
Earlier, this was broken as we introduce a new stack slot aside of the preload
stack slot, which remained uninitialized and caused our escaping loads to
contain garbage. This happened due to us clearing the pre-populated values
in EscapeMap after kernel code generation. We address this issue by preserving
the original host values and restoring them after kernel code generation.
EscapeMap is not expected to be used during kernel code generation, hence we
clear it during kernel generation to make sure that any unintended uses are
noticed.
llvm-svn: 314894
Iterate over statement instructions instead over basic block
instructions when creating MemoryAccesses. It allows making the creation
of MemoryAccesses independent of how the basic blocks are split into
multiple ScopStmts.
llvm-svn: 314665
Nikita Popov