[MLIR][OpenMP] Host lowering of distribute-parallel-do/for #127819
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
skatrak
requested review from
Meinersbur,
jsjodin,
tblah,
kiranchandramohan and
DominikAdamski
|
@llvm/pr-subscribers-flang-openmp @llvm/pr-subscribers-mlir-openmp Author: Sergio Afonso (skatrak) ChangesThis patch adds support for translating composite Existing translation rules take care of creating a parallel region to hold the workshared and workdistributed loop. Full diff: https://github.com/llvm/llvm-project/pull/127819.diff 3 Files Affected:
diff --git a/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp b/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
index c8221a9f9854a..7e8a9bdb5b133 100644
--- a/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
+++ b/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
@@ -260,10 +260,6 @@ static LogicalResult checkImplementationStatus(Operation &op) {
LogicalResult result = success();
llvm::TypeSwitch<Operation &>(op)
.Case([&](omp::DistributeOp op) {
- if (op.isComposite() &&
- isa_and_present<omp::WsloopOp>(op.getNestedWrapper()))
- result = op.emitError() << "not yet implemented: "
- "composite omp.distribute + omp.wsloop";
checkAllocate(op, result);
checkDistSchedule(op, result);
checkOrder(op, result);
@@ -1993,6 +1989,14 @@ convertOmpWsloop(Operation &opInst, llvm::IRBuilderBase &builder,
bool isSimd = wsloopOp.getScheduleSimd();
bool loopNeedsBarrier = !wsloopOp.getNowait();
+ // The only legal way for the direct parent to be omp.distribute is that this
+ // represents 'distribute parallel do'. Otherwise, this is a regular
+ // worksharing loop.
+ llvm::omp::WorksharingLoopType workshareLoopType =
+ llvm::isa_and_present<omp::DistributeOp>(opInst.getParentOp())
+ ? llvm::omp::WorksharingLoopType::DistributeForStaticLoop
+ : llvm::omp::WorksharingLoopType::ForStaticLoop;
+
llvm::OpenMPIRBuilder::LocationDescription ompLoc(builder);
llvm::Expected<llvm::BasicBlock *> regionBlock = convertOmpOpRegions(
wsloopOp.getRegion(), "omp.wsloop.region", builder, moduleTranslation);
@@ -2008,7 +2012,8 @@ convertOmpWsloop(Operation &opInst, llvm::IRBuilderBase &builder,
ompLoc.DL, loopInfo, allocaIP, loopNeedsBarrier,
convertToScheduleKind(schedule), chunk, isSimd,
scheduleMod == omp::ScheduleModifier::monotonic,
- scheduleMod == omp::ScheduleModifier::nonmonotonic, isOrdered);
+ scheduleMod == omp::ScheduleModifier::nonmonotonic, isOrdered,
+ workshareLoopType);
if (failed(handleError(wsloopIP, opInst)))
return failure();
@@ -3792,6 +3797,12 @@ convertOmpDistribute(Operation &opInst, llvm::IRBuilderBase &builder,
return regionBlock.takeError();
builder.SetInsertPoint(*regionBlock, (*regionBlock)->begin());
+ // Skip applying a workshare loop below when translating 'distribute
+ // parallel do' (it's been already handled by this point while translating
+ // the nested omp.wsloop).
+ if (isa_and_present<omp::WsloopOp>(distributeOp.getNestedWrapper()))
+ return llvm::Error::success();
+
// TODO: Add support for clauses which are valid for DISTRIBUTE constructs.
// Static schedule is the default.
auto schedule = omp::ClauseScheduleKind::Static;
diff --git a/mlir/test/Target/LLVMIR/openmp-llvm.mlir b/mlir/test/Target/LLVMIR/openmp-llvm.mlir
index a5a490e527d79..d85b149c66811 100644
--- a/mlir/test/Target/LLVMIR/openmp-llvm.mlir
+++ b/mlir/test/Target/LLVMIR/openmp-llvm.mlir
@@ -3307,3 +3307,68 @@ llvm.func @distribute() {
// CHECK: store i64 1, ptr %[[STRIDE]]
// CHECK: %[[TID:.*]] = call i32 @__kmpc_global_thread_num({{.*}})
// CHECK: call void @__kmpc_for_static_init_{{.*}}(ptr @{{.*}}, i32 %[[TID]], i32 92, ptr %[[LASTITER]], ptr %[[LB]], ptr %[[UB]], ptr %[[STRIDE]], i64 1, i64 0)
+
+// -----
+
+llvm.func @distribute_wsloop(%lb : i32, %ub : i32, %step : i32) {
+ omp.parallel {
+ omp.distribute {
+ omp.wsloop {
+ omp.loop_nest (%iv) : i32 = (%lb) to (%ub) step (%step) {
+ omp.yield
+ }
+ } {omp.composite}
+ } {omp.composite}
+ omp.terminator
+ } {omp.composite}
+ llvm.return
+}
+
+// CHECK-LABEL: define void @distribute_wsloop
+// CHECK: call void{{.*}}@__kmpc_fork_call({{.*}}, ptr @[[OUTLINED_PARALLEL:.*]],
+
+// CHECK: define internal void @[[OUTLINED_PARALLEL]]({{.*}})
+// CHECK: %[[ARGS:.*]] = alloca { i32, i32, i32, ptr, ptr, ptr, ptr }
+// CHECK: %[[LASTITER_ALLOC:.*]] = alloca i32
+// CHECK: %[[LB_ALLOC:.*]] = alloca i32
+// CHECK: %[[UB_ALLOC:.*]] = alloca i32
+// CHECK: %[[STRIDE_ALLOC:.*]] = alloca i32
+// CHECK: %[[LB_ARG:.*]] = getelementptr {{.*}}, ptr %[[ARGS]], i32 0, i32 3
+// CHECK: store ptr %[[LB_ALLOC]], ptr %[[LB_ARG]]
+// CHECK: %[[UB_ARG:.*]] = getelementptr {{.*}}, ptr %[[ARGS]], i32 0, i32 4
+// CHECK: store ptr %[[UB_ALLOC]], ptr %[[UB_ARG]]
+// CHECK: %[[STRIDE_ARG:.*]] = getelementptr {{.*}}, ptr %[[ARGS]], i32 0, i32 5
+// CHECK: store ptr %[[STRIDE_ALLOC]], ptr %[[STRIDE_ARG]]
+// CHECK: %[[LASTITER_ARG:.*]] = getelementptr {{.*}}, ptr %[[ARGS]], i32 0, i32 6
+// CHECK: store ptr %[[LASTITER_ALLOC]], ptr %[[LASTITER_ARG]]
+// CHECK: call void @[[OUTLINED_DISTRIBUTE:.*]](ptr %[[ARGS]])
+
+// CHECK: define internal void @[[OUTLINED_DISTRIBUTE]](ptr %[[ARGS_STRUCT:.*]])
+// CHECK: %[[LB_PTR:.*]] = getelementptr {{.*}}, ptr %[[ARGS_STRUCT]], i32 0, i32 3
+// CHECK: %[[LB:.*]] = load ptr, ptr %[[LB_PTR]]
+// CHECK: %[[UB_PTR:.*]] = getelementptr {{.*}}, ptr %[[ARGS_STRUCT]], i32 0, i32 4
+// CHECK: %[[UB:.*]] = load ptr, ptr %[[UB_PTR]]
+// CHECK: %[[STRIDE_PTR:.*]] = getelementptr {{.*}}, ptr %[[ARGS_STRUCT]], i32 0, i32 5
+// CHECK: %[[STRIDE:.*]] = load ptr, ptr %[[STRIDE_PTR]]
+// CHECK: %[[LASTITER_PTR:.*]] = getelementptr {{.*}}, ptr %[[ARGS_STRUCT]], i32 0, i32 6
+// CHECK: %[[LASTITER:.*]] = load ptr, ptr %[[LASTITER_PTR]]
+// CHECK: br label %[[DISTRIBUTE_BODY:.*]]
+
+// CHECK: [[DISTRIBUTE_BODY]]:
+// CHECK-NEXT: br label %[[DISTRIBUTE_REGION:.*]]
+
+// CHECK: [[DISTRIBUTE_REGION]]:
+// CHECK-NEXT: br label %[[WSLOOP_REGION:.*]]
+
+// CHECK: [[WSLOOP_REGION]]:
+// CHECK: %omp_loop.tripcount = select {{.*}}
+// CHECK-NEXT: br label %[[PREHEADER:.*]]
+
+// CHECK: [[PREHEADER]]:
