Re: [flashrom] [commit] r806 - trunk

[Luc Verhaegen]

On Thu, Dec 17, 2009 at 04:20:02PM +0100, [e-mail] wrote:
> Set board-specific max decode size for Elitegroup K7VTA3.
> Set board-specific max decode size for Shuttle AK38N.
>
> Signed-off-by: Carl-Daniel Hailfinger 
> Acked-by: Uwe Hermann
> Modified: trunk/board_enable.c
>
> --- trunk/board_enable.c	2009-12-17 04:22:40 UTC (rev 805)
> +++ trunk/board_enable.c	2009-12-17 15:20:01 UTC (rev 806)
> @@ -987,16 +987,24 @@
>  }
>
> /**
> Suited for:
> Shuttle AK38N: VIA KT333CF + VIA VT8235 + ITE IT8705F
> Elitegroup K7VTA3: VIA Apollo KT266/A/333 + VIA VT8235 + ITE IT8705F
> Suited for: Elitegroup K7VTA3: VIA Apollo KT266/A/333 + VIA VT8235 + ITE IT8705F
> */
>
> -static int it8705f_write_enable_2e(const char *name)
> +static int elitegroup_k7vta3(const char *name)
>  {
> max_rom_decode.parallel = 256 * 1024;
> return it8705f_write_enable(0x2e, name);
>
> }
>
> /**
> Suited for: Shuttle AK38N: VIA KT333CF + VIA VT8235 + ITE IT8705F
> */
>
> +static int shuttle_ak38n(const char *name)
> +{
> max_rom_decode.parallel = 256 * 1024;
> return it8705f_write_enable(0x2e, name);
>
> +}
>
> +/**

What happened here, and why do we require two different entries here?

Luc Verhaegen.

[commit] r806 - trunk

[Carl-Daniel Hailfinger]

On 18.12.2009 09:31, Luc Verhaegen wrote:
> On Thu, Dec 17, 2009 at 04:20:02PM +0100, [e-mail] wrote:
> > Set board-specific max decode size for Elitegroup K7VTA3.
> > Set board-specific max decode size for Shuttle AK38N.
> > Modified: trunk/board_enable.c
> >
> > --- trunk/board_enable.c	2009-12-17 04:22:40 UTC (rev 805)
> > +++ trunk/board_enable.c	2009-12-17 15:20:01 UTC (rev 806)
> > @@ -987,16 +987,24 @@
> > Suited for:
> > Shuttle AK38N: VIA KT333CF + VIA VT8235 + ITE IT8705F
> > Elitegroup K7VTA3: VIA Apollo KT266/A/333 + VIA VT8235 + ITE IT8705F
> > Suited for: Elitegroup K7VTA3: VIA Apollo KT266/A/333 + VIA VT8235 + ITE IT8705F
> > */
> >
> > -static int it8705f_write_enable_2e(const char *name)
> > +static int elitegroup_k7vta3(const char *name)
> >  {
> > max_rom_decode.parallel = 256 * 1024;
> > return it8705f_write_enable(0x2e, name);
> >
> > }
> >
> > /**
> > Suited for: Shuttle AK38N: VIA KT333CF + VIA VT8235 + ITE IT8705F
> > */
> >
> > +static int shuttle_ak38n(const char *name)
> > +{
> > max_rom_decode.parallel = 256 * 1024;
> > return it8705f_write_enable(0x2e, name);
> >
> > +}
>
> What happened here, and why do we require two different entries here?

The decode size is board specific and independent of the board write
enable function. We could have renamed this function to
it8705f_write_enable_2e_can_only_decode_256k (IMHO ugly) because the old
name doesn't match the new contents. Instead, we created two different
board specific entries.

If you prefer the long alternative name, get Uwe to agree and send a
patch. I won't object and I'm not willing to start a flamewar over this.
There's too much work to do to to risk a bad mood for any of the developers.

Regards,
Carl-Daniel