Re: flash bbt broken due to unitialized bitflip_threshold?

[Shmulik Ladkani <shmulik.ladkani@gmail.com>]

Hi,

On Wed, 06 Jun 2012 18:46:15 +0300 Artem Bityutskiy <artem.bityutskiy@linux.intel.com> wrote:
> On Wed, 2012-06-06 at 18:15 +0300, Shmulik Ladkani wrote:
> > - By 'add_mtd_device'. Here it defaults to 'ecc_strength' if NOT
> >   previously set by the driver.
> 
> But this is wrong. If I use the old doc2000 driver, with ecc_strength =
> 2, and it works fine for me, and I am happy that UBI scrubs for a single
> bit-flip, why should my system become broken because someone decided
> that now UBI should start scrubbing on 2 bit-flips?

As I remember, the motivation was to reduce unnecessary scrubbing for
devices exposing high rate of bitflips but having strong ECC to
compensate for.
For these users, the system was "broken" in their perspective, as it
performed too many scrubbing...

Anyway, I understand your point preserving the old behavior.
I thought you were aware of this change.

> We should not change the defaults - if I do not set the threshold via
> sysfs of in the driver, it should be 1.

Fair point. This is safer, backwords compatible.

But eventually, wouldn't we end up with all the drivers assigning
bitflip_threshold to ecc_strength?

> Unless I am completely confused, we should change this, CC -stable if
> needed, and ask dwmw2 to merge that.

No you're not confused.
If we'd like to preserve old behavior, the default assignment needs to
be changed.

BTW this is independent of the 'scan_bbt' bug spotted by Sascha.

Regards,
Shmulik