+// CHECK: store i32 0, ptr %[[LB]]
+// CHECK: %[[TRIPCOUNT:.*]] = sub i32 %omp_loop.tripcount, 1
+// CHECK: store i32 %[[TRIPCOUNT]], ptr %[[UB]]
+// CHECK: store i32 1, ptr %[[STRIDE]]
+// CHECK: %[[TID:.*]] = call i32 @__kmpc_global_thread_num({{.*}})
+// CHECK: %[[DIST_UB:.*]] = alloca i32
+// CHECK: call void @__kmpc_dist_for_static_init_{{.*}}(ptr @{{.*}}, i32 %[[TID]], i32 34, ptr %[[LASTITER]], ptr %[[LB]], ptr %[[UB]], ptr %[[DIST_UB]], ptr %[[STRIDE]], i32 1, i32 0)
diff --git a/mlir/test/Target/LLVMIR/openmp-todo.mlir b/mlir/test/Target/LLVMIR/openmp-todo.mlir
index 71dbc061c3104..d1c745af9bff5 100644
--- a/mlir/test/Target/LLVMIR/openmp-todo.mlir
+++ b/mlir/test/Target/LLVMIR/openmp-todo.mlir
@@ -66,25 +66,6 @@ llvm.func @do_simd(%lb : i32, %ub : i32, %step : i32) {
// -----
-llvm.func @distribute_wsloop(%lb : i32, %ub : i32, %step : i32) {
- // expected-error@below {{LLVM Translation failed for operation: omp.parallel}}
- omp.parallel {
- // expected-error@below {{not yet implemented: composite omp.distribute + omp.wsloop}}
- // expected-error@below {{LLVM Translation failed for operation: omp.distribute}}
- omp.distribute {
- omp.wsloop {
- omp.loop_nest (%iv) : i32 = (%lb) to (%ub) step (%step) {
- omp.yield
- }
- } {omp.composite}
- } {omp.composite}
- omp.terminator
- } {omp.composite}
- llvm.return
-}
-
-// -----
-
llvm.func @distribute_allocate(%lb : i32, %ub : i32, %step : i32, %x : !llvm.ptr) {
// expected-error@below {{not yet implemented: Unhandled clause allocate in omp.distribute operation}}
// expected-error@below {{LLVM Translation failed for operation: omp.distribute}}
|
|
@llvm/pr-subscribers-mlir Author: Sergio Afonso (skatrak) ChangesThis patch adds support for translating composite Existing translation rules take care of creating a parallel region to hold the workshared and workdistributed loop. Full diff: https://github.com/llvm/llvm-project/pull/127819.diff 3 Files Affected:
diff --git a/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp b/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
index c8221a9f9854a..7e8a9bdb5b133 100644
--- a/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
+++ b/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
@@ -260,10 +260,6 @@ static LogicalResult checkImplementationStatus(Operation &op) {
LogicalResult result = success();
llvm::TypeSwitch<Operation &>(op)
.Case([&](omp::DistributeOp op) {
- if (op.isComposite() &&
- isa_and_present<omp::WsloopOp>(op.getNestedWrapper()))
- result = op.emitError() << "not yet implemented: "
- "composite omp.distribute + omp.wsloop";
checkAllocate(op, result);
checkDistSchedule(op, result);
checkOrder(op, result);
@@ -1993,6 +1989,14 @@ convertOmpWsloop(Operation &opInst, llvm::IRBuilderBase &builder,
bool isSimd = wsloopOp.getScheduleSimd();
bool loopNeedsBarrier = !wsloopOp.getNowait();
+ // The only legal way for the direct parent to be omp.distribute is that this
+ // represents 'distribute parallel do'. Otherwise, this is a regular
+ // worksharing loop.
+ llvm::omp::WorksharingLoopType workshareLoopType =
+ llvm::isa_and_present<omp::DistributeOp>(opInst.getParentOp())
+ ? llvm::omp::WorksharingLoopType::DistributeForStaticLoop
+ : llvm::omp::WorksharingLoopType::ForStaticLoop;
+
llvm::OpenMPIRBuilder::LocationDescription ompLoc(builder);
llvm::Expected<llvm::BasicBlock *> regionBlock = convertOmpOpRegions(
wsloopOp.getRegion(), "omp.wsloop.region", builder, moduleTranslation);
@@ -2008,7 +2012,8 @@ convertOmpWsloop(Operation &opInst, llvm::IRBuilderBase &builder,
ompLoc.DL, loopInfo, allocaIP, loopNeedsBarrier,
convertToScheduleKind(schedule), chunk, isSimd,
scheduleMod == omp::ScheduleModifier::monotonic,
- scheduleMod == omp::ScheduleModifier::nonmonotonic, isOrdered);
+ scheduleMod == omp::ScheduleModifier::nonmonotonic, isOrdered,
+ workshareLoopType);
if (failed(handleError(wsloopIP, opInst)))
return failure();
@@ -3792,6 +3797,12 @@ convertOmpDistribute(Operation &opInst, llvm::IRBuilderBase &builder,
return regionBlock.takeError();
builder.SetInsertPoint(*regionBlock, (*regionBlock)->begin());
+ // Skip applying a workshare loop below when translating 'distribute
+ // parallel do' (it's been already handled by this point while translating
+ // the nested omp.wsloop).
+ if (isa_and_present<omp::WsloopOp>(distributeOp.getNestedWrapper()))
+ return llvm::Error::success();
+
// TODO: Add support for clauses which are valid for DISTRIBUTE constructs.
// Static schedule is the default.
auto schedule = omp::ClauseScheduleKind::Static;
diff --git a/mlir/test/Target/LLVMIR/openmp-llvm.mlir b/mlir/test/Target/LLVMIR/openmp-llvm.mlir
index a5a490e527d79..d85b149c66811 100644
--- a/mlir/test/Target/LLVMIR/openmp-llvm.mlir
+++ b/mlir/test/Target/LLVMIR/openmp-llvm.mlir
@@ -3307,3 +3307,68 @@ llvm.func @distribute() {
// CHECK: store i64 1, ptr %[[STRIDE]]
// CHECK: %[[TID:.*]] = call i32 @__kmpc_global_thread_num({{.*}})
// CHECK: call void @__kmpc_for_static_init_{{.*}}(ptr @{{.*}}, i32 %[[TID]], i32 92, ptr %[[LASTITER]], ptr %[[LB]], ptr %[[UB]], ptr %[[STRIDE]], i64 1, i64 0)
+
+// -----
+
+llvm.func @distribute_wsloop(%lb : i32, %ub : i32, %step : i32) {
+ omp.parallel {
+ omp.distribute {
+ omp.wsloop {
+ omp.loop_nest (%iv) : i32 = (%lb) to (%ub) step (%step) {
+ omp.yield
+ }
+ } {omp.composite}
+ } {omp.composite}
+ omp.terminator
+ } {omp.composite}
+ llvm.return
+}
+
+// CHECK-LABEL: define void @distribute_wsloop
+// CHECK: call void{{.*}}@__kmpc_fork_call({{.*}}, ptr @[[OUTLINED_PARALLEL:.*]],
+
+// CHECK: define internal void @[[OUTLINED_PARALLEL]]({{.*}})
+// CHECK: %[[ARGS:.*]] = alloca { i32, i32, i32, ptr, ptr, ptr, ptr }
+// CHECK: %[[LASTITER_ALLOC:.*]] = alloca i32
+// CHECK: %[[LB_ALLOC:.*]] = alloca i32
