Pointer arithmetic rules in the tree optimizers?

[James E Wilson]

Diego Novillo wrote:
>      1. In testsuite/gfortran.fortran-torture/execute/forall_5.f90 we
>         are generating P = &VAR + R, where both P and R are of a pointer
>         type.  Is that correct?

This looks wrong to me.  I think the Fortran front end is just trying to 
make the types match, without realizing that pointer arithmetic is 
different.  They probably don't see a lot of variables of pointer type 
in the Fortran front end.  Arrays yes, pointers no.

>      2. In cp/parser.c we are generating 'P = X + Y', where 'P' is of
>         pointer type, '+' is of pointer type.  But *both* 'X' and 'Y'
>         are plain integers.

Not enough info to go on, but it does look suspiciously wrong.
-- 
Jim Wilson, GNU Tools Support, http://www.SpecifixInc.com