[llvm-branch-commits] [flang] [Flang][OpenMP] Update flang with changes to the OpenMP dialect (PR #92524)
Sergio Afonso via llvm-branch-commits
llvm-branch-commits at lists.llvm.org
Mon Jul 1 02:27:35 PDT 2024
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 <safonsof at amd.com>
Date: Fri, 17 May 2024 11:38:36 +0100
Subject: [PATCH] [Flang][OpenMP] Update flang with changes to the OpenMP
dialect
This patch applies fixes after the updates to OpenMP clause operands, as well
as updating some tests that were impacted by changes to the ordering or
assembly format of some clauses in MLIR.
---
flang/lib/Lower/OpenMP/ClauseProcessor.cpp | 4 ++--
flang/lib/Lower/OpenMP/ClauseProcessor.h | 4 ++--
flang/lib/Lower/OpenMP/OpenMP.cpp | 19 ++++++++-------
flang/test/Lower/OpenMP/atomic-capture.f90 | 2 +-
flang/test/Lower/OpenMP/copyin-order.f90 | 2 +-
flang/test/Lower/OpenMP/parallel-wsloop.f90 | 2 +-
flang/test/Lower/OpenMP/parallel.f90 | 24 +++++++++----------
flang/test/Lower/OpenMP/simd.f90 | 2 +-
flang/test/Lower/OpenMP/target.f90 | 24 +++++++++----------
.../use-device-ptr-to-use-device-addr.f90 | 2 +-
10 files changed, 43 insertions(+), 42 deletions(-)
diff --git a/flang/lib/Lower/OpenMP/ClauseProcessor.cpp b/flang/lib/Lower/OpenMP/ClauseProcessor.cpp
index f78cd0f9df1a1..d507e58b164dd 100644
--- a/flang/lib/Lower/OpenMP/ClauseProcessor.cpp
+++ b/flang/lib/Lower/OpenMP/ClauseProcessor.cpp
@@ -1073,7 +1073,7 @@ bool ClauseProcessor::processEnter(
}
bool ClauseProcessor::processUseDeviceAddr(
- mlir::omp::UseDeviceClauseOps &result,
+ mlir::omp::UseDeviceAddrClauseOps &result,
llvm::SmallVectorImpl<mlir::Type> &useDeviceTypes,
llvm::SmallVectorImpl<mlir::Location> &useDeviceLocs,
llvm::SmallVectorImpl<const semantics::Symbol *> &useDeviceSyms) const {
@@ -1085,7 +1085,7 @@ bool ClauseProcessor::processUseDeviceAddr(
}
bool ClauseProcessor::processUseDevicePtr(
- mlir::omp::UseDeviceClauseOps &result,
+ mlir::omp::UseDevicePtrClauseOps &result,
llvm::SmallVectorImpl<mlir::Type> &useDeviceTypes,
llvm::SmallVectorImpl<mlir::Location> &useDeviceLocs,
llvm::SmallVectorImpl<const semantics::Symbol *> &useDeviceSyms) const {
diff --git a/flang/lib/Lower/OpenMP/ClauseProcessor.h b/flang/lib/Lower/OpenMP/ClauseProcessor.h
index 53571ae5abc20..43795d5c25399 100644
--- a/flang/lib/Lower/OpenMP/ClauseProcessor.h
+++ b/flang/lib/Lower/OpenMP/ClauseProcessor.h
@@ -129,12 +129,12 @@ class ClauseProcessor {
mlir::omp::ReductionClauseOps &result) const;
bool processTo(llvm::SmallVectorImpl<DeclareTargetCapturePair> &result) const;
bool processUseDeviceAddr(
- mlir::omp::UseDeviceClauseOps &result,
+ mlir::omp::UseDeviceAddrClauseOps &result,
llvm::SmallVectorImpl<mlir::Type> &useDeviceTypes,
llvm::SmallVectorImpl<mlir::Location> &useDeviceLocs,
llvm::SmallVectorImpl<const semantics::Symbol *> &useDeviceSyms) const;
bool processUseDevicePtr(
- mlir::omp::UseDeviceClauseOps &result,
+ mlir::omp::UseDevicePtrClauseOps &result,
llvm::SmallVectorImpl<mlir::Type> &useDeviceTypes,
llvm::SmallVectorImpl<mlir::Location> &useDeviceLocs,
llvm::SmallVectorImpl<const semantics::Symbol *> &useDeviceSyms) const;
diff --git a/flang/lib/Lower/OpenMP/OpenMP.cpp b/flang/lib/Lower/OpenMP/OpenMP.cpp
index 22c41fce31723..d8679fb693659 100644
--- a/flang/lib/Lower/OpenMP/OpenMP.cpp
+++ b/flang/lib/Lower/OpenMP/OpenMP.cpp
@@ -244,7 +244,8 @@ createAndSetPrivatizedLoopVar(lower::AbstractConverter &converter,
// clause. Support for such list items in a use_device_ptr clause
// is deprecated."
