[PATCHv2] drivers: mtd: spinand: Add generic spinand frameowrk.

[Florian Fainelli]

Hello,

Le jeudi 4 juillet 2013 10:13:43 Sourav Poddar a écrit :
> > 
> > Can this somehow be made a runtime thing?
> 
> Ahh..I think we might opt for a device tree entry and based on that
> check for ECC.

Ok, sounds good too.

> 
> > [snip]
> > 
> >> +               if (count<  oob_num&&  ops->oobbuf&&  chip->oobbuf) {
> >> +                       int size;
> >> +                       int offset, len, temp;
> >> +
> >> +                       /* repack spare to oob */
> >> +                       memset(chip->oobbuf, 0,
> >> info->ecclayout->oobavail);
> >> +
> >> +                       temp = 0;
> >> +                       offset = info->ecclayout->oobfree[0].offset;
> >> +                       len = info->ecclayout->oobfree[0].length;
> >> +                       memcpy(chip->oobbuf + temp,
> >> +                               chip->buf + info->page_main_size +
> >> offset, len);> 
> > Sounds like a for look might be useful here
> 
> I dont think so, there is a while loop above under which it happens.
> We are increasing count at the bottom of the while loop. So, I think
> this should work fine.

What I meant here, is that you could use a for loop to repeat 4 times the same 
following pattern, such that it becomes:

for (j = 0; j < 4; j++0 {
	temp += len;
	offset = info->ecclayout->oobfree[j].offset;
	len = info->ecclayout->oobfree[j].length;
	memcpy(chip->oobbuf + temp,
		chip->buf + info->page_main_size + offset, len);
}

Or even make it a helper function which is inlined if that is deemed more 
elegant.
-- 
Florian