This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: broken arithmetic operations on vectors (c and c++)


> I think we should be checking that we need a predicate for 
> same_type_ignoring_signedness.  That's not hard to implement.

This may be hairier than I thought.

I can code something up for the "sameness" I'm interested in, but it
makes sense to make something more generic.  How generic is too generic?

For example, should we compare alignment?  User alignment?  The readonly_flag?
How about the slew of other flags like: type.no_force_blk_flag,
type.no_force_blk_flag, type.restrict_flag, etc.  How about the
language flags?

For this bug I only care about type modifiers that would change the
validity of arithmetic operations, so only things like size, unit size, and
perhaps attributes attract me.

Thoughts?


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]