Renaud Kauffmann 652ff20140 [flang][cuda] Adding atomicadd as a cudadevice intrinsic and converting it LLVM dialect (#123840) ...

With these changes, CUF atomic operations are handled as cudadevice
intrinsics and are converted straight to the LLVM dialect with the
`llvm.atomicrw` operation.

I am only submitting changes for `atomicadd` to gather feedback. If we
are to proceed with these changes I will add support for all other
applicable atomic operations following this pattern.

2025-01-22 19:06:28 -08:00

..

__cuda_builtins.f90

…

__cuda_device.f90

[CUF] Moving __fadd_rd and __fadd_ru to cudadevice.f90 as they are symbol already known upstream. (#123127)

2025-01-16 11:53:18 -08:00

__fortran_builtins.f90

[flang] Add default component initialization to some built-in types (#121416)

2025-01-08 13:14:28 -08:00

__fortran_ieee_exceptions.f90

[flang] Modifications to ieee floating point environment procedures (#121949)

2025-01-15 10:55:09 -05:00

__fortran_type_info.f90

[flang][cuda] Add c_devloc as intrinsic and inline it during lowering (#120648)

2025-01-08 11:23:05 -08:00

__ppc_intrinsics.f90

…

__ppc_types.f90

…

.clang-format

…

cudadevice.f90

[flang][cuda] Adding atomicadd as a cudadevice intrinsic and converting it LLVM dialect (#123840)

2025-01-22 19:06:28 -08:00

ieee_arithmetic.f90

[flang] IEEE_REM (#115936)

2024-11-13 13:02:20 -05:00

ieee_exceptions.f90

…

ieee_features.f90

…

iso_c_binding.f90

[flang] Add UNSIGNED (#113504)

2024-12-18 07:02:37 -08:00

iso_fortran_env_impl.f90

[flang] Add UNSIGNED (#113504)

2024-12-18 07:02:37 -08:00

iso_fortran_env.f90

[flang] Add UNSIGNED (#113504)

2024-12-18 07:02:37 -08:00

mma.f90

…