Re: LTO vs zero length bit fields

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



FYI I've reported it here: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78472

On 2 March 2016 at 16:05, Jay Foad <jay.foad@xxxxxxxxx> wrote:
> Hi,
>
> I'm trying to use LTO to build a large C/C++ application, and I'm
> getting some warnings that I don't understand:
>
> warning: type of ‘s’ does not match original declaration
>
> Here's a reduced test case:
>
> $ cat common.h
> extern struct S
> {
> unsigned i:4;
> unsigned :0;
> } s;
> static void *f(void)
> {
> return &s;
> }
>
> $ cat c.c
> #include "common.h"
>
> $ cat cpp.cpp
> #include "common.h"
>
> $ gcc -o /dev/null -flto c.c cpp.cpp
> common.h:5:3: warning: type of ‘s’ does not match original declaration
>  } s;
>    ^
> common.h:5:3: note: previously declared here
>  } s;
>    ^
>
> It looks like the zero-length bitfield is causing a problem, and
> somehow makes the structure incompatible in C vs C++. Why? Is there a
> better fix than just not using zero-length bitfields?
>
> I'm using GCC 5.2.1 on Ubuntu 15.10, but I get the same warning if I
> use a GCC 6 built from trunk sources about a week ago.
>
> Thanks,
> Jay.




[Index of Archives]     [Linux C Programming]     [Linux Kernel]     [eCos]     [Fedora Development]     [Fedora Announce]     [Autoconf]     [The DWARVES Debugging Tools]     [Yosemite Campsites]     [Yosemite News]     [Linux GCC]

  Powered by Linux