This is an archive of the discontinued LLVM Phabricator instance.

Differential D132215

[NFC][OpenMP] Add test for simd directive with nested loop
ClosedPublic

Authored by domada on Aug 19 2022, 3:14 AM.

Details

Reviewers
ftynse
jdoerfert
dcaballe
kiranchandramohan
Commits
rGb736179e3449: [NFC][OpenMP] Add test for simd directive with nested loop
Summary

This test is a follow up of the review: https://reviews.llvm.org/D131402

Co-Author of the patch: Kiran Chandramohan

Diff Detail

Event Timeline

domada created this revision.Aug 19 2022, 3:14 AM
Herald added a reviewer: ftynse. · View Herald TranscriptAug 19 2022, 3:14 AM
Herald added a project: Restricted Project. · View Herald Transcript
Herald added subscribers: bzcheeseman, awarzynski, sdasgup3 and 21 others. · View Herald Transcript
domada requested review of this revision.Aug 19 2022, 3:14 AM
Herald added a reviewer: jdoerfert. · View Herald TranscriptAug 19 2022, 3:14 AM
Herald added a reviewer: dcaballe. · View Herald Transcript
Herald added a project: Restricted Project. · View Herald Transcript
Herald added subscribers: sstefan1, stephenneuendorffer, nicolasvasilache. · View Herald Transcript
domada added a reviewer: kiranchandramohan.Aug 19 2022, 3:15 AM
kiranchandramohan accepted this revision.Aug 19 2022, 3:28 AM

Thanks @domada. LG.

This revision is now accepted and ready to land.Aug 19 2022, 3:28 AM
Harbormaster completed remote builds in B182184: Diff 453937.Aug 19 2022, 3:28 AM
Closed by commit rGb736179e3449: [NFC][OpenMP] Add test for simd directive with nested loop (authored by domada). · Explain WhyAug 19 2022, 3:46 AM
This revision was automatically updated to reflect the committed changes.
domada added a commit: rGb736179e3449: [NFC][OpenMP] Add test for simd directive with nested loop.

Revision Contents

PathSize
mlir/
test/
Conversion/
OpenMPToLLVM/
29 lines

Diff 453937

mlir/test/Conversion/OpenMPToLLVM/convert-to-llvmir.mlir

Show First 20 Lines • Show All 110 Lines • ▼ Show 20 Lines
// CHECK-LABEL: @threadprivate // CHECK-LABEL: @threadprivate
// CHECK: (%[[ARG0:.*]]: !llvm.ptr<i32>) // CHECK: (%[[ARG0:.*]]: !llvm.ptr<i32>)
// CHECK: %[[VAL0:.*]] = omp.threadprivate %[[ARG0]] : !llvm.ptr<i32> -> !llvm.ptr<i32> // CHECK: %[[VAL0:.*]] = omp.threadprivate %[[ARG0]] : !llvm.ptr<i32> -> !llvm.ptr<i32>
func.func @threadprivate(%a: !llvm.ptr<i32>) -> () { func.func @threadprivate(%a: !llvm.ptr<i32>) -> () {
%1 = omp.threadprivate %a : !llvm.ptr<i32> -> !llvm.ptr<i32> %1 = omp.threadprivate %a : !llvm.ptr<i32> -> !llvm.ptr<i32>
return return
} }
// -----
// CHECK: llvm.func @simdloop_block_arg(%[[LOWER:.*]]: i32, %[[UPPER:.*]]: i32, %[[ITER:.*]]: i64) {
// CHECK: omp.simdloop for (%[[ARG_0:.*]]) : i32 =
// CHECK-SAME: (%[[LOWER]]) to (%[[UPPER]]) inclusive step (%[[LOWER]]) {
// CHECK: llvm.br ^[[BB1:.*]](%[[ITER]] : i64)
// CHECK: ^[[BB1]](%[[VAL_0:.*]]: i64):
// CHECK: %[[VAL_1:.*]] = llvm.icmp "slt" %[[VAL_0]], %[[ITER]] : i64
// CHECK: llvm.cond_br %[[VAL_1]], ^[[BB2:.*]], ^[[BB3:.*]]
// CHECK: ^[[BB2]]:
// CHECK: %[[VAL_2:.*]] = llvm.add %[[VAL_0]], %[[ITER]] : i64
// CHECK: llvm.br ^[[BB1]](%[[VAL_2]] : i64)
// CHECK: ^[[BB3]]:
// CHECK: omp.yield
func.func @simdloop_block_arg(%val : i32, %ub : i32, %i : index) {
omp.simdloop for (%arg0) : i32 = (%val) to (%ub) inclusive step (%val) {
cf.br ^bb1(%i : index)
^bb1(%0: index):
%1 = arith.cmpi slt, %0, %i : index
cf.cond_br %1, ^bb2, ^bb3
^bb2:
%2 = arith.addi %0, %i : index
cf.br ^bb1(%2 : index)
^bb3:
omp.yield
}
return
}