[PATCH 0/2] CFI fixups for quirks in some new NOR flash chips (try 3)

Trent Piepho tpiepho at freescale.com
Wed Apr 23 01:02:06 EDT 2008


Maybe the third time's the charm.  I'd like to get this in 2.6.26.  I
first posted these patchs last month.

The first patch extends the CFI probe code to get the extended ID from
AMD/Spansion flash chips.  I checked with Spansion's support and for AMD
or Spansion devices, a flash device id of 0x7e is always used to
indicates an extended ID is present later on in the CFI query data. 
There are no AMD/Spansion chips with a non-extended ID that is the same
as another chip's extended ID.  So, once can just replace the normal
device id (0x7e) with the extended one without losing any information. 
There's nothing to gain by adding a new field to the flash info struct
and then adding code to the fixup routines to also matches the new field,
etc.

The second patch is just a couple simple fixup entries that change the
sector count for the affected flash chips.  It uses the extended ID to
get the right chips.



More information about the linux-mtd mailing list