If no initializer-clause is specified, the private variables will be
initialized following the rules for initialization of objects with static
storage duration.
Need to adjust the implementation to the current version of the
standard.
Differential Revision: https://reviews.llvm.org/D99539
Added basic parsing/sema/serialization support to extend the
existing 'destroy' clause for use with the 'interop' directive.
Differential Revision: https://reviews.llvm.org/D98834
Added basic parsing/sema/serialization support for interop directive.
Support for the 'init' clause.
Differential Revision: https://reviews.llvm.org/D98558
emission
Ensure that we are in a function declaration context before checking
the diagnostic emission status, to avoid dereferencing a NULL function
declaration.
Differential Revision: https://reviews.llvm.org/D97573
Initial support for using the OpenMPIRBuilder by clang to generate loops using the OpenMPIRBuilder. This initial support is intentionally limited to:
* Only the worksharing-loop directive.
* Recognizes only the nowait clause.
* No loop nests with more than one loop.
* Untested with templates, exceptions.
* Semantic checking left to the existing infrastructure.
This patch introduces a new AST node, OMPCanonicalLoop, which becomes parent of any loop that has to adheres to the restrictions as specified by the OpenMP standard. These restrictions allow OMPCanonicalLoop to provide the following additional information that depends on base language semantics:
* The distance function: How many loop iterations there will be before entering the loop nest.
* The loop variable function: Conversion from a logical iteration number to the loop variable.
These allow the OpenMPIRBuilder to act solely using logical iteration numbers without needing to be concerned with iterator semantics between calling the distance function and determining what the value of the loop variable ought to be. Any OpenMP logical should be done by the OpenMPIRBuilder such that it can be reused MLIR OpenMP dialect and thus by flang.
The distance and loop variable function are implemented using lambdas (or more exactly: CapturedStmt because lambda implementation is more interviewed with the parser). It is up to the OpenMPIRBuilder how they are called which depends on what is done with the loop. By default, these are emitted as outlined functions but we might think about emitting them inline as the OpenMPRuntime does.
For compatibility with the current OpenMP implementation, even though not necessary for the OpenMPIRBuilder, OMPCanonicalLoop can still be nested within OMPLoopDirectives' CapturedStmt. Although OMPCanonicalLoop's are not currently generated when the OpenMPIRBuilder is not enabled, these can just be skipped when not using the OpenMPIRBuilder in case we don't want to make the AST dependent on the EnableOMPBuilder setting.
Loop nests with more than one loop require support by the OpenMPIRBuilder (D93268). A simple implementation of non-rectangular loop nests would add another lambda function that returns whether a loop iteration of the rectangular overapproximation is also within its non-rectangular subset.
Reviewed By: jdenny
Differential Revision: https://reviews.llvm.org/D94973
If the mapped structure has data members, which have 'default' mappers,
need to map these members individually using their 'default' mappers.
Differential Revision: https://reviews.llvm.org/D92195
OpenMP 5.0 removed a lot of restriction for overlapped mapped items
comparing to OpenMP 4.5. Patch restricts the checks for overlapped data
mappings only for OpenMP 4.5 and less and reorders mapping of the
arguments so, that present and alloc mappings are processed first and
then all others.
Differential Revision: https://reviews.llvm.org/D86119
The tile directive is in OpenMP's Technical Report 8 and foreseeably will be part of the upcoming OpenMP 5.1 standard.
This implementation is based on an AST transformation providing a de-sugared loop nest. This makes it simple to forward the de-sugared transformation to loop associated directives taking the tiled loops. In contrast to other loop associated directives, the OMPTileDirective does not use CapturedStmts. Letting loop associated directives consume loops from different capture context would be difficult.
A significant amount of code generation logic is taking place in the Sema class. Eventually, I would prefer if these would move into the CodeGen component such that we could make use of the OpenMPIRBuilder, together with flang. Only expressions converting between the language's iteration variable and the logical iteration space need to take place in the semantic analyzer: Getting the of iterations (e.g. the overload resolution of `std::distance`) and converting the logical iteration number to the iteration variable (e.g. overload resolution of `iteration + .omp.iv`). In clang, only CXXForRangeStmt is also represented by its de-sugared components. However, OpenMP loop are not defined as syntatic sugar. Starting with an AST-based approach allows us to gradually move generated AST statements into CodeGen, instead all at once.