static void promoteNonCPtrUseDevicePtrArgsToUseDeviceAddr(
- mlir::omp::UseDeviceClauseOps &clauseOps,
+ llvm::SmallVectorImpl<mlir::Value> &useDeviceAddrVars,
+ llvm::SmallVectorImpl<mlir::Value> &useDevicePtrVars,
llvm::SmallVectorImpl<mlir::Type> &useDeviceTypes,
llvm::SmallVectorImpl<mlir::Location> &useDeviceLocs,
llvm::SmallVectorImpl<const semantics::Symbol *> &useDeviceSymbols) {
@@ -256,10 +257,9 @@ static void promoteNonCPtrUseDevicePtrArgsToUseDeviceAddr(
// Iterate over our use_device_ptr list and shift all non-cptr arguments into
// use_device_addr.
- for (auto *it = clauseOps.useDevicePtrVars.begin();
- it != clauseOps.useDevicePtrVars.end();) {
+ for (auto *it = useDevicePtrVars.begin(); it != useDevicePtrVars.end();) {
if (!fir::isa_builtin_cptr_type(fir::unwrapRefType(it->getType()))) {
- clauseOps.useDeviceAddrVars.push_back(*it);
+ useDeviceAddrVars.push_back(*it);
// We have to shuffle the symbols around as well, to maintain
// the correct Input -> BlockArg for use_device_ptr/use_device_addr.
// NOTE: However, as map's do not seem to be included currently
@@ -267,11 +267,11 @@ static void promoteNonCPtrUseDevicePtrArgsToUseDeviceAddr(
// future alterations. I believe the reason they are not currently
// is that the BlockArg assign/lowering needs to be extended
// to a greater set of types.
- auto idx = std::distance(clauseOps.useDevicePtrVars.begin(), it);
+ auto idx = std::distance(useDevicePtrVars.begin(), it);
moveElementToBack(idx, useDeviceTypes);
moveElementToBack(idx, useDeviceLocs);
moveElementToBack(idx, useDeviceSymbols);
- it = clauseOps.useDevicePtrVars.erase(it);
+ it = useDevicePtrVars.erase(it);
continue;
}
++it;
@@ -1024,7 +1024,7 @@ static void genCriticalDeclareClauses(lower::AbstractConverter &converter,
llvm::StringRef name) {
ClauseProcessor cp(converter, semaCtx, clauses);
cp.processHint(clauseOps);
- clauseOps.nameAttr =
+ clauseOps.criticalNameAttr =
mlir::StringAttr::get(converter.getFirOpBuilder().getContext(), name);
}
@@ -1196,8 +1196,9 @@ 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, useDeviceTypes,
- useDeviceLocs, useDeviceSyms);
+ promoteNonCPtrUseDevicePtrArgsToUseDeviceAddr(
+ clauseOps.useDeviceAddrVars, clauseOps.useDevicePtrVars, useDeviceTypes,
+ useDeviceLocs, useDeviceSyms);
}
static void genTargetEnterExitUpdateDataClauses(
diff --git a/flang/test/Lower/OpenMP/atomic-capture.f90 b/flang/test/Lower/OpenMP/atomic-capture.f90
index 32d8cd7bbf328..667ae8ed7a133 100644
--- a/flang/test/Lower/OpenMP/atomic-capture.f90
+++ b/flang/test/Lower/OpenMP/atomic-capture.f90
@@ -35,7 +35,7 @@ program OmpAtomicCapture
