public inbox for u-boot@lists.denx.de
 help / color / mirror / Atom feed
* [U-Boot] [RFC] - Relocating i386 Code - Permission to break orphan boards
@ 2009-08-30 10:44 Graeme Russ
  2009-08-31  7:35 ` Graeme Russ
  0 siblings, 1 reply; 2+ messages in thread
From: Graeme Russ @ 2009-08-30 10:44 UTC (permalink / raw)
  To: u-boot

Hi Wolfgang,

I have been playing around with gcc's -fpic and ld's -PIE and I think that
I can get 'proper' relocation happening on the i386 port. I've get the
basics down, and I've been looking at /lib_ppc/board.c for how the PPC
relocation scheme works and would like to replicate it (i.e. board_init_f,
board_init_r and relocate_code.

Unfortunately this leaves two options:

 1) Tangle the i386 port up in a heap of #ifdefs, or;
 2) Do a clean re-write for the eNET and break the other i386 boards
    (sc520_cdp and sc520_spunk).

The other two i386 boards are OLD and i have my doubts as to if they ever
did work completely, and have really strong doubts as to them working now

I would really like to just throw all the old code out and do it all
cleanly, but not without strong support from the community.

What are your thoughts?

Regards,

G

^ permalink raw reply	[flat|nested] 2+ messages in thread

* [U-Boot] [RFC] - Relocating i386 Code - Permission to break orphan boards
  2009-08-30 10:44 [U-Boot] [RFC] - Relocating i386 Code - Permission to break orphan boards Graeme Russ
@ 2009-08-31  7:35 ` Graeme Russ
  0 siblings, 0 replies; 2+ messages in thread
From: Graeme Russ @ 2009-08-31  7:35 UTC (permalink / raw)
  To: u-boot

On Sun, Aug 30, 2009 at 8:44 PM, Graeme Russ<graeme.russ@gmail.com> wrote:
> Hi Wolfgang,
>
> I have been playing around with gcc's -fpic and ld's -PIE and I think that
> I can get 'proper' relocation happening on the i386 port. I've get the
> basics down, and I've been looking at /lib_ppc/board.c for how the PPC
> relocation scheme works and would like to replicate it (i.e. board_init_f,
> board_init_r and relocate_code.
>
> Unfortunately this leaves two options:
>
>  1) Tangle the i386 port up in a heap of #ifdefs, or;
>  2) Do a clean re-write for the eNET and break the other i386 boards
>    (sc520_cdp and sc520_spunk).
>

Actually, I have a 3rd (better) plan - I'll create new source files for
the relocatable version and conditionally compile based on a new
#define in the board header.

Looking forward to getting into it :)

> The other two i386 boards are OLD and i have my doubts as to if they ever
> did work completely, and have really strong doubts as to them working now
>
> I would really like to just throw all the old code out and do it all
> cleanly, but not without strong support from the community.
>
> What are your thoughts?
>
> Regards,
>
> G
>

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2009-08-31  7:35 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-08-30 10:44 [U-Boot] [RFC] - Relocating i386 Code - Permission to break orphan boards Graeme Russ
2009-08-31  7:35 ` Graeme Russ

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox