skatrak wrote:
@tblah thanks for giving this a look. Basically the issue here with the
reduction clause is that its description is currently different for each
operation that accepts it. They are generally quite similar, just not the same.
I think the clause is already consistent in terms of h
@@ -1086,8 +1086,9 @@ static void genTargetDataClauses(
// ordering.
// TODO: Perhaps create a user provideable compiler option that will
// re-introduce a hard-error rather than a warning in these cases.
- promoteNonCPtrUseDevicePtrArgsToUseDeviceAddr(clauseOps, useDevi
skatrak wrote:
Quick update about reduction clause descriptions @tblah: I updated the previous
PR on the stack (#92521) to add a shared description that basically
incorporates all of the information that was spread across the various
operations using it. I updated the description of `omp.taskl
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/92524
>From 33e726ed0d41dd2abb7b99eb413ea9bc014184d1 Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Fri, 17 May 2024 11:38:36 +0100
Subject: [PATCH] [Flang][OpenMP] Update flang with changes to the OpenMP
dialect
skatrak wrote:
> @skatrak Is this ready for final review?
Yes, it may need very minor changes when rebasing due to recent additions to
the main branch, but this should be it for the most part.
https://github.com/llvm/llvm-project/pull/92524
___
llvm-
skatrak wrote:
> > I guess fixing byref is on me (#92244). Unfortunately I can't work on this
> > immediately so I won't hold up this PR for it.
>
> @skatrak does #96215 cover everything you need?
Thank you for the heads-up, that certainly helps. My plan is to update the PR
stack after yours
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/92524
>From 441b83ca7fdc31b8717207529b6e86e33a1f130a Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Fri, 17 May 2024 11:38:36 +0100
Subject: [PATCH] [Flang][OpenMP] Update flang with changes to the OpenMP
dialect
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/92524
>From 95d6f3446201d2b3162b694887e3fd888b628e96 Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Fri, 17 May 2024 11:38:36 +0100
Subject: [PATCH] [Flang][OpenMP] Update flang with changes to the OpenMP
dialect
https://github.com/skatrak created
https://github.com/llvm/llvm-project/pull/97563
This patch updates the `FirOpBuilder::getAllocaBlock()` method to avoid
returning blocks which are part of a region owned by a loop wrapper operation.
This avoids introducing `fir.alloca` operations inside of lo
https://github.com/skatrak created
https://github.com/llvm/llvm-project/pull/97564
This patch removes the `outerCombined` argument from `genTargetOp()` and the
`processReduction` argument from `genTargetClauses()`, as they aren't used.
>From 83d53c3299de2e7b9874ee0083d180674ab51c30 Mon Sep 17
https://github.com/skatrak created
https://github.com/llvm/llvm-project/pull/97565
This patch moves the logic associated with the creation of a
`DataSharingProcessor` instance for loop-associated OpenMP leaf constructs to
the `genOMPDispatch` function, avoiding code duplication for standalone
https://github.com/skatrak created
https://github.com/llvm/llvm-project/pull/97566
This patch splits the lowering for `omp.loop_nest` into its own function and
updates lowering for all supported loop wrappers to stop creating this
operation themselves.
Lowering functions for loop constructs a
https://github.com/skatrak created
https://github.com/llvm/llvm-project/pull/97718
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.
Th
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/97718
>From 27490fbac8e08ed6c11437ef28efd8298fb120b1 Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Thu, 4 Jul 2024 12:56:43 +0100
Subject: [PATCH 1/2] [Flang][OpenMP] Add lowering support for DO SIMD
This patch a
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/97718
>From 27490fbac8e08ed6c11437ef28efd8298fb120b1 Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Thu, 4 Jul 2024 12:56:43 +0100
Subject: [PATCH 1/3] [Flang][OpenMP] Add lowering support for DO SIMD
This patch a
https://github.com/skatrak created
https://github.com/llvm/llvm-project/pull/97819
This patch adds support for lowering 'DISTRIBUTE SIMD' constructs to MLIR.