!CHECK: %[[SUB:.*]] = arith.subi %[[VAL_8]], %[[VAL_X_LOADED]] : i32
!CHECK: %[[NO_REASSOC:.*]] = hlfir.no_reassoc %[[SUB]] : i32
!CHECK: %[[ADD:.*]] = arith.addi %[[VAL_20]], %[[NO_REASSOC]] : i32
-!CHECK: omp.atomic.capture memory_order(acquire) hint(nonspeculative) {
+!CHECK: omp.atomic.capture hint(nonspeculative) memory_order(acquire) {
!CHECK: omp.atomic.read %[[VAL_X_DECLARE]]#1 = %[[VAL_Y_DECLARE]]#1 : !fir.ref<i32>, i32
!CHECK: omp.atomic.write %[[VAL_Y_DECLARE]]#1 = %[[ADD]] : !fir.ref<i32>, i32
!CHECK: }
diff --git a/flang/test/Lower/OpenMP/copyin-order.f90 b/flang/test/Lower/OpenMP/copyin-order.f90
index 0620d89ffb678..8999c247eebdc 100644
--- a/flang/test/Lower/OpenMP/copyin-order.f90
+++ b/flang/test/Lower/OpenMP/copyin-order.f90
@@ -2,7 +2,7 @@
!https://github.com/llvm/llvm-project/issues/91205
-!CHECK: omp.parallel if(%{{[0-9]+}} : i1) {
+!CHECK: omp.parallel if(%{{[0-9]+}}) {
!CHECK: %[[THP1:[0-9]+]] = omp.threadprivate %{{[0-9]+}}#1
!CHECK: %[[DCL1:[0-9]+]]:2 = hlfir.declare %[[THP1]] {uniq_name = "_QFcopyin_scalar_arrayEx1"}
!CHECK: %[[LD1:[0-9]+]] = fir.load %{{[0-9]+}}#0
diff --git a/flang/test/Lower/OpenMP/parallel-wsloop.f90 b/flang/test/Lower/OpenMP/parallel-wsloop.f90
index e5c303d7bb2ed..e0257533af4f5 100644
--- a/flang/test/Lower/OpenMP/parallel-wsloop.f90
+++ b/flang/test/Lower/OpenMP/parallel-wsloop.f90
@@ -35,7 +35,7 @@ subroutine parallel_do_with_parallel_clauses(cond, nt)
! CHECK: %[[COND:.*]] = fir.load %[[COND_DECL]]#0 : !fir.ref<!fir.logical<4>>
! CHECK: %[[COND_CVT:.*]] = fir.convert %[[COND]] : (!fir.logical<4>) -> i1
! CHECK: %[[NT:.*]] = fir.load %[[NT_DECL]]#0 : !fir.ref<i32>
- ! CHECK: omp.parallel if(%[[COND_CVT]] : i1) num_threads(%[[NT]] : i32) proc_bind(close)
+ ! CHECK: omp.parallel if(%[[COND_CVT]]) num_threads(%[[NT]] : i32) proc_bind(close)
! CHECK: %[[WS_LB:.*]] = arith.constant 1 : i32
! CHECK: %[[WS_UB:.*]] = arith.constant 9 : i32
! CHECK: %[[WS_STEP:.*]] = arith.constant 1 : i32
diff --git a/flang/test/Lower/OpenMP/parallel.f90 b/flang/test/Lower/OpenMP/parallel.f90
index 6b4307b4d4357..07f4680f79ee2 100644
--- a/flang/test/Lower/OpenMP/parallel.f90
+++ b/flang/test/Lower/OpenMP/parallel.f90
@@ -24,63 +24,63 @@ subroutine parallel_if(alpha, beta, gamma)
logical(4) :: logical4
logical(8) :: logical8
- !CHECK: omp.parallel if(%{{.*}} : i1) {
+ !CHECK: omp.parallel if(%{{.*}}) {
!$omp parallel if(alpha .le. 0)
!CHECK: fir.call
call f1()
!CHECK: omp.terminator
!$omp end parallel
- !CHECK: omp.parallel if(%{{.*}} : i1) {
+ !CHECK: omp.parallel if(%{{.*}}) {
!$omp parallel if(.false.)
