Observed issues › afar-nontemporal-assumed-shape-ice
Appears to behave incorrectlyCompiler crash · Fortran · OpenMP · Open
amdflang: "verification of lowering to FIR failed" for nontemporal on an assumed-shape dummy
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 | behaves incorrectly | behaves incorrectly | behaves incorrectly | behaves incorrectly | behaves incorrectly | n/a | n/a |
| CCE 19.0.0 | correct | correct | correct | unknown not run | unknown not run | n/a | n/a |
| CCE 20.0.2 | correct | correct | correct | unknown not run | unknown not run | n/a | n/a |
| CCE 21.0.3 | correct | correct | correct | unknown not run | unknown not run | n/a | n/a |
| CCE 21.0.4 | correct | correct | correct | unknown not run | unknown not run | n/a | n/a |
| NVHPC 25.11 | n/a | n/a | n/a | n/a | n/a | fails another way | fails another way |
| NVHPC 26.9 | n/a | n/a | n/a | n/a | n/a | fails another way | fails another way |
| 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 | AFAR 24.3.0, MI300A unified AFAR 24.3.0, MI300A copy AFAR 24.3.0, MI250X AFAR 24.3.0, MI300X AFAR 24.3.0, MI350X | showerror: loc("afar-nontemporal-assumed-shape-ice.f90":16:13): 'omp.simd' op operand #0 must be variadic of OpenMP-compatible variable type, but got '!fir.box<!fir.array<?x?xi32>>'
error: verification of lowering to FIR failed |
| fails another way | NVHPC 25.11, A100 separate NVHPC 25.11, A100 managed | showNVFORTRAN-S-0034-Syntax error at or near identifier nontemporal (afar-nontemporal-assumed-shape-ice.f90: 16) NVFORTRAN/x86-64 Linux 25.11-0: compilation completed with severe errors |
| fails another way | NVHPC 26.9, A100 separate NVHPC 26.9, A100 managed | showNVFORTRAN-S-0034-Syntax error at or near identifier nontemporal (afar-nontemporal-assumed-shape-ice.f90: 16) NVFORTRAN/x86-64 Linux 26.9-0: compilation completed with severe errors |
Reproduce
AFAR 24.3.0, MI300A unified
$AFAR/bin/amdflang -fopenmp --offload-arch=gfx942 -O2 repro.f90 -o repro HSA_XNACK=1 OMP_TARGET_OFFLOAD=mandatory ./repro
CCE 19.0.0, MI300A unified (runs correctly)
module load cce/19.0.0 craype-accel-amd-gfx942 rocm/6.4.3 module unload cray-mpich ftn -fopenmp -O2 repro.f90 -o repro HSA_XNACK=1 OMP_TARGET_OFFLOAD=mandatory ./repro
Program (33 lines)
program p
implicit none
integer, parameter :: n = 16, m = 4
integer :: o1(n, m), i, j
integer(8) :: cs
o1 = 0
call kp(o1)
cs = sum(int(o1, 8))
print '(a,i0,a,i0)', 'RESULT checksum=', cs, ' ran_on_host=', 0
contains
subroutine kp(o1)
integer, intent(inout) :: o1(:, :)
integer :: i, j
!$omp target teams distribute parallel do map(tofrom: o1)
do i = 1, n
!$omp simd nontemporal(o1)
do j = 1, m
o1(i, j) = i + j
end do
end do
end subroutine kp
end program p
! spec (requirement): The nontemporal clause on a simd construct takes a list of variables, and its text states no
! restriction on their form, so an assumed-shape dummy array is a valid list item.
! "The nontemporal clause specifies that accesses to the storage locations to which the list items refer have low
! temporal locality across the iterations in which those storage locations are accessed." (5.2 §10.4.1 p.236
! l.21-22)
! spec (requirement): The list items of nontemporal may also appear in data-environment attribute clauses.
! "The list items of the nontemporal clause may also appear as list items of data-environment attribute clauses." (5.2
! §10.4.1 p.236 l.23-24)
! spec (this program): nontemporal(o1) names an assumed-shape dummy that is also mapped by the enclosing target
! construct; the spec text has no restriction against assumed-shape list items (it does not mention them), and
! amdflang stops with a failed lowering to FIR.
Something to add? Dispute its legality · Report it fixed in a newer release · Report a mistake on this page
Expected and observed
Observed where it behaves incorrectly
error: loc("afar-nontemporal-assumed-shape-ice.f90":16:13): 'omp.simd' op operand #0 must be variadic of OpenMP-compatible variable type, but got '!fir.box<!fir.array<?x?xi32>>'
error: verification of lowering to FIR failedWhat the specification requires
The specification text this program relies on and the compiler is expected to support.
OpenMP 5.2 §10.4.1 p.236 l.21-22
The nontemporal clause specifies that accesses to the storage locations to which the list items refer have low temporal locality across the iterations in which those storage locations are accessed.
OpenMP 5.2 §10.4.1 p.236 l.23-24
The list items of the nontemporal clause may also appear as list items of data-environment attribute clauses.
Evidence
Same compiler, one change
- nontemporal(o1) removed from the simd directive: 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
Control program
program p implicit none integer, parameter :: n = 16, m = 4 integer :: o1(n, m), i, j integer(8) :: cs o1 = 0 call kp(o1) cs = sum(int(o1, 8)) print '(a,i0)', 'RESULT checksum=', cs contains subroutine kp(o1) integer, intent(inout) :: o1(:, :) integer :: i, j !$omp target teams distribute parallel do map(tofrom: o1) do i = 1, n !$omp simd do j = 1, m o1(i, j) = i + j end do end do end subroutine kp end program p
Other compilers run it correctly
CCE 19.0.0, CCE 20.0.2, CCE 21.0.3, CCE 21.0.4, GCC 16.2.0