Gary Thomas wrote:
Jim Seymour wrote:
I'm curious, though. The Toshiba drivers seem to assume the flash is
in code space - although NAND flash (as I understand it) cannot be
used to directly execute code. Is this a legacy from the NOR flash
driver it was based on? Or is there some eCos weirdness that I need
to be aware of?
NAND FLASH is treated differently - it cannot be accessed
directly and the 'flash_read_buf()' function is used to stage
a portion of the FLASH into RAM. [...]
What prompted my question is the function "flash_code_overlaps" in
.../devs/flash/toshiba/tc58xxx/current/include/flash_tc58xxx.inl
It compares the flash address with "stext" and "etext" to determine
whether it "overlaps currently running code".
However, since we cannot execute out of NAND flash (and, in fact, do not
map the NAND flash into the normal memory address space), it seems that
this function should always return false.
Am I misunderstanding this? Or is it a (minor) bug?
--
Jim Seymour, Cipher Systems, Inc., 503-617-7447, http://www.cipher.com
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
