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]

Re: one-bit alignment (was Re: [PATCH] Fix for the C++ testcase submitted today)


    It doesn't really make sense to talk about the alignment, size, or
    mode of a function.  We could just make all of these NULL, or
    equivalent.

This isn't a function, but a decl whose type was incomplete when it
was defined.  Functions never get passed to get_inner_reference.

    You're probably right that there's a deeper question here.  But, I
    don't really think this is papering over the problem; I think that
    using 1 byte-alignment, rather that 1 bit-alignment, brings us to a
    monotonically saner state.

FUNCTION_DECLs don't even *have* a DECL_ALIGNMENT field!

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