!CHECK: fir.call
call f2()
!CHECK: omp.terminator
!$omp end parallel
- !CHECK: omp.parallel if(%{{.*}} : i1) {
+ !CHECK: omp.parallel if(%{{.*}}) {
!$omp parallel if(alpha .ge. 0)
!CHECK: fir.call
call f3()
!CHECK: omp.terminator
!$omp end parallel
- !CHECK: omp.parallel if(%{{.*}} : i1) {
+ !CHECK: omp.parallel if(%{{.*}}) {
!$omp parallel if(.true.)
!CHECK: fir.call
call f4()
!CHECK: omp.terminator
!$omp end parallel
- !CHECK: omp.parallel if(%{{.*}} : i1) {
+ !CHECK: omp.parallel if(%{{.*}}) {
!$omp parallel if(beta)
!CHECK: fir.call
call f1()
!CHECK: omp.terminator
!$omp end parallel
- !CHECK: omp.parallel if(%{{.*}} : i1) {
+ !CHECK: omp.parallel if(%{{.*}}) {
!$omp parallel if(logical1)
!CHECK: fir.call
call f1()
!CHECK: omp.terminator
!$omp end parallel
- !CHECK: omp.parallel if(%{{.*}} : i1) {
+ !CHECK: omp.parallel if(%{{.*}}) {
!$omp parallel if(logical2)
!CHECK: fir.call
call f1()
!CHECK: omp.terminator
!$omp end parallel
- !CHECK: omp.parallel if(%{{.*}} : i1) {
+ !CHECK: omp.parallel if(%{{.*}}) {
!$omp parallel if(logical4)
!CHECK: fir.call
call f1()
!CHECK: omp.terminator
!$omp end parallel
- !CHECK: omp.parallel if(%{{.*}} : i1) {
+ !CHECK: omp.parallel if(%{{.*}}) {
!$omp parallel if(logical8)
!CHECK: fir.call
call f1()
@@ -173,7 +173,7 @@ subroutine parallel_multiple_clauses(alpha, num_threads)
integer, intent(inout) :: alpha
integer, intent(in) :: num_threads
- !CHECK: omp.parallel if({{.*}} : i1) proc_bind(master) {
+ !CHECK: omp.parallel if({{.*}}) proc_bind(master) {
!$omp parallel if(alpha .le. 0) proc_bind(master)
!CHECK: fir.call
call f1()
@@ -187,14 +187,14 @@ subroutine parallel_multiple_clauses(alpha, num_threads)
!CHECK: omp.terminator
!$omp end parallel
- !CHECK: omp.parallel if({{.*}} : i1) num_threads({{.*}} : i32) {
+ !CHECK: omp.parallel if({{.*}}) num_threads({{.*}} : i32) {
!$omp parallel num_threads(num_threads) if(alpha .le. 0)
!CHECK: fir.call
call f3()
!CHECK: omp.terminator
!$omp end parallel
- !CHECK: omp.parallel if({{.*}} : i1) num_threads({{.*}} : i32) allocate(
+ !CHECK: omp.parallel if({{.*}}) num_threads({{.*}} : i32) allocate(
!CHECK: %{{.+}} : i64 -> %{{.+}} : !fir.ref<i32>
!CHECK: ) {
!$omp parallel num_threads(num_threads) if(alpha .le. 0) allocate(omp_high_bw_mem_alloc: alpha) private(alpha)
diff --git a/flang/test/Lower/OpenMP/simd.f90 b/flang/test/Lower/OpenMP/simd.f90
index e98136dd57b0a..66b7e463b2357 100644
--- a/flang/test/Lower/OpenMP/simd.f90
+++ b/flang/test/Lower/OpenMP/simd.f90
@@ -148,7 +148,7 @@ subroutine simd_with_simdlen_safelen_clause(n, threshold)
! CHECK: %[[LB:.*]] = arith.constant 1 : i32
! CHECK: %[[UB:.*]] = fir.load %[[ARG_N]]#0
! CHECK: %[[STEP:.*]] = arith.constant 1 : i32
- ! CHECK: omp.simd simdlen(1) safelen(2) {
+ ! CHECK: omp.simd safelen(2) simdlen(1) {
! CHECK-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]]) {
do i = 1, n
! CHECK: fir.store %[[I]] to %[[LOCAL:.*]]#1 : !fir.ref<i32>
diff --git a/flang/test/Lower/OpenMP/target.f90 b/flang/test/Lower/OpenMP/target.f90
index 0fcbad7e58fee..0500b21a84cb0 100644
