2015-09-15 13:32 GMT+03:00 Richard Biener <richard.guent...@gmail.com>: > On Tue, Sep 15, 2015 at 11:28 AM, Ilya Enkovich <enkovich....@gmail.com> > wrote: >> 2015-09-13 16:36 GMT+03:00 Richard Biener <richard.guent...@gmail.com>: >>> On Mon, Sep 7, 2015 at 2:39 PM, Ilya Enkovich <enkovich....@gmail.com> >>> wrote: >>>> 2015-09-02 15:35 GMT+03:00 Richard Biener <richard.guent...@gmail.com>: >>>>> >>>>> DECL_FIELD_BIT_OFFSET should be never NULL. Whoever created that >>>>> FIELD_DECL created an invalid one. >>>>> >>>>> Richard. >>>>> >>>> >>>> layout_class_type doesn't place fields with no type >>> >>> Err - that's because fields should also have a type. >> >> Sure. But we are talking about a wrong code and still want to continue >> compilation to some point even if some field misses a type. It means >> everything possibly invoked at this stage should check type against >> error_mark_node. Thus I need to handle it gracefully in >> chkp_find_bound_slots I suppose. > > I see. I wonder why we even call chkp_find_bound_slots if seen_errors().
Even with errors we still gimplify function. Function parameters gimplification checks where parameters are passed to possibly copy some of them. It triggers ix86_function_arg_advance which uses chkp_find_bound_slots to skip required amount of bounds registers. > I suppose only recursing for COMPLETE_TYPE_P () would work? Yep, it should work. I'll rework my fix. Thanks, Ilya > > Richard. >