Translation of `omp.distribute` operations to LLVM IR is still not supported,
so its composition with `omp.simd` isn't either.
>From 41
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/97718
>From 69253a2e1d7a53f4cae907eed30a1c2bdf12c223 Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Thu, 4 Jul 2024 12:56:43 +0100
Subject: [PATCH] [Flang][OpenMP] Add lowering support for DO SIMD
This patch adds
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/97819
>From 7fc2ea1416a502dae450e16d4da61b2bfbf06a99 Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Fri, 5 Jul 2024 12:49:46 +0100
Subject: [PATCH] [Flang][OpenMP] Add lowering support for DISTRIBUTE SIMD
This pat
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/97819
>From a61b3069a400f43eaa4f71e443b8a85c90aea0f0 Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Fri, 5 Jul 2024 12:49:46 +0100
Subject: [PATCH] [Flang][OpenMP] Add lowering support for DISTRIBUTE SIMD
This pat
https://github.com/skatrak created
https://github.com/llvm/llvm-project/pull/99506
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)
https://github.com/skatrak created
https://github.com/llvm/llvm-project/pull/99507
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 operati
https://github.com/skatrak created
https://github.com/llvm/llvm-project/pull/99508
This patch adds the "gen-openmp-clause-ops" `mlir-tblgen` generator to produce
the structure definitions previously in OpenMPClauseOperands.h automatically
from the information contained in OpenMPOps.td and Open
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/99508
>From 1d99939c020aab8650cd20df24e0b1e71726ae90 Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Wed, 17 Jul 2024 13:26:09 +0100
Subject: [PATCH 1/2] [MLIR][OpenMP] Automate operand structure definition
This pa
@@ -12,11 +12,43 @@
#include "mlir/TableGen/GenInfo.h"
+#include "mlir/TableGen/CodeGenHelpers.h"
+#include "llvm/ADT/StringExtras.h"
+#include "llvm/ADT/TypeSwitch.h"
#include "llvm/TableGen/Error.h"
#include "llvm/TableGen/Record.h"
using namespace llvm;
+/// The cod
@@ -148,6 +169,110 @@ static void verifyClause(Record *op, Record *clause) {
"or explicitly skipping this field.");
}
+/// Translate the type of an OpenMP clause's argument to its corresponding
+/// representation for clause operand structures.
+///
+/// All kinds
@@ -23,303 +23,31 @@
#define GET_ATTRDEF_CLASSES
#include "mlir/Dialect/OpenMP/OpenMPOpsAttributes.h.inc"
+#include "mlir/Dialect/OpenMP/OpenMPClauseOps.h.inc"
+
namespace mlir {
namespace omp {
//===--=
https://github.com/skatrak created
https://github.com/llvm/llvm-project/pull/100154
The `OffloadModuleInterface` holds getter/setter methods to access OpenMP
dialect module-level discardable attributes used to hold general OpenMP
compilation information.
This patch adds the `omp.target_triple
https://github.com/skatrak created
https://github.com/llvm/llvm-project/pull/100155
This patch adds support for the `-fopenmp-targets` option to the `bbc` and
`flang -fc1` tools. It adds an `OMPTargetTriples` property to the `LangOptions`
structure, which is filled with the triples represented
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/100155
>From 54e52e8a37fd725976e157cd0f9e0221a355dead Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Tue, 23 Jul 2024 16:40:18 +0100
Subject: [PATCH 1/2] [Flang][OpenMP] Add frontend support for -fopenmp-targets
T
@@ -894,6 +894,7 @@ static bool parseDiagArgs(CompilerInvocation &res,
llvm::opt::ArgList &args,
/// options accordingly. Returns false if new errors are generated.
