[tuples] Fix invalid GIMPLE D.1234_5 = (long int) .omp_data_i->x created by OpenMP lowering

Jakub Jelinek jakub@redhat.com
Fri Jul 11 17:20:00 GMT 2008


On Fri, Jul 11, 2008 at 09:25:32AM -0700, Diego Novillo wrote:
> OK, I guess.  Would this be a non-issue if I come up with a general
> mechanism for handling regimplification of assignments?

Yes.  lower_omp_1 will then just decide whether any of the operands
might need regimplification and just call your smart regimplification
routine on the whole GIMPLE_ASSIGN or GIMPLE_CALL.

	Jakub



More information about the Gcc-patches mailing list