Jan Hubicka <hubi...@ucw.cz> wrote: >> On 09/04/2013 06:04 PM, Jan Hubicka wrote: >> > this is third fallout of my change to remove >DECL_ARGUMENTS/DECL_RESULT for functions w/o >> > bodies I did not really anticipate. >> [...] >> > I would like to basically ask if it seems to make sense to go this >route and >> > try to get rid of those declarations. >> >> I'm currently working on a new target, ptx, which uses a >> pseudo-assembler where functions (even extern ones) need to be >declared >> with their arguments and return types. With my current code I have to >> look at DECL_ARGUMENTS fairly late in the compilation. I'm not quite >> sure yet whether the change to delete them will break the backend. > >How do you support K&R functions here? My basic idea was that >TYPE_ARG_TYPES >should give enough information about external function calling >convention >anyone will ever need. I would hope that this will be sufficient for >your >use, too, despite the fact you no longer have parameter names at hand >and you also lose info about external inline K&R-style delcared >functions >that has been optimized out. > >If not that indeed, you will not see DECL_ARGUMENTS for external >function >anytime after cgraph_remove_unreachable_functions is called.
In fact it has to work because of indirect calls and how we now handle gimple call abi via gimple_call_fntype. Richard. >Honza