static bool parseDialectArgs(CompilerInvocation &res, llvm::opt::ArgList &args,
skatrak wrote:
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/99508
>From 1d99939c020aab8650cd20df24e0b1e71726ae90 Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Wed, 17 Jul 2024 13:26:09 +0100
Subject: [PATCH 1/3] [MLIR][OpenMP] Automate operand structure definition
This pa
@@ -12,11 +12,52 @@
#include "mlir/TableGen/GenInfo.h"
+#include "mlir/TableGen/CodeGenHelpers.h"
+#include "llvm/ADT/StringExtras.h"
+#include "llvm/ADT/TypeSwitch.h"
#include "llvm/TableGen/Error.h"
#include "llvm/TableGen/Record.h"
using namespace llvm;
+/// The cod
@@ -12,11 +12,52 @@
#include "mlir/TableGen/GenInfo.h"
+#include "mlir/TableGen/CodeGenHelpers.h"
+#include "llvm/ADT/StringExtras.h"
+#include "llvm/ADT/TypeSwitch.h"
#include "llvm/TableGen/Error.h"
#include "llvm/TableGen/Record.h"
using namespace llvm;
+/// The cod
@@ -408,17 +408,26 @@ class ElementsAttrBase :
let storageType = [{ ::mlir::ElementsAttr }];
let returnType = [{ ::mlir::ElementsAttr }];
let convertFromStorage = "$_self";
+
+ // The underlying C++ value type of each element.
+ string elementReturnType = ?;
---
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/99506
>From 96ddba7f30440a83602e8f89d1193813a03dceee Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Tue, 16 Jul 2024 15:48:00 +0100
Subject: [PATCH 1/2] [MLIR][OpenMP] Create `LoopRelatedClause`
This patch introdu
skatrak wrote:
@Meinersbur I have updated this PR according to our discussion in #99505. Now
the loop bounds and steps are now part of the `LoopRelatedClause` and named
`loopLowerBounds`, `loopUpperBounds` and `loopSteps`.
https://github.com/llvm/llvm-project/pull/99506
___
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/99507
>From c80e6f18eabfc05804c1ac5320bf12af756f5b79 Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Tue, 16 Jul 2024 17:18:13 +0100
Subject: [PATCH] [MLIR][OpenMP] Add missing clauses to OpenMP op definitions
This
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/99508
>From 893b8de7264be5c290db38adee66b7e00d4b6fe6 Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Wed, 17 Jul 2024 13:26:09 +0100
Subject: [PATCH] [MLIR][OpenMP] Automate operand structure definition
This patch
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/100154
>From 81fddab5c1b5043a477787872c9f18307efaacf6 Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Tue, 23 Jul 2024 16:32:16 +0100
Subject: [PATCH] [MLIR][OpenMP] Add omp.target_triples attribute to the
OffloadM
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/100155
>From f16bc33e0ac3af52aa33400e3402b1931d16dc33 Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Tue, 23 Jul 2024 16:40:18 +0100
Subject: [PATCH] [Flang][OpenMP] Add frontend support for -fopenmp-targets
This
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/100156
>From 9094873e1176b0e8af5065cfc9a7c3fbc12794c0 Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Tue, 23 Jul 2024 16:53:40 +0100
Subject: [PATCH] [MLIR][OpenMP][OMPIRBuilder] Add lowering support for
omp.targe
@@ -2370,6 +2370,24 @@ static void processMapMembersWithParent(
assert(memberDataIdx >= 0 && "could not find mapped member of structure");
+if (checkIfPointerMap(memberClause)) {
skatrak wrote:
Nit: It would be nice to add a short comment of what's a
https://github.com/skatrak edited
https://github.com/llvm/llvm-project/pull/96265
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
https://github.com/skatrak approved this pull request.
Thank you Andrew, this looks reasonable to me, though I'd suggest waiting for
another approval before merging.
