On Wed, Sep 27, 2017 at 8:04 AM, Richard Biener <rguent...@suse.de> wrote:
>
> Another thing I notice is that we don't handle the multi-dimensional
> accesses the fortran frontend produces:
>
> (gdb) p debug_data_reference (dr)
> #(Data Ref:
> #  bb: 18
> #  stmt: _43 = *a_141(D)[_42];
> #  ref: *a_141(D)[_42];
> #  base_object: *a_141(D);
> #  Access function 0: {{(_38 + stride.88_115) + 1, +, 1}_4, +,
> stride.88_115}_5
>
> ultimatively we fail here because we try to build a constraint for
>
> {{(_38 + stride.88_115) + 1, +, 1}_4, +, stride.88_115}_5
>
> which ends up computing isl_pw_aff_mul (A, stride.88_115) with
> A being the non-constant constraint generated for
> {(_38 + stride.88_115) + 1, +, 1}_4 and stride.88_115 being
> a parameter.  ISL doesn't like that multiplication as the result
> isn't affine (well - it is, we just have parameters in there).
>
> I suppose ISL doesn't handle this form of accesses given the
> two "dimensions" in this scalarized form may overlap?  So we'd
> really need to turn those into references with different access
> functions (even if that's not 100% a valid semantic transformation
> as scalarization isn't reversible without extra information)?

You are right.
This multivariate memory access would be better handled in
delinearized form:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66981
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61000
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=14741

There are two ways to handle this issue:
- fix the FORTRAN front-end to emit multi dimensions ARRAY_REFs,
- implement an array delinearization pass, as I implemented in LLVM
http://llvm.org/doxygen/Delinearization_8cpp_source.html
"On Recovering Multi-Dimensional Arrays in Polly"
http://impact.gforge.inria.fr/impact2015/papers/impact2015-grosser.pdf
"Optimistic Delinearization of Parametrically Sized Arrays"
https://dl.acm.org/citation.cfm?id=2751248
LLVM does not have an equivalent for multi-dim ARRAY_REF description
it only reasons about linearized memory accesses like in GCC's RTL:
gep = Get Element Pointer, so we had no other option than to delinearize.

Sebastian

Reply via email to