--- a/flang/test/Lower/OpenMP/target.f90
+++ b/flang/test/Lower/OpenMP/target.f90
@@ -30,7 +30,7 @@ subroutine omp_target_enter_depend
!$omp end task
!CHECK: %[[BOUNDS:.*]] = omp.map.bounds lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
!CHECK: %[[MAP:.*]] = omp.map.info var_ptr({{.*}}) map_clauses(to) capture(ByRef) bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
- !CHECK: omp.target_enter_data map_entries(%[[MAP]] : !fir.ref<!fir.array<1024xi32>>) depend(taskdependin -> %[[A]]#1 : !fir.ref<!fir.array<1024xi32>>)
+ !CHECK: omp.target_enter_data depend(taskdependin -> %[[A]]#1 : !fir.ref<!fir.array<1024xi32>>) map_entries(%[[MAP]] : !fir.ref<!fir.array<1024xi32>>)
!$omp target enter data map(to: a) depend(in: a)
return
end subroutine omp_target_enter_depend
@@ -84,7 +84,7 @@ subroutine omp_target_enter_if
!CHECK: %[[VAL_5:.*]] = arith.cmpi slt, %[[VAL_3]], %[[VAL_4]] : i32
!CHECK: %[[BOUNDS:.*]] = omp.map.bounds lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
!CHECK: %[[MAP:.*]] = omp.map.info var_ptr({{.*}}) map_clauses(to) capture(ByRef) bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
- !CHECK: omp.target_enter_data if(%[[VAL_5]] : i1) map_entries(%[[MAP]] : !fir.ref<!fir.array<1024xi32>>)
+ !CHECK: omp.target_enter_data if(%[[VAL_5]]) map_entries(%[[MAP]] : !fir.ref<!fir.array<1024xi32>>)
!$omp target enter data if(i<10) map(to: a)
end subroutine omp_target_enter_if
@@ -169,7 +169,7 @@ subroutine omp_target_exit_depend
!$omp end task
!CHECK: %[[BOUNDS:.*]] = omp.map.bounds lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
!CHECK: %[[MAP:.*]] = omp.map.info var_ptr({{.*}}) map_clauses(from) capture(ByRef) bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
- !CHECK: omp.target_exit_data map_entries(%[[MAP]] : !fir.ref<!fir.array<1024xi32>>) depend(taskdependout -> %[[A]]#1 : !fir.ref<!fir.array<1024xi32>>)
+ !CHECK: omp.target_exit_data depend(taskdependout -> %[[A]]#1 : !fir.ref<!fir.array<1024xi32>>) map_entries(%[[MAP]] : !fir.ref<!fir.array<1024xi32>>)
!$omp target exit data map(from: a) depend(out: a)
end subroutine omp_target_exit_depend
@@ -190,7 +190,7 @@ subroutine omp_target_update_depend
!CHECK: %[[BOUNDS:.*]] = omp.map.bounds
!CHECK: %[[MAP:.*]] = omp.map.info var_ptr(%[[A]]#0 : !fir.ref<!fir.array<1024xi32>>, !fir.array<1024xi32>) map_clauses(to) capture(ByRef) bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
- !CHECK: omp.target_update motion_entries(%[[MAP]] : !fir.ref<!fir.array<1024xi32>>) depend(taskdependin -> %[[A]]#1 : !fir.ref<!fir.array<1024xi32>>)
+ !CHECK: omp.target_update depend(taskdependin -> %[[A]]#1 : !fir.ref<!fir.array<1024xi32>>) map_entries(%[[MAP]] : !fir.ref<!fir.array<1024xi32>>)
!$omp target update to(a) depend(in:a)
end subroutine omp_target_update_depend
@@ -209,7 +209,7 @@ subroutine omp_target_update_to
!CHECK-SAME: map_clauses(to) capture(ByRef)
!CHECK-SAME: bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
- !CHECK: omp.target_update motion_entries(%[[TO_MAP]] : !fir.ref<!fir.array<1024xi32>>)
