On 03/04/2013 07:27:40 PM, Tom Rini wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 03/04/2013 08:12 PM, Scott Wood wrote:
On 03/04/2013 04:17:10 PM, Tom Rini wrote:
The flag changed from WITH_INLINE_OOB to WITH_YAFFS_OOB by accident in 418396e.
Signed-off-by: Tom Rini trini@ti.com --- Changes in v4: - Add patch to fix CONFIG_CMD_NAND_YAFFS
Changes in v3: None Changes in v2: None
common/cmd_nand.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/common/cmd_nand.c b/common/cmd_nand.c index 76f4d3f..d9010d2 100644 --- a/common/cmd_nand.c +++ b/common/cmd_nand.c @@ -673,7 +673,7 @@ static int do_nand(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[]) } ret = nand_write_skip_bad(nand, off, &rwsize, NULL, maxsize, (u_char *)addr, - WITH_INLINE_OOB); + WITH_YAFFS_OOB); #endif
Oops. Probably a leftover from an attempt to share code between yaffs and raw accesses.
BTW, it looks like there is no board that selects CONFIG_CMD_NAND_YAFFS, so it doesn't get compile tested... ...and smdk6400 defines CONFIG_SYS_NAND_YAFFS_WRITE, which nothing ever tests. :-P
Indeed. We lack some real users of YAFFS in U-Boot right now, and I don't wish to open a can of worms on why that might be.
I see some other things in nand_write_skip_bad() that look broken for WITH_YAFFS_OOB (but have been there since before WITH_YAFFS_OOB was introduced...), in particular check_skip_len() and the call to nand_write() when need_skip is 0.
Sorry if some worms just popped out. :-)
-Scott