[PATCH] [flashrom] Numony support

[Stefan Reinauer]

See patch!

[Peter Stuge]

On Sat, Aug 02, 2008 at 05:04:40PM +0200, Stefan Reinauer wrote:
> erased blockwise

..
> +int spi_chip_erase(struct flashchip *flash);

Please change the function names to match the functionality.

Maybe spi_chip_erase_all_blocks ?

With that changed, it's

Acked-by: Peter Stuge

[Stefan Reinauer]

Peter Stuge wrote:
> On Sat, Aug 02, 2008 at 05:04:40PM +0200, Stefan Reinauer wrote:
> > erased blockwise
>
> ..
> > +int spi_chip_erase(struct flashchip *flash);
>
> Please change the function names to match the functionality.

It erases the chip?

Since it does not do blockwise erase on all chipsets, pretending it did 
would be bad.

[Peter Stuge]

On Sat, Aug 02, 2008 at 05:04:40PM +0200, Stefan Reinauer wrote:
> Add support for Numonyx M25PE80. This chip requires to be erased
> blockwise, as the c7 method is unreliable, either for ICH7 or for
> this chip.
>
> Signed-off-by: Stefan Reinauer

Acked-by: Peter Stuge

On Sat, Aug 02, 2008 at 05:12:19PM +0200, Stefan Reinauer wrote:
> > > erased blockwise
> >
> > ..
> > > +int spi_chip_erase(struct flashchip *flash);
> >
> > Please change the function names to match the functionality.
>
> It erases the chip?
>
> Since it does not do blockwise erase on all chipsets, pretending it
> did would be bad.

Right you are. This is the best solution at the moment!

//Peter

[Carl-Daniel Hailfinger]

On 02.08.2008 17:04, Stefan Reinauer wrote:
> Add support for Numonyx M25PE80. This chip requires to be erased blockwise, as
> the c7 method is unreliable, either for ICH7 or for this chip.
>
> Signed-off-by: Stefan Reinauer

Please don't commit yet. There's another chip erase command which may be
more reliable. More later.

Regards,
Carl-Daniel

[Stefan Reinauer]

Carl-Daniel Hailfinger wrote:
> On 02.08.2008 17:04, Stefan Reinauer wrote:
> > Add support for Numonyx M25PE80. This chip requires to be erased blockwise, as
> > the c7 method is unreliable, either for ICH7 or for this chip.
> >
> > Signed-off-by: Stefan Reinauer
>
> Please don't commit yet. There's another chip erase command which may be
> more reliable. More later.

The method I sent can be considered 100% reliable. A script measuring 
data in 200 flash writes has not revealed a single failure.

What other method are you suggesting? Are you coming up with a patch? 
I'd gladly test anything as long as it's more generic than what I sent.

Stefan