Also, it looks like the flang/test/Integration/OpenMP/map-types-and-sizes.f90
test needs to be updated.
https://github.com/llvm
https://github.com/skatrak edited
https://github.com/llvm/llvm-project/pull/96266
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
@@ -189,85 +295,60 @@ generateMemberPlacementIndices(const Object &object,
indices = llvm::SmallVector{llvm::reverse(indices)};
}
-void addChildIndexAndMapToParent(
-const omp::Object &object,
-std::map> &parentMemberIndices,
-mlir::omp::MapInfoOp &mapOp, semanti
@@ -953,6 +954,22 @@ bool ClauseProcessor::processMap(
if (origSymbol && fir::isTypeWithDescriptor(origSymbol.getType()))
symAddr = origSymbol;
+ if (object.sym()->owner().IsDerivedType()) {
+omp::ObjectList objectList = gatherObjects
@@ -203,10 +201,11 @@ bool
ClauseProcessor::processMotionClauses(lower::StatementContext &stmtCtx,
for (const omp::Object &object : objects) {
llvm::SmallVector bounds;
std::stringstream asFortran;
+ std::optional parentObj;
- lo
https://github.com/skatrak commented:
Thank you Andrew for this work, and sorry for the delay getting to it. I've got
some comments, most of them nits as usual.
https://github.com/llvm/llvm-project/pull/96266
___
llvm-branch-commits mailing list
llvm-
@@ -52,12 +52,22 @@ using DeclareTargetCapturePair =
struct OmpMapMemberIndicesData {
// The indices representing the component members placement in its derived
// type parents hierarchy.
- llvm::SmallVector memberPlacementIndices;
+ llvm::SmallVector> memberPlacementIndi
@@ -953,6 +954,22 @@ bool ClauseProcessor::processMap(
if (origSymbol && fir::isTypeWithDescriptor(origSymbol.getType()))
symAddr = origSymbol;
+ if (object.sym()->owner().IsDerivedType()) {
+omp::ObjectList objectList = gatherObjects
@@ -280,75 +361,66 @@ void insertChildMapInfoIntoParent(
// precedes the children. An alternative, may be to do
// delayed generation of map info operations from the clauses and
// organize them first before generation.
- mapOp->moveAfter(indices.second.b
@@ -14,14 +14,15 @@
#define FORTRAN_LOWER_SUPPORT_UTILS_H
#include "flang/Common/indirection.h"
+#include "flang/Optimizer/Builder/FIRBuilder.h"
#include "flang/Parser/char-block.h"
#include "flang/Semantics/tools.h"
#include "mlir/Dialect/Arith/IR/Arith.h"
#include "mlir/
@@ -55,6 +55,13 @@ struct IdTyTemplate {
return designator == other.designator;
}
+ // defining an "ordering" which allows types derived from this to be
+ // utilised in maps and other containers that require comparison
+ // operators for ordering
ska
@@ -189,85 +295,60 @@ generateMemberPlacementIndices(const Object &object,
indices = llvm::SmallVector{llvm::reverse(indices)};
}
-void addChildIndexAndMapToParent(
-const omp::Object &object,
-std::map> &parentMemberIndices,
-mlir::omp::MapInfoOp &mapOp, semanti
@@ -85,67 +135,227 @@ class OMPMapInfoFinalizationPass
descriptor = alloca;
}
+return descriptor;
+ }
+
+ /// Simple function that will generate a FIR operation accessing
+ /// the descriptors base address (BoxOffsetOp) and then generate a
+ /// MapInfoOp for
@@ -85,67 +135,227 @@ class OMPMapInfoFinalizationPass
descriptor = alloca;
}
+return descriptor;
+ }
+
+ /// Simple function that will generate a FIR operation accessing
+ /// the descriptors base address (BoxOffsetOp) and then generate a
+ /// MapInfoOp for
@@ -189,85 +295,60 @@ generateMemberPlacementIndices(const Object &object,
indices = llvm::SmallVector{llvm::reverse(indices)};
}
-void addChildIndexAndMapToParent(
-const omp::Object &object,
-std::map> &parentMemberIndices,
-mlir::omp::MapInfoOp &mapOp, semanti
@@ -671,4 +672,51 @@ static inline bool isEqual(const Fortran::lower::SomeExpr
*x,
}
} // end namespace Fortran::lower
+// OpenMP utility functions used in locations outside of the
+// OpenMP lowering.