+// CHECK: %[[UB_ALLOC:.*]] = alloca i32
+// CHECK: %[[STRIDE_ALLOC:.*]] = alloca i32
+// CHECK: %[[LB_ARG:.*]] = getelementptr {{.*}}, ptr %[[ARGS]], i32 0, i32 3
+// CHECK: store ptr %[[LB_ALLOC]], ptr %[[LB_ARG]]
+// CHECK: %[[UB_ARG:.*]] = getelementptr {{.*}}, ptr %[[ARGS]], i32 0, i32 4
+// CHECK: store ptr %[[UB_ALLOC]], ptr %[[UB_ARG]]
+// CHECK: %[[STRIDE_ARG:.*]] = getelementptr {{.*}}, ptr %[[ARGS]], i32 0, i32 5
+// CHECK: store ptr %[[STRIDE_ALLOC]], ptr %[[STRIDE_ARG]]
+// CHECK: %[[LASTITER_ARG:.*]] = getelementptr {{.*}}, ptr %[[ARGS]], i32 0, i32 6
+// CHECK: store ptr %[[LASTITER_ALLOC]], ptr %[[LASTITER_ARG]]
+// CHECK: call void @[[OUTLINED_DISTRIBUTE:.*]](ptr %[[ARGS]])
+
+// CHECK: define internal void @[[OUTLINED_DISTRIBUTE]](ptr %[[ARGS_STRUCT:.*]])
+// CHECK: %[[LB_PTR:.*]] = getelementptr {{.*}}, ptr %[[ARGS_STRUCT]], i32 0, i32 3
+// CHECK: %[[LB:.*]] = load ptr, ptr %[[LB_PTR]]
+// CHECK: %[[UB_PTR:.*]] = getelementptr {{.*}}, ptr %[[ARGS_STRUCT]], i32 0, i32 4
+// CHECK: %[[UB:.*]] = load ptr, ptr %[[UB_PTR]]
+// CHECK: %[[STRIDE_PTR:.*]] = getelementptr {{.*}}, ptr %[[ARGS_STRUCT]], i32 0, i32 5
+// CHECK: %[[STRIDE:.*]] = load ptr, ptr %[[STRIDE_PTR]]
+// CHECK: %[[LASTITER_PTR:.*]] = getelementptr {{.*}}, ptr %[[ARGS_STRUCT]], i32 0, i32 6
+// CHECK: %[[LASTITER:.*]] = load ptr, ptr %[[LASTITER_PTR]]
+// CHECK: br label %[[DISTRIBUTE_BODY:.*]]
+
+// CHECK: [[DISTRIBUTE_BODY]]:
+// CHECK-NEXT: br label %[[DISTRIBUTE_REGION:.*]]
+
+// CHECK: [[DISTRIBUTE_REGION]]:
+// CHECK-NEXT: br label %[[WSLOOP_REGION:.*]]
+
+// CHECK: [[WSLOOP_REGION]]:
+// CHECK: %omp_loop.tripcount = select {{.*}}
+// CHECK-NEXT: br label %[[PREHEADER:.*]]
+
+// CHECK: [[PREHEADER]]:
+// CHECK: store i32 0, ptr %[[LB]]
+// CHECK: %[[TRIPCOUNT:.*]] = sub i32 %omp_loop.tripcount, 1
+// CHECK: store i32 %[[TRIPCOUNT]], ptr %[[UB]]
+// CHECK: store i32 1, ptr %[[STRIDE]]
+// CHECK: %[[TID:.*]] = call i32 @__kmpc_global_thread_num({{.*}})
+// CHECK: %[[DIST_UB:.*]] = alloca i32
+// CHECK: call void @__kmpc_dist_for_static_init_{{.*}}(ptr @{{.*}}, i32 %[[TID]], i32 34, ptr %[[LASTITER]], ptr %[[LB]], ptr %[[UB]], ptr %[[DIST_UB]], ptr %[[STRIDE]], i32 1, i32 0)
diff --git a/mlir/test/Target/LLVMIR/openmp-todo.mlir b/mlir/test/Target/LLVMIR/openmp-todo.mlir
index 71dbc061c3104..d1c745af9bff5 100644
--- a/mlir/test/Target/LLVMIR/openmp-todo.mlir
+++ b/mlir/test/Target/LLVMIR/openmp-todo.mlir
@@ -66,25 +66,6 @@ llvm.func @do_simd(%lb : i32, %ub : i32, %step : i32) {
// -----
-llvm.func @distribute_wsloop(%lb : i32, %ub : i32, %step : i32) {
- // expected-error@below {{LLVM Translation failed for operation: omp.parallel}}
- omp.parallel {
- // expected-error@below {{not yet implemented: composite omp.distribute + omp.wsloop}}
