This patch updates the use_device_ptr and use_device_addr clauses to use
the mapInfoOps for lowering. This allows all the types that are handle
by the map clauses such as derived types to also be supported by the
use_device_clauses.
This is patch 2/2 in a series of patches.
The intention of this change is to ensure that allocas end up in the
entry block not spread out amongst complex reduction variable
initialization code.
The tests we have are quite minimized for readability and
maintainability, making the benefits less obvious. The use case for this
is when there are multiple reduction variables each will multiple blocks
inside of the init region for that reduction.
2/3
Part 1: https://github.com/llvm/llvm-project/pull/102522
Part 3: https://github.com/llvm/llvm-project/pull/102525
Fixes#102935
Updates matching logic for finding the LLVM value that corresponds to an
MLIR value. We need that matching to find the delayed privatizer for an
LLVM value being privatized.
The issue occures when there is an "indirect" correspondence between
MLIR and LLVM values: in some cases the values we are trying to match
stem from a pair of load/store ops that point to the same memref. This
PR adds such matching logic.
This patch modifies MLIR to LLVM IR lowering of the OpenMP dialect to take into
consideration the contents of the `omp.target_triples` module attribute while
generating code for `omp.target` operations.
It adds the `OpenMPIRBuilderConfig::TargetTriples` field and initializes it
using the `amendOperation` flow of the `OpenMPToLLVMIRTranslation` pass. Some
changes are introduced into the `OpenMPIRBuilder` to allow passing the
information about whether a target region is intended to be offloaded from
outside.
The result of this change is that offloading calls are only generated when the
`--offload-arch` or `-fopenmp-targets` options are given to the compiler.
Otherwise, only the host fallback code is generated. This fixes linker errors
currently triggered by `flang-new` if a source file containing a `target`
construct is compiled without any of the aforementioned options.
Several unit tests impacted by these changes, which are intended to check host
code generated for `omp.target` operations, are updated to contain the new
attribute. Without it, no calls to `__tgt_target_kernel` and associated control
flow operations are generated.
Fixes#100209.
This patch adds the missing `OpenMP_Clause` definitions to all existing
`OpenMP_Op`s and updates their operand structure based builders to
initialize the new arguments.
The result of this change is that operation operand structures are now
based in the same list of clauses as their tablegen counterparts. This
means that all of the information needed is now in place to
automatically generate OpenMP operand structures from tablegen
defitions.
Since this change doesn't involve the introduction of actual support for
these clauses, new arguments are not initialized from values stored in
the corresponding operand structure fields but rather set to empty or
null. Those should be updated when support for these clauses on the
corresponding operation is added.
This patch introduces a new OpenMP clause definition not defined by the spec.
Its main purpose is to define the `loop_inclusive` (previously "inclusive",
renamed according to the parent of this PR in the stack) argument of
`omp.loop_nest` in such a way that a followup implementation of a tablegen
backend to automatically generate clause and operation operand structures
directly from `OpenMP_Op` and `OpenMP_Clause` definitions can properly generate
the `LoopNestOperands` structure.
`collapse` clause arguments are also moved into this new definition, as they
represent information on the loop nests being collapsed rather than the
`collapse` clause itself.
Currently, there are some inconsistencies to how clause arguments are
named in the OpenMP dialect. Additionally, the clause operand structures
associated to them also diverge in certain cases. The purpose of this
patch is to normalize argument names across all `OpenMP_Clause` tablegen
definitions and clause operand structures.
This has the benefit of providing more consistent representations for
clauses in the dialect, but the main short-term advantage is that it
enables the development of an OpenMP-specific tablegen backend to
automatically generate the clause operand structures without breaking
dependent code.
The main re-naming decisions made in this patch are the following:
- Variadic arguments (i.e. multiple values) have the "_vars" suffix.
This and other similar suffixes are removed from array attribute
arguments.
- Individual required or optional value arguments do not have any suffix
added to them (e.g. "val", "var", "expr", ...), except for `if` which
would otherwise result in an invalid C++ variable name.