+namespace Fortran::lower::omp {
+
+[[maybe_unused]] static void fillMembe
@@ -85,67 +135,227 @@ class OMPMapInfoFinalizationPass
descriptor = alloca;
}
+return descriptor;
+ }
+
+ /// Simple function that will generate a FIR operation accessing
+ /// the descriptors base address (BoxOffsetOp) and then generate a
+ /// MapInfoOp for
@@ -280,75 +361,66 @@ void insertChildMapInfoIntoParent(
// precedes the children. An alternative, may be to do
// delayed generation of map info operations from the clauses and
// organize them first before generation.
- mapOp->moveAfter(indices.second.b
@@ -51,21 +55,66 @@ class OMPMapInfoFinalizationPass
: public fir::impl::OMPMapInfoFinalizationPassBase<
OMPMapInfoFinalizationPass> {
- void genDescriptorMemberMaps(mlir::omp::MapInfoOp op,
- fir::FirOpBuilder &builder,
-
@@ -51,21 +55,66 @@ class OMPMapInfoFinalizationPass
: public fir::impl::OMPMapInfoFinalizationPassBase<
OMPMapInfoFinalizationPass> {
- void genDescriptorMemberMaps(mlir::omp::MapInfoOp op,
- fir::FirOpBuilder &builder,
-
@@ -85,67 +135,227 @@ class OMPMapInfoFinalizationPass
descriptor = alloca;
}
+return descriptor;
+ }
+
+ /// Simple function that will generate a FIR operation accessing
+ /// the descriptors base address (BoxOffsetOp) and then generate a
+ /// MapInfoOp for
@@ -7053,13 +7053,30 @@ OpenMPIRBuilder::InsertPointTy
OpenMPIRBuilder::emitTargetTask(
<< "\n");
return Builder.saveIP();
}
+
static void emitTargetCall(
OpenMPIRBuilder &OMPBuilder, IRBuilderBase &Builder,
OpenMPIRBuilder::InsertPointTy Alloc
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/100156
>From 366b716e3a192265aed4a1328c49d0ffeef0166d Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Tue, 23 Jul 2024 16:53:40 +0100
Subject: [PATCH] [MLIR][OpenMP][OMPIRBuilder] Add lowering support for
omp.targe
https://github.com/skatrak edited
https://github.com/llvm/llvm-project/pull/102341
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
https://github.com/skatrak commented:
Thank you for picking up this work. In general it looks good, but apart from
some nits I think some improvements can be made to the op verifiers as well.
Also, please update mlir/test/Dialect/OpenMP/ops.mlir to add the
`omp.composite` attribute to all ops
@@ -1796,6 +1810,10 @@ LogicalResult SimdOp::verify() {
if (getNestedWrapper())
return emitOpError() << "must wrap an 'omp.loop_nest' directly";
+ if (llvm::cast(getOperation()).isComposite())
skatrak wrote:
This check is incorrect, since `omp.simd` mi
@@ -1546,6 +1546,9 @@ LogicalResult ParallelOp::verify() {
if (!isWrapper())
return emitOpError() << "must take a loop wrapper role if nested inside "
"of 'omp.distribute'";
+if (!llvm::cast(getOperation()).isComposite())
@@ -1749,10 +1755,18 @@ LogicalResult WsloopOp::verify() {
return emitOpError() << "must be a loop wrapper";
if (LoopWrapperInterface nested = getNestedWrapper()) {
+if (!llvm::cast(getOperation()).isComposite())
+ return emitError()
+ << "'omp.compo
@@ -2063,10 +2063,14 @@ static void genCompositeDistributeSimd(
// TODO: Populate entry block arguments with private variables.
