[EMAIL PROTECTED] writes:
[...]
> Well, why not ? We have a functional driver, it seems. Rushing things now
> and making changes before we have validation for what is there now isn't
> good.
But we know what we have is broken -- invalid when used with an old
scanimage.
> As for the principle course of development we can discuss that.
>
> 1. Error handling, the neverending story.
IMHO current situation acceptable, assuming that the oops is caused by
the ohci driver, not by us. Is there something else horribly wrong
with it?
> 2. Module unloading
Why, is that broken?
> 3. Devfs naming
We implement our SCSI host controller. Is there anything that isn't
taken care of?
4. Proc interface.
--
http://altern.org/vii
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
