[Patch, Fortran, committed] Add missing fold_convert (PR36316)

Toon Moene toon@moene.indiv.nluug.nl
Wed May 28 19:24:00 GMT 2008


Tobias Burnus wrote:

> gfortran was generating the following tree:
> 
>    integer(kind=8) D.1026;
>    integer(kind=4) D.1025;
>    integer(kind=4) D.1024;
>    D.1024 = D.1022->local_start;
>    D.1025 = D.1022->local_end;
>    D.1026 = D.1025 - D.1024;
> 
> The problem is that "D.1025-D.1024" is integer(4) while D.1026 is 
> integer(8). With tree-checking enabled, this caused an ICE as Toon found 
> out (thanks!).

Thanks for fixing this - it not only fixed the test case, but also the 
(rather large) original HIRLAM code (http://hirlam.org).

I tested it not only by compiling, but also running a couple of our test 
cases; all went well.

Thanks, Tobias !

-- 
Toon Moene - e-mail: toon@moene.indiv.nluug.nl - phone: +31 346 214290
Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands
At home: http://moene.indiv.nluug.nl/~toon/
Progress of GNU Fortran: http://gcc.gnu.org/ml/gcc/2008-01/msg00009.html



More information about the Gcc-patches mailing list