auto distributeOp = genWrapperOp(
converter, loc, distributeClauseOps, /*blockArgTypes=*/{});
+ llvm::cast(distributeOp.getOperation())
+
@@ -1825,11 +1843,17 @@ LogicalResult DistributeOp::verify() {
return emitOpError() << "must be a loop wrapper";
if (LoopWrapperInterface nested = getNestedWrapper()) {
+if (!llvm::cast(getOperation()).isComposite())
+ return emitError()
+ << "'omp.c
https://github.com/skatrak created
https://github.com/llvm/llvm-project/pull/102613
After decomposition of OpenMP compound constructs and assignment of applicable
clauses to each leaf construct, composite constructs are then combined again
into a single element in the construct queue. This hel
https://github.com/skatrak edited
https://github.com/llvm/llvm-project/pull/102613
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
https://github.com/skatrak updated
https://github.com/llvm/llvm-project/pull/102613
>From aa0403a8a137295345b066cebaab9635e4b9886f Mon Sep 17 00:00:00 2001
From: Sergio Afonso
Date: Fri, 9 Aug 2024 12:58:27 +0100
Subject: [PATCH 1/2] [Flang][OpenMP] Prevent re-composition of composite
construc
https://github.com/skatrak edited
https://github.com/llvm/llvm-project/pull/102613
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
@@ -2141,13 +2154,50 @@ static void genCompositeTaskloopSimd(
semantics::SemanticsContext &semaCtx, lower::pft::Evaluation &eval,
mlir::Location loc, const ConstructQueue &queue,
ConstructQueue::const_iterator item, DataSharingProcessor &dsp) {
+ assert(std::distan
@@ -90,38 +83,33 @@ ConstructQueue buildConstructQueue(
Fortran::lower::pft::Evaluation &eval, const parser::CharBlock &source,
llvm::omp::Directive compound, const List &clauses) {
- List constructs;
-
ConstructDecomposition decompose(modOp, semaCtx, eval, compoun
https://github.com/skatrak commented:
Thank you Krzysztof for your comments, they should be addressed now.
https://github.com/llvm/llvm-project/pull/102613
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/c
@@ -2263,24 +2321,13 @@ static void genOMPDispatch(lower::AbstractConverter
&converter,
// Composite constructs
case llvm::omp::Directive::OMPD_distribute_parallel_do:
-genCompositeDistributeParallelDo(converter, symTable, semaCtx, eval, loc,
-
https://github.com/skatrak edited
https://github.com/llvm/llvm-project/pull/102341
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
https://github.com/skatrak approved this pull request.
Thank you again Akash, this LGTM. I have a couple of minor nits left, but
there's no need for another review by me after addressing them. Can you add to
ops.mlir an instance of `distribute parallel do/for simd` after line 117?
https://gith
@@ -2383,3 +2383,165 @@ func.func @masked_arg_count_mismatch(%arg0: i32, %arg1:
i32) {
}) : (i32, i32) -> ()
return
}
+
+// -
+func.func @omp_parallel_missing_composite(%lb: index, %ub: index, %step:
index) -> () {
+ omp.distribute {
+ // expected-error@+1 {{'omp.