- // expected-error@below {{LLVM Translation failed for operation: omp.distribute}}
- omp.distribute {
- omp.wsloop {
- omp.loop_nest (%iv) : i32 = (%lb) to (%ub) step (%step) {
- omp.yield
- }
- } {omp.composite}
- } {omp.composite}
- omp.terminator
- } {omp.composite}
- llvm.return
-}
-
-// -----
-
llvm.func @distribute_allocate(%lb : i32, %ub : i32, %step : i32, %x : !llvm.ptr) {
// expected-error@below {{not yet implemented: Unhandled clause allocate in omp.distribute operation}}
// expected-error@below {{LLVM Translation failed for operation: omp.distribute}}
|
This was referenced Feb 19, 2025
This was referenced Feb 19, 2025
skatrak
force-pushed
the
users/skatrak/target-spmd-04-ompirbuilder-composite
branch
from
cb7ae2d to
dbe0d70
Compare
skatrak
force-pushed
the
users/skatrak/target-spmd-05-mlir-composite
branch
from
38ba269 to
33d5af4
Compare
skatrak
force-pushed
the
users/skatrak/target-spmd-04-ompirbuilder-composite
branch
from
dbe0d70 to
ba9ea8c
Compare
skatrak
force-pushed
the
users/skatrak/target-spmd-05-mlir-composite
branch
from
33d5af4 to
aad04fa
Compare
jsjodin
approved these changes
Feb 22, 2025
skatrak
force-pushed
the
users/skatrak/target-spmd-04-ompirbuilder-composite
branch
from
ba9ea8c to
d164acf
Compare
skatrak
force-pushed
the
users/skatrak/target-spmd-05-mlir-composite
branch
from
1172e9b to
e61c797
Compare
skatrak
force-pushed
the
users/skatrak/target-spmd-04-ompirbuilder-composite
branch
from
d164acf to
6bd8d01
Compare
Base automatically changed from
users/skatrak/target-spmd-04-ompirbuilder-composite
to
main
February 25, 2025 10:30
This patch adds support for translating composite `omp.parallel` + `omp.distribute` + `omp.wsloop` loops to LLVM IR on the host. This is done by passing an updated `WorksharingLoopType` to the call to `applyWorkshareLoop` associated to the lowering of the `omp.wsloop` operation, so that `__kmpc_dist_for_static_init` is called at runtime in place of `__kmpc_for_static_init`. Existing translation rules take care of creating a parallel region to hold the workshared and workdistributed loop.
skatrak
force-pushed
the
users/skatrak/target-spmd-05-mlir-composite
branch
from
e61c797 to
ac8b967
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This patch adds support for translating composite
omp.parallel+omp.distribute+omp.wslooploops to LLVM IR on the host. This is done by passing an updatedWorksharingLoopTypeto the call toapplyWorkshareLoopassociated to the lowering of theomp.wsloopoperation, so that__kmpc_dist_for_static_initis called at runtime in place of__kmpc_for_static_init.Existing translation rules take care of creating a parallel region to hold the workshared and workdistributed loop.