[PATCH] Fix Winbond W25X{10, 20, 40, 80} incorrect erase blocks

[Sean Nelson]

Fix erase blocks for Winbond W25X* SPI chips. The Winbond W25X10 and 
related chips only have 4k and 64k blocks and only accept erase 
commands: 20h, d8h, and c7h.

Signed-off-by: Sean Nelson

[PATCH] Fix Winbond W25X{10, 20, 40, 80} incorrect erase blocks

[David Hendricks]

On Sun, Jan 31, 2010 at 6:31 PM, Sean Nelson  wrote:
> Fix erase blocks for Winbond W25X* SPI chips. The Winbond W25X10 and
> related chips only have 4k and 64k blocks and only accept erase commands:
> 20h, d8h, and c7h.
>
> Signed-off-by: Sean Nelson

Tested on a W25x16 flash chip using an NM10 southbridge. Verbose flashrom -E
log attached. Following an erase with a read, the image produced turns up
all 0xFF bytes. The warnings reported seem benign, and probably have more to
do with the NM10 than the flash chip...

Acked by: David Hendricks

[PATCH] Fix Winbond W25X{10, 20, 40, 80} incorrect erase blocks

[David Hendricks]

D'oh! I just realized that this patch doesn't affect the x16. I really ought
to read the patch more carefully...

On Thu, Feb 11, 2010 at 3:50 PM, David Hendricks wrote:
> On Sun, Jan 31, 2010 at 6:31 PM, Sean Nelson wrote:
> > Fix erase blocks for Winbond W25X* SPI chips. The Winbond W25X10 and
> > related chips only have 4k and 64k blocks and only accept erase commands:
> > 20h, d8h, and c7h.
> >
> > Signed-off-by: Sean Nelson
>
> Tested on a W25x16 flash chip using an NM10 southbridge. Verbose flashrom
> -E log attached. Following an erase with a read, the image produced turns up
> all 0xFF bytes. The warnings reported seem benign, and probably have more to
> do with the NM10 than the flash chip...
>
> Acked by: David Hendricks

[PATCH] Fix Winbond W25X{10, 20, 40, 80} incorrect erase blocks

[David Hendricks]

Okay, so I applied the same changes to the W25x16 and got good results. It
looks like the same applies to the W25x{16,32,64} products in addition to
the W25x{10,20,40,80}. Please update the patch to include the additional
chips.

W25x{16,32,64} datasheet:
http://www.winbond-usa.com/products/Nexflash/pdfs/datasheets/W25X16_32_64h.p...

On Thu, Feb 11, 2010 at 3:56 PM, David Hendricks wrote:
> D'oh! I just realized that this patch doesn't affect the x16. I really
> ought to read the patch more carefully...
>
> On Thu, Feb 11, 2010 at 3:50 PM, David Hendricks wrote:
> > On Sun, Jan 31, 2010 at 6:31 PM, Sean Nelson wrote:
> > > Fix erase blocks for Winbond W25X* SPI chips. The Winbond W25X10 and
> > > related chips only have 4k and 64k blocks and only accept erase commands:
> > > 20h, d8h, and c7h.
> > >
> > > Signed-off-by: Sean Nelson
> >
> > Tested on a W25x16 flash chip using an NM10 southbridge. Verbose flashrom
> > -E log attached. Following an erase with a read, the image produced turns up
> > all 0xFF bytes. The warnings reported seem benign, and probably have more to
> > do with the NM10 than the flash chip...
> >
> > Acked by: David Hendricks

[Sean Nelson]

On 2/11/10 4:08 PM, David Hendricks wrote:
> Okay, so I applied the same changes to the W25x16 and got good 
> results. It looks like the same applies to the W25x{16,32,64} products 
> in addition to the W25x{10,20,40,80}. Please update the patch to 
> include the additional chips.
>
> W25x{16,32,64} datasheet: 
> http://www.winbond-usa.com/products/Nexflash/pdfs/datasheets/W25X16_32_64h.p...
>
> On Thu, Feb 11, 2010 at 3:56 PM, David Hendricks <[e-mail] 
> > wrote:
> D'oh! I just realized that this patch doesn't affect the x16. I
> really ought to read the patch more carefully...
>
>
> On Thu, Feb 11, 2010 at 3:50 PM, David Hendricks
> <[e-mail] <mailto:[e-mail]>> wrote:
>
>     On Sun, Jan 31, 2010 at 6:31 PM, Sean Nelson
>     <[e-mail] <mailto:[e-mail]>> wrote:
>
>         Fix erase blocks for Winbond W25X* SPI chips. The Winbond
>         W25X10 and related chips only have 4k and 64k blocks and
>         only accept erase commands: 20h, d8h, and c7h.
>
>         Signed-off-by: Sean Nelson <[e-mail]
>         <mailto:[e-mail]>>
>
>
>     Tested on a W25x16 flash chip using an NM10 southbridge.
>     Verbose flashrom -E log attached. Following an erase with a
>     read, the image produced turns up all 0xFF bytes. The warnings
>     reported seem benign, and probably have more to do with the
>     NM10 than the flash chip...
>
>     Acked by: David Hendricks <[e-mail]
>     <mailto:[e-mail]>>
>
>     -- 
>     David Hendricks (dhendrix)
>     Systems Software Engineer, Google Inc.
>
>
>
>
> -- 
> David Hendricks (dhendrix)
> Systems Software Engineer, Google Inc.

The W25x16 and the W25x16BV use the exact same chip ids but the W25x16BV 
supports the already defined erase commands.

[PATCH] Fix Winbond W25X{10, 20, 40, 80} incorrect erase blocks

[David Hendricks]

On Fri, Feb 12, 2010 at 10:12 AM, Sean Nelson  wrote:
> The W25x16 and the W25x16BV use the exact same chip ids but the W25x16BV
> supports the already defined erase commands.

Thanks for clarifying. For those who weren't on IRC, basically what this
means is that Flashrom will spit out some benign warnings for non-BV parts
when one of the unsupported commands is used.

[Sean Nelson]

On 2/12/10 10:21 AM, David Hendricks wrote:
> On Fri, Feb 12, 2010 at 10:12 AM, Sean Nelson <[e-mail] 
> > wrote:
> The W25x16 and the W25x16BV use the exact same chip ids but the
> W25x16BV supports the already defined erase commands.
>
>
> Thanks for clarifying. For those who weren't on IRC, basically what 
> this means is that Flashrom will spit out some benign warnings for 
> non-BV parts when one of the unsupported commands is used.

Thanks, committed in r897.