Defaultmap(firstprivate) drops the copy-back of explicitly mapped variables

Legality: independently confirmed. 2 independent readings find the program legal and the result required. How this is checked

Affected variants

Compiler	MI300A gfx942 unified memory (XNACK 1)	MI300A gfx942 copy (XNACK 0)	MI250X gfx90a copy (XNACK 0)	MI300X gfx942 copy (XNACK 0)	MI350X gfx950 copy (XNACK 0)	A100 sm_80 separate memory	A100 sm_80 managed memory
AFAR 24.3.0	fails another way	fails another way	unknown not run	fails another way	fails another way	n/a	n/a
CCE 19.0.0	behaves incorrectly	behaves incorrectly	unknown not run	unknown not run	unknown not run	n/a	n/a
CCE 20.0.2	correct	correct	unknown not run	unknown not run	unknown not run	n/a	n/a
CCE 21.0.3	behaves incorrectly	behaves incorrectly	unknown not run	unknown not run	unknown not run	n/a	n/a
CCE 21.0.4	behaves incorrectly	behaves incorrectly	unknown not run	unknown not run	unknown not run	n/a	n/a
NVHPC 25.11	n/a	n/a	n/a	n/a	n/a	behaves incorrectly	behaves incorrectly
NVHPC 26.9	n/a	n/a	n/a	n/a	n/a	behaves incorrectly	behaves incorrectly
GCC 16.2.0	correct	correct	correct	correct	correct	correct	n/a

"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	Compiler and GPU	Message
behaves incorrectly	CCE 19.0.0, MI300A unified CCE 19.0.0, MI300A copy	show sum 0 host_flag 0
behaves incorrectly	CCE 21.0.3, MI300A unified CCE 21.0.3, MI300A copy CCE 21.0.4, MI300A unified CCE 21.0.4, MI300A copy	show sum 2144 host_flag 1
behaves incorrectly	NVHPC 25.11, A100 separate NVHPC 25.11, A100 managed NVHPC 26.9, A100 separate NVHPC 26.9, A100 managed	show sum 0 host_flag 1
fails another way	AFAR 24.3.0, MI300A unified AFAR 24.3.0, MI300A copy AFAR 24.3.0, MI300X AFAR 24.3.0, MI350X	show error: loc("nvhpc-defaultmap-firstprivate-explicit-map.f90":8:9): /__w/TheRock/TheRock/compiler/amd-llvm/flang/lib/Lower/OpenMP/OpenMP.cpp:1797: not yet implemented: Firstprivate is currently unsuppor LLVM ERROR: aborting

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 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

GCC 16.2.0, MI300A unified (runs correctly)

export LD_LIBRARY_PATH=$GCC_16_2/lib64:/opt/rocm-7.2.1/lib
$GCC_16_2/bin/gfortran -fopenmp -foffload=amdgcn-amdhsa -foffload-options=amdgcn-amdhsa=-march=gfx942 -O2 repro.f90 -o repro
HSA_XNACK=1 OMP_TARGET_OFFLOAD=mandatory ./repro

Program (40 lines)

! NVHPC 25.11, NVIDIA A100 80GB PCIe (wingtip, sm_80), both memory modes:
! with defaultmap(firstprivate) on a target construct, the arrays and
! scalars the construct maps EXPLICITLY (map(from: b), map(tofrom:
! host_flag)) are not copied back: the host sees b and host_flag unchanged.
! Spec: "The defaultmap clause determines the implicit data-mapping or
! data-sharing attribute of certain variables" (5.2 §5.8.7 p.161); "When
! specified, explicit data-environment attribute clauses on target
! directives determine the attributes" of the variables they name (5.2
! §5.8.1 p.148).
! Vendor: "The defaultmap clause is supported using OpenMP 5.0 semantics"
! (HPC Compilers User's Guide 25.11 §7.4, 2.12.5).
! Expected: "sum 2144 host_flag 0".
! Observed (2026-10-24, nvfortran 25.11-0, -O2, OMP_TARGET_OFFLOAD=
! mandatory, two runs, the same): "sum 0 host_flag 1".
! Found by the catalog (omp50/52/60-clause-defaultmap-firstprivate: their
! cells read ran_on_host=1 check=fail; confirmer R20-runtime-bug); reduced
! by hand.
! NVHPC 26.9 (26.9-0, the latest release; rerun 2026-10-24 on wingtip with
! the same flags): still present -- the same: "sum 0 host_flag 1", both modes.
! Fix (2026-10-24): rerun on 25.11 and 26.9 in both memory modes,
! each with its own runtime libraries (LD_LIBRARY_PATH = the release's
! compilers/lib; the first 26.9 runs had loaded 25.11's): still present
! on both releases, the same output.
! NVHPC: nvfortran -mp=gpu -gpu=cc80 -gpu=mem:separate -O2
! Run:   OMP_TARGET_OFFLOAD=mandatory ./a.out
program p
  use omp_lib, only: omp_is_initial_device
  implicit none
  integer :: b(64), host_flag, i, c
  b = 0
  host_flag = 1
  c = 1
  !$omp target map(from: b) map(tofrom: host_flag) defaultmap(firstprivate)
    host_flag = merge(1, 0, omp_is_initial_device())
    do i = 1, 64
      b(i) = i + c
    end do
  !$omp end target
  print *, 'sum', sum(b), 'host_flag', host_flag
end program p

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

Expected

sum 2144 host_flag 0

Observed where it behaves incorrectly

 sum 0 host_flag 0

What the specification requires

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

OpenMP 5.2 §5.8.7 p.161

The defaultmap clause determines the implicit data-mapping or data-sharing attribute of certain variables

OpenMP 5.2 §5.8.1 p.148

Evidence

Same compiler, one change

defaultmap(firstprivate) removed: AFAR 24.3.0: correct; CCE 19.0.0: correct; CCE 20.0.2: correct; CCE 21.0.3: correct; CCE 21.0.4: correct; NVHPC 25.11: correct; NVHPC 26.9: correct

Control program

program p
  use omp_lib, only: omp_is_initial_device
  implicit none
  integer :: b(64), host_flag, i, c
  b = 0
  host_flag = 1
  c = 1
  !$omp target map(from: b) map(tofrom: host_flag)
    host_flag = merge(1, 0, omp_is_initial_device())
    do i = 1, 64
      b(i) = i + c
    end do
  !$omp end target
  print *, 'sum', sum(b), 'host_flag', host_flag
end program p

Other compilers run it correctly

GCC 16.2.0