- The associated clause's name is prepended to argument names that don't
already contain it as part of its name. This avoids future collisions
between arguments named the same way on different clauses and adding
both clauses to the same operation.
- Privatization and reduction related arguments that contain lists of
symbols pointing to privatizer/reducer operations use the "_syms"
suffix. This removes the inconsistencies between the names for
"copyprivate_funcs", "[in]reductions", "privatizers", etc.
- General improvements to names, replacement of camel case for snake
case everywhere, etc.
- Renaming of operation-associated operand structures to use the
"Operands" suffix in place of "ClauseOps", to better differentiate
between clause operand structures and operation operand structures.
- Fields on clause operand structures are sorted according to the
tablegen definition of the same clause.
The assembly format for a few arguments is updated to better reflect the
clause they are associated with:
- `chunk_size` -> `dist_schedule_chunk_size`
- `grain_size` -> `grainsize`
- `simd` -> `par_level_simd`
This patch handles dependencies specified by the `depend` clause on an
OpenMP target construct. It does this much the same way clang does it by
materializing an OpenMP `task` that is tagged with the dependencies.
The following functions are relevant to this patch -
1) `createTarget` - This function itself is largely unchanged except
that it now accepts a vector of `DependData` objects that it simply
forwards to `emitTargetCall`
2) `emitTargetCall` - This function has changed now to check if an outer
target-task needs to be materialized (i.e if `target` construct has
`nowait` or has `depend` clause). If yes, it calls `emitTargetTask` to
do all the heavy lifting for creating and dispatching the task.
3) `emitTargetTask` - Bulk of the change is here. See the large comment
explaining what it does at the beginning of this function
This shares code with WsloopOp (the changes to Wsloop should be NFC).
OpenMPIRBuilder basically implements SECTIONS as a wsloop over a case
statement with each SECTION as a case for a particular loopiv value.
Unfortunately it proved very difficult to share code between these and
ParallelOp. ParallelOp does quite a few things differently (doing more
work inside of the bodygen callback and laying out blocks differently).
Aligning reduction implementations for wsloop and parallel will probably
involve functional changes to both, so I won't attempt that in this
commit.
This patch adds support for lowering 'DO SIMD' constructs to MLIR. SIMD
information is now stored in an `omp.simd` loop wrapper, which is
currently ignored by the OpenMP dialect to LLVM IR translation stage.
The end result is that runtime behavior of compiled 'DO SIMD' constructs
does not change after this patch, so 'DO SIMD' still runs like 'DO'
(i.e. SIMD width = 1). However, all of the required information is now
present in the resulting MLIR representation.
To avoid confusion, the previous wsloop-simd.f90 lit test is renamed to
wsloop-schedule.f90 and a new wsloop-simd.f90 test is created to check
the addition of SIMD clauses to the `omp.simd` operation produced when a
'DO SIMD' construct is lowered to MLIR.
This patch updates `OpenMP_Op` definitions to be based on the new set of
`OpenMP_Clause` definitions, and to take advantage of clause-based
automatically-generated argument lists, descriptions, assembly format
and class declarations.
There are also changes introduced to the clause operands structures to
match the current set of tablegen clause definitions. These two are very
closely linked and should be kept in sync. It would probably be a good
idea to try generating clause operands structures from the tablegen
`OpenMP_Clause` definitions in the future.
As a result of this change, arguments for some operations have been
reordered. This patch also addresses this by updating affected operation
build calls and unit tests. Some other updates to tests related to the
order of arguments in the resulting assembly format and others due to
certain previous inconsistencies in the printing/parsing of clauses are
addressed.
The printer and parser functions for the `map` clause are updated, so
that they are able to handle `map` clauses linked to entry block
arguments as well as those which aren't.
This PR causes a build failure in the flang subproject. This is addressed
by the next PR in the stack.
This patch migrates the CGOpenMPRuntimeGPU::emitReduction and related functions to the OpenMPIRBUilder. In future patches MLIR OpenMP translation would be making use of these functions.
