restoring ov7670/cafe_ccic to previous state

[Jim Gettys]

Note that the blood sprinkling and goat entrails talking to the camera
chip makes CAFE look good... We had to go to omnivision N times to get
new incantations of register read/writes during the development of the
driver.
                       - Jim


On Mon, 2008-06-23 at 14:24 -0700, Deepak Saxena wrote:
> On Jun 23 2008, at 17:58, Daniel Drake was caught saying:
> > Hi,
> > 
> > As you may have heard, olpc3 builds occasionally hang on boot during
> > cafe_ccic/ov7670 initialization.
> > 
> > Erik and myself tracked down the problem to this upstream commit, which
> > has never been shipped in an olpc stream before olpc3:
> > http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff_plain;h=6d77444
> > On boots where the bug appears, the do..while loop added by the patch
> > iterates indefinitely.
> > 
> > Jon Corbet is looking into the problem, but I feel that we should revert
> > the driver to the update1 state for the time being. We should be
> > starting to stabilise olpc3.
> 
> I agree. I'll pull your commits into master and testing and hopefuly
> we'll see a more stable fix in time for 8.2 (though my experience
> so far with the CAFE chip makes me not very optimistic).
> 
> ~Deepak
> 
-- 
Jim Gettys <jg at laptop.org>
One Laptop Per Child