+ !CHECK: omp.target_update map_entries(%[[TO_MAP]] : !fir.ref<!fir.array<1024xi32>>)
!$omp target update to(a)
end subroutine omp_target_update_to
@@ -228,7 +228,7 @@ subroutine omp_target_update_from
!CHECK-SAME: map_clauses(from) capture(ByRef)
!CHECK-SAME: bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
- !CHECK: omp.target_update motion_entries(%[[FROM_MAP]] : !fir.ref<!fir.array<1024xi32>>)
+ !CHECK: omp.target_update map_entries(%[[FROM_MAP]] : !fir.ref<!fir.array<1024xi32>>)
!$omp target update from(a)
end subroutine omp_target_update_from
@@ -245,7 +245,7 @@ subroutine omp_target_update_if
!CHECK-DAG: %[[BOUNDS:.*]] = omp.map.bounds
!CHECK-DAG: %[[COND:.*]] = fir.convert %{{.*}} : (!fir.logical<4>) -> i1
- !CHECK: omp.target_update if(%[[COND]] : i1) motion_entries
+ !CHECK: omp.target_update if(%[[COND]]) map_entries
!$omp target update from(a) if(i)
end subroutine omp_target_update_if
@@ -262,7 +262,7 @@ subroutine omp_target_update_device
!CHECK-DAG: %[[BOUNDS:.*]] = omp.map.bounds
!CHECK-DAG: %[[DEVICE:.*]] = arith.constant 1 : i32
- !CHECK: omp.target_update device(%[[DEVICE]] : i32) motion_entries
+ !CHECK: omp.target_update device(%[[DEVICE]] : i32) map_entries
!$omp target update from(a) device(1)
end subroutine omp_target_update_device
@@ -278,7 +278,7 @@ subroutine omp_target_update_nowait
!CHECK-DAG: %[[A_DECL:.*]]:2 = hlfir.declare %{{.*}}(%{{.*}})
!CHECK-DAG: %[[BOUNDS:.*]] = omp.map.bounds
- !CHECK: omp.target_update nowait motion_entries
+ !CHECK: omp.target_update nowait map_entries
!$omp target update from(a) nowait
end subroutine omp_target_update_nowait
@@ -373,7 +373,7 @@ subroutine omp_target_depend
!CHECK: %[[UBOUND_A:.*]] = arith.subi %c1024, %c1 : index
!CHECK: %[[BOUNDS_A:.*]] = omp.map.bounds lower_bound(%[[LBOUND_A]] : index) upper_bound(%[[UBOUND_A]] : index) extent(%[[EXTENT_A]] : index) stride(%[[STRIDE_A]] : index) start_idx(%[[STRIDE_A]] : index)
!CHECK: %[[MAP_A:.*]] = omp.map.info var_ptr(%[[A]]#0 : !fir.ref<!fir.array<1024xi32>>, !fir.array<1024xi32>) map_clauses(tofrom) capture(ByRef) bounds(%[[BOUNDS_A]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
- !CHECK: omp.target map_entries(%[[MAP_A]] -> %[[BB0_ARG:.*]] : !fir.ref<!fir.array<1024xi32>>) depend(taskdependin -> %[[A]]#1 : !fir.ref<!fir.array<1024xi32>>) {
+ !CHECK: omp.target depend(taskdependin -> %[[A]]#1 : !fir.ref<!fir.array<1024xi32>>) map_entries(%[[MAP_A]] -> %[[BB0_ARG:.*]] : !fir.ref<!fir.array<1024xi32>>) {
!$omp target map(tofrom: a) depend(in: a)
a(1) = 10
!CHECK: omp.terminator
@@ -512,7 +512,7 @@ subroutine omp_target_device_ptr
type(c_ptr) :: a
integer, target :: b
!CHECK: %[[MAP:.*]] = omp.map.info var_ptr({{.*}}) map_clauses(tofrom) capture(ByRef) -> {{.*}} {name = "a"}
- !CHECK: omp.target_data map_entries(%[[MAP]]{{.*}}
+ !CHECK: omp.target_data use_device_ptr({{.*}}) map_entries(%[[MAP]]{{.*}}
!$omp target data map(tofrom: a) use_device_ptr(a)
!CHECK: ^bb0(%[[VAL_1:.*]]: !fir.ref<!fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>>):