Co-authored-by: Jan Leyonberg <jan.leyonberg@amd.com>
This PR attempts to fix common block mapping for regular mapping of
these types as well as when they have been marked as "declare target
link". This PR should allow correct mapping of both the members of a
common block and the full common block via its block symbol.
The main changes were some adjustments to the Fortran OpenMP lowering to
HLFIR/FIR, the lowering of the LLVM+OpenMP dialect to LLVM-IR and
adjustments to the way the we handle target kernel map argument
rebinding inside of the OMPIRBuilder.
For the Fortran OpenMP lowering were two changes, one to prevent the
implicit capture of common block members when the common block symbol
itself has been marked and the other creates intermediate member access
inside of the target region to be used in-place of those external to the
target region, this prevents external usages breaking the
IsolatedFromAbove pact.
In the latter case, there was an adjustment to the size calculation for
types to better handle cases where we pass an array as the type of a map
(as opposed to the bounds and the type of the element), which occurs in
the case of common blocks. There is also some adjustment to how
handleDeclareTargetMapVar handles renaming of declare target symbols in
the module to the reference pointer, now it will only apply to those
within the kernel that is currently being generated and we also perform
a modification to replace constants with instructions as necessary as we
cannot replace these with our reference pointer (non-constant and
constants do not mix nicely).
In the case of the OpenMPIRBuilder some changes were made to defer
global symbol rebinding to kernel arguments until all other arguments
have been rebound. This makes sure we do not replace uses that may refer
to the global (e.g. a GEP) but are themselves actually a separate
argument that needs bound.
Currently "declare target to" still needs some work, but this may be the
case for all types in conjunction with "declare target to" at the
moment.
Uses the new InsertPosition class (added in #94226) to simplify some of
the IRBuilder interface, and removes the need to pass a BasicBlock
alongside a BasicBlock::iterator, using the fact that we can now get the
parent basic block from the iterator even if it points to the sentinel.
This patch removes the BasicBlock argument from each constructor or call
to setInsertPoint.
This has no functional effect, but later on as we look to remove the
`Instruction *InsertBefore` argument from instruction-creation
(discussed
[here](https://discourse.llvm.org/t/psa-instruction-constructors-changing-to-iterator-only-insertion/77845)),
this will simplify the process by allowing us to deprecate the
InsertPosition constructor directly and catch all the cases where we use
instructions rather than iterators.
Fix for Fujitsu test suite test: 0275_0032.f90. The MLIR to LLVM
translation logic assumed that reduction arguments to an `omp.parallel`
op are always the last set of arguments to the op. However, this is a
wrong assumption since private args come afterward.
Fixes a crash uncovered by test 0007_0019.f90 in the Fujitsu test suite.
Previously, in the `PrivCB`, we cloned the `omp.private` op without
inserting it in the parent module of the original op. This causes issues
whenever there is an op that needs to lookup the parent module (e.g. for
symbol lookup). This PR fixes the issue by cloning in the parent module
instead of creating an orphaned op.
It was incorrect to set the insertion point to the init block after
inlining the initialization region because the code generated in the
init block depends upon the value yielded from the init region. When
there were multiple reduction initialization regions each with multiple
blocks, this could lead to the initilization region being inlined after
the init block which depends upon it.
Moving the insertion point to before inlining the initialization block
turned up further issues around the handling of the terminator for the
initialization block, which are also fixed here.
This fixes a bug in #92430 (but the affected code couldn't compile
before #92430 anyway).
Some of the OpenMP code can change the instruction pointed at by the
insertion point. This leads to an assert in the compiler about
BB->getParent() and IP->getParent() not matching.
The fix is to rebuild the insertionpoint from the block, rather than use
builder.restoreIP.
Also, move some of the alloca generation, rather than skipping back and
forth between insert points (and ensure all the allocas are done before
their users are created).
A simple test, mainly to ensure the minimal reproducer doesn't fail to
compile in the future is also added.
