[PATCH 0/2, fortran] Better code generation for DO loops with +-1 step

[Martin Liška]

On 07/07/2016 04:40 PM, Jan Hubicka wrote:
>>
>> Why is the behavior only undefined for step 1 if the last iteration IV
>> increment overflows?
>> Doesn't this apply to all step values?
> 
> This is what Fortran standard says:
> 
>   The iteration count is established and is the value of the expression (m2-m1+m3)/m3 unless that value is negative,
>   in which case the iteration count is 0.
> 
> My reading of this is that the do statement is undefined whenever the expression above is undefined
> (m1 is lower bound, m2 is upper bound, m3 is step) and because I think the evaulation order of
> m2-m1+m3 is not fixed, I think the statement is not defined whethever (m2-m1), (m1+m3) or (m2-m1)+m3
> overflows or underflows as signed integer.
> 
> For example it is not valid to iterate from -10 to INT_MAX with step 1.
> Honza
> 

Hi.

I'm attaching a candidate patch that emits the warnings. Problem with current implementation of loop generation
(w/ step different than 1) is that it utilizes unsigned type, thus the calculation of iteration count works
even though the expression overflows:

  do i = array(1), array(2), 17 
      block(i) = block(i) + 10
  end do

is transformed to:

    D.3428 = (*array)[0];
    D.3429 = (*array)[1];
    i = D.3428;
    countm1.0 = ((unsigned int) D.3429 - (unsigned int) D.3428) / 17;, if (D.3429 < D.3428)
      {
        goto L.2;
      };

Thoughts about the patch?
Martin
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Enhance-warning-message-for-DO-loops-with-step-1.patch
Type: text/x-patch
Size: 4058 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/gcc-patches/attachments/20160708/f6d7a3d4/attachment.bin>