[PATCH 1/2] m25p80: add additional sanity checks for erasure

[Sascha Hauer]

Hi Paul,

Sorry for the delay.

On Wed, Sep 21, 2011 at 12:29:01AM +0400, Paul Fertser wrote:
> On Tue, Sep 20, 2011 at 09:31:21PM +0200, Sascha Hauer wrote:
> > On Thu, Sep 15, 2011 at 03:27:36PM +0400, Paul Fertser wrote:
> > > This guards for the cases where the initial offset or byte count is
> > > not aligned with regard to erase block size, thus making it impossible
> > > for erase to do any harm to the nearby sectors.
> > 
> > I'm unsure about this one. The other flash drivers allow to erase areas
> > which are not eraseblock aligned. Maybe we should instead add a
> > cdev->erasesize field. Then we can add this check in the generic code
> > and fix this for all drivers.
> 
> I guess i was too surprised by seeing an endless loop when i tried to
> erase an area of the wrong size (the counter underflowed) to think
> about it in a less narrow-minded way, sorry, you're right of course.
> 
> Talking about generalisation, shouldn't the m25p80 driver be hooked
> into the mtd subsystem to allow running ubi on top of it? By the
> cursory look it seems to be doable without much effort.

Yes, it probably should. The cfi driver also is not a mtd driver but
has some mtd glue code. Maybe the same approach (even better the same
code) can be used here.

> 
> Another question: should i repost this early (questionable) series you
> seem to have missed[1]?
> 
> [1] http://lists.infradead.org/pipermail/barebox/2011-August/004483.html

Not necessary, I just applied it.

Thanks
 Sascha



-- 
Pengutronix e.K.                           |                             |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |