On Fri, Mar 01, 2013 at 10:57:40AM -0500, Tom Rini wrote:
> On Thu, Feb 28, 2013 at 07:37:51PM -0600, Scott Wood wrote:
> > On 02/28/2013 01:09:05 PM, Tom Rini wrote:
[snip]
> > >@@ -526,16 +543,28 @@ int nand_write_skip_bad(nand_info_t *nand,
> > >loff_t offset, size_t *length,
> > > if ((offset & (nand->writesize - 1)) != 0) {
> > > printf("Attempt to write non page-aligned data\n");
> > > *length = 0;
> > >+ if (actual)
> > >+ *actual = 0;
> > > return -EINVAL;
> > > }
> >
> > Again, what about the returns in the WITH_YAFFS_OOB section? Or if
> > we document that "actual" is undefined for error returns we can not
> > worry about this.
>
> OK. Currently we don't set length to 0 on WITH_YAFFS_OOB errors, but we
> ought to. And we can deal with actual the same way.OK, I'm going to do a follow-up patch to deal with length, and that CONFIG_CMD_NAND_YAFFS is broken as well. -- Tom
signature.asc
Description: Digital signature
_______________________________________________ U-Boot mailing list [email protected] http://lists.denx.de/mailman/listinfo/u-boot