@@ -1748,11 +1754,27 @@ LogicalResult WsloopOp::verify() {
if (!isWrapper())
return emitOpError() << "must be a loop wrapper";
+ auto wrapper =
+ llvm::dyn_cast_if_present((*this)->getParentOp());
+ bool isCompositeWrapper = wrapper && wrapper.isWrapper() &&
+
@@ -1748,11 +1754,27 @@ LogicalResult WsloopOp::verify() {
if (!isWrapper())
return emitOpError() << "must be a loop wrapper";
+ auto wrapper =
+ llvm::dyn_cast_if_present((*this)->getParentOp());
+ bool isCompositeWrapper = wrapper && wrapper.isWrapper() &&
-
@@ -210,7 +210,7 @@ func.func @invalid_nested_wrapper(%lb : index, %ub : index,
%step : index) {
omp.terminator
}
skatrak wrote:
Nit: Add attribute to `omp.distribute` too.
https://github.com/llvm/llvm-project/pull/102341
___
@@ -36,9 +36,9 @@ func.func @invalid_nested_wrapper(%lb : index, %ub : index,
%step : index) {
omp.terminator
}
skatrak wrote:
Nit: I suppose it doesn't matter because the parent op's verifier fails before
checking this one, but I think it makes
@@ -2173,7 +2173,7 @@ func.func @omp_distribute_nested_wrapper(%lb: index, %ub:
index, %step: index) -
"omp.terminator"() : () -> ()
}) : () -> ()
skatrak wrote:
Nit: Add attribute to `omp.wsloop` too.
https://github.com/llvm/llvm-project/pull/10234
@@ -1965,7 +1965,7 @@ func.func @taskloop(%lb: i32, %ub: i32, %step: i32) {
omp.terminator
}
skatrak wrote:
Nit: Add attribute to `omp.distribute` too.
https://github.com/llvm/llvm-project/pull/102341
___
l
@@ -1064,16 +1064,15 @@ static void printMapClause(OpAsmPrinter &p, Operation
*op,
}
static ParseResult parseMembersIndex(OpAsmParser &parser,
- DenseIntElementsAttr &membersIdx) {
- SmallVector values;
+
https://github.com/skatrak edited
https://github.com/llvm/llvm-project/pull/96265
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
@@ -2541,6 +2541,31 @@ static void processMapMembersWithParent(
assert(memberDataIdx >= 0 && "could not find mapped member of structure");
+// If we're currently mapping a pointer to a block of data, we must
+// initially map the pointer, and then attatch/bind the
https://github.com/skatrak approved this pull request.
Thank you Andrew for working on my previous comments. I have a couple of
suggestions to hopefully help simplify things a bit further, but it LGTM. No
need to wait for another look by me before merging.
https://github.com/llvm/llvm-project/
@@ -2261,47 +2261,47 @@ static int getMapDataMemberIdx(MapInfoData &mapData,
static mlir::omp::MapInfoOp
getFirstOrLastMappedMemberPtr(mlir::omp::MapInfoOp mapInfo, bool first) {
- mlir::DenseIntElementsAttr indexAttr = mapInfo.getMembersIndexAttr();
-
+ mlir::ArrayAttr inde
@@ -2541,6 +2541,31 @@ static void processMapMembersWithParent(
assert(memberDataIdx >= 0 && "could not find mapped member of structure");
+// If we're currently mapping a pointer to a block of data, we must
+// initially map the pointer, and then attatch/bind the
@@ -1087,50 +1086,31 @@ static ParseResult parseMembersIndex(OpAsmParser
&parser,
if (failed(parser.parseRSquare()))
return failure();
-// Only set once, if any indices are not the same size
-// we error out in the next check as that's unsupported
-if (s
https://github.com/skatrak edited
https://github.com/llvm/llvm-project/pull/96266
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
@@ -267,6 +267,24 @@ mlir::Block *fir::FirOpBuilder::getAllocaBlock() {
return getEntryBlock();
}
+static mlir::ArrayAttr makeI64ArrayAttr(llvm::ArrayRef values,
+mlir::MLIRContext *context) {
+ llvm::SmallVector attrs;
+ for (auto &
@@ -175,99 +271,63 @@ getComponentObject(std::optional object,
return getComponentObject(baseObj.value(), semaCtx);
}
-static void
-generateMemberPlacementIndices(const Object &object,
- llvm::SmallVectorImpl &indices,
-
1 - 100 of 645 matches
Mail list logo
