RE:Re: GDB 5.1/Core files and ptids

[Takis Psarogiannakopoulos <takis@XFree86.Org>]

Hi Kevin,

On Wed, 16 Jan 2002, Kevin Buettner wrote:

> The context here is that we're trying to fetch the appropriate
> .reg (.reg2, etc) section from bfd, right?

That is correct. We are trying to get a section with name "name".

> > Especially when, even in the new gdb-5.1/bfd we find:
> > 
> > ======
> > static int
> > elfcore_make_pid (abfd)
> >      bfd *abfd;
> > {
> >   return ((elf_tdata (abfd)->core_lwpid << 16)
> >           + (elf_tdata (abfd)->core_pid));
> > }
> > =======
> 
> This will need to change.  It is incorrect to attempt to represent
> both the pid and lwp as a single integer whose size isn't large
> enough to hold all of the bits.
>   

I have no problem with the ptid especially when it is less confusing
from the old mixed pid's! 
My problem is that a core DG/UX file has lot of info sections that 
I would like to read and wasnt sure if gdb would be able now taking 
in account that elfcore_make_pid had the above form.

>     So, to fetch a .reg section in a multithreaded core dump where
>     the pid is 14 and the lwp is 42, GDB would need to ask BFD for
>     the .reg/14+42 (pseudo) section.


>      
>       else if (TIDGET (inferior_ptid))
>       sprintf (section_name, "%s/%d+%d", name, PIDGET (inferior_ptid),
>                TIDGET (inferior_ptid));
>       else if (PIDGET (inferior_ptid))
>       sprintf (section_name, "%s/%d", name, PIDGET (inferior_ptid));
>       else
>       strcpy (section_name, name);  
>        


OK. That is indeed a solution to what I was saying about reading sections
using ptid's. Thanks Kevin.

Regards,
Takis