VGA console endian bug

Hollis Blanchard hollis at austin.ibm.com
Sat Aug 11 07:13:35 EST 2001


On Fri, Aug 10, 2001 at 10:41:39AM +0200, Geert Uytterhoeven wrote:
>
> On Thu, 9 Aug 2001, Hollis Blanchard wrote:
> > These two patches came from Daniel Berlin on July 9. They completely resolve
> > the VGA console backwards-endian problem for me on PPC. If they have Geert's
> > seal of approval ;) can they be committed?
> >
> > -Hollis
> >
> > --- linuxppc_2_4_devel/include/asm-ppc/vga.h.old	Thu Aug  9 11:49:23 2001
> > +++ linuxppc_2_4_devel/include/asm-ppc/vga.h	Thu Aug  9 11:49:42 2001
> > @@ -37,6 +37,9 @@
> >
> >  #define VT_BUF_HAVE_MEMCPYW
> >  #define scr_memcpyw	memcpy
> > +#define VT_BUF_HAVE_MEMCPYF
> > +#define scr_memcpyw_to memcpy
> > +#define scr_memcpyw_from memcpy
> >
> >  #endif /* !CONFIG_VGA_CONSOLE && !CONFIG_MDA_CONSOLE */
>
> The first I cannot approve: if scr_{write,read}w() swap bytes, how can it work
> if scr_memcpyw_{to,from}() don't swap bytes?

I have no idea. Perhaps those functions weren't used, and that's why it
appeared to work. I have not tried with only the second patch; perhaps that
will be enough...

> And where's the 3rd patch? I once posted a patch to fix a similar bug in
> drivers/char/console.c.

I can't find it in the linuxppc-dev archives. :(

-Hollis

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/





More information about the Linuxppc-dev mailing list