Erasing more than one sector at a time?

Erasing more than one sector at a time?

[Stephane Laroche]

The erase functions in cfi_cmd_*.c seem to support erasing more than 1
sector per call.

I'm wondering why the parameter validation at the start of those
function specifically prohibits erasing more than one sector.  Is this
because the interface is defined this way or is it an oversight?

I'm thinking of removing the validation so my user programs don't have
to loop when erasing more than one sector.

-Stephane



To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org

Re: Erasing more than one sector at a time?

[David Woodhouse]

stephane.laroche@colubris.com said:
>  I'm wondering why the parameter validation at the start of those
> function specifically prohibits erasing more than one sector.  Is this
> because the interface is defined this way or is it an oversight?

Unless I'm smoking crack, the CFI command set code _does_ support erases 
longer than one erase block. The three checks which could cause it to 
return EINVAL are:
	1. start address must be erase-block-aligned
	2. length needs to be a multiple of erase block size.
	3. start + length must not exceed the _total_ size of the device.


The DiskOnChip code is different, but that's been/being fixed - depending 
on whether the patch has hit the CVS tree yet or not.




--
dwmw2




To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org

Re: Erasing more than one sector at a time?

[Stephane Laroche]

Obviously, my mistake.  I didn't realize I had to have the erase size be a
multiple of a sector size.  Sometimes, I just can't read code properly (an
it wasn't even 3 AM).

Thanks,

-Stephane


David Woodhouse wrote:

> stephane.laroche@colubris.com said:
> >  I'm wondering why the parameter validation at the start of those
> > function specifically prohibits erasing more than one sector.  Is this
> > because the interface is defined this way or is it an oversight?
>
> Unless I'm smoking crack, the CFI command set code _does_ support erases
> longer than one erase block. The three checks which could cause it to
> return EINVAL are:
>         1. start address must be erase-block-aligned
>         2. length needs to be a multiple of erase block size.
>         3. start + length must not exceed the _total_ size of the device.
>
> The DiskOnChip code is different, but that's been/being fixed - depending
> on whether the patch has hit the CVS tree yet or not.
>
> --
> dwmw2



To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org