Re: [OG10] Fortran: delinearize multi-dimensional array accesses

[Thomas Koenig <tkoenig@netcologne.de>]

Hi Sandra,

> The attached patch implements delinearization of array accesses in the 
> Fortran front end, something that has been discussed for a long time.

Definitely - among others, this is the subject of PR 14741, which is by
now quite historic.

> I've been asked to try to get this patch committed on the OG10 branch 
> since it is blocking some further optimization work with Graphite for 
> OpenACC kernels regions.  I have a mainline version of this patch as 
> well that I can send to anyone interested in trying it out, but TBH, I 
> don't think this is ready for mainline yet.

That would be indeed interesting.  Could you post that to the list as
well?

The current status is that
> there are still two gfortran tests that are regressing 
> (gfortran.dg/graphite/id-9.f and 
> gfortran.dg/vect/fast-math-mgrid-resid.f), and while it's been confirmed 
> that this helps with Graphite optimizations as intended, we haven't yet 
> run any benchmarks to confirm that it doesn't make other things slower. 

It is probably too late; this could go in for the next stage 1.

> (It might, for instance, be appropriate to only delinearize when 
> Graphite optimizations are also enabled.)

Makes sense.

Does the patch actually make loop interchange for matrix multiplication
work (done with C for loops or Fortran DO loops), or is there additional
work required?

Regarding scalarized loops: We still to not collapse loops for

   subroutine foo(a)
     real, dimension(:,:), contiguous :: a
     a = 5.
   end subroutine foo

so an extension to scalarized loops would be quite valuable.

It would be interesting to see if

   subroutine foo(a,n,m)
     real, dimension(n,m) :: a
     do j=1,m
       do i=1,n
         a(i,j) = 5.
       end do
     end do
   end subroutine foo

is collapsed to a single loop with the patch and Graphite.

Best regards

	Thomas