Re: GCC/gfortran: asking for permission to merge devel/gfortran-test into master, adding a shared memory implementation of coarrays.

On 2/8/26 22:15, Thomas Koenig wrote:

> Jerry,
> 
> you know my concerns about this patch set - race conditions which I
> do not know were fixed.
> 
> Is there a record somewhere of the race conditions that were found
> and fixed?
> 
> Can you describe the testing that was done?  Were test
> programs run on machines which which would likely provoke
> race conditions, like a 400-process parallel run on cfarm120?

We tested the lock race condition you found (test program):

program memain
    use, intrinsic :: iso_fortran_env, only : lock_type
    type(lock_type), codimension[*] :: lck
    integer, codimension[*] :: count
    integer :: i, j, s
    do j=0,5  ! To be adjusted, dunno how long this runs
      if (this_image() == 1) count = 0
      sync all
      do i=1,10**j
         lock (lck[1])
         count[1] = count[1] + 1
         unlock (lck[1])
      end do
      sync all
      if (this_image() == 1) then
         print *,"Expected: ",10**j*num_images(), "Found: ", count
      end if
      sync all
    end do
end program

It turned out to be a problem that was genuine to the coarray 
implementation in the front-end (tested by Andre himself with OpenCoarrays).

He fixed it, and we moved on.

Kind regards,

-- 
Toon Moene - e-mail: [email protected] - phone: +31 346 214290
Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands