This is an archive of the discontinued LLVM Phabricator instance.

Differential D132023

[NFC][Flang] Add simd collapse test case
ClosedPublic

Authored by domada on Aug 17 2022, 4:18 AM.

Download Raw Diff

Details

Reviewers

peixin
kiranchandramohan
jdoerfert

Commits

rGcc8c746f8df5: [NFC][Flang] Add simd collapse test case

Summary

Flang supports lowering collapse clause to MLIR for worksharing loops and simd loops. Simd collapse clause is represented in MLIR code as a simd-loop having a list of indices, bounds and steps where the size of the list is equal to the collapse value.

Support for simd collapse clause was added by several patches:

https://reviews.llvm.org/D118065 -> basic support for simd-loop in MLIR. Loop collapsing is done in the same way as for worksharing loops: https://reviews.llvm.org/D105706
https://reviews.llvm.org/D125282 -> support for lowering simd clause from Fortran into MLIR.
https://reviews.llvm.org/D125302 -> lowering collapse clause from Fortran to MLIR. Lowering collapse clause is done before simd-specific function is called.
https://reviews.llvm.org/D128338 -> modified the MLIR representation of collapse clause. Removed collapse keyword in OpenMP MLIR dialect. Use loop list to represent collapse clause. This loop list is created by changes from patch: https://reviews.llvm.org/D125302 and it is passed to function responsible for lowering of simd directive which was implemented in patch: https://reviews.llvm.org/D125282

Diff Detail

Repository: rG LLVM Github Monorepo

Event Timeline

domada created this revision.Aug 17 2022, 4:18 AM

Herald added projects: Restricted Project, Restricted Project. · View Herald TranscriptAug 17 2022, 4:18 AM

Herald added subscribers: bzcheeseman, rriddle. · View Herald Transcript

domada requested review of this revision.Aug 17 2022, 4:18 AM

Herald added a reviewer: jdoerfert. · View Herald TranscriptAug 17 2022, 4:18 AM

Herald added subscribers: sstefan1, stephenneuendorffer, jdoerfert. · View Herald Transcript

domada mentioned this in D131674: [NFC][OpenMP] Update simd loop collapse support description.Aug 17 2022, 4:26 AM

LGTM.

flang/test/Lower/OpenMP/simd.f90
97–98	Nit: These two are not used and can be removed.
106–108	Nit: Use CHECK-SAME

This revision is now accepted and ready to land.Aug 17 2022, 4:27 AM

Harbormaster completed remote builds in B181716: Diff 453249.Aug 17 2022, 4:34 AM

Closed by commit rGcc8c746f8df5: [NFC][Flang] Add simd collapse test case (authored by domada). · Explain WhyAug 17 2022, 5:48 AM

This revision was automatically updated to reflect the committed changes.

domada added a commit: rGcc8c746f8df5: [NFC][Flang] Add simd collapse test case.

Revision Contents

Path

Size

flang/

test/

Lower/

OpenMP/

simd.f90

23 lines

Diff 453263

flang/test/Lower/OpenMP/simd.f90

Show First 20 Lines • Show All 83 Lines • ▼ Show 20 Lines	integer, parameter :: simdlen = 2;
do i = 1, n		do i = 1, n
! CHECK: fir.store %[[I]] to %[[LOCAL:.*]] : !fir.ref<i32>		! CHECK: fir.store %[[I]] to %[[LOCAL:.*]] : !fir.ref<i32>
! CHECK: %[[LD:.*]] = fir.load %[[LOCAL]] : !fir.ref<i32>		! CHECK: %[[LD:.*]] = fir.load %[[LOCAL]] : !fir.ref<i32>
! CHECK: fir.call @_FortranAioOutputInteger32({{.*}}, %[[LD]]) : (!fir.ref<i8>, i32) -> i1		! CHECK: fir.call @_FortranAioOutputInteger32({{.*}}, %[[LD]]) : (!fir.ref<i8>, i32) -> i1
print*, i		print*, i
end do		end do
!$OMP END SIMD		!$OMP END SIMD
end subroutine		end subroutine

		!CHECK-LABEL: func @_QPsimdloop_with_collapse_clause
		subroutine simdloop_with_collapse_clause(n)
		integer :: i, j, n
		integer :: A(n,n)
		! CHECK: %[[LOWER_I:.*]] = arith.constant 1 : i32
		! CHECK: %[[UPPER_I:.]] = fir.load %[[PARAM_ARG:.]] : !fir.ref<i32>
		kiranchandramohanUnsubmitted Not Done Reply Inline Actions Nit: These two are not used and can be removed. kiranchandramohan: Nit: These two are not used and can be removed.
		! CHECK: %[[STEP_I:.*]] = arith.constant 1 : i32
		! CHECK: %[[LOWER_J:.*]] = arith.constant 1 : i32
		! CHECK: %[[UPPER_J:.]] = fir.load %[[PARAM_ARG:.]] : !fir.ref<i32>
		! CHECK: %[[STEP_J:.*]] = arith.constant 1 : i32
		! CHECK: omp.simdloop for (%[[ARG_0:.]], %[[ARG_1:.]]) : i32 = (
		! CHECK-SAME: %[[LOWER_I]], %[[LOWER_J]]) to (
		! CHECK-SAME: %[[UPPER_I]], %[[UPPER_J]]) inclusive step (
		! CHECK-SAME: %[[STEP_I]], %[[STEP_J]]) {
		!$OMP SIMD COLLAPSE(2)
		do i = 1, n
		kiranchandramohanUnsubmitted Not Done Reply Inline Actions Nit: Use CHECK-SAME kiranchandramohan: Nit: Use CHECK-SAME
		do j = 1, n
		A(i,j) = i + j
		end do
		end do
		!$OMP END SIMD
		end subroutine