This PR attempts to consolidate the different topological sort utilities
into one place. It adds them to the analysis folder because the
`SliceAnalysis` uses some of these.
There are now two different sorting strategies:
1. Sort only according to SSA use-def chains
2. Sort while taking regions into account. This requires a much more
elaborate traversal and cannot be applied on graph regions that easily.
This additionally reimplements the region aware topological sorting
because the previous implementation had an exponential space complexity.
I'm open to suggestions on how to combine this further or how to fuse
the test passes.
This commit renames the name of the block sorting utility function to
`getBlocksSortedByDominance`. A topological order is not defined on a
general directed graph, so the previous name did not make sense.
Fixes#88935
Toggling reduction by-ref broke when multiple reduction clauses were
used. Decisions made for the by-ref status for later clauses could then
invalidate decisions for earlier clauses. For example,
```
reduction(+:scalar,scalar2) reduction(+:array)
```
The first clause would choose by value reduction and generate by-value
reduction regions, but then after this the second clause would force
by-ref to support the array argument. But by the time the second clause
is processed, the first clause has already had the wrong kind of
reduction regions generated.
This is solved by toggling whether a variable should be reduced by
reference per variable. In the above example, this allows only `array`
to be reduced by ref.
If there is only one non-terminator operation in the update region then
the update operation can be found and we can try to generate an
atomicrmw instruction. Otherwise use the cmpxchg loop.
Fixes#91929
The current comparator doesn't work correctly when two identical entries
with -1 are compared. The comparator returns `first` is case when
`aIndex == -1 && bIndex == -1`, but it should `continue` as those
indexes are the same.
llvm::sort requires the comparator to return `false` for equal elements,
otherwise it triggers `Your comparator is not a valid strict-weak
ordering` assert.
This patch seeks to refactor slightly and extend the current record type map
support that was put in place for Fortran's descriptor types to handle explicit
member mapping for record types at a single level of depth (the case of explicit
mapping of nested record types is currently unsupported).
This patch seeks to support this by extending the OpenMPToLLVMIRTranslation phase
to more generally support record types, building on the prior groundwork in the
Fortran allocatables/pointers patch. It now supports different kinds of record type
mapping, in this case full record type mapping and then explicit member mapping
in which there is a special case for certain types when mapped individually to not
require any parent map link in the kernel argument structure. To facilitate this
required:
* The movement of the setting of the map flag type "ptr_and_obj" to respective
frontends, now supporting it as a possible flag that can be read and printed
in mlir form. Some minor changes to declare target map type setting was
neccesary for this.
* The addition of a member index array operand, which tracks the position
of the member in the parent, required for caclulating the appropriate size
to offload to the target, alongside the parents offload pointer (always the
first member currently being mapped).
* A partial mapping attribute operand, to indicate if the entire record type is
being mapped or just member components, aiding the ability to lower
record types in the different manners that are possible.
* Refactoring bounds calculation for record types and general arrays to one
location (as well as load/store generation prior to assigning to the kernel
argument structure), as a side affect enter/exit/update/data mapping
should now be more correct and fully support bounds mapping, previously
this would have only worked for target.
Pull Request: https://github.com/llvm/llvm-project/pull/82852
Extends current support for delayed privatization during translation to
LLVM IR. This adds support for materlizaing the `dealloc` region in
`omp.private` ops when this region contains clean-up/deallocation logic
that needs to be executed at the end of the parallel region.
This changes the `OMPIRBuilder` slightly to execute the finalization
callback **after** the privatization callback. This allows us to collect
information about privatized variables on the MLIR and LLVM sides so
that we can properly emit deallocation logic.
This patch introduces minimal changes to the MLIR to LLVM IR translation
of `omp.wsloop` to support the loop wrapper approach.
There is `omp.loop_nest` related translation code that should be
extracted and shared among all loop operations (e.g. `omp.simd`). This
would possibly also help in the addition of support for compound
constructs later on. This first approach is only intended to keep things
running after the transition to loop wrappers and not to add support for
other use cases enabled by that transition.
