Lastprivate on distribute in a target region loses the final value, even with an explicit map

Affected variants

"behaves incorrectly" / "not supported": this issue, confirmed in up to four environments (another build directory, another node, a repeat). "fails another way": a different failure, not counted as this issue. "unknown": not run there, or the runs disagree.

Messages

RESULT checksum=0

Reproduce

CCE 19.0.0, MI300A unified

module load cce/19.0.0 craype-accel-amd-gfx942 rocm/6.4.3
module unload cray-mpich
ftn -fopenmp -O2 repro.f90 -o repro
HSA_XNACK=1 OMP_TARGET_OFFLOAD=mandatory ./repro

CCE 20.0.2, MI300A unified

module load cce/20.0.2 craype-accel-amd-gfx942 rocm/6.4.3
module unload cray-mpich
ftn -fopenmp -O2 repro.f90 -o repro
HSA_XNACK=1 OMP_TARGET_OFFLOAD=mandatory ./repro

CCE 21.0.3, MI300A unified

module load cce/21.0.3 craype-accel-amd-gfx942 rocm/7.2.0
module unload cray-mpich
ftn -fopenmp -O2 repro.f90 -o repro
HSA_XNACK=1 OMP_TARGET_OFFLOAD=mandatory ./repro

CCE 21.0.4, MI300A unified

module load cce/21.0.4 craype-accel-amd-gfx942 rocm/7.2.0
module unload cray-mpich
ftn -fopenmp -O2 repro.f90 -o repro
HSA_XNACK=1 OMP_TARGET_OFFLOAD=mandatory ./repro

NVHPC 25.11, A100 separate

export LD_LIBRARY_PATH=$NVHPC_25_11/compilers/lib
nvfortran -mp=gpu -gpu=cc80 -gpu=mem:separate -O2 repro.f90 -o repro
OMP_TARGET_OFFLOAD=mandatory ./repro

NVHPC 26.9, A100 separate

export LD_LIBRARY_PATH=$NVHPC_26_9/compilers/lib
nvfortran -mp=gpu -gpu=cc80 -gpu=mem:separate -O2 repro.f90 -o repro
OMP_TARGET_OFFLOAD=mandatory ./repro

AFAR 24.3.0, MI300A unified (runs correctly)

$AFAR/bin/amdflang -fopenmp --offload-arch=gfx942 -O2 repro.f90 -o repro
HSA_XNACK=1 OMP_TARGET_OFFLOAD=mandatory ./repro

! Minimized by compliance.fuzz_min from fuzz seed 14 (fuzz-1).
! toolchain: cce20 (-O2, HSA_XNACK=1, OMP_TARGET_OFFLOAD=mandatory)
! failure: wrong_code
! expected: the host reference checksum (host reference (cce20/afar) checksum=24)
! observed: device checksum=0 ran_on_host=0
! Legal OpenMP 5.2: offload.engine.prerun.check accepts it (rerun: offload.engine.rules with the
!   race rule and the restriction checks, and the AFAR flang front end, not under test here); no
!   compiler under test, gfortran or gcc is a legality witness.
! also fails on: cce19, cce21
program fuzz_prog
  implicit none
  integer, parameter :: n = 24, m = 8
  integer :: i, j, jj, lp1, acc1, t1
  logical :: lsimd
  integer(8) :: cs
  lp1 = 0
  lsimd = .false.
  ! fuzz: kernels
    !$omp target teams map(tofrom: lp1)
      !$omp distribute parallel do simd lastprivate(lp1) private(t1, acc1, jj) num_threads(8)
      do i = 1, n
        t1 = 0
        acc1 = 0
        lp1 = i
      end do
      !$omp end distribute parallel do simd
    !$omp end target teams
  ! fuzz: checksum
  cs = 0
  cs = modulo(cs * 31_8 + int(lp1, 8), 2147483629_8)
  print '(a,i0,a,i0)', 'RESULT checksum=', cs, ' ran_on_host=', 0
end program fuzz_prog

Source in the repository

Something to add? Dispute its legality · Report it fixed in a newer release · Report a mistake on this page

Expected and observed

RESULT checksum=0

What the specification requires

The specification text this program relies on and the compiler is expected to support.

The same requirement as the issue stress/cce-c-distribute-lastprivate-lost.

Evidence