This is the mail archive of the gcc-bugs@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: ARM/THUMB long calls stubs


Hi,

On Mon, Aug 09, 2004 at 01:20:49PM +0100, Nick Clifton wrote:
> 
> >>>Would it be hard to fix this? So that gcc would take __attribute__
> >>>((section())) in account when determining if functions are in different
> >>>sections or not. Feels bad as all functions are called with the long 
> >>>call..
> 
> I think that this is a real bug in gcc.
>
> The problem appears to be that the short-call attribute is being set for 
> any function that has already been compiled in a compilation unit, 
> regardless of whether that function has a section attribute or not.  In 
> effect this is saying that the compiler thinks that any function it has 
> already compiled must be within a short-call range of any function that 
> it is about to compile.
> 
> The attached patch removes this assumption and tidies up some of the 
> comments in the code associated with the arm_is_longcall_p().  I have 
> tested it with an arm-elf build and found no regressions, so unless 
> Richard has any objections I plan to check it in later this week.

Thanks, seemd to work!


     -sk


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