This PR on its own will not pass premerge tests. All patches in the
stack are needed before it can be compiled and passes tests.
This patch updates the definition of `omp.simdloop` to enforce the
restrictions of a wrapper operation. It has been renamed to `omp.simd`,
to better reflect the naming used in the spec. All uses of "simdloop" in
function names have been updated accordingly.
Some changes to Flang lowering and OpenMP to LLVM IR translation are
introduced to prevent the introduction of compilation/test failures. The
eventual long term solution might be different.
I missed this before. For by-ref reductions, the private reduction
variable is a pointer to the pointer to the variable. So an extra load
is required to get the right value.
See the "red.private.value.n" loads in the reduction combiner region for
reference.
This patch separates the lowering dispatch for host and target devices.
For the target device, if the current operation is not a top-level
operation (e.g. omp.target) or is inside a target device code region it
will be ignored, since it belongs to the host code.
This is an alternative approach to #84611, the new test in this PR was
taken from there.
Currently, by-ref reductions will allocate the per-thread reduction
variable in the initialization region. Adding a cleanup region allows
that allocation to be undone. This will allow flang to support reduction
of arrays stored on the heap.
This conflation of allocation and initialization in the initialization
should be fixed in the future to better match the OpenMP standard, but
that is beyond the scope of this patch.
The argument to the initialization region of reduction declarations was
never mapped. This meant that if this argument was accessed inside the
initialization region, that mlir operation would be translated to an
llvm operation with a null argument (failing verification).
Adding the mapping ensures that the right LLVM value can be found when
inlining and converting the initialization region.
We have to separately establish and clean up these mappings for each use
of the reduction declaration because repeated usage of the same
declaration will inline it using a different concrete value for the
block argument.
This argument was never used previously because for most cases the
initialized value depends only upon the type of the reduction, not on
the original variable. It is needed now so that we can read the array
extents for the local copy from the mold.
Flang support for reductions on assumed shape arrays patch 2/3
This PR refactors bounds offsetting by combining the two differing
implementations (one applying to initial derived type member map
implementation for descriptors and the other for regular arrays,
effectively allocatable array vs regular array in fortran) now that it's
a little simpler to do.
The PR also moves the utilization of createAlteredByCaptureMap into
genMapInfoOp, where it will be correctly applied to all MapInfoData,
appropriately offsetting and altering Pointer data set in the kernel
argument structure on the host. This primarily means bounds offsets will
now correctly apply to enter/exit/update map clauses as opposed to just
the Target directive that is currently the case. A few fortran runtime
tests have been added to verify this new behavior.
This PR depends on: https://github.com/llvm/llvm-project/pull/84328 and
is an extraction of the larger derived type member map PR stack (so a
requirement for it to land).
Previously reduction variables were always passed by value into and out
of the initialization and combiner regions of the OpenMP reduction
declare operation.
This worked well for reductions of primitive types (and might perform
better than passing by reference). But passing by reference will be
useful for array and derived type reductions (e.g. to move allocation
inside of the init region).
Passing reductions by reference requires different LLVM-IR generation
when lowering from MLIR because some of the loads/stores/allocations
will now be moved inside of the init and combiner regions. This
alternate code generation is requested using a new attribute to
omp.wsloop and omp.parallel.
Existing lowerings from mlir are unaffected (these will continue to use
the by-value argument passing.
Flang will continue to pass by-value argument passing for trivial types
unless a (hidden) command line argument is supplied. Non-trivial types
will always use the by-ref lowering.
Array reductions are not ready yet (but are coming very soon). In the
meantime, this is tested by forcing existing reductions to use by-ref.
Commit series for by-ref OpenMP reductions 3/3
---------
Co-authored-by: Mats Petersson <mats.petersson@arm.com>
Use the new copyprivate list from omp.single to emit calls to
__kmpc_copyprivate, during the creation of the single operation
in OMPIRBuilder.
This is patch 4 of 4, to add support for COPYPRIVATE in Flang.
Original PR: https://github.com/llvm/llvm-project/pull/73128
