Re: Problem with 2.21 ld when linking HelenOS ia64 binaries

[jakub]

Hi Nick,

Quoting Nick Clifton <address@hidden>:

> > The following instruction:
> >
> >         addl r36 = @ltoff(@fptr(__entry#)), gp
> >
> > gets linked as either (ld 2.20):
> >
> > 28806:       40 02 04 00 48 20                   addl r36=0,r1
> >
> > or (ld 2.21):
> >
> > 24846:       40 02 07 8c 48 20                   addl r36=9056,r1
>
> Have you verified that the object files produced by the 2.20 and  
> 2.21 assemblers are the same ?  (Including the relocs).  If not then  
> this could be a gas bug rather than an ld bug...

No actually. What I did was that I built with 2.21 assembler (and 4.6 gcc) and
obtained a set of .o files. When I start the linking phase from the these
.o files, I get two different results depending on ld version.

(Please ignore the program counter addresses as I unfortunately copy-n-pasted
older disassembly where the instruction was located at different address. This
does not change anything about the above observation though.)

> It sounds like the evaluation of the relocation that computes the gp  
> offset is going wrong.  Without more context it is hard to theorize  
> any further.  Things that you could try include:
>
>   * Submit a bug report to http://sourceware.org/bugzilla/
>     If you include a small testcase to reproduce the problem that  
> would really help.

In the worst case, I could attach my .o files, the linker script and  
the linker
command line. I will try to see if I can limit the test case to only one .o.

Thanks,
Jakub

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.