!CHECK: {{.*}} = fir.coordinate_of %[[VAL_1:.*]], {{.*}} : (!fir.ref<!fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>>, !fir.field) -> !fir.ref<i64>
@@ -533,7 +533,7 @@ subroutine omp_target_device_addr
!CHECK: %[[VAL_0_DECL:.*]]:2 = hlfir.declare %0 {fortran_attrs = #fir.var_attrs<pointer>, uniq_name = "_QFomp_target_device_addrEa"} : (!fir.ref<!fir.box<!fir.ptr<i32>>>) -> (!fir.ref<!fir.box<!fir.ptr<i32>>>, !fir.ref<!fir.box<!fir.ptr<i32>>>)
!CHECK: %[[MAP_MEMBERS:.*]] = omp.map.info var_ptr({{.*}} : !fir.ref<!fir.box<!fir.ptr<i32>>>, i32) var_ptr_ptr({{.*}} : !fir.llvm_ptr<!fir.ref<i32>>) map_clauses(tofrom) capture(ByRef) -> !fir.llvm_ptr<!fir.ref<i32>> {name = ""}
!CHECK: %[[MAP:.*]] = omp.map.info var_ptr({{.*}} : !fir.ref<!fir.box<!fir.ptr<i32>>>, !fir.box<!fir.ptr<i32>>) map_clauses(tofrom) capture(ByRef) members(%[[MAP_MEMBERS]] : [0] : !fir.llvm_ptr<!fir.ref<i32>>) -> !fir.ref<!fir.box<!fir.ptr<i32>>> {name = "a"}
- !CHECK: omp.target_data map_entries(%[[MAP_MEMBERS]], %[[MAP]] : {{.*}}) use_device_addr(%[[VAL_0_DECL]]#1 : !fir.ref<!fir.box<!fir.ptr<i32>>>) {
+ !CHECK: omp.target_data use_device_addr(%[[VAL_0_DECL]]#1 : !fir.ref<!fir.box<!fir.ptr<i32>>>) map_entries(%[[MAP_MEMBERS]], %[[MAP]] : {{.*}}) {
!$omp target data map(tofrom: a) use_device_addr(a)
!CHECK: ^bb0(%[[VAL_1:.*]]: !fir.ref<!fir.box<!fir.ptr<i32>>>):
!CHECK: %[[VAL_1_DECL:.*]]:2 = hlfir.declare %[[VAL_1]] {fortran_attrs = #fir.var_attrs<pointer>, uniq_name = "_QFomp_target_device_addrEa"} : (!fir.ref<!fir.box<!fir.ptr<i32>>>) -> (!fir.ref<!fir.box<!fir.ptr<i32>>>, !fir.ref<!fir.box<!fir.ptr<i32>>>)
diff --git a/flang/test/Lower/OpenMP/use-device-ptr-to-use-device-addr.f90 b/flang/test/Lower/OpenMP/use-device-ptr-to-use-device-addr.f90
index 90eede4f84108..0ad06f73a8ce9 100644
--- a/flang/test/Lower/OpenMP/use-device-ptr-to-use-device-addr.f90
+++ b/flang/test/Lower/OpenMP/use-device-ptr-to-use-device-addr.f90
@@ -47,7 +47,7 @@ subroutine only_use_device_addr
end subroutine
!CHECK: func.func @{{.*}}mix_use_device_ptr_and_addr_and_map()
-!CHECK: omp.target_data map_entries(%{{.*}}, %{{.*}} : !fir.ref<i32>, !fir.ref<i32>) use_device_ptr(%{{.*}} : !fir.ref<!fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>>) use_device_addr(%{{.*}}, %{{.*}} : !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>>, !fir.ref<!fir.box<!fir.ptr<!fir.array<?xf32>>>>) {
+!CHECK: omp.target_data use_device_ptr(%{{.*}} : !fir.ref<!fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>>) use_device_addr(%{{.*}}, %{{.*}} : !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>>, !fir.ref<!fir.box<!fir.ptr<!fir.array<?xf32>>>>) map_entries(%{{.*}}, %{{.*}} : !fir.ref<i32>, !fir.ref<i32>) {
!CHECK: ^bb0(%{{.*}}: !fir.ref<!fir.box<!fir.ptr<!fir.array<?xf32>>>>, %{{.*}}: !fir.ref<!fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>>, %{{.*}}: !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>>):
subroutine mix_use_device_ptr_and_addr_and_map
use iso_c_binding
More information about the llvm-branch-commits
mailing list