I would also like to refactor `checkOpenMPLoop` into its functionalities in a follow-up. In this patch it is used twice. Once for checking proper nesting and emitting diagnostics, and additionally for deriving the logical iteration space per-loop (instead of for the loop nest).
Differential Revision: https://reviews.llvm.org/D76342
Even code in target and declare target regions might not be emitted.
With this patch we delay more diagnostics and use laziness and linkage
to determine if a function is emitted (for the device). Note that we
still eagerly emit diagnostics for target regions, unfortunately, see
the TODO for the reason.
This hopefully fixes PR48933.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D95928
Type errors in function declarations were not (always) diagnosed prior
to this patch. Furthermore, certain remarks did not get associated
properly which caused them to be emitted multiple times.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D95912
The number of iterations calculation was failing in some cases with more
than two collpased loops. Now the LoopIterationSpace selected matches
InitDependOnLC and CondDependOnLC.
Differential Revision: https://reviews.llvm.org/D95834
The `assumes` directive is an OpenMP 5.1 feature that allows the user to
provide assumptions to the optimizer. Assumptions can refer to
directives (`absent` and `contains` clauses), expressions (`holds`
clause), or generic properties (`no_openmp_routines`, `ext_ABCD`, ...).
The `assumes` spelling is used for assumptions in the global scope while
`assume` is used for executable contexts with an associated structured
block.
This patch only implements the global spellings. While clauses with
arguments are "accepted" by the parser, they will simply be ignored for
now. The implementation lowers the assumptions directly to the
`AssumptionAttr`.
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D91980
There is a build error with gcc-5 [0], investigating now.
[0] https://reviews.llvm.org/D91980#2456526
This reverts commit a5a14cbe7f87e01882ecaa14df5d596cbf38823a.
The `assumes` directive is an OpenMP 5.1 feature that allows the user to
provide assumptions to the optimizer. Assumptions can refer to
directives (`absent` and `contains` clauses), expressions (`holds`
clause), or generic properties (`no_openmp_routines`, `ext_ABCD`, ...).
The `assumes` spelling is used for assumptions in the global scope while
`assume` is used for executable contexts with an associated structured
block.
This patch only implements the global spellings. While clauses with
arguments are "accepted" by the parser, they will simply be ignored for
now. The implementation lowers the assumptions directly to the
`AssumptionAttr`.
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D91980
Support present modifier in defaultmap by adding an extra dimension
for `ImplicitMap`. Therefore, we now create OMPMapClause in `ActOnOpenMPExecutableDirective`
based on both `maptype` and `maptype-modifier`.
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D92427
The variables used in atomic construct should be captured in outer
task-based regions implicitly. Otherwise, the compiler will crash trying
to find the address of the local variable.
Differential Revision: https://reviews.llvm.org/D92682
Compiler needs to convert some of the loop iteration
variables/conditions to different types for better codegen and it may
lead to spurious warning messages about implicit signed/unsigned
conversions.
Differential Revision: https://reviews.llvm.org/D92655
Reviewed by aaron.ballman, rsmith, wchilders
Highlights of review:
- avoid specifying an underlying type (unless such an enum is stored (or part of an abi?))
- avoid using enums as bit-fields, preferring unsigned bit-fields that we static_cast enumerators to. (MS's abi laysout enum bit-fields differently).
- clang-format, clang-format, clang-format.
https://reviews.llvm.org/D91035
Thank you!
If the variable is implicitly firstprivatized in the inner task-based
region, it also must be firstprivatized in outer task-based regions.
Previously firstprivates were captured in tasks but later it was
optimized to reduce the memory usage. But still need to mark such
variables as implicit firstprivate in outer tasks.
Differential Revision: https://reviews.llvm.org/D91627
In order not to modify the `tgt_target_data_update` information but still be
able to pass the extra information for non-contiguous map item (offset,
count, and stride for each dimension), this patch overload `arg` when
the maptype is set as `OMP_MAP_DESCRIPTOR`. The origin `arg` is for
passing the pointer information, however, the overloaded `arg` is an
array of descriptor_dim:
struct descriptor_dim {
int64_t offset;
int64_t count;
int64_t stride
};
and the array size is the same as dimension size. In addition, since we
have count and stride information in descriptor_dim, we can replace/overload the
`arg_size` parameter by using dimension size.
For supporting `stride` in array section, we use a dummy dimension in
descriptor to store the unit size. The formula for counting the stride
in dimension D_n: `unit size * (D_0 * D_1 ... * D_n-1) * D_n.stride`.
Demonstrate how it works:
```
double arr[3][4][5];
D0: { offset = 0, count = 1, stride = 8 } // offset, count, dimension size always be 0, 1, 1 for this extra dimension, stride is the unit size
D1: { offset = 0, count = 2, stride = 8 * 1 * 2 = 16 } // stride = unit size * (product of dimension size of D0) * D1.stride = 4 * 1 * 2 = 8
D2: { offset = 2, count = 2, stride = 8 * (1 * 5) * 1 = 40 } // stride = unit size * (product of dimension size of D0, D1) * D2.stride = 4 * 5 * 1 = 20
D3: { offset = 0, count = 2, stride = 8 * (1 * 5 * 4) * 2 = 320 } // stride = unit size * (product of dimension size of D0, D1, D2) * D3.stride = 4 * 25 * 2 = 200
// X here means we need to offload this data, therefore, runtime will transfer
// data from offset 80, 96, 120, 136, 400, 416, 440, 456
// Runtime patch: https://reviews.llvm.org/D82245
// OOOOO OOOOO OOOOO
// OOOOO OOOOO OOOOO
// XOXOO OOOOO XOXOO
// XOXOO OOOOO XOXOO
```
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D84192
This recommits 7f1f89ec8d9944559042bb6d3b1132eabe3409de and
40df06cdafc010002fc9cfe1dda73d689b7d27a6 with bug fixes for
memory sanitizer failure and Tensile build failure.
folding to not constant folding.
Constant folding of ICEs is done as a GCC compatibility measure, but new
code was picking it up, presumably by accident, due to the bad default.
While here, also switch the flag from a bool to an enum to make it more
obvious what it means at call sites. This highlighted a couple of places
where our behavior is different between C++11 and C++14 due to switching
from checking for an ICE to checking for a converted constant
expression (where there is no 'fold' codepath).
This reapplies D88384 with the minor modification that an assertion was
changed to a regular conditional and graceful exit from
ASTContext::mergeTypes.
Especially for templates we need to check at some point if the base
function matches the specialization we might call instead. Before this
lead to the replacement of `std::sqrt(int(2))` calls with one that
converts the argument to a `std::complex<int>`, clearly not the desired
behavior.
Reported as PR47655
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D88384
Need to fix a check for the variable if it is declared in the inner
OpenMP region to be able to firstprivatize it.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D88240
Need to fix a check for the variable if it is declared in the inner
OpenMP region to be able to firstprivatize it.
Differential Revision: https://reviews.llvm.org/D88240
No need to make final copy from the firsptrivate/lastprivate copy to the original item if the item is a data memeber.
Firstprivate copy creates a copy by reference and the original item gets
updated correctly when updating the lastprivate shared variable.
Differential Revision: https://reviews.llvm.org/D88179
In CUDA/HIP a function may become implicit host device function by
pragma or constexpr. A host device function is checked in both
host and device compilation. However it may be emitted only
on host or device side, therefore the diagnostics should be
deferred until it is known to be emitted.
Currently clang is only able to defer certain diagnostics. This causes
false alarms and limits the usefulness of host device functions.
This patch lets clang defer all overloading resolution diagnostics for host device functions.
An option -fgpu-defer-diag is added to control this behavior. By default
it is off.
It is NFC for other languages.
Differential Revision: https://reviews.llvm.org/D84364
With this extension the effects of `omp begin declare variant` will be
applied to template function declarations. The behavior is opt-in and
controlled by the `extension(allow_templates)` trait. While generally
useful, this will enable us to implement complex math function calls by
overloading the templates of the standard library with the ones in
libc++.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D85735
This extension allows to declare variants in between `omp begin/end
declare variant` that do not match the type of the existing function
with that name. Without this extension we would not find a base function
(with a compatible type), therefore create a new one, which would
cause conflicting declarations. With this extension we will not create
"missing" base functions, which basically renders these specializations
harmless. They will be generated but never called.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D85878
Due to `omp begin/end declare variant`, OpenMP context selectors can be
nested. This patch adds initial support for this so we can use it for
target math variants. We should improve the detection of "equivalent"
scores and user conditions, we should also revisit the data structures
of the OMPTraitInfo object, however, both are not pressing issues right
now.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D85877
The base must be shared between the threads, threadprivates are not
allowed to be bases for array